Application Of Improved Particle Swarm Optimization In The Optimal Operation Of Hydropower Stations

The optimal operation of hydropower stations mainly studies the optimization ofscientific management and dispatching decision. At present, a lot of further researchhas been carried through and many results gained both at home and abroad, but thereare still incompletenesses in this theory needing improving. It needs further researchhow both hydraulic and electric contacts are reflected in the short-term optimaloperation model of cascade hydropower stations and so does how the vibrationfeatures of the units are taken in account in the inner-plant economical operation ofhydropower stations. What is more, conventional optimization algorithms for thesolution of the optimal models have many limitations in computing speed andconvergence. Particle swarm optimization (PSO) is a new simple and practical overalloptimization algorithm based on swarm intelligence. Because of its advantages, it hasgot great development and been used extensively. Firstly, to overcome somedisadvantages of standard PSO, catastrophe mechanism simulating the effects of thecircumstance is introduced into the PSO to form the catastrophe PSO, with itsperformance greatly improved. Secondly, the operation pattern of cascadehydropower stations is studied, whose overall water discharge is fixed during theoperation period, and the short-term optimal operation model of cascade hydropowerstations is established with many factors taken into account. Also, giving attention tooperational efficiency, vibration state of units and many other factors, the problems ofthe inner-plant economical operation of hydropower stations are studied, with theinner-plant economical operation model of hydropower station constructed. Next, inorder to make sure the two models' validity, they are applied to Shuoduogang CascadeHydropower Stations and Lijiaxia Hydropower Station respectively, and are solvedusing catastrophe PSO whereas the disadvantages of the conventional optimizationalgorithms. The analysis of the results shows the validity and practicability of the twomodels constructed in this paper, and the applicability of the PSO to the field ofoptimal operation of hydropower stations which provides a new way of solving theoptimal operation models.