Gradient Crossover Real-coded Genetic Algorithm And Its Application In Optimal Operation Of Cascade Reservoirs

Optimal operation of cascade reservoirs is a high-dimensional complex nonlinear problem.With the development of modern computer technology and the rapid rise of biomimetic biology,a series of intelligent optimization algorithms have been widely used in reservoir optimal operation because of their good robustness,high flexibility and strong organization.Intelligent algorithm uses group information to guide key search areas and generate new points.Taking genetic algorithm as an example,it guides key search areas according to fitness,and generates offspring individuals through crossover operation.However,only according to the fitness value to guide the search of key areas,the amount of individual information that can be used is extremely limited,and the solution produced by crossover operation has a search blind area.In order to make better use of hydropower resources and maximize social and economic benefits,this paper improves the genetic algorithm as follows,and uses it to solve the optimal operation model of cascade reservoirs.(1)Increase gradient information to guide the search of key areas.Guide the search of key areas together with fitness information to improve the algorithm optimization;Gradient information is used to guide crossover,which avoids searching blind areas and reduces randomness and blindness of crossover.(2)Introducing ε adaptive constraints.The equality constraints and inequality constraints of the objective function are treated separately,which improves the diversity of population and accelerates the convergence of the algorithm.(3)Combining the above improvement measures,a new genetic algorithm-Gradient Cross Real-coded Genetic Algorithm(GC-RCGA)is merged,and the convergence of the improved algorithm and the effectiveness of solving the optimal value are tested by using classical function.The results show that GC-RCGA is obviously superior to other selected reference algorithms.The improved genetic algorithm is applied to solve the problem of the lowest thermal power cost in the short-term operation of hydropower system and the maximum long-term power generation of cascade reservoirs,and the calculated results are915,193 $ and 9,141.94 million k W·h respectively.Compared with the calculated results of other intelligent algorithms,GC-RCGA has better calculation results,faster solution speed and better stability,which shows that GC-RCGA has good performance and has certain applicability and superiority in solving the optimal operation of cascade reservoirs.