Artificial Electric Field Algorithm Applied to the Economic Load Dispatch Problem With Valve Point Loading Effect: AEFA Applied to ELD With VPLE

1. Introduction

In recent scenarios, the electrical energy market has become liberal and highly competitive because of increasing load demand. Economic load dispatch (ELD)s beneficial in the operation and planning of power system management (Soni et al., 2020). ELD is used to maintain the economy of the power system by reducing production costs and increasing reliability by maximizing the capability of the thermal unit (Soni & Pandya, 2018). The main aim of ELD is to predict variables for sharing all load to make the system economical by considering equal and unequal constraints. In practical ELD problems, other constraints should consider, like valve point effect, ramp rate, and prohibited operating zones(POZ). This ELD problem is initially solved by some classical methods like quadratic programming (Shah et al., 2019), Dynamic Programming, Linear Programming, gradient method, Lagrangian relaxation, and Hopfield framework (Dieu & Schegner, 2012). The main issue with this method is that they are susceptible to starting points and mostly converge and diverge at a local optimum solution. The solution to the ELD problem by DP technique makes large dimensions that require more computational efforts. These methods are not feasible due to nonlinear characteristics like ramp rate limits, discontinue POZ, and non-smooth cost function. Therefore, classical calculus-based methods are not used. To overcome these drawbacks, robust and reliable techniques are developed. Hence, some new optimization techniques like artificial intelligence (AI) were found to overcome the disadvantages of classical techniques. Hopfield neural network (HNN) is an example of an AI-based algorithm used to solve non-convex and non-differentiable ELD optimization problems. However, they require a large number of iterations to reach global optima. Hence, it takes more time to reach the solution (Bhattacharjee & Patel, 2020).