Optimizing Charging Strategies for Electric Vehicles in Residential Areas Based on Halton Sequence Randomized Sled Dog Optimizer

Faced with the increasing demand for electric vehicle (EV) charging, the pressure on the power grid to control the load has also risen. Currently, many scholars are exploring optimization strategies for EV charging to address issues such as delayed charging time and insufficient charging efficiency. However, the definition of the charging objective function in related research is relatively simple, which transforms the EV charging optimization problem into a smooth optimization problem; However, the objective functions corresponding to actual charging demand scenarios are often complex, and solving these objective functions to find the best charging strategy often fails to achieve the expected results. Specifically, when the optimization objective function is non-smooth, traditional gradient-dependent methods will fail. Therefore, in this paper, a swarm intelligence based as well as a derivative-free optimization method called Halton sequence randomized sled dog optimizer (HSSDO) has been proposed. The experimental results have demonstrated that this method can effectively handle complex situations where the objective function is non-smooth.