138. Wu, Z. Y. and Xu, G. Q. (2014). "A Novel Formulation for Finite Element Model-based Damage Detection" International Conference on Bridge Maintenance, Safety and Management, July, 2014, Shanghai, China.
137. Wu, Z. Y. and Roshani, E. (2014). "Sensor Placement Optimization for Water Quality Model Calibration" ASCE World Environmental and Water Resources Congress, June 3-5, 2014, Portland, U.S.A.
ABSTRACT: Several sophisticated methods have been developed for water quality (WQ) sensor placement in water distribution system analysis, but most of them are geared toward mitigating water security concerns, including but not limited to contaminant detection, chemical intrusion or terroristic attacks. The WQ sensor or logger placement has been less concerned for the water quality monitoring or field data collection in order to conduct WQ model calibration. The sensor locations are conventionally determined in an ad-hoc manner, based on geographic coverage, pipe diameter, pipe material, distance to the source, and accessibility. This paper presents a new methodology for helping engineers to identify the near optimal locations of WQ sensors for WQ model calibration. The approach maximizes the sensory network efficiency, the coverage of the pipes due to wall reaction coefficient adjustments that are the primary model parameters for WQ model calibration. This new method allows us to collect good and sensible data to calibrate a WQ model.
136. Wu, Z. Y. and Quader S. (2014). "Parallelizing Water Quality Analysis Solver with Portable GPU Computing Paradigm" ASCE World Environmental and Water Resources Congress, June 3-5, 2014, Portland, U.S.A.
ABSTRACT:Water Quality (WQ) analysis model is a modeling tool widely adopted by water utilities. In order to gain accurate and systematic understanding on the water quality dynamics in a large water distribution system, a WQ simulation is required to be performed for a long period of time (e.g., 168 hours) with a small time step (e.g., less than 5 minutes). It is a very time consuming computation that is currently implemented as a sequential program on the Central Processing Unit (CPU). In the meantime, Graphics Process Unit (GPU) technology, originally developed and used for video games, has evolved into a powerful yet affordable device for general purpose computing. It offers massively parallel computing capability on personal computers. GPU computing capability is far from being fully utilized for engineering computing in general and water distribution system analysis in particular.This paper presents a heterogeneous parallel computing WQ model with CPU and GPU. The GPU computing functions for WQ analysis are implemented by using OpenCL, the industry-standard for heterogeneous computing, to ensure the portability of the parallelized solver on GPU hardware from different vendors. The parallelized WQ solver is tested on the large water distribution WQ models with various simulation durations. The results are first compared with those obtained with the original WQ solver on CPU to ensure that the same accuracy is achieved by the GPU-based WQ solver. The computation performance of the parallelized WQ solver is then compared against the performance of the conventional sequential solver. The results obtained show that the GPU-based WQ solver is 4 times faster than the conventional solver. The speedup of the WQ analysis will facilitate a wide range of WQ applications for water distribution system design, management and operation.
135. Walski, T., Sage, P. and Wu, Z. Y. (2014). "What Does It Take to Make Automated Calibration Find Closed Valves and Leaks?" ASCE World Environmental and Water Resources Congress, June 3-5, 2014, Portland, U.S.A.
ABSTRACT: In many water distribution systems, the hed loss from the boundary nodes to the measurement points is small and, thus, extra head loss must be generated. The easiest way to accomplish this is by opening one or more hydrants. This magnifies the discrepancies between the model and field data and makes the model much more sensitive to changes in parameters. Even with these high flows, it is important to measure the resulting pressure and the elevation of the pressure gage to a high degree of accuracy (and remember that the elevation of the gage is not necessarily the elevation of the model node).
134. Wu, Z. Y. and ElMaghraby M. (2014). "Portable GPU-based Artificial Neural networks for Accelerated Data-Driven Modeling" HIC 2014 - the 11th International Conference on Hydroinformatics, Aug. 17- 21, 2014 new York, U.S.A.
132. Naga, D., Wu, Z. Y., Allen, T., Croxton, N., Lavery, D. and Powell, J. (2014) "Pressure Logger Placement for Smart Water System Management" IWA Water Loss 2014, March 30 - April 2, 2014, Vienna, Austria.
ABSTRACT: The essence of smart water system management is to improve planning, scheduling and strategic decision-making for efficient and effective operation. Following the introduction by OFWAT of the Service Incentive Mechanism (SIM) Water companies in the UK are under increasing pressure to improve customer service. In order to achieve this there is a drive to increase network monitoring and analysis capabilities to proactively manage distribution systems and prevent failures in customer service occurring, thereby reducing unwanted telephone customer contacts and improving customer satisfaction levels. There is no doubt that being smart or intelligent at managing water systems requires good metering, monitoring technology and data analysis, but determining what kind of data and where to collect the data are among the foremost important issues to be addressed to be truly smart. This paper applies the latest development of pressure logger placement method and software tool to optimize the number of data loggers and their locations in a number of the selected DMAs. The field data is then collected and used for leakage hotspot detection via hydraulic model calibration. The results show that the best practice of water system monitoring and leakage detection can be established with the innovative methods and modelling tools.
131. Solarczyk, A. and Wu, Z. Y. (2014) "Case Studies in Applying Advanced Modelling Analysis to NRW" IWA Water Loss 2014, March 30 - April 2, 2014, Vienna, Austria.
ABSTRACT: The main purpose of the project was to gain a better understanding of the practical processes involved in using hydraulic model calibration tool Darwin Calibrator in conjunction with hydrant flushing and multi – logger data collection, in a dynamic way for the purposes of localising leakage hotspots for detection teams to target. The project was quite sucessful in the number of confirmed leakage hot spots. Out of 14 identified leakage hotspots 8 leaks were detected (in one of the analysed DMAs 5 leaks were detected in the vicinity of 3 identified leakage hot spots), 2 locations could not be accessed and no leaks were identified in the remaining 2 locations.
129. Behandish, M. and Wu, Z. Y. (2013) “Generalized GPU-Based Artificial Neural Network Surrogate Model for Extended Period of Hydraulic Simulation.” ASCE World Environmental and Water Resources Congress, May 20-23, 2013, Cincinnati, OH, USA.
ABSTRACT
This paper presents a generalized methodology and implementation aimed at surrogating the extended period hydraulic simulation with Artificial Neural Networks (ANNs) for a large water distribution system. Each snapshot of the hydraulic simulation is modeled as a set of input/output relationships, replicated by a collection of ANNs trained on the commodity Graphics Processing Unit (GPU). Previous research by the authors on a particular demand monitoring zone (DMZ) system with the limited hydraulic properties is generalized to include a comprehensive account of state variables, and their dependencies are captured by a systematic sensitivity analysis that guides the modular assembly of multi-ANN meta-model. The trained networks are validated for successive calls over extended periods of time (as long as a week), and tremendous improvements in terms of accuracy and reliability are obtained and compared with the results in previous publications. The improved metamodel manifests as a powerful tool for fast and accurate hydraulic response prediction for operation optimization.
118. Wu, Z. Y. (2012) "Roles of Advanced Information Technology in Water Loss Management."IWA Water Loss 2012, Manila, Philippines.
ABSTRACT
Pressure-dependent leakage detection (PDLD) approach has been developed by the authors and successfully applied to a number of water systems around the world. PDLD as part of model calibration is often challenged by the unknown valve status, which often cause the problems during hydraulic model calibration. The number of occurrences of such valves is probably increasing as a result of extensive mains rehabilitation work, district meter area (DMA) and discrete pressure area (DPA) boundary design. Accordingly, the leakage detection, as part of model calibration, is also required to determine the valve settings including open, close and partially open. This paper presents the optimization model and solution method for optimizing valve settings. The optimization model is formulated to search for the locations of a given number of uncertain valves and the corresponding settings (open, close and partially open). With the valve identification method, working together with the PDLD for leakage hotspot detection, the model calibration capability is improved for identifying a wide range of uncertain parameters (leakage hotspots, valve settings, pipe roughness and node demand etc.). A case study is conducted and presented in the paper to demonstrate the efficacy of the developed valve identification method.
