Development of a real-time, near-optimal control process for water-distribution networks

Abstract

This paper presents a new approach for the real-time, near-optimal control of water-distribution networks, which forms an integral part of the POWADIMA research project. The process is based on the combined use of an artificial neural network for predicting the consequences of different control settings and a genetic algorithm for selecting the best combination. By this means, it is possible to find the optimal, or at least near-optimal, pump and valve settings for the present time-step as well as those up to a selected operating horizon, taking account of the short-term demand fluctuations, the electricity tariff structure and operational constraints such as minimum delivery pressures, etc. Thereafter, the near-optimal control settings for the present time-step are implemented. Having grounded any discrepancies between the previously predicted and measured storage levels at the next update of the monitoring facilities, the whole process is repeated on a rolling basis and a new operating strategy is computed. Contingency measures for dealing with pump failures, pipe bursts, etc., have also been included. The novelty of this approach is illustrated by the application to a small, hypothetical network. Its relevance to real networks is discussed in the subsequent papers on case studies.

Optimizing the operation of the Haifa-A water-distribution network

“…Other features relate to the modification of the fitness scaling which is applied to adjust the range of fitness values during the search procedure, in order to avoid extreme values wielding too much influence Another was the introduction of an adaptive penalty function which ensures the penalty coefficients are neither too large or too small at each stage of the optimization process. A more detailed explanation of these modifications can be found in the third paper of this series (Rao & Salomons 2007). …”

“…In this application, a genetic algorithm (GA) optimizer and an artificial neural network (ANN) predictor have been combined, using a software package referred to as DRAGA-ANN (Dynamic, Real-time, Adaptive Genetic AlgorithmArtificial Neural Network), which was developed as part of the POWADIMA research project (see Rao & Salomons 2007). The approach is based on replicating a conventional hydraulic simulation model by means of an ANN, which is significantly more computationally efficient than using the simulation model directly.…”

“…Having shown that it was possible to formulate a dynamic, near-optimal control process for a small, hypothetical water-distribution network (Rao & Salomons 2007), the next major challenge was to apply the methodology developed to a real network. This would not only provide experience of scale-related issues but also give exposure to some of the idiosyncrasies in network design that are found in practice.…”

Incorporation of Variable-Speed Pumping in Multiobjective Genetic Algorithm Optimization of the Design of Water Transmission Systems

“…VSPs have been incorporated in the optimization of the operation of existing WDSs in previous studies (Wegley et al, 2000;Rao and Salomons, 2007;da Costa Bortoni et al, 2008;Wu et al, 2009). However, for the optimal design of WDSs involving pumping, FSPs have often been used, despite the advantages of VSPs discussed above.…”

Trading-off Constraints in the Pump Scheduling Optimization of Water Distribution Networks

“…Boulos et al (2001) developed the H2ONET tool, using genetic algorithms for minimal operation costs and pump scheduling. Zyl et al (2004), Yu et al (2005), Farmani et al (2006) and Rao & Salomons (2007) are other researchers who have employed evolutionary algorithms for the solution of the operational optimization problem of Water Distribution Networks (WDN).…”