Optimal Operation of Microgrids Based on a Radial Basis Function Metamodel

Abstract

The optimal operation of a microgrid (MG) is a nonlinear multiconstraint problem. In addition to optimizing the output of different distributed generations (DGs) at the same time, the output of DGs at different times must be coordinated to achieve a balance between power supply and load demand. To solve the optimal operation problem of an MG, an optimal operation model to minimize user electricity cost and MG operation cost is proposed in this article. This model refines the load types and considers the scheduling elasticity of multiple types of home loads of end-users. It also quantifies the scheduling elasticity as the demand response based on the time-of-use price. Furthermore, to solve the problems of a complex solving process and long computation times for the MG optimal operation model, a global optimization algorithm based on a radial basis function (RBF) metamodel is proposed. The novelty of the proposed algorithm is that it adopts Latin hypercube sampling and an RBF to construct a metamodel. The constructed RBF metamodel is used to replace the proposed MG optimal operation model to estimate the function values, thereby avoiding calling the original complex objective function repeatedly. The proposed model and algorithm are tested on the modified IEEE 37-bus and 300-bus systems. The test results show that, the proposed model can effectively reduce the electricity cost of the end-users and enable the MG to obtain more electricity selling revenue. Meanwhile, the proposed algorithm can effectively reduce computational burden and time and has higher search efficiency of the global optimal solution and computational accuracy.