116. Behandish, M. and Wu, Z.Y. “GPU-based Artificial Neural Network Configuration and Training for Water Distribution System Analysis ”, ASCE Annual World Environmental and Water Resources Congress, May 21 – 24, 2012, Albuquerque, New Mexico, USA.
ABSTRACT
Previous research by the authors has confirmed that GPU-based Artificial Neural Network (ANN) construction offers significant computational speed-up over the CPU-based ANN implementation for water distribution system meta-modeling. Further research showed that for real-world scale systems, however, the hydraulic prediction accuracy achieved by the previously trained ANN is inadequate when the ANN is employed as a surrogate model to replace the hydraulic simulator for pump scheduling optimization. This paper improves the GPU-based ANN meta-modeling approach for constructing the ANN model of adequate accuracy in an efficient manner. This includes system decomposition, ANN configuration and training specifications to achieve a reliable surrogate model for pump scheduling optimization. The technique is successfully applied to a real-world water distribution system, and the GPU-based ANN predicted the hydraulic responses with a Root-Mean-Square Error (RMSE) of less than 1% for tank levels and pump energy consumption.
115. Wu, Z. Y. and Khaliefa M. “21 – 24, 2012, Albuquerque, New Mexico, USA. "Cloud Computing for High Performance Optimization of Water Distribution Systems” ASCE Annual World Environmental and Water Resources Congress, May
ABSTRACT
Cloud computing is quickly becoming an innovative model for delivering IT infrastructure, applications and data management. It shifts the emphasis from static, stand-alone applications to dynamic, shared environments, dynamically allocated among various tasks and accessed via a network. In this paper, we investigate the use of cloud computing for high performance optimization of water distribution systems. The paper covers the general survey of leading commercial cloud computing services, high performance computing (HPC) cloud differentiators and demonstration of the improved HPC cloud implementation. With necessary background information on cloud computing, a prototype of the high performance computing (HPC) cloud is proposed and developed for water system optimization. The prototyped HPC cloud is constructed by using many-core machines that form the cloud platform for running parallel applications. Finally, as an example of cloud-based water distribution optimization, a pump scheduler has been deployed onto the HPC cloud with web-based user interface, through which a user could submit, execute and retrieve optimization analysis jobs.
114. Syed J. L. and Wu, Z. Y. "Effect of Surge Vessel Sizing with Respect to its Location in the Water Transmission Main" ASCE Annual World Environmental and Water Resources Congress, May 21 – 24, 2012, Albuquerque, New Mexico, USA.
ABSTRACT
Water transmission main is subjected to hydraulic transients due to sudden pipe failure, valve closure or any other transient phenomenon. These hydraulic transients might cause adverse effect within the pumps, bursting of the transmission main and detachment of the pipe from the joint locations such as air valve or control valve. The negative transient pressures can cause water quality issues especially from intrusion of contaminants from at the air valves and loose joints. In order to avoid both unacceptable minimum and maximum transient pressures, surge protection is imperative for a water transmission main. It is a common practice to locate the surge vessel near the pumps in order to dampen the generated surge due to sudden pump failure. In this study, surge analysis results considering two different topology configurations (Option 1 & 2) with respect to different surge vessel locations are compared for a water transmission main connected to two pumps (each having a capacity of 14.5 l/s @ 110 m head ) in parallel. The transmission main is serving residential consumers which are located on 50 m higher elevation from the pump station at a distance of about 12.5 KM. For Option 1, it is considered that two surge vessels are placed on a 150 mm DI line, 2.5 m far at the downstream side of each pump. For Option 2, only one surge vessel is placed on a 300 mm transmission main originating just after the parallel connection of the 150 mm DI lines at the downstream of the two pumps. Transient analysis is undertaken for different vessel sizes of two configurations. The results obtained are compared for the Minimum Transient Pressures in the modelled pipeline. Although, comparative difference between the two configurations is not much but gives a clear indication that surge vessel location and sizes could play a vital role in controlling the transient pressures. It is also observed that the pre-charge pressure in the surge vessel is also very important for controlling the transient pressures and must be precisely determined with respect to the mechanism of the working of the selected surge vessel.
113. Wu, Z. Y. “Comment on “Robust optimization Methodologies for Water Supply Systems Design” by J. Marques et al” Drinking Water Engineering and Science Discussion, 5, C89-91, 2012.
ABSTRACT
A discussion note published for commenting on Robust Optimization Methodologies for Water Supply System Design by Marques et al.
112. Wu, Z. Y., Xu, G., Mi, T. and Zhao, J. (2012) "High performance computing for damage detection of civil infrastructural systems", proceeding of 6th on Bridge International ConferenceMaintenance, Safety and Management.
ABSTRACT
Previously developed solution method by authors, based on finite element model updating or calibration, has proved to be effective at locating structure damage elements. The model is formulated to simultaneously search for the given number of the damaged elements and the corresponding damage indicators. It is a mixed integer and continuous optimization problem that is solved by applying a competent genetic algorithm to minimize the discrepancy between the field monitored and model analyzed responses. It is computationally costly to apply it to a large structure system. In this paper, a high performance computing (HPC) framework has been presented and applied to efficiently solve the problem. The HPC framework takes advantage of well-developed FE analysis models and software design patterns. The method is implemented by parallelizing the solution evaluations on a cluster of many-core machines. The parallel optimization essentially speeds up the computation and enables fast convergence of the integrated method. The developed method has been tested on the identification of the damage scenario for a benchmark system. The results obtained show that the proposed method is efficient and effective at detecting damage in a large infrastructure system.
111. Wu, Z. Y, Elsayed, M. and Song Y. (2012) "HPC Optimization for Solving BWN-II Problem", WDSA 2012, Adelaide, Australia
ABSTRACT
Previously published works prove that evolutionary algorithms outperform the conventional optimization techniques in optimizing the design of water distribution systems. However, due to the high dimensionality (large number of optimization variables) and intensive computation (large number of solution evaluations), it is becoming a challenge to optimize the design of a system e.g. the Battle of the Water Network II (BWN-II), which is required to consider to schedule the pump operation over a long period of 168 hours and also compute water age. In this paper, a high performance optimization tool, based on competent genetic algorithm (GA), is employed for solving the problem. The solution method is formulated to allow optimization runs to be taken in multiple steps with the selected decision variables with the corresponding objective and constraints. The approach enable iterative optimization of parallel pipes, new pipe sizes, tank volumes to be expanded, and pump operation controls under normal operation conditions, and also the number of standby diesel engines to be installed at pump stations for power outage scenarios. With the parallelized competent GA framework, which can be run on single many-core computer and a cluster of many-core computers, the integrated method ensures computation efficiency speedup and achieves the optimized solution of good quality for BWN-II problem.
110. Wu, Z. Y and Behandish M. (2012) "Comparing Methods of Parallel Genetic Optimization for Pump Scheduling using Hydraulic Model and GPU-Based Ann Meta-Model", WDSA2012, Adelaide, Australia
ABSTRACT
This paper presents the comparison of two different approaches for solving the computationally intensive problem of near-optimal pump scheduling for large water distribution systems. The optimization problem is formulated to minimize the pump operation cost subject to water supply service requirements. The first method, utilizes the hydraulic model integrated with a parallel genetic algorithm (GA), which can run on either a many-core machine or a cluster of many-core machines. The second method, on the other hand, uses a GPU-accelerated artificial neural network (ANN) meta-model to surrogate the hydraulic model in GA optimization. The study shows that the GPU-based ANN is capable of rapidly predicting the energy rates with adequate accuracy and robustness, as well as the tank levels which can be used for online optimization on a rolling-forward basis. GA+ANN is capable of reducing the optimization run time from several hours to a few minutes, thus enables real-time or online pump scheduling for large water distribution systems.
