Entetprise Load Optimization Based On Multiobjective Algorithm And Transformer Optimal Contorl

Iron and Steel enterprises as the most typical large electricity consumers in the power industry and the consumption of electricity has a large proportion in the cost of production.With the downturn of economy,the production dropped and load shrinks heavily in the steel industry,this resulted in the operating losses increase of distribution transformer.This study aims at the problem that the enterprise is in the state of non-economic operation,load forecasting model is built by combining the actual load of the enterprise,loads of steel south,steel south west and steel north are redistributed,then the transformer economic operation control strategy is put forwarded according to load fluctuating to realize the optimal operation of transformer and achieve the purpose of reducing the operating losses of the transformer.The research of this paper includes the following aspects:(1)The existing load forecasting methods are summarized and analyzed in this paper,the advantages and disadvantages of the traditional least squares support vector machine(LSSVM)and neural network in load forecasting are compared,the shortages of LSSVM which is used to forecast load and the characteristics of existing optimal LSSVM model are analyzed specifically,the new load forecasting model is constructed which is combined with LSSVM and global artificial fish swarm algorithm(GAFSA)that has high precision for the optimization of two dimensional space,and the loads of the steel enterprise are used to illustrate the accuracy of the prediction model.Finally,load of Changzhi steel for next five years is forecasted and this load forecasting model also provides a new way to solve the problem of load distribuation.(2)In this paper,the current load status of Shougang Changzhi Iron and Steel enterprise are investigated,the transformer economic load coefficient and economic operation region are analyzed.Considering the situation that the new lines are constructed in the processes of load distribution because of load geographical factors,the number of overhead lines for each line is selected according to the investigation situation of each line load,Shanxi state grid overhead line and tower selection condition,and the cost is included into the optimum load distribution.In order to meet the economic operation of the transformer and the minimum of load distribution cost,multi objective NASG-II algorithm is proposed to distribute the load of steel south,steel south west and steel north,the Parteo optimal solution set is obtained and the solution set is deeply analyzed,the result shows that the operating loss of the transformer is reduced obviously after the load distribution.Finally,according to the results of the load forecasting in the next five years,a practical load distribution scheme is formulated for the enterprise.(3)The transformer substation in the iron and steel enterprise often keeps two transformers running in parallel in order to ensure the reliability of the operation of the enterprise.However,the load of enterprise is of great fluctuation,and the non-economic operation is inevitable.Aiming at this problem,by analyzing the economic operation mode of transformer,the power saving effect of transformer under different operation modes is obtained,and the optimal switching strategy based on GAFSA is determined.Considering the feasible region of the traditional GAFSA is contiguous,however,in practice,the feasible region of project is discrete,an improved GAFSA is proposed to achieve the purpose of determining the optimal switching strategy in discrete spaces.Finally,the method is applied to the transformer operation of steel east station to illustrate the effectiveness of the proposed approach.