Optimization Methods For Conventional Scheduling Of Drainage Pumps In Plain Polders

Plain polders are widely distributed in Shanghai and southern part of Jiangsu province in China. Within these polder areas along the dikes, there are many pumping station that have small installation capacity and simple equipment. Therefore, conducting routine scheduling optimization study for the pumping stations in the polder area has a great significance and practical value for preparing response plan, reducing operating costs of the pumping stations, and improving management level of the polder areas.Based on a review of the existing theories and methods on optimal operation of parallel pumping stations in China and abroad, considering the hydrological connections and distribution characteristics of the pumping stations in plain polder areas, this dissertation presents a comprehensive study on conventional scheduling optimization that targets minimizing the total energy consumption, using the orthogonal test selection method, dynamic programming and large system decomposition-polymerization analysis method, as well as the unsteady flow dynamic model; a preparation method for conventional scheduling optimization of the pumping stations was proposed. Major findings of this research are as below:(1)Based on thorough analysis on water system and pumping station distribution characteristics, we conducted feasibility study on optimizing conventional scheduling of pumping station for polder drainage, proposed technical route for optimization of the conventional scheduling of pump stations, i.e., combinations of limited working conditions that commonly exist, routine scheduling optimization using the method of optimization, allowing the averagely more efficient pumps operate more during the high efficiency running time, therefore achieving the optimization goal of minimizing the total energy consumption under specific drainage requirement.(2)On the basis of analyzing the main factors on drainage conditions of polder areas, and summarizing historical scheduling record, we established a classification standard that covers typical characteristics of all possible working conditions in conventional scheduling of the study area. For all kinds of drainage condition, we proposed corresponding target and general requirements respectively for optimal operation of the pumping stations, which in turn facilitate introduction of system analysis model for preparation of the conventional scheduling optimization program.(3)According to the distribution characteristics of pump stations in plain polder area, we proposed a preparation method based on "minimizing the total energy consumption, optimizing single objective test, excluding the worst results from multi-objective simulation as limited by the water level difference inside and outside the polder ". For each group of typical drainage condition,we conducted an orthogonal test with the groups of pumping stations as test factors, and different operation rules of each group of pumping stations as test levels and the minimum operation energy consumption as a test result; carried out the orthogonal analysis on the test results to obtain a combinational sequence of the different operation rules of the pumping stations in descending order.To get the minimum power consumption of each orthogonal test,we also proposed a one dimensional dynamic programming model and solving method for polders that are not under tidal influence, and a large system decomposition aggregation model and solution for polders that are under tidal influence. With a case study, we presented the whole procedure of the conventional scheduling optimization for given conditions under both with and without tidal influences; the results were evaluated both qualitatively and quantitatively.(4)Considering the difference in technical level of the management staff, we proposed a simplified optimization method that targets "minimizing the total energy consumption, optimizing single objective test, excluding the worst results from multi-objective simulation as limited by the actual pumping capacity and the water level difference inside and outside the polder"". After the success in obtaining the optimum scheduling sequence, the actual catchment ability was compared with the calculated drainage volume; alternative plans were selected considering the water level difference inside and outside the polder, the left one with mimimum total energy consumption was the optimal scheduling scheme for the pumping stations in the polder area. The applicability and practicability of the simplified optimization method was proved with a case study.