109. Wu, Z. Y and Behandish M. (2012) "Real-Time Pump Scheduling using Genetic Algorithm and Artificial Neural Network Based on Graphics Processing Unit", WDSA2012, Adelaide, Australia
ABSTRACT
This paper presents a real-time pump scheduling (RTPS) methodology and the case study of a large water distribution system. The method employs the artificial neural network (ANN) based on graphics processing unit (GPU), a meta-model or surrogate solver for evaluating the hydraulic responses of pump scheduling solutions. A genetic algorithm (GA) is used to search for near-optimal pump operation policy (POP) subject to water supply service requirements. The resulting POP is applied to the system operation for a predefined period or so-called rolling-forward time step that is much less than a typical operation cycle (e.g. 24 hours). A rolling-forward time step is determined such that the surrogate model is capable of predicting sufficiently accurate hydraulic responses for the purpose of GA solution evaluations. At the end of each rolling forward step, new boundary conditions, e.g. tank levels obtained from the field monitoring system, are used as the initial conditions for the next GA+ANN optimization process. This RTPS methodology has been successfully applied to identify the tank operation range for real-time operation and minimize the pumping energy cost of a large demand monitoring zone (DMZ) in the UK. The results indicate that significant saving in energy costs can be achieved in comparison with the current operation policies.
108. Wu, Z. Y. and Song, Y. (2012) "Optimizing Pressure Logger Placement for Leakage Detection and Model Calibration",WDSA2012, Adelaide, Australia
ABSTRACT
Pressure loggers are used to record field pressure data, which are the essential information for water utilities to achieve sound model calibration. Thus determining where and how many pressure loggers to be placed in a distribution system is an important task. In this paper, an effective method is presented for optimizing pressure logger placement. It is developed into a two-stage solution method. At the first stage, a Mont Carlo method is used to generate a large number of random events, each of them represents either demand change and/or leakage flow. All the events are simulated by conducting hydraulic model analysis. The simulated nodal pressures are compared with the baseline condition. The pressure change at each node is evaluated with respect of logger accuracy. If the pressure change is greater than the logger accuracy, a value of 1 is assigned to the node for an event, otherwise a 0 is a assigned to the node. Thus a binary database is constructed for the randomized event for a given distribution system. The constructed binary database is employed to optimize the pressure logger locations in the second stage. The pressure logger locations are optimized for a given number of loggers such that the randomized events are detected or covered as many as possible. The method is tested on two systems. The results are compared with the pressure logger locations previously designed by the experienced engineers. It shows the method is able to achieve greater coverage with less number of pressure loggers than the sampling design by experienced engineers.
107. Wu, Z. Y. and Song, Y. (2012) "Optimization Model for Identifying Unknown Valve Statuses and Settings", WDSA2012
ABSTRACT
Hundreds of valves are installed in a system, even as small as in the size of a District Meter Area (DMA). It is not uncommon that the status and settings are unknown for some valves due to various reasons. Correctly identifying the valves and the corresponding statuses (open or close) and/or settings (degree of opening) is imperative to adequately simulate the network hydraulics, especially for the valves on critical flow paths. In this paper, an optimization model, based on the leakage detection and model calibration framework previously developed by the author, is proposed for identifying unknown valve statuses and settings. It is formulated to search for the given number of unknown valves (locations and settings) such that the difference between the observed and simulated values (flows and pressures) is minimized. The new valve identification method is implemented as an extension of the GA-based model calibration tool, and allows modellers to progressively identify unknown valves, along with calibrating many model parameters including leakage hotspots, pipe roughness and nodal demand. The results obtained from the case studies of real water systems show that the proposed method is effective at identifying valves with unknown status.
106. Sayed, J. L. and Wu, Z. Y. (2012) "Sufficiency of Data Loggers for Optimized Model Calibration of Water Distribution Networks", WDSA 2012
ABSTRACT
The degree of accuracy of any hydraulic model’s output is primarily dependent upon the extent of calibration conducted with the observed flow and pressure data, which are mostly recorded by using data loggers at different locations of the modelled area. In General, the more loggers are placed and used for data collection, the better the model can be calibrated, but the greater the cost is for the modelling project. In this study, the optimized calibration has been performed to investigate the sufficiency of the field data with respect to the numbers and locations of data loggers. The comparative analysis has been made based on the root mean square error (RMSE) of flows and hydraulic grades, and also the number of detected leakage nodes that are compared with the leakage locations found in the field. It is indicated that a good trade-off must be achieved between the number of data loggers and sound solution in terms of detected leakage nodes and RMSE. The outcomes of this study would help engineers to minimize CAPEX (Capital Expenditures) and OPEX (Operating Expenditures) required for the placement and maintenance of data loggers in a municipal water network.
105. Wu, Z. Y. and Walski T. (2012) "Effective Approach for Solving Battle of Water Calibration" ASCE journal of water resources planning and management, Sept/Oct. Issue
ABSTRACT
This paper presents an effective calibration approach for constructing robust extended simulation model. Using the benchmark model calibration problem, set up for Battle of Water Calibration Networks, the EPS model calibration is solved in a progressive manner of optimizing model parameters plus engineering judgments. The model calibration approach consists of multiple steps, including (1) constructing the initial EPS model with the given model and SCADA data; (2) calibrating for static and fire flow test data; and (3) calibrating EPS model for given SCADA data over 167 hours. The calibration tasks were iteratively conducted per District Meter Area (DMA), verified and then fine tuned for the whole system. While the field data contained some noise, good model calibration has been achieved for BWCN. The performance of the calibrated model is compared with those calibrated by fourteen teams around the world participated in the BWCN.
104. Wu, Z. Y., Walski T. et al. (2012) "Battle of Water Calibration Networks" ASCE journal of water resources planning and management, Sept/Oct. Issue
102. Wu, Z. Y. and Xu, G. (2011). "Integrated Evolutionary Optimization Framework for Finite Element Model Identification" Proceedings of SMAR 2011, the 1st Middle East Conference on Smart Monitoring, Assessment and Rehabilitation of Civil Structure, Feb. 8-10, 2011, Dubai, UAE
ABSTRACT
An integrated evolutionary optimization method is developed for finite element (FE) model identification and structural damage detection. The method is formulated to optimize FE model parameters such that the difference between the observed and analyzed responses is minimized. It is implemented as a generalized framework by coupling the well-developed FE analysis model with the evolutionary optimization technique. The implemented framework is designed by using Extensible Markup Language (XML) to ensure its compatibility and extensibility. A competent genetic algorithm (GA) is employed to search for the optimal and near-optimal solutions. Each solution is evaluated by the defined error function taking into account both static and modal responses. The approach is tested with an example of truss structure damage detection. The results obtained show that the proposed method is effective at detecting damage and that the framework is generic at facilitating FE model parameter identification.
101. Zheng Yi Wu, Pratik Deb, Sudip Chakraborty, Qiang Gao and Dru Crawley (2011). "Parallel Optimization of Structural Design and Building Energy Performance", proceeding of ASCE Structure Congress 2011, Las Vagas, NV, USA
ABSTRACT
An optimization model is formulated and developed for simultaneously minimizing structure cost and energy consumption cost while satisfying all the design constraints. The method is capable of optimizing any combination of user specified structural design variables such as cross section areas, member sizes, truss topology and connectivity, as well as energy performance related design variables e.g. windows, insulations. Genetic algorithm is applied to generate and optimize the design solutions that are analyzed by performing structure analysis and energy analysis. The optimization model is parallelized across a network of machines to speed up the computation. The parallel optimization ensures scalable performance of the implemented method.
100. Wu, Z. Y. (2011). "High Performance Optimization of Civil Infrastructure Systems on a Cloud." proceeding of ICCES11, Nanjing, China
ABSTRACT
Infrastructure system optimization is usually an implicit and nonlinear programming problem. The problem has to be carefully formulated in the way that classic mathematic programming can be applied to. In comparison, a genetic algorithm (GA) based on natural evolution mechanics and genetic reproduction is found flexible at solving the engineering problem as is, namely without special mathematic treatment. A competent GA has been successfully applied to water distribution optimization problems including model calibration, network design and pump scheduling. It has been decoupled and generalized as Darwin optimization framework, incrementally developed as a general parallel optimization solver for infrastructure system analysis. The framework is further developed and deployed as thin-client architecture on a high performance computing (HPC) cloud.
