Research On Intelligent Optimization Methods For Power Generation,Peak Shaving And Ecological Operation Of Cascade Hydropower Reservoirs

China is a country with a severe shortage of water resources,and the runoff interannual variation is very uneven.Water shortages have worsened in many parts of the country in recent years due to climate change.How to effectively improve the utilization of water resources has become a scientific problem that domestic water resources and environmental protection agencies urgently need to solve.In recent years,with the large-scale commissioning and operation of hydro plant,managers have gradually realized that the joint operation of cascade reservoirs is an important method for the optimal allocation of water resources.However,the cascade reservoirs are connected by hydraulic connections,and each reservoir needs to meet a series of complicated equations and inequality constraints.Efficient solutions and scientific theories are urgently needed to deal with the cascade reservoirs operation.Thus,this paper focuses on the cascade reservoirs operation,the main research contents and achievements as follows:(1)An adaptive sine cosine algorithm(QSCA)based on the sine cosine algorithm for power generation problem of cascade reservoirs is proposed.In QSCA,the elite-guided evolutionary strategy to increase the position information of outstanding individuals in the population,which improves the balance between exploration and exploitation;the quasi-oppositional learning strategy to increase the search range of individuals in the feasible space and improve SCA’s search efficiency;the non-equal probability mutation strategy to increases the information sharing of elite individuals in the population,preventing a rapid decrease in population diversity during the search process.The results of 24 CEC2005 functions show that the proposed algorithm has better search performance and robustness than sine cosine algorithm in numerical function optimization.Finally,the cascaded reservoir in the Wu hydro plant are used as the engineering background and the power generation model is adopted,then the adaptive sine cosine algorithm is used to solve engineering problem.The results show that the adaptive sine cosine algorithm can maximize the power generation efficiency of the cascade reservoirs,and has good feasibility and engineering practicability.(2)An elite grey wolf optimizer(IGWO)based on the grey wolf optimizer for peak shaving problem of cascade reservoirs is proposed.The quasi-oppositional learning strategy to find a better individual position and improves the efficiency of the algorithm;The elite individual mutation strategy to retains the elite individuals during the search process and increases the diversity of the population;The elastic ball boundary strategy eliminates the effect of infeasible solutions on population evolution and improves the convergence accuracy of the algorithm.The results of the 12 benchmark functions show that the proposed algorithm has better convergence performance,feasibility and robustness in numerical functions than other classical algorithms.Finally,a commonly complex constraint processing strategy is introduced for the peak shaving problem of cascade reservoirs.The large-scale cascade hydropower and Wu hydro plant are used as the engineering background and the peak shaving model is adopted to solve the engineering problem by the elite grey wolf optimizer.The results show that the elite grey wolf optimizer can provide an effective scheduling plan for the power grid and provide a scientific basis for the peak shaving operation of cascade reservoirs.(3)A hybrid gravitational search algorithm(MGSA)based on the gravitational search algorithm for ecological operation of cascade reservoirs is proposed.The Levi’s local search strategy is introduced to search around the globall optimal position,which improves the algorithm’s convergence speed;The mixed mutation strategy is introduced to avoid population "premature convergence" and increase population diversity;The elite selection strategy is introduced to retain excellent individuals and improve the individual quality of the offspring.The 24 benchmark functions show that the proposed algorithm can overcome the premature convergence defect of traditional optimization algorithms in numerical functions.Finally,several ecological flow calculation methods are summarized,Wu hydro plant are used as the engineering background and the ecological operation model is used to solve the engineering problem by hybrid gravitational search algorithm.The results show that the hybrid gravity search algorithm is of practical engineering value and can provide a reasonable and effective solution for the ecological operation of cascade reservoirs.