The electrical power generation from fossil fuel releases several contaminants into the air, and these become excrescent if the generating unit is fed by multiple fuel sources (MFS). The ever more stringent environmental regulations have forced the utilities to produce electricity at the cheapest price and the minimum level of pollutant emissions. The restriction in generator operations increases the complexity in plant operations. The cost effective and environmental responsive operations in MFS environment can be recognized as a multi-objective constrained optimization problem. The ant lion optimizer (ALO) has been chosen as an optimization tool for solving the MFS dispatch problems. The fuzzy decision-making mechanism is integrated in the search process of ALO to fetch the best compromise solution (BCS). The intended algorithm is implemented on the standard test systems considering the prevailing operational constraints such as valve-point loadings, CO2 emission, prohibited operating zones and tie-line flow limits.
TopSolution Methods
The solution approaches addressing this problem can be categorized into mathematical and heuristic methods. The research reports addressed the multi-fuel power dispatch problems are briefed in this section.
Figure 1. Solution procedure to multi-objective multi-fuel generation dispatch
The classical optimization methods, including Hierarchical Method (HM) and artificial neural network models such as Hopfield Neural Network (HNN) and Adaptive HNN (AHNN) models have been reported to address the economic operation of MFS (Shoults & Mead, 1984; Lin & Viviani, 1984; Park et al., 1993; Lee et al., 1998). The main drawback of these methods is the exponentially growing time for large scale systems with non-convex constraints.
The meta-heuristic search techniques such as Genetic Algorithm (GA) (Baskar et al., 2003), Evolutionary Programming (EP) (Jayabarathi et al., 2005), Particle Swarm Optimization (PSO) (Park et al., 2005), Artificial Immune System (AIS) (Panigrahi et al., 2007), Differential Evolution (DE) (Noman & Iba, 2008), Artificial Bee Colony Algorithm (ABC) (Hemamalini & Simon, 2010) and Biogeography Based Optimization (BBO) (Bhattacharya & Chattopadhyay, 2011) have been reported for solving ED with PQCF. The modified versions of heuristic search techniques such as hybrid Real Coded GA (RCGA), fast EP, improved fast EP, Improved GA – Multiplier Updating (IGA-MU), New PSO-Local Random Search (NPSO-LRS), penalty parameter less PSO/DE and New Adaptive PSO (NAPSO) have been reported to solve multi-fuel power dispatch problem (Baskar et al., 2003, Jayabarathi et al., 2005; Park et al., 2005; Chiang, 2005; Selvakumar & Thanushkodi, 2007; Manoharan et al., 2008; Niknam et al., 2011). The improved version of PSO has been reported to solve the ED problem considering the valve-point effects (Polprasert et al., 2013). Further, improved versions of HNN and mathematical methods such as Augmented Lagrange HNN (ALHNN), Enhanced ALHNN (EALHNN), Quadratic Programming – Augmented Lagrange Hopfield Network (QP-ALHN), Hopfield Lagrange Network (HLN), Auction based Algorithm (AA) and Dimensional Steepest Decline (DSD) have also been reported to determine cost effective dispatch schedules (Vo & Ongsakul, 2012; Dieu et al., 2013; Dieu & Schenger, 2013; Thang, 2013; Binetti et al., 2014; Zhan et al., 