The HPC cloud computing platform is comprised of a HPC cluster and the parallelized optimization framework. It enables engineers to efficiently and cost-effectively undertake computation-intensive tasks of large infrastructure system optimization. Based on the outcomes of the previous research projects, a wide range of optimization problems, including, but not limited to, water pipeline leakage detection, pump scheduling, geometry design, structure damage detection and building performance-based design optimization, have been solved by applying the HPC optimization.
99. Wu, Z. Y and Eftekharian A.A. (2011). "Parallel Artificial Neural Network Using CUDA-enabled GPU for Extracting Hydraulic Domain Knowledge of Large Water Distribution Systems", in proceeding of 2011 World Environmental & Water Resources Congress on May 22-26, 2011, Palm Springs, California, USA.
ABSTRACT
This paper presents an Artificial Neural Network (ANN) implemented on Graphic Processing Unit (GPU) using Compute Unified Device Architecture (CUDA) for parallel computation and its application in water distribution systems. First of all, an overview is given for underlying CUDA architecture and state of the art techniques in parallel computing. Due to ANN’s structure and CUDA-GPU’s high performance computing architecture, artificial neural network is found one of the best candidates for GPU implementations. The implementation and testing of a parallel CUDA-ANN are conducted for capturing domain knowledge of water distribution systems. CUDA-ANN is trained and retrained in a very efficient manner. Instead of taking days or weeks to train CUDA-ANN for a large system, it only takes a few minutes to achieve the same accuracy as CPU-based ANN. The results obtained from the trained CUDA-ANN show good agreement with the hydraulic model. It is demonstrated that CUDA-ANN can be an effective surrogate solver for solving real time optimization problems of large dimensions
98. Wu, Z. Y., Sage, P. and Croxton, N. (2011). "Identifying Leakage Hotspots and Valve Status Using Night Fire Flow Testing Data" presented at 2011 World Environmental & Water Resources Congress on May 22-26, 2011, Palm Springs, California, USA
ABSTRACT
This talk presents the method and the case study for simultaneously identifying leakage hotspots and valve status using fire flow test data undertaken at minimum night flow hours. The parameter identification is improved with field data collected under the hydraulic conditions that induce more head losses at low demand hours. Collaborated with United Utilities in UK, night fire flow testing (NFFT) was conducted for collecting the field data of pressures and flows by flushing the selected hydrants. This talk will present the results of leakage and valve status detection of a district meter area (DMA) using PDLD approach with NFFT data.
97. Z Y Wu & I Lee, "Lessons for Parallelizing Linear Equation Solvers", CCWI 2011, Exeter UK
ABSTRACT
Efficiently solving a linear equation system is crucial for many engineering modelling systems, e.g. water distribution analysis model. In this paper, a number of linear equation system solution methods are closely reviewed for their convergence rate and general applicability to a wide range of modelling systems. Numerical experiment has been undertaken in parallelizing the solution method that is selected for its fast convergence rate among others. A portable and extensible linear equation solver toolkit for scientific computation, originally developed by Argonne National Labs, has been employed and tested out with a number of numerical benchmark problems. The parallelized linear equation solver has been applied to water distribution analysis. The results have been demonstrated for benchmarking the computation time on different sizes of the water system models. A detailed analysis has been undertaken on computation time used by different tasks for solving large water distribution model such as more than 250,000 pipe elements. Lessons are learned and summarized for developing an effective approach for achieving performance gain for a broad range of water distribution analysis.
97. Z Y Wu, Q Wang, S Butala & T Mi, "Generalized Framework for High Performance", CCWI 2011, exeter, UK
ABSTRACT
This paper summarizes the research in parallel computing techniques for genetic evolutionary optimization. Different paradigms are compared for portable parallel computation on single multi-core PC and a cluster of PCs. The strength and weakness are reported for various parallel and distributed computation models. An extensible parallel optimization framework has been developed for solving optimization problems in infrastructure system analysis. The framework encapsulates the core methodology for solving single and multiple objective optimization problems, the techniques for handling equality and inequality constraints, and also provides the options for executing solution evaluation/fitness in multiple processes as needed. It decouples the parallel optimization solver from domain applications and enables rapid implementation of infrastructure optimization projects with thin-client architecture. The parallel optimization is based on the competent genetic algorithm that has successfully applied to water distribution optimization problems including model calibration, network design and pump scheduling. The framework has recently been applied to geometry design optimization, building energy performance-based design optimization, finite element analysis model updating (calibration) and damage detection for civil infrastructure health monitoring.
96. J L Syed & Z Y Wu, "Applying Genetic Algorithm Optimization to Identify Leakage Locations inApplying Genetic Algorithm Optimization to Identify Leakage Locations in Distract Meter Area with Roof Tanks and Underground Reservoirs " CCWI 2011, Exeter UK
ABSTRACT
Minimizing and controlling leakages in Water Distribution pipes are currently one of the greatest concerns for every public water utility due to depleting water resources. District Area Meters (DMAs) play a significant and vital role in minimizing leakages and controlling losses as DMA is usually supplied from a single feeding point and the difference between the inflow and the flow actually consumed by customers within the DMA gives an approximation of the water losses. DMAs with roof-tank and underground reservoir are different in demand pattern from those directly supplied by distribution pipelines. Water is supplied to and stored in the underground reservoirs during the night while actual consumptions are supplied from the roof tanks. The distribution demand is usually high at night and low during the day. In this study, the optimized calibration of the hydraulic model using Genetic Algorithm is carried out to detect the leakage nodes within the DMA based on the field measured data. Four scenarios of the optimized calibration have been conducted for the combinations of leakage node detection, demand adjustments and valve setting identification. The hydraulic results of both un-calibrated and calibrated models are compared and checked with the actual field leakage conditions. It is observed that the leakage detection and valve setting identification produce the best goodness-of-fit. Some of the critical leakage nodes identified by the optimized model calibration supports the actual leakage conditions. Since pipe water leakages are directly proportional to the nodal pressures, therefore, this study is helpful in minimizing leakage by setting the appropriate PRV pressures at the DMA feeding point during minimum, average and maximum water consumptions.
95. P Sage, N Croxton & Z Y Wu, "Recent developments in Leak Hotspot Detection using Network Model Optimization" CCWI 2011, Exeter, UK
ABSTRACT
This paper reports on a recent application of a pressure-dependent leakage detection (PDLD) method for water network model calibration and associated leakage detection. The study has been undertaken on a United Utilities’ water distribution system for which a District Meter Area (DMA) and its associated Discrete Pressure Areas (DPAs) have already been established. In addition to the conventional 15-minute interval data, a novel feature of the method has been the introduction of a Night Fire Flow Field Test (NFFFT) in which a number of hydrants were discharged in a planned sequence at night while the pressures and flows were recorded every minute during the field test. Both types of field data have been used for leakage detection and model calibration optimization. The results obtained have compared well with the subsequent leakage detection and repairs in the field. Benefit analysis has also been performed for the PDLD approach as applied to the leakage detection undertaken. Several valuable lessons were learnt during the application of the NFFFT, the ensuing interpretation of the captured data and processing of it for presentation to the optimization engine. Using a further hydrant, a leak was fabricated in one of the DPAs and in the other a line valve was closed. These two interventions were ignored in the model presented to the optimizer to test whether they could be found during the optimization process. They both were. It has been found that the combination of network model optimization and NFFFT provides a useful addition to the existing techniques for leakage management. The method discussed enables engineers to undertake leakage hotspot optimization as an independent task or combine the task with hydraulic model calibration, subject to suitably varied field data. Initial estimates, based on task durations, indicate that there are likely to be positive cost benefits derived from the subsequent detection of ‘hard-to-find’ leaks as well as those to be gained from making the model detected hotspots available to the leakage teams to accelerate their routine searches for unreported leaks.
