A5001288130- Water Systems and Optimization
- Water resources management and optimization
- Urban Stormwater Management Solutions
- Hydraulic flow and structures
- Infrastructure Maintenance and Monitoring
- Structural Health Monitoring Techniques
- Water Treatment and Disinfection
- Water Quality Monitoring Technologies
- Geotechnical Engineering and Underground Structures
- Smart Grid Energy Management
- High voltage insulation and dielectric phenomena
- Anomaly Detection Techniques and Applications
- Concrete Corrosion and Durability
- Hydraulic and Pneumatic Systems
- Handwritten Text Recognition Techniques
- Energy Load and Power Forecasting
- Underwater Vehicles and Communication Systems
- Optical measurement and interference techniques
- 3D Surveying and Cultural Heritage
- Advanced Image and Video Retrieval Techniques
- Reservoir Engineering and Simulation Methods
- Geophysical Methods and Applications
- Neural Networks and Applications
- Electricity Theft Detection Techniques
- Time Series Analysis and Forecasting
Harbin Engineering University
2025
Shanghai Customs College
2025
Chongqing University of Science and Technology
2025
Wenzhou Polytechnic
2025
Jiangsu University of Science and Technology
2025
Bentley Systems (United States)
2015-2024
Bentley Systems (United Kingdom)
2005-2022
University of Michigan
2022
Stanford University
2022
Sichuan University
2021
Following the events of September 11, 2001, in United States, world public awareness for possible terrorist attacks on water supply systems has increased dramatically. Among different threats a distribution system, most difficult to address is deliberate chemical or biological contaminant injection, due both uncertainty type injected and its consequences, time location injection. An online monitoring system considered as major opportunity protect against impacts intrusion. However, although...
Crack detection has been an active research topic for civil infrastructure inspection. Over the last few years, many efforts have focused on applying deep learning-based techniques to automatically detect cracks in images. Good results reported with bounding boxes around detected However, there is no accurate crack segmentation, quantitative assessment, or integrated visualization context of engineering structures. In addition, most previously developed models trained homogenous images...
Cost-effective reduction of water loss is a compelling but challenging task for utilities. This paper presents model-based optimization method leakage detection distribution systems. Leakage hotspots are assumed to exist at the model nodes identified. represented as pressure-dependent demand simulated emitter flows selected nodes. The formulated optimize node locations and their associated coefficients such that differences between predicted field observed values pressure flow minimized....
Calibration is a process of comparing model results with field data and making the appropriate adjustments so that both agree. methods can involve formal optimization or manual in which modeler informally examines alternative parameters. The development calibration framework typically involves following: (1) definition variables, coefficients, equations; (2) selection an objective function to measure quality calibration; (3) set be used for process; (4) optimization/manual scheme altering...
The Battle of the Water Networks II (BWN-II) is latest a series competitions related to design and operation water distribution systems (WDSs) undertaken within Distribution Systems Analysis (WDSA) Symposium series. BWN-II problem specification involved broadly defined for an existing network that has be upgraded increased future demands, addition new development area. decisions parallel pipes, storage, operational controls pumps valves, sizing backup power supply. Design criteria hydraulic,...
This paper presents a new method for identifying leakage zones of water distribution systems. A large network is first divided into number zones. The zone used as the category label multiclass support vector machine (M-SVM), which trained with data set generated by simulation possible leakages using hydraulic model. M-SVM identification model and applied to determine likely observed field data. Two case studies are presented in this demonstrate effectiveness method. results indicate that has...
Timely assessment of damages induced to buildings due an earthquake is critical for ensuring life safety, mitigating financial losses, and expediting the rehabilitation process as well enhancing structural resilience where measured by infrastructure's capacity restore full functionality post extreme events. Since manual inspection expensive, time consuming risky, low-cost unmanned aerial vehicles or robots can be leveraged a viable alternative quick reconnaissance. Visual data captured...
A genetic algorithm has been applied to the optimal design and rehabilitation of a water distribution system. Many previous applications have limited small systems, where computer time used for solving problem relatively small. In order apply evolutionary optimization technique large-scale system, this paper employs one competent genetic-evolutionary algorithms—a messy enhance efficiency an procedure. maximum flexibility is ensured by formulation string solution representation scheme,...
