Data Mining And Optimal Control On PV-Storage Hybrid Generation System

With the construction and development of the smart grid, the large-scale renewable energy are being connected to the grid. Power output with random fluctuations of renewable energy has a great impact to the safe operation of the grid. Solar energy sources is the greatest potential in renewable energy, photovoltaic power generation is the main form of solar energy utilization. However, there are different spans in power generation data collection stage, in the process of science research and worksite application. It’s lacking of theoretical basis. As a useful complement to renewable energy, energy storage system can smooth power output and help with the control strategy of the energy management system in the power generation side and assist renewable energy with power generation.In response to this situation, this paper analyzed data mining and optimal control on PV-storage hybrid generation system based on large number of historical data. Main contents are as follows:Firstly,aiming at the actuality of the lack of theoretical basis and mature conclusions in photovoltaic power generation operating data, the paper used probability and statistics principles to take an optimum sample size estimating for photovoltaic plant’s power output data with the autocorrelation analysis and a clusteranalysis of output data based on weather features.Secondly, a moving average method based on Markov chain prediction method is propoesd to determine the target power for energy storage system. A typical operation curve mining algorithm based on cloud model of energy storage system is proposed aiming at the problem of typical operation curve. The capacity configuration of the energy storage system is obtained based on the typical operation curve.Finally, a bidirectional complementary control strategy of battery energy storage system to complete the target power is proposed aiming at the problem of battery energy storage system loss. The comparative analysis is obtained under the bidirectional complementary control strategy and the traditional one.