94. H Gao, D Diaz & Z Y Wu, "Applying i-Model Technology for Enabling Interopability of Water Distribution System Analysis", CCWI 2011, Exeter UK
ABSTRACT
Interoperability in infrastructure community is primarily concerned about sharing data between systems, particularly across discipline or department boundaries, engaging in a project with a new partner, downstream from design in the infrastructure asset’s lifecycle, and integration of information systems. In this paper, an information model, named i-model, is presented as one of the interoperability approaches. It is featured with (1) a published rendition in a secure read-only container, (2) a portable, self-describing, and semantically rich data file, (3) an interoperable deliverable tailored for change management, (4) "currency" of information exchange for all phases of a project lifecycle, and (5) an industry concept (like e-mail) rather than a Bentley brand. In this paper, the i-model, which has been implemented as one of the backbone components for infrastructure software solutions, is integrated with water distribution modelling package. There are various possible applications of imodel for water distribution analysis, such as using i-model for sharing the desired modelling attributes, updating asset management systems with the latest model information and providing model-based operation strategy. In particular, we illustrate that i-model is an effective means for constructing 3D water distribution model within 3D city. It helps to leverage the usage of 2D GIS and conventional hydraulic model information.
93. Wu. Z. Y. & Xu G. (2011). "Effective Method for Locating Damage Elements by Parallel Optimization of Model Updating" Proceeding of SHMII-5, Cancun, Mexico.
ABSTRACT
Many different methods including physics-based damage detection, data-driven damage detection and statistics-based damage detection have been developed, but a limited number of successful applications are achieved. With large investment made in developing smart sensors and monitoring the aged civil infrastructure systems, it is essential to develop effective method to leverage the usage of the large scale of the observed data for detecting possible damage at early stage. This paper presented an effective damage detection model and the solution method for identifying structure damage elements. The model is formulated to simultaneously search for the given number of the damaged elements and the corresponding damage indicators. It is a mixed integer and continuous optimization problem that is solved by applying a competent genetic algorithm to minimize the discrepancy between the field monitored and model analyzed responses. A software framework previously prototyped for FE model parameter identification is extended for structural damage detection. The framework takes advantage of well-developed FE analysis models and software design patterns. The method is implemented by parallelizing the solution evaluations on a cluster of many-core machines. The parallel optimization is essentially speed up the computation and enable fast convergence of the integrated method. The developed method has been tested on the identification of the damage scenario for a large truss structure. The results obtained show that the proposed method is effective at detecting damage in a large infrastructure system.
92. Wu, Z. Y, Sage, P. and Turtle, D. (2010) “Pressure Dependent Leakage Detection Approach and its Application to District Water Systems.” ASCE Journal of Water Resources Planning and Management, Vol. 136, No. 1, pp. 116-128.
ABSTRACT
Cost-effective reduction of water loss is a compelling but challenging task for water utilities. This paper presents a model based optimization method for leakage detection of water distribution systems. Leakage hotspots are assumed to exist at the model nodes identified. Leakage is represented as pressure dependent demand simulated as emitter flows at selected model nodes. The leakage detection method is formulated to optimize the leakage node locations and their associated emitter coefficients such that the difference between the model predicted and the field observed values for pressure and flow are minimized. The optimization problem is solved by using a competent genetic algorithm. The leakage detection method is developed as an add-on feature of the optimization-based model calibration tool. This enables engineers to undertake leakage hotspot optimization as an independent task or combine the task with hydraulic model calibration, subject to suitably varied field data. Two case studies are discussed in this paper including an example from literature and a district water system in United Kingdom. The results obtained illustrate that the optimization model for predicting leakage hotspots can be effective despite the recognized challenges of model calibration and the physical measurement limitations from the pressure and flow surveys also referred to as field tests. It is found that the method is effective at being applied for hydraulic conditions that occur in the early hours of the morning, often on water networks with excess design capacity and where hydraulic gradients are slack and loggers may sometimes be working close to their limits of accuracy.
91. Wu, Z. Y. and Yan, X. G. (2010) “Gene Expression Approach for Short-Term Water Demand Forecasting”, WDSA2010, Sept 11 – 13, 2010, Tucson, Arizona, USA
ABSTRACT
Rapidly increased concern in energy efficiency has focused attention more sharply on the potential for energy savings and carbon footprint reduction. For water distribution networks, pumping operation cost represents the major portion of overall cost in order to meet supply requirement for adequate water quantity and water quality. It is often the single largest cost element for a water system. In order to improve operation efficiency, pumps need to be scheduled and operated in an optimal way. A fundamental requirement for pump scheduling is the forecast of what water consumption/demand is going to be in a short term such as 24 hours. Therefore, developing a short-term demand forecasting model is of great importance to ensure efficient water system operation. Prediction modeling has attracted tremendous effort from industry sectors although it is not easy to predict a future event. Water utilities, like electricity and gas industry sectors, have maintained some sort of demand forecasters based on previous consumption patterns and operation experiences. The resulting forecasting is frequently quite adequate in terms of meeting the supply requirements. The disadvantage with the approach lies in the fact that the method is highly subjective and liable to human errors/biases. In order to develop a robust and effective approach for practical application, this research has focused on the development of a short-term water demand method using genetic programming. The developed method is tested with the historical flow data for a real water system in UK to illustrate its applicability and effectiveness of the approach.
90. Wu, Z. Y. et al. (2010) “Pressure-dependent Leakage Detection Method Compared with Conventional Detection Techniques”, WDSA2010, Sept 11 – 13, 2010, Tucson, Arizona, USA.
ABSTRACT
Cost-effective reduction of water loss is a compelling but challenging task for water utilities. This paper presents a model based optimization method for leakage detection of water distribution systems. Leakage hotspots are assumed to exist at the model nodes identified. Leakage is represented as pressure dependent demand simulated as emitter flows at selected model nodes. The leakage detection method is formulated to optimize the leakage node locations and their associated emitter coefficients such that the difference between the model predicted and the field observed values for pressure and flow are minimized. The optimization problem is solved by using a competent genetic algorithm. The leakage detection method is developed as an add-on feature of the optimization-based model calibration tool. This enables engineers to undertake leakage hotspot optimization as an independent task or combine the task with hydraulic model calibration, subject to suitably varied field data. Two case studies are discussed in this paper including an example from literature and a district water system in United Kingdom. The results obtained illustrate that the optimization model for predicting leakage hotspots can be effective despite the recognized challenges of model calibration and the physical measurement limitations from the pressure and flow surveys also referred to as field tests. It is found that the method is effective at being applied for hydraulic conditions that occur in the early hours of the morning, often on water networks with excess design capacity and where hydraulic gradients are slack and loggers may sometimes be working close to their limits of accuracy.
89. Wu Z. Y. and Walski, T. (2010) “Progressive Optimization Approach for Calibrating Extended Period Simulation Mo
del” WDSA2010, Sept 11 – 13, 2010, Tucson, Arizona, USA
ABSTRACT Hydraulic model calibration is an important task for engineers to construct a robust and effective model. Although many calibration methods and tools (Walski et al. 2003; Savic et al., 2009 and many others) have been developed by applying different optimization methods, challenges still remain in handling a variety of the model parameters, effectively utilizing multiple field observed datasets and supporting a complete process of model calibration. Optimization tools work well for calibration once the user gains an understanding of model dependencies i.e. which variable accounts for the differences between the model and the field data. Having the ability to use a variety of different parameters for adjustment enables the user to explore and eliminate different sources of error and focus on the most important sensitivities. However, there is still a need for sound judgment on the part of the user to drive the process. Therefore, calibration is unlikely to be completed in one single optimization run. In this paper, a progressive optimization procedure, based on the insights and the methods developed by the authors (Walski 1990; Wu et al. 2002, Wu 2009), together with the data processing and visualization tools in the modeling package (Bentley 2009a), is proposed and applied to solve BWCN problem.
88. Wu, Z. Y., Butala, S. and Yan, X. G. (2010) “High Performance Cloud Computing for Optimizing Water Distribution Pump Operation” the 10th Hydroinformatic Conference, Tianjin, China.
ABSTRACT
The enterprise landscape is rapidly changing. Corporations today demand speed and flexibility from their applications. They demand for the services that allow them to make better business decisions, create more satisfied customers, and react ever more quickly to evolving market conditions. Current economic circumstances and increased competition are also driving the demand for a more effective model to deliver applications and services to the end users. This relentless push for a faster, better and more cost-effective technology delivery model has set the stage for new approaches to application development, deployment and management. Enterprises require applications with greater scalability, agility and easier management capabilities.
