Research On Robust Optimization Method For Joint Photovoltaic And Energy Storage In Microgrid Based On Adaptive Dynamic Programming

At present,the shortage of fossil fuel resources and the environmental pollution caused by the use of fossil energy to generate electricity have become increasingly serious.The development and utilization of renewable energy has become the primary task of sustainable development of society.As the cleanest renewable energy source with the highest current utilization rate,photovoltaic power generation is affected by uncertainties such as solar irradiance and temperature,making photovoltaic power generation difficult to predict accurately,which leads to the impact of unstable photovoltaic power generation on the stable operation of the power grid,and it will have a greater impact on grid optimization scheduling.The microgrid can effectively avoid the impact of photovoltaic grid connection on the large grid and reduce the impact of uncertainty in photovoltaic power generation on the grid.By combining the photovoltaic power generation system with the large power grid through the microgrid,the photovoltaic power generation system can be fully utilized and the large power grid can be stably and reliably operated.In view of the inaccurate prediction of photovoltaic power generation and the uneconomical problems of microgrid scheduling caused by inaccurate prediction of photovoltaic power generation,this paper mainly does some work:(1)Aiming at the problem of inaccurate prediction of photovoltaic power generation,a short-term prediction method of photovoltaic power based on principal component analysis(PCA)and particle swarm optimization(PSO)for radial basis function neural network(RBF neural network)is proposed.First,the historical data of photovoltaic power generation and corresponding meteorological data are used as raw prediction inputs.PCA is then used to reduce the original variables of most dimensions to a few independent variables as the input to the neural network.Finally,the particle swarm optimization(PSO)algorithm is used to optimize the RBF neural network to predict the photovoltaic power generation,improve the prediction accuracy of photovoltaic power generation,and reduce the impact of the volatility and uncertainty on the safe and stable operation of the grid.(2)In view of the uncertainty of photovoltaic generation and load in microgrid,robust optimization is an effective way to suppress the uncertainty of photovoltaic generation and load,but it also brings strong conservativeness.This paper firstly mathematically models the distributed energy in the microgrid and builds a microgrid scheduling model.Then a multi-energy storage scheduling method based on weak robust optimization is proposed.Weak robust optimization can adjust the penalty coefficient against the constraint condition according to the actual optimization scheduling problem,and weigh the safety and economy of the scheduling scheme.In the weak robust optimization model,the uncertainty of photovoltaic power generation and load is fully considered.The slack variable that symbolizes the load deficiency is introduced into the load balance formula.This method can effectively improve the conservation of traditional robust models.Using the equivalence transformation method,the uncertain constraints in the model are transformed into the determined constraints.Finally,the PSO algorithm is used to solve the optimal scheduling scheme of the weak robust optimization model.(3)The mathematical model of distributed energy in microgrid is built under the condition of known photovoltaic,wind power and load in microgrid operation,and interruptible load is introduced as optimization variable to fully excavate the regulation energy of power supply side and demand side.An iterative adaptive dynamic programming(ADP)algorithm is proposed.The real-time capacity of energy storage at each time is taken as the state variable of the system,the charge and discharge of energy storage,load interruption and power purchase at each time are taken as the control variables,and the operation cost of microgrid is taken as the utility function.According to Bellman principle,the Bellman difference equation is established.According to the derivation of iterative adaptive dynamic programming,the optimal control is obtained.