Deep learning adaptive dynamic programming for real time energy management and control strategy of micro-grid

With the widespread application of distributed renewable energy in power systems, energy management problems in micro-grids have become increasingly significant. Offline or static optimization methods are frequently employed to solve the problems which are typically discrete and nonlinear. There are few online optimization methods and they are not only complex but also normally consider the distributed generations and loads in micro-grid as a whole. The outcome is that the online methods fail to reflect the composing characteristic of distributed multi-energy and the contribution made by developing renewable energy to reduce the consumption of traditional fossil energy. Moreover, results obtained by either the offline or the on-line methods can deviate to certain extents from the true values. This paper treats the management of distributed energy in micro-grids as an optimal control problem. Using the system control theory, a framework of real-time management of distributed energy in micro-grids is proposed. A deep learning adaptive dynamic programming is proposed for this framework. Due to the introduction of the concept of closed-loop feedback, the proposed management and control strategy is a real-time algorithm. Furthermore, the accuracy of managing and controlling the objective function can be improved. The gap of the flexible load after the optimization can be helpful in terms of guiding the flexible load consumers to change their habits of energy consumption, thereby reducing the coal-fired power generation and providing room for reducing carbon emissions. Because it is real-time, micro-grid operators can also realize intra-day scheduling, which can be done through combining this optimized management and control strategy with the mechanism of energy operating and managing. Moreover, some data sets of the real-time optimal control strategy are obtained in this paper. These intuitive and accurate data sets can be used to further optimize energy operating and managing of micro-grids, thereby realizing effective energy management for micro-grids. Finally, the real-time and effectiveness of the proposed management and control strategy are proved by the simulations. Some positive conclusions are drawn.