Environmental Economic Hydrothermal System Dispatch by Using a Novel Differential Evolution

Abstract

This paper proposes the Novel Differential Evolution (NDE) method for solving the environmental economic hydrothermal system dispatch (EEHTSD) problem with the aim to reduce electricity generation fuel costs and emissions of thermal units. The EEHTSD problem is constrained by limitations on generations, active power balance, and amount of available water. NDE applies two modified techniques. The first one is modified mutation, which is used to balance global and local search. The second one is modified selection, which is used to keep the best solutions. When performing this modified selection, the proposed method completely reduces the impact of crossover by setting it to one. Moreover, the task of tuning this factor can be canceled. Original Differential Evolution (ODE), ODE with the first modification (MMDE), and ODE with the second modification (MSDE), and NDE were tested on two different hydrothermal systems for comparison and evaluation purposes. The performance of NDE was also compared to existing methods. It was indicated that the proposed NDE is a very promising method for solving the EEHTSD problem.

Minimizing Electricity Fuel Cost of Thermal Generating Units by Using Improved Firefly Algorithm

“…In order to further investigate the performance of IFA, comparisons were carried out of results obtained by IFA and other optimization tools, such as EALHN [1], HNN [3], HRCGA [4], RCGA [4], DE [5], HNUM [6], AHNN [7], ELANN [8], IEP [9], AIS [10], HICDEDP [11], Lambda [15], HM [15], TS [18], IGA [20], BBO [21][22], CGA [23] FCGA [23], CSA [24][25], and ORCSA [26]. In addition to the comparison of best cost, another comparison criterion was considered, i.e.…”

“…9is similar to the mutation operation of the differential evolution algorithm (DEA) in which β acts as mutation factor, ranging from 0 to 2. Some previous studies [26] have pointed out disadvantages of DEA, such as low convergence to a global optimum or easily getting trapped in a local optimum. Consequently, the proposed improvement aims to tackle the limitations of FA by using Eqs.…”

“…In addition, it has a small number of control parameters that lie within a predetermined range. However, the task of finding the best values for these control parameters by tuning is time-consuming and needs a large number of evaluations for different results from different sets of control parameters [26]. In fact, DE has two main factors, the crossover factor and the mutation factor, where the first is from zero to 2 while the latter is from zero to 1.…”

artificial bee colony optimization algorithm for adaptive power scheduling in an isolated system

“…In that work optimization related neural network formulation is used to determine the hydrothermal scheduling. A method -differential evolution -is proposed by Le et al [11] to handle the dispatch problems in the hydrothermal power system. In that work, they aim to reduce electricity generation costs.…”