Conventional water distribution models are formulated under the assumption that consumptions or demands defined at nodes known values, allowing nodal hydraulic heads and pipe flows to be determined by solving a set of quasi-linear equations. This formulation is well developed valid for scenarios in which pressures throughout system adequate delivering required demands. However, there some where pressure not sufficient supplying demand. These cases may include planned maintenances, unplanned...
The expansion of scale and the increase complexity construction projects puts higher requirements on level collaboration among different stakeholders. How to realize better information interoperability multiple disciplines software platforms becomes a key problem in collaborative process. openBIM (building model), as common approach exchange, can meet needs interaction well improve efficiency accuracy collaboration. To best our knowledge, there is currently no comprehensive survey context...
Nowadays, GPUs significantly boost rendering performance. However, the high memory requirements limit their use, especially on low-end mobile platforms. Compression techniques have been widely adopted to reduce consumption but face two primary issues when applied GPUs: 1) low repetition ratio caused by small raw data sizes and concurrency, 2) locality unpredictable behaviors. These limitations result in a compression compressors are devices. This paper introduces gCom , fine-grained...
Identifying how much water is being lost from networks and where the losses are occurring of great importance to utilities both for operational planning reasons as well reputation. In this paper, an optimization-based approach presented simultaneously quantifying locating via process hydraulic model calibration. The calibration formulated a nonlinear optimization problem that solved by using genetic algorithm. method developed integrated framework simulation modeling. Case studies...
Partitioning water distribution networks into district metered areas (DMAs) is a challenge for utilities due to the complexity of network topology, especially large-scale and highly looped supply systems. This paper presents novel approach, using complex theory multiobjective optimization, efficiently partitioning networks. In this first mapped on weighted undirected graph, its strength measured by an extended modularity random walk strategy. Then, Louvain algorithm employed based...
The success of the application genetic algorithms (GA) or evolutionary optimization methods to design and rehabilitation water distribution systems has been shown be an innovative approach for industry. optimal is a constrained non-linear problem. Constraints (for example, minimum pressure requirements) are generally handled within algorithm by introducing penalty cost function. near solution found when pressures at some nodes close required boundary critical constraints. This paper presents...
Placement of water quality sensor has received an increasing concern/or timely providing the warning possible contamination in a system. Due to large dimension distribution network and difficulty for predicting where event occurs, it is great challenge engineers come up with good locations any confidence effectively detect events. The problem complicated by fact that location evaluated against number objective criteria may include detection likelihood, expected time, affected population...
Deep Learning (DL) is the state-of-art paradigm of Artificial Neural Network (ANN) computing. It a new breakthrough in machine learning, and differentiates from conventional or shallow learning algorithms by emulating six-layer human neocortex, which unique for brain containing billions interconnected neurons. Unlike canonical ANN, DL capable self-learning data features mimicking functions layer cortex creating data-driven model with given dataset. This paper reports initial applications...
Abstract With increasing adoption of advanced meter infrastructure, smart sensors together with SCADA (Supervisory Control and Data Acquisition) systems, it is imperative to develop novel data analytics couple the results hydraulic modeling improve quality efficiency water services. One important task timely detect localize anomaly events, which may include, but not be limited to, pipe bursts unauthorized usages. In this paper, a comprehensive solution framework has been developed for...
Optimal design and rehabilitation of a water distribution system is constrained nonlinear optimization problem. A penalty function often employed to transform into nonconstrained problem within the framework genetic algorithm search. factor used for defining calculating cost solutions with constraint violation. Effective factors vary from one model another. This paper introduces self-adaptive approach artificially evolve both solutions. Solution robustness proposed quantified, along solution...
In spite of the growing computational power offered by commodity hardware, fast pump scheduling complex water distribution systems is still a challenge. this paper, Artificial Neural Network (ANN) meta-modeling technique has been employed with Genetic Algorithm (GA) for simultaneously optimizing operation and tank levels at ends cycle. The generalized GA+ANN algorithm tested on real system in UK. Comparing to existing operation, daily cost reduced about 10-15%, while number switches are kept...
Many US water supply infrastructure systems are nearly 100 years old. Where as some working efficiently when they were first installed, others showing their age with clogging, leaks, breaks, and service disruptions. To pinpoint the best solutions for each particular project, past practices often involved a tedious trial‐and‐evaluation procedure that seldom led to most effective or economical upgrading pipe networks. Because these upgrades modifications usually very expensive, it is...