Microgrid Energy Optimization Management Based On Multi-time Scale Method

In recent years, energy shortages and environmental pollution problems have become increasingly severe, and the main solution to both problems is to develop the technology for the use of renewable energy. As a result, The micro-grid technology has great prospects in receiving large-scale wind and solar power to the grid. To make a better economy, reliability, autonomy, openness and flexibility for micro-grid when accepting a large-scale of renewable energy, this paper proposed a multi-time scale energy optimization management. The main work is as follows:(1) Firstly, the main sources of uncertainty in the micro-grid operation and their impact on the micro-grid tie-line exchanging power and micro-grid power quality. A uncertainty distribution model for wind turbine power fluctuations, photovoltaic power fluctuations, fluctuations in load power, unit operation failures and price fluctuations in the electricity market has been established in this paper.(2) Secondly, based on uncertainty distribution model, this paper established a multi-time scale energy optimization model which contained chance constraints. According to the prediction error level of different time scales, we analyzed the objective functions and operating constraints of three kinds of time scale optimization model, and finally when the event type of error occurs, start the real-time scheduling model to eliminate errors to realize the coordinated goal for micro-grid energy optimization.(3) Finally, for the chance constrained model, we proposed a hybrid intelligent algorithm which contains stochastic simulation, PSO and BP neural network to solve the problem. using stochastic simulation method to simulated microgrid uncertain variables, then the particle swarm algorithm to calculate the result. In order to speed up the calculation, we used BP neural network to approach functions with uncertain variables.