A Novel Selective Particle Swarm Optimization Approach for Combined Heat and Power Economic Dispatch

Abstract

This article presents a new, improved particle swarm optimization algorithm with a selection operator for the solution of the combined heat and power economic dispatch problem. In this technique, starting with a large swarm of particles, only those particles whose fitness is above the scaled average fitness are selected in successive iterations, using a selection factor that is adjustable depending on the nature of the problem. The method is illustrated using a test case. The result compares favorably with other particle swarm optimization variants and other existing non-conventional methods.

Improved Particle Swarm Optimization for Combined Heat and Power Economic Dispatch

“…Tables 7-9. For load demand 1 in Table 7, the proposed IPSO algorithms can obtain better total cost than LR [4], GA [5], IACSA [6], MADS-LHS and MADS-PSO [14], EP [7], and IGA-MU [8] and the same total cost with other methods except HSA [12,13] and SPSO [18]. Note that the solution obtained by HSA in [12][13] is infeasible since 9,257.10 ----GA [5] 9,267.20 ----IACSA [6] 9,452.20 ----EP [7] 9,257.10 ----IGA-MU [8] 9,257.08 ----HSA [12] 8606.07 ----HSA [13] 8606.07 ----MADS-LHS [14] 9277 The solution violated the feasible operating zone of cogeneration unit 3.…”

“…(18). During the process, each new velocity and position cannot always satisfy their limits and the following de nitions are useful to x them: The power output of the slack pure power unit and heat output of the slack pure heat unit are then obtained as in Section 3.2.…”

“…Global vision of PSO with inertia weight (GWPSO) By inserting an inertia weight factor into the original PSO, an improved version of PSO is developed by Shi and Eberhart [30]. The suggestion aims to improve the new velocity of each particle meanwhile the way to obtain the new position is retained as in (18). In the paper, the improvement is named global vision of PSO with inertia weight (GWPSO) and the velocity for the GWPSO is as follows [30]: [32] indicates the usefulness of constriction factor for improving the convergence characteristic of PSO; Eberhart and Shi have applied this parameter to PSO to improve the velocity of particles as follows [31]: …”

“…The best way to minimize the operating cost of the thermal units is to use both heat and electricity under working condition. Consequently, the optimal operation of combined heat and power units has played a very important role in enhancement of the e ciency of cial Immune System (AIS) [16], Lagrangian Relaxation with Surrogate Subgradient Multiplier Updates (LR-SSMU) [17], Selective Particle Swarm Optimization (SPSO) [18], Particle Swarm Optimization with Time Varying Acceleration Coe cients (TVAC-PSO) [19], Improved Group Search Optimization (IGSO) [20], and Oppositional Teaching Learning Based Method (OTLBO) [2]. Generally, the conventional methods such as Newton Method [3] and LR [4] su er diculty when dealing with the problems with nonconvex functions.…”

Differential evolution with Gaussian mutation for combined heat and power economic dispatch

“…Improved ant colony search algorithm (Song et al 1999), evolutionary programming (Wong and Algie 2002), genetic algorithm (Su and Chiang 2004), harmonic search algorithm (Vasebi et al 2007;Karami et al 2013), multi-objective particle swarm optimization (Wang and Singh 2008), self adaptive real-coded genetic algorithm (Subbaraj et al 2009), novel selective particle swarm optimization (Ramesh et al 2009), mesh adaptive direct search algorithm (Sadat Hosseini et al 2011), particle swarm optimization with time varying acceleration coefficients (Mohammadi-Ivatloo et al 2013), and oppositional teaching-learning-based optimization (Roy et al 2014) have been applied to solve CHPED problem.…”