Enter cloud computing, an innovative model for delivering IT infrastructure, applications and data that shifts the emphasis from static, stand-alone application silos to dynamic, shared environments, dynamically allocated among various tasks and accessed via a world wide web connected network.
The concept of cloud computing can be traced back to ideas such as utility computing, which envisioned a future where packaged information technology services would be metered and delivered to customers much like electricity, gas and water. SaaS companies like Salesforce.com deliver Customer Relation management (CRM) services, so clients can manage their customer information without installing specialized software.
87. Wu, Z. Y. and Yan, X. G. (2010) “Genetic Programming Approach for Water Distribution Demand Forecasting” the 10th Hydroinformatic Conference, Tianjing, China.
ABSTRACT
Rapidly increased concern in energy efficiency has focused attention more sharply on the potential for energy savings and carbon footprint reduction. For water distribution networks, pumping operation cost represents the major portion of overall cost in order to meet supply requirements for adequate water quantity and water quality. It is often the single largest cost element for a water system. In order to improve operation efficiency, pumps need to be scheduled and operated in an optimal way. A fundamental requirement for pump scheduling is the forecast of what water consumption/demand is going to be in a short term such as 24 hours. Therefore, developing a short-term demand forecasting model is of great importance to ensure efficient operation.
Prediction modeling has attracted tremendous effort from industry sectors although it is not easy to predict a future event. Water utilities, like electricity and gas industry sectors, have used some sort of demand forecasters based on previous consumption patterns and operation experiences. The resulting forecasting is frequently quite adequate in terms of meeting the supply requirements. The disadvantage with the current practice of demand forecasting lies in the fact that the method is highly subjective and liable to human errors/biases. In order to develop a robust and effective approach for practical application, this research has focused on the development of a short-term water demand method using genetic programming. The developed method is tested with the historical flow data for a real water system in UK to illustrate its applicability and effectiveness of the approach.
86. Sage, P, Wu, Z. Y. and Croxton, N. (2010) “Improves Leakage Detection and Model Calibration Using Fast Field Testing” EWRI Conference 2010, Providence, RI.
ABSTRACT
A pressure-dependent leakage detection method has been developed. The results indicate that the method is effective despite model calibration problems and field data limitations. Optimisation predictions can be improved when based on hydraulic conditions that occur at night that have been impacted by hydrant and/or other planned interventions. This is even so for networks with excess design capacity and where loggers may be working close to their limits of accuracy.The paper considers the impact of more interventionist field testing on water mains networks and its beneficial influence on model calibration and optimisation techniques. The approach includes planned opening and shutting of hydrants during and around minimum night-flow. Line valves are also operated in a planned way. The water industry has been averse to such interventions because of concerns about discoloured water complaints. Mitigation of such risks is addressed. Due to the need to collect most of the data at night including hydrant discharges the new approach has been referred to as Night Fire Flow Testing (NFFT). Using fabricated NFFT, two simple desk top models, one with two pressure zones, have been reviewed as have two all-mains models supplied by United Utilities.
85. Wu, Z. Y. (2010) “Parallel Optimization for Minimizing Energy Cost of Water Distribution Pump Operation”, AWWA ACE2010, June 20- 24, 2010, Chicago, IL, USA
ABSTRACT
Pump scheduling has been a research topic over the last 30 years and many optimization methods have been applied to pump scheduling. However, few successful applications are achieved in the water industry. This is mainly due to (1) lack of a robust, effective and efficient solution method developed for practical application and (2) no flexible off-the-shelf control software that is readily available and applicable for water utilities. In the mean time, parallel computing has been a research topic for decades. It is employed to achieve the speedup in computation efficiency for solving engineering and scientific problems that requires for intensive computation. Since 1980’s, the faster a PC is built, the faster the sequential computing programs could run. Such a ‘free lunch’ is over when the CPU speed is reaching the limit at which energy consumption is the primary concern so that computer industry has to shift from single-core paradigm to multi-core PCs configuration. It is highly unlikely a faster CPU will be made for PC but multi-core machines are becoming the mainstream of PC industry. Therefore, parallel computing is not only an approach for solving computation-intensive tasks but also essential for making use of the computing power offered by multi-core PCs that are readily accessible by engineers. This paper presents a variety of parallel computing models. Computation efficiency is enhanced by developing a scalable parallel computing framework, which allows many-core machines to be employed for the maximum computing efficiency. The improved pump scheduling method is applied to the optimization of a water system in the UK. It ensures scalable, portable and robust high performance optimization of water distribution operations.
84. Wu, Z. Y. (2009) “A Unified Approach for Leakage Detection and Extended Period Model Calibration of Water Distribution Systems”, Urban Water Journal, 6(1), 53-67.
83. Wu, Z. Y., Tryby, M., Todini, E. and Walski, T. (2009) “Modeling Variable Speed Pumps for Target Hydraulic Characteristics”, Journal of America Water Works Association, Vol. 101, No.1, pp54-64.
82. Wu, Z. Y., Walski, T., Wang, R.H. Boulder, D., Yang, S. Y. (2009) “Extended Global Gradient Algorithm for Pressure Dependent Demand Analysis of Water Distribution Systems”, ASCE Journal of Water Resources Planning and Management., Vol.135, No.1, pp13-22.
81. Wu, Z. Y. and Clark, C. (2009) “Evolving Robust Hydraulic Model for Municipal Water Systems”, EWRA International Journal of Water Resources Management, 23:117-136.
80. Wu, Z. Y. (2009) “Cost Effective Water Quality Monitoring for Maximum Protection of Water Distribution Systems” Proceeding of the 3rd IWA-ASPIRE Conference, Oct. 18-22 2009, Taipei, Taiwan.
79. Wu, Z. Y. (2009) “Optimizing Target Hydraulic Head for Energy Efficient Operation of Variable Speed Pumps”, CCWI2009, Sept.1 – 3, 2009, Sheffield, UK
78. Wu, Z. Y. Woodward, K. Allen, T. (2009) “Pump Scheduling Study on a Demand Monitoring Zone System”, CCWI2009, Sept.1 – 3, 2009, Sheffield, UK
77. Wu, Z. Y. (2009) “Bridging Gaps Between Theory and Practice for Optimizing Urban Water Systems” World City water Forum 2009, Aug 18-22, Incheon, South Korea
76. Wu, Z. Y. and Zhu, Q. (2009) “Scalable Parallel Computing Framework for Pump Scheduling Optimization, ASCE, Proc. of EWRI2009, May 17-21, 2009, Kansa, Missouri.
75. Sethaputra, S.; Limanond S.; Wu, Z. Y.; Thungkanapak, P.; Areekul, K. (2009), “Experiences Using Water Network Analysis Modeling for Leak Localization”, Proc. of IWA Water Loss Conference, April 26-30, 2009,Cape Town, South Africa.
74. Wu, Z. Y. (2009) “Essential Steps for Applying Optimization Modeling Tool for Cost Effective Leakage Detection”, Journal of India Water Works Association.
73. Wu, Z. Y. (2008) “Data Usage Protocol for Leakage Detection and EPS Model Calibration”, WDSA2008, Aug. 17-20, 2008, Skukuza, South Africa.
72. Ostfeld A. J, Uber, E. Salomons, J. W. Berry, W. E. Hart, C. A. Phillips, J.-P. Watson, G. Dorini, P. Jonkergouw, Z. Kapelan, F. di Pierro, S.-T. Khu, D. Savic ; D. Eliades, M. Polycarpou, S. R. Ghimire, B. D. Barkdoll, R. Gueli, J. J. Huang, E. A. McBean, W. James, A. Krause, J. Leskovec, S. Isovitsch, J. Xu, C. Guestrin, J. VanBriesen, M. Small, P. Fischbeck, A. Preis1, M. Propato11, O. Piller, G. B. Trachtman, Z. Y. Wu and T. Walsk (2008). "Battle of the Water Sensor Networks (BWSN): A Design Challenge for Engineers and Algorithms." ASCE Journal of Water Resources Planning Management, Vol. 134, No.6, pp556-568.
71. Wu, Z. Y.; Sage, P.; Turtle, D.; Wheeler, M.; Hayuti, M.; Velickov, S.; Gomez, C. and Hartshorn, J. (2008) “Leakage Detection Case Study by Means of Optimizing Emitter Locations and Flows” WDSA2008, Skukuza, South Africa.
