Surrogate-Based Constraint-Handling Technique for Multi-Area Combined Economic/Emission Dispatch Problems Within Bi-Level Programming Framework

Due to long optimization time and limited dispatching cycle, multi-area combined economic/emission dispatch (MACEED) problems are converted into computationally expensive MACEED (CEMACEED) problems. Existing centralized and distributed method for CEMACEED problems include time-consuming and inaccurate disadvantages. The drawbacks, i.e., low convergence rate and unaffordable computing time may make scheduling cycles can hardly be met. In this paper, a novel surrogate-based bi-level programming framework is proposed. CEMACEED problem is decoupled to obtain completely independent sub-area dispatching problems. Then, a novel surrogate-based constraint-handling technique is proposed to handle tie line constraint while realizing decoupling. Original tie line constraint is replaced by improved polynomial regression surrogate models to save more than 75% of operating time. Finally, a novel meta-heuristic algorithm named emotional accumulation game evolutionary algorithm (EAGEA) is proposed to execute CEMACEED problems. The proposed method obtains the minimum cost, emission, and operating time over all simulation cases.