Research On Short-term Hydrothermal Scheduling Based On Improved Krill Herd Algorithm

Short-term hydrothermal scheduling(STHS)is a nonlinear,multi-dimensional,multiconstrained and time-varying optimization problem.When the valve point effect is introduced,the problem becomes non-convex and more complicated.The research objective of thesis is to obtain the optimal fuel cost of thermal power plants,and two aspects of proceeding the valve point and improving the performance of algorithms are researched for this problem in this thesis.In order to solve the STHS problem that the valve point effect makes it difficult,a thermal power plant generation distribution model based on the valve point effect is established.In the load balance of distribution model,when the transmission loss is not considered,the method of adding the balance deviation of the system to a thermal power plant is more economical;when the transmission loss is considered,the method of distributing the balance deviation to each power plant in random ratio is proposed to approximate the transmission loss of the system,and an evaluation model of the transmission loss deviation of the system is established.In order to improve the optimization ability of krill herd algorithm(KHA)in solving the STHS problem,the hybrid chaotic maps are introduced to improving the global convergence speed of KHA;in order to avoid premature convergence of the algorithm,by recording the number of times that the target value of each generation of the optimal krill herd remains unchanged.A decision is made to mutate the position of the non-position dominant krill herd in its feasible region,and a hybrid chaotic krill herd algorithm is proposed(HCKHA),which is the first improvement of KHA in this thesis.Due to the non-memory characteristics of gravitational search algorithm(GSA),combined with the update method of particle swarm optimization(PSO)and the position mutation strategy of KHA,an improved gravitational search algorithm based on krill herd and particle swarm algorithm is proposed(KHA_PSO_GSA),which is the second improvement based on KHA in this thesis.The power generations of thermal power plants in the recent literatures are redistributed,and the established generation distribution model is evaluated by comparing the fuel cost value,transmission loss value and transmission loss deviation value before and after distribution.The performance of the proposed algorithms is analyzed through five test functions with different characteristics,and the proposed two algorithms are combined with the thermal power plants generation distribution model and used in the four-hydro three-thermal and four-hydro ten-thermal test systems.The non-parameter Friedman test is used to evaluate the difference of the algorithms.The results show that the optimization capabilities of HCKHA and KHA_PSO_GSA,the fuel cost value and transmission loss value of system are better than the optimization methods of KHA,CKHA and other related literatures.