Optimization Of Operation Of Cascade Reservoirs By Artificial Bee Colony Algorithm

The cascade reservoir scheduling problem is a typical large-scale,multi-constrained nonlinear optimization problem.The traditional dynamic programming algorithm will have a “dimension disaster” with the increase of the solution dimension,and the intelligent optimization algorithm can overcome this problem,and its principle Simple,quick to solve.This paper uses the Artificial Bee Colony Algorithm(ABC)to solve this problem.The ABC algorithm has better robustness,and has no requirement for the continuity and concavity of the solution function.Under the premise of low precision requirements,it can provide a satisfactory solution for the problem of optimal scheduling of cascade reservoirs.The ABC algorithm cooperates according to the division of labor between the lead bee,the follower bee and the scout bee,and the local and global search optimal solutions can be simultaneously performed in the iterative process.Under the premise of analyzing the components of ABC algorithm,this paper proposes a new bee colony algorithm(Elite Artificial Bee Colony,EABC)for the characteristics of cascade reservoir scheduling problem.The EABC algorithm is used to detect bees.The global search strategy has been improved,and the elite population has been introduced.The scrambling point of the center of gravity is used as a guide to improve the global search efficiency.In order to investigate the performance of the EABC algorithm,the paper first tests its optimization ability for high-dimensional functions in several classical standard functions.Then,the paper tests the optimization performance of the EABC algorithm on the simulation of the classic hydro-thermal power combined operation of the cascade reservoir group,and compares it with the genetic algorithm,the bird group algorithm and the standard bee colony algorithm.The results show that When EABC rounds up the number of elite populations,the optimal and standard deviation of all algorithms are obtained,which shows the superiority of EABC in the short-term optimal scheduling model of reservoirs.Finally,the paper applies the EABC algorithm to the actual large-scalemodel solving task of the six-pool hybrid long-term scheduling power generation in the Qinshui Basin,and selects representative typical years of abundant water,typical flat water years and typical dry years as experiments.Object.The paper has carried out many experiments and compared with other algorithms.The results show that the proposed EABC algorithm has better search accuracy and stability,and the obtained reservoir scheduling optimization results are also in line with the scheduling reality.This paper confirms the applicability and superiority of the proposed EABC algorithm in solving such problems as cascade reservoir optimal dispatching,and provides a feasible optimization method for the optimal model of cascaded reservoir optimal power generation.