Research And Implementation Of The Remaining Power Estimation Method Of The Energy Storage Power Station In Micro-grid Systems

In recent years, many bottlenecks in the development of renewable energy prompt a growing number of companies and researchers to turn their attention to the study of energy storage power station. The research technology of the energy storage power station can improve randomness, volatility and other issues on the renewable energy generation to a certain extent. It can achieve a smooth output of new energy power generation. The energy storage batteries are a very important part of the energy storage power station, which can achieve load shifting, load compensation, and improve power quality. We select group of lead-acid batteries as energy storage batteries, for further research on energy storage batteries, the establishment of a mathematical model in line with the dynamic characteristics of batteries, and the accurate prediction of the remaining charge of the storage battery have a very important significance to the development of battery management system and energy storage power station.Firstly, we make a series of experiments to test the performance of the batteries for understanding the performance of the lead-acid batteries deeply. And on this basis, we use the Thevenin model as an equivalent circuit model of the battery which can be well simulated the dynamic characteristics of the battery. Secondly, it determines the initial value of each parameter by using the intermittent discharge test with a constant current, and fits the function relationships between the parameters and the SOC, and simulates the voltage response data by combining with the electrical characteristic equations of the Thevenin model, and then searches out the best parameter online according with unconstrained nonlinear optimization methods by the difference between the two kinds of data which is the simulated voltage response data and the actual experiment voltage data. Finally, on the basis of the parameter identification of the mathematical model, we establish the state space equations, and use the extended Kalman estimation method to estimate the SOC. In the realization of the algorithms and the validation of the mathematical model, you can see the Thevenin model and the extended Kalman algorithm have higher estimation accuracy to meet the basic needs of the actual simulation and engineering applications.