| Inter-basin Water Transfer Project (IWTP) is an important project to solve the uneven distribution of water resources in China, and improving the economy of IWTP is the key to ensure the sustainable construction of the project. Pumping station system has a large proportion in the cost of the whole project, therefore, it is of practical significance to study the pumping station system and reduce the cost of the operation of the pumping station. In this paper, a single pumping station as the research object, researching on the optimization schedule of multiple pump units with adjustable blade-angle. The research work of this paper mainly includes:(1)In this paper, improved hierarchical genetic algorithm (The improved HGA) is proposed. Dynamic elite preservation strategy is proposed, the strategy selects individuals whose fitness is greater than the average fitness and collect them in a set. Then selecting individuals whose fitness is greater than the set’s average fitness, these individuals are known as the elite individuals of the population. Five principles of crossover and mutation probability are summarized, and a new adaptive crossover and mutation probability is proposed based on these five principles. Based on the previous studies, directed crossover operator based discrete variable is proposed. The improved HGA model is used to solve the optimization schedule model of multiple pump units with adjustable blade-angle, the encoding method and the appropriate fitness function are proposed according to the characteristics of the model, and the solving process is described in detail.(2)In this paper, teaching and learning optimization algorithm based on multi Agent (MATLBO) is proposed. We analyze the convergence of TLBO. For TLBO is easy to fall into the local optimum, the multi Agent technology is introduced to construct the individual’s living environment and competitive environment Based on this, the paper puts forward the group learning operator and the self-learning operator. Using MATLBO to solve the optimization schedule model of multiple pump units with adjustable blade-angle. According to the characteristics of the model, presenting the two-dimensional vector coding, improving "teaching" and "learning" operation. The discrete directed crossover operator is used to realize the "teaching" operation, the single point crossover and bit mutation is used to realize the "learning" operation, and the solution process of the algorithm is described in detail.(3)In this paper, the improved HGA and MATLBO are used to study the optimal scheduling model of JiangDu No.4 multiple pump units with adjustable blade-angle. From the theoretical and experimental, the performance of the two algorithms are analyzed, the results show that MATLBO has a better ability to solve the problem, but it cost more time. The results of this paper are compared with the results of the literature, showing that the two algorithms in this paper have a better ability to solve the problem.(4)On the basis of the above work, this paper constructs and implements a optimization schedule system of multiple pump units with adjustable blade-angle based on the development platform of Studio 2012+SQL Server Visual 2012. The system can provide scientific and efficient management support for the management of pumping station. |
