A comprehensive review of soft computing algorithms for optimal generation scheduling

Abstract

Summary Strategic planning for optimal operation of modern power systems has always been a significant task to satisfy the techno‐social aspects in the sustainable energy scenario. In the light of restructured power systems and decentralized generation, the economic dispatch (ED) problem has remained the utmost important economic assignment for the network operators. The conventional techniques have failed to address the nonlinear characteristics of generators for solving the practical ED problem. The complications identified in the literature are valve point loading, prohibited operating zones, multiple fuels, ramp rate limits, and spinning reserve. Therefore, a diverse range of traditional and modern optimization techniques have been adopted to tackle those complexities and nonlinearities. With the recent advancements in centralized optimization techniques, several research works were reported in the past decade to accomplish better results. This review paper endeavors to cover the pioneering research on centralized approaches as well as present‐day research trends on decentralized and distributed approaches by focusing on the economic aspects. First, a detailed survey of several deterministic, stochastic, nature‐inspired, and meta‐heuristic based centralized approaches for solving the ED problem is reviewed. A brief analysis based on the performance evaluation of centralized algorithms on six standard test systems is presented. Then, the recent literature on decentralized and distributed optimization approach and the consensus protocols with respect to various network topologies is briefly reviewed. The prime objective of this review is to summarize the centralized, decentralized, and distributed approaches to solve the classic dispatch, dynamic dispatch, economic emission dispatch, and multi‐area ED problems.

Customer‐orientedenergy demand management of grid connected microgrids

“…Nature‐inspired metaheuristic algorithms have gained popularity in many research fields and have a proven track record in solving many complex engineering problems. A brief taxonomy of nature‐inspired algorithms in solving optimal scheduling problems with a detailed algorithm performance analysis can be found in Lolla et al 34 As described earlier, to solve the proposed EMS problem involving the stochastic nature of renewable sources and DRPs with a non‐linear price responsive model, a recently reported novel metaheuristic algorithm named “Black Widow Optimizer” (BWO) is employed in this research work. The BWO is a nature‐inspired algorithm first proposed by Hayyolalam and Kazem 32 .…”

A swarm intelligence approach for energy management of grid‐connected microgrids with flexible load demand response

“…Owing to the fact that deterministic algorithms can easily get trapped in the local optimum, the application of numerous nature-inspired optimization algorithms has gained popularity in past decades. 43 As per the taxonomy listed in the above reference, the swarm intelligencebased optimization algorithms are essential and solve global optimization problems with high efficiency. The recently reported Sparrow Search Algorithm proposed by Xue and Shen 44 falls under the same category and proved to be robust and stable.…”

Optimal energy management of microgrids‐integrated nonconvex distributed generating units with load dynamics

“…Moreover, the stochastic frameworks designed in the past literature confines optimizing DGs' performance with convex cost function only 23 . On the other hand, the throttling valves of the DGs incur losses are supposed to be considered while modeling their cost function, and this phenomenon is termed as “valve‐point effect” in the literature 24 . With the involvement of VPE, the DG costs can be modeled as a recurring rectified sinusoidal function with nonconvex and nonsmooth characteristics.…”