Comprehensive Evaluation Method of Energy Efficiency of Hydropower Station Based on SSA-LSTM Model
With the deepening of China's power system reform,hydropower has bid farewell to the traditional exten-sive development model,so it is urgent to support a more mature and general energy efficiency evaluation system to guide hydropower operation and dispatching.In this paper,a comprehensive evaluation method of energy efficiency of hydro-power stations based on deep learning was proposed.A long short-term memory(LSTM)network was introduced to construct a theoretical power generation model of hydropower stations.For the given original power generation sequence,the trend term,period term and noise were extracted by singular spectrum analysis(SSA).The LSTM network was built separately for the first two terms,and the theoretical power generation was obtained by superimposing the simulation re-sults.On this basis,the relative increase benefit index and the relative energy efficiency improvement rate index were proposed.The comprehensive score value of the hydropower station was obtained by using the entropy weight method.Finally,12 power stations in a Southern Province were taken as examples to evaluate energy efficiency.Experiments show that this method can fully reflect the energy efficiency characteristics of hydropower in dispatching operation.This study has reference significance for optimizing the dispatching strategy of hydropower stations and improving the level of hydropower dispatching.
万方数据
维普
