A novel filled function based closest static load margin computation method

Due to the variety and complexity of constraints in power systems, the optimization problem of system operation can be nonconvex and multimodal. The traditional optimization method is very likely to be trapped in local optimum. The filled function method is an effective global optimization method for multimodal problems. In this paper, this method is used for the computation of the closest load margin for power system operations. The distance in the load power parameter space to the locally closest loading boundary is an index of system security and can provide useful information for arranging load sites and making full utilization of the generation sources. The traditional load power margin is computed by assuming a direction of load increase in load parameter space. In this paper, a closest load power margin computation model considering various load increase directions is proposed. While the load increase power factor is provided, the closest load margin from the loading boundary to the given operating load power can be computed. The test results show that the model and filled function method is effective. The introduction of filled function can help the optimization search process to jump out of the local optimum and more likely obtain the global optimal solution.