Study Of Optimal Operation Of Large Scale Multi-Source Water Supply System

The problem that the spatial arrangement of city does not meet the bearing capacity municipal of water environment emerges as a major issue in our country. water shortage and environmental pollution stand out to be the two primary obstacles for water distribution system development. To relieve the contradictions between supply and demand and improve the water quality, many cities carried out the project of long distance water transfer and water source switch. During the process of water sources switching, the problem that how to operate the new and old water sources to make the water distribution system economical, reasonabl and safe is to be settled urgently in the world. This paper researches the theory and practice of optimal operation and safety analysis in water distribution system during the water source switch.The conventioanl hydraulic model in water distribuiton system is demand driven model, which can not model the pressure driven demand and represent the leakge characteristic. An improved pipe leakage model was developed and integrated into pressure driven demand hydraulic model to simulate the water supply network. In this paper, a new approach is proposed for calibrating pressure driven demand hydraulic and leakage model. This procedures uses a new node demand distribution method and Genetic Algorithm to solve the calibration objective, which involves comparing meaured system flows, pressures, and leakages with computer simulations. Pipe roughness and leakage coefficient are chosed as adjusting parameters. This paper integrates the dynamic storage tank model with the hydraulic model to simulate the water age in outflow of storage tank. Under the condition of different inflow pattern of storage tank, the corresponding pattern of water age in outflow is analyzed and compared.Optimal operation of multi-source water distribution system during the process of water source switch is a large-scale non-linear optimization problem with discrete and continuous variables, which represents one of the most difficult problems to solve. The proposed optimal operation model inclueds three objectives: minimization of operation cost (energy cost +treatment cost), minimization of leakage flow, and minimizaiton of avage age in water network nodes. A mixed coding methodology and a new crossover operator are developed according to the characteristics of decision variables. the constraints are handled with dirrent level.according to its importance. To improve the quality of optimal solutions, a Hybrid MOEA is proposed which combined with a local search algorithm and multiple attribute decision methoto solve the water source scheduling problem.A two-stage method is proposed to solve a new class of multi-storage tank multi-source system. In the first stage, the optimal storage policy of each tank is determined according to the electricity tariff, and the ground-level storage tank is modeled as a node. In the second stage, multi-objective evolutionary algorithm is applied to solve pump scheduling problem under the condition of meeting the water network demand and storage volume.This paper discusses the influence of water source switching on water quality in drinking water distribution system. The influence function of water source switching is proposed to assement water quality safety. Physical control measures are optimized to guarantee water quality safety. An optimal rehabilitation model considering water qualtiy is developed, which involves economic objective and effect conference of water quality.Taking shenyang water supply system as an example, the optimal schedule model of multi-tank multi-source water distribution system was solved by multi-objective evolutionary algorithm and multiple attribute decision method. Four kinds of MOEAS (NSGA-II, epsilon-MOEA, SPEA2, Hybrid MOEA) are implemented and compared.. By application of the proposed method in Shenyang water distribution system, daily operation cost savings of approximately 6.7% is obtained, and the leakage rate and aveage age of network nodes are improved greatly. Practical application of this method shows that proposed optimal operation and decision model can make efficient decision to support the operators.