70. Walski, T., Wu, Z. Y., Hartell, W. and Culin, K. (2008) “Determining the Best Way to Model Distribution Flushing”, World Environmental and Water Resources Congress, May 13-15, 2008, Ahupua’a, HI, USA.
69. Wu, Z. Y. (2008) “Innovative Optimization Model for Water Distribution System Leakage Detection.” International Water Loss Workshop, Jan. 28-30, Marbella, Spain.
68. Wu, Z. Y. (2007) “Discussion of ‘Solution for Water Distribution Systems under Pressure-Deficient Conditions by Wah Khim Ang and Paul W. Jowitt” ASCE Journal of Water Resources Planning and Management, Vol.133, No. 6, pp567-568.
67. Wu, Z. Y and Gulterrize, A. (2007) “Integrated Modeling, Data Warehousing and Web Publishing for Water Asset Management”, IWA Water Asset Management International, Vol. 3 No.2, pp27-31.
66. Wu, Z. Y. (2007) “Improving Water Utility Revenue with State-of-the-Art Modeling Paradigm” Asian Water, Vol 23, No.4, pp16-19.
65. Wu, Z. Y. (2007) “Technical Perspective of Water Asset Management for Water Industry” Journal of Water and Wastewater Engineering, (in Chinese), Vol 144, No.6, pp 26-30.
64.Wu, Z. Y. (2007) “Parameter Optimization Applied to Large Water Distribution Models” Journal of Water and Wastewater Engineering, (in Chinese), Vol.144, No.4 , pp27-30.
63. Wu, Z. Y. (2007) “Genetic Algorithm Plays a Role for Municipal Water Systems”, ACM SIGEVOlution, Vol.1 No.4.
62. Wu, Z. Y. and Todini, E. (2007) “Extended Global Gradient Algorithm for Modeling Fixed-flow Variable Speed Pumps and Pump Battery” Water Management Challenges in Global Change: CCWI2007 and SUWM2007 Conference, Sept. 3-5, 2007, De Montfort University, Leicester, UK
61. Todini, E., Wu, Z. Y. and Walski, T. (2007) “Direct Calculation of Variable Pump Speed Coefficient for Water Distribution System Analysis” Water Management Challenges in Global Change: CCWI2007 and SUWM2007 Conference, Sept. 3-5, 2007, De Montfort University, Leicester, UK
60. Wu, Z. Y. and Sage P. (2007) “Pressure Dependent Demand optimization for Leakage Detection in Water Distribution Systems” Water Management Challenges in Global Change: CCWI2007 and SUWM2007 Conference, Sept. 3-5, 2007, De Montfort University, Leicester, UK, pp. 353-361.
59. Wu, Z. Y. (2007) “Integrated Simulation and Optimization Models for Detecting Water Distribution Leakage” AWWA ACE2007, June 24-28, Toronto, Ontario, Canada.
58. Wu, Z. Y. (2007) “From Theory to Practice on Evolutionary Optimization of Urban Water Systems” Workshop on Advances in Hydroinformatics (HIW07), June 4 – 7, 2007, Niagara Falls, Canada.
57. Wu, Z. Y. (2007) “A Benchmark Study for Minimizing Energy Cost of Constant and Variable Speed Pump Operation”, 2007 World Water and Environmental Resource Congress, May 15-19, Tampa, FL. USA
56. Wu, Z. Y. Todini, E. and Walski, T. (2007) “Enhancements for Modeling target Hydraulic Head by Automatic Calculation of Variable Pump Speed”, 2007 World Water and Environmental Resource Congress, May 15-19, Tampa, FL.
55. Wu, Z. Y. (2006) “Asset Management for Asian Water Utilities”, Asian Water, Vol.22, No.10, pp10-12.
54. Wu, Z. Y. (2006) “Technological Requirements for Water System Asset Management” Water Science and Technology: Water Supply, Vol 6 No 5 pp 123–128.
53. Wu, Z. Y. (2006) “Automatic Model Calibration Method for Water Distribution Water Quality Model.” Journal of Environmental Science and Health (A), Vol 41, No.7, pp1363-1378.
52. Walski T. and Wu, Z. Y. (2006) “Energy Saving Tips” AWWA DSS 2006, Sept, 17-20, 2006, Phoenix, Arizona.
51. Wu, Z. Y. (2006) “Technological Requirement for Water Distribution System Asset Management”, IWA World water Congress 2006, September 10 – 14, Beijing, China.
50. Wu, Z. Y. (2006) “Pressure Dependent Demand Modeling of Water Distribution System under Abnormal Conditions”, IWA World water Congress 2006, September 10 – 14, Beijing, China.
49. Wu, Z. Y. and Walski T. (2006) “How not to Use Optimization Based Model Calibration” AWWA ACE 2006, June, 17-20, 2006, San Antonio, Texas.
48. Walski T. and Wu, Z. Y. (2006) “Finding Energy Thieves in Your Water System” AWWA ACE 2006, June, 17-20, 2006, San Antonio, Texas.
47. Wu, Z. Y. and Walski T. (2006) “Simulate What Happen When Pressure is Low” AWWA DSS 2006, September, 17-20, 2006, Phoenix, Arizona.
46. Wu, Z. Y. and Clark, C. (2006) “Integrated Simulation and Optimization Approach for Practical Water System Analysis” the 8th Annual International Symposium on Water Distribution Systems Analysis, Cincinnati, Ohio, August 27-30, 2006.
45. Wu, Z. Y. and Sage P. (2006) “Water Loss Detection via Genetic Algorithm Optimization-based Model Calibration” ASCE 8th Annual International Symposium on Water Distribution Systems Analysis, Cincinnati, Ohio, August 27-30, 2006.
44. Wu, Z. Y. and Walski, T. (2006) “Multi Objective Optimization of Sensor Placement in Water Distribution Systems” the 8th Annual International Symposium on Water Distribution Systems Analysis, Cincinnati, Ohio, August 27-30, 2006.
43. Wu, Z. Y. and Walski, T. (2006) “Efficient Pressure Dependent Demand Model for Large Water Distribution System Analysis” the 8th Annual International Symposium on Water Distribution Systems Analysis, Cincinnati, Ohio, August 27-30, 2006.
42. Wu, Z. Y. (2005) “Asset Management: A New Perspective for Water Industry.” Proc. of the 1st International Conference for China Urban Water Development Strategy, Oct. 29-31, Beijing, China.
41. Walski, T. and Wu, Z. Y. (2005) “How Well Does Automated Water Distribution Model Calibration Work” AWWA DSS Conference, Sept. 18-21, 2005, Tampa, FL, USA.
40. Wu Z. Y. (2005) “Optimizing water system improvement for a growing community” International Conference of Computing and Control in the Water Industry, Sept. 5-7 2005, Exeter, UK.
39. Wu Z. Y. and Walski T. (2005) “Diagnosing error prone application of optimal model calibration.” International Conference of Computing and Control in the Water Industry, Sept. 5-7 2005, Exeter, UK.
38. Wu Z. Y. (2005) “Calibrating water quality model by means of fast messy genetic algorithm.” 2005 World Water and Environmental Resource Congress, May 15-19, Anchorage, AK.
37. Wu Z. Y. and Walski T. M. (2005) “Self-adaptive penalty method compared to other constraint handling techniques for pipeline optimization.” ASCE Journal of Water Resources Planning and Management May/Jun, pp. 181-192.
36. Wu, Z. Y. (2005) “Maximizing energy cost saving for pump operation”, Journal of Water and Wastewater Engineering, July issue, in Chinese.
35. Wu, Z. Y., Elio F. A. and Ernesto G. (2004) “Darwin Calibrator ¾ Improving Project Productivity and Model Quality for Large Water Systems” Journal of AWWA, Vol. 96, No.10, pp27-34.
34. Wu Z, Y. (2004) “A Benchmark Study on Maximizing Energy Cost Saving for Pump Operation”, the 4th International Conference on Watershed Management and Urban Water Supply, Dec. 13-15, 2004, Shenzhen, China.
33. Wu Z, Y. (2004) “Constructing Accurate Water Quality Model for Complying Public Health”, the 4th International Conference on Watershed Management and Urban Water Supply, Dec. 13-15, 2004, Shenzhen, China.
32. Wu, Z. Y, (2004) “Self-adaptive Penalty Cost for Optimal Design of Water Distribution Systems”, 2004 World Water and Environmental Resource Congress, June 22-26, Salt Lake City, UT.
31. Walski, T. Wu, Z. Y., and Hartell, W. (2004) “Performance of Automated Calibration For Water Distribution Systems”, 2004 World Water and Environmental Resource Congress, June 22-26, Salt Lake City, UT.
30. Wu Z. Y. and Walski T. (2003) “Water Quality and Geospatial Modeling for Water Security Management”, Water Security In The 21st Century, July 30 - August 1, Washington, D.C.
29. Walski, T., Baranowski, T., Wu, Z. Y., Mankowski R. & Hartell W. “Trading Off Relaibility In Cost In Optimal Water Distribution System Design”, in Proceeding of 2003 World Water and Environmental Resource Congress, June 22-26, Philadelphia, Pennsylvania.
28. Wu, Z. Y. Wang, R. H., Diezo, D., Walski, T. (2003) “Mining Water Consumption and GIS-based Data for Loading Water Distribution Models”, in Proceeding of 2003 World Water and Environmental Resource Congress, June 22-26, Philadelphia, Pennsylvania.
27. Wu, Z. Y, Walski, T., Mankowski, R., Cook, J. Tryby, M. and Herrin G. (2002) “Calibrating Water Distribution Model Via Genetic Algorithms”, in Proceedings of the AWWA IMTech Conference, April 16-19, Kansas City, MI.
26. Wu, Z. Y, Walski, T., Mankowski, R., Cook, J. Tryby, M. and Herrin G. (2002) “Impact Of Measurement Errors On Optimal Calibration Of Water Distribution Models”, in Proceeding of International Conference on Technology Automation and Control of Wastewater and Drinking Water Systems TiASWiK'02, Gdansk - Sobieszewo, June 19 - 21, Poland.
25. Wu, Z. Y, Walski, T., Mankowski, R., Cook, J. Tryby, M. and Herrin G. (2002) “Optimal Capacity of Water Distribution Systems”, in Proceeding of 1st Annual Environmental and Water Resources Systems Analysis (EWRSA) Symposium, May 19-22, Roanoke, VA, USA.
24. Wu, Z. Y. et al. (2001) “Optimal pump operation of water distribution systems using genetic algorithms.” Proc. of AWWA Water Distribution Symposium, San Diego, 23-25.
23. Wu, Z. Y. and Simpson A. R. (2002) “Self-adaptive boundary search of genetic algorithms and application to water distribution systems.” Journal of Hydraulic Research, IAHR, Vol. 40, No. 2.
22. Wu, Z. Y. and Simpson A. R. (2001) “Competent Genetic Algorithm Optimization of Water Distribution Systems.” Journal of Computing in Civil Engineering, ASCE, Vol 15, No. 2, pp89-101.
21. Wu, Z. Y. et al. (2001) “Rehabilitation of water distribution system using genetic algorithm” Journal of AWWA, Vol.93, No.11, 74-85.
20. Wu, Z. Y. et al. (2001) “Using genetic algorithm for water distribution system optimization.” in Proc. of World Water & Environmental Resources Congress, May 20-24, Orlando, Florida.
19. Wu Z. Y. (2001) “Discrete optimal design of water distribution systems by means of evolution.” In Proc. of the 29th IAHR World Congress, Sept. Beijing, China.
18. Boulos, P.F., Wu Z. Y., Heath E., and Hauffen, P. (2000) “Optimal design and rehabilitation of water distribution systems." In Proceedings of the ASCE Joint Conference on Water Resource Engineering, July 30-August 2, Minneapolis, MN, USA.
17. Boulos, P.F., Wu, Z. Y., Orr, C.H., and Ro, J.J, (2000) “Least Cost Design and Rehabilitation of Water Distribution Systems Using Genetic Algorithms.” Proceedings of the AWWA IMTech Conference, April 16-19, Seattle, WA.
16. Boulos, P.F., Wu, Z. Y., Orr, C.H., and Ro, J.J (2000) “Optimal design and rehabilitation of water distribution piping systems." In Proceedings of the ALES 2000 Annual Convention, July 20-22, American University of Beirut, Beirut, Lebanon.
15. Wu, Z. Y., Boulos, P.F., Orr, C.H., and Ro, J.J (2000) “An Efficient Genetic Algorithms Approach to an Intelligent Decision Support System for Water Distribution Networks.” in Proceedings of the Hydroinformatics 2000 Conference, July 26-29, Iowa, IW.
14. Wu, Z. Y. and Simpson A. R. (2000) “Evaluation of Critical Transient Loading for Optimal Design of Water Distribution Systems.” in Proceedings of the Hydroinformatics 2000 Conference, July 26-29, Iowa, IW.
13. Wu, Z. Y. (1998) Messy genetic algorithms for optimization of water distribution systems including water hammer, Ph.D thesis, 350 pages, the University of Adelaide, South Australia.
12. Simpson A. R and Wu, Z. Y. (1998) “Computer modeling of hydraulic transients in pipe networks and the associated design criteria.” International Congress on Modeling and Simulation, 8-11 December, 1997, Hobart, Tasmania, Australia.
11. Wu, Z. Y. and Simpson A. R. (1997) “An efficient genetic algorithm paradigm for discrete optimization of pipeline networks.” International Congress on Modeling and Simulation, 8-11 December, 1997, Hobart, Tasmania, Australia.
10. Simpson A. R and Wu, Z. Y. (1997) “Optimal Rehabilitation of Water Distribution Systems Using a Messy Genetic Algorithm.” AWWA 17th Federal Convention Water in the Balance, 16-21 March 1997, Melbourne, Australia
9. Wu, Z. Y. and Simpson A. R. (1996) “Messy genetic algorithm for optimal design of water distribution systems.” Research Report, No. 140, Department of Civil & Environmental Engineering, University of Adelaide.
8. Wu, Z. Y. and Chr. L. Larsen (1996) “Verification of Hydrological and Hydrodynamic Models Calibrated by Genetic Algorithms.” International Conference of Water Resources & Environment Research: Towards the 21st Century, Kyoto from Oct. 29-31, 1996, Japan.
7. Wu, Z. Y. (1994) Automatic model calibration by simulating evolution. M.Sc. Thesis, HH 149, 150 pages, International Institute for Infrastructure, Hydraulic and Environmental Engineering, Delft, The Netherlands.
6. Babovic V, Wu Z. Y. and Chr. L. Larsen (1994) “Hydrodynamic model calibration by means of simulating evolution.” Proceeding of 1st Hydroinformatics Conference, Sept. 1994, International Inst. for Infrastructure, Hydraulic and Environmental Engineering, Delft, The Netherlands.
5. Wu Z. Y. (1994) “Optimal operation of urban sewer system by using genetic algorithm and expert system.” research report, 50 pages, International Institute for Infrastructure, Hydraulic and Environmental Engineering, Delft, The Netherlands.
4. Wu Z. Y. and Wang Y. T. (1992) “Arch Dam optimisation design under strength fuzziness and fuzzy safety measure.” Proc. of Int. Conf. on Arch Dam, Hehai University, Najing, China pp 129 - 131.
3. Wu Z. Y. and Wang Y. T. (1991) “Fuzzy Optimization of Arch Dam.” Proc. of the 1st National Conf. on Hydraulic Structures for Young Scientists and Engineers, Oct. 25 - 29, 1991, Zhengzhou, National Press of Literatures.
2. Wu Z. Y. and Wang Y. T. (1987) “Structure fuzzy optimization and arch dam optimization design.” Proc. of the 1st provincial conf. on science and technology , Guizhou Guiyang.
1. Wu Z. Y. (1986) Structure fuzzy optimization and its application in arch dam design. M.Sc. Thesis, 120 pages, Guizhou Inst. of Tech., Guizhou, P. R. China.