Improved Membrane Computing Optimization Algorithm And Its Application In The Micro-grid Of Load Forecast And Economic Operation

Membrane computing, namely P system, is a distributed and parallel computing model. The nature of it is aiming at abstract computing model or computing idea from the functioning and structures of living cells, as well as from the way the cells are organized in tissues or higher order structures. P system becomes a new research area quickly, since it was proposed. The research area is mainly concentrated in three aspects, including theory research, implementation of software and hardware and the applied research.Membrane algorithm is an approximate optimization algorithm, which is inspired by P system. The rules and dynamic membrane structure of P system is effective to improve some basic intelligence algorithm. Therefore, membrane algorithm combined with some intelligent algorithm is applied in micro-grid because of the above thought in this paper. The algorithm is seen as a new solution plan for economic operation of micro-grid and load forecasting, but also as an advantage of extending the application of membrane algorithm. This main research contents are shown as follows:(1) On the one side, the basic principle of membrane computing optimization algorithm is discussed. On the other side, the structures of membrane algorithms are analyzed. Finally, a genetic algorithm based on P system was studied and it is applied in simulation experiments on several test functions.(2) The principle and model of distributed power applied in micro-grid model is simply introduced, so as to determine the mathematical model of micro-grid in interconnection and isolated operation. Then the genetic algorithm based on P system is applied in the simulation of the model. The results show that the advantage of micro-grid with battery by comparing two different models for micro-grid without battery and micro-grid with battery. At the same time, when comparing optimization results of the algorithm with basic genetic algorithm and improved genetic algorithm, the effectiveness of adopted algorithm is also highlighted in this paper.(3) In view of the advantages and disadvantages of commonly used algorithm in short-term load forecasting of power system, the algorithm which combining particle swarm optimization based on P system with BP neural network has been used in short-term load forecast of power system this paper. Where, the prediction process and steps are introduced, and the results of training model and prediction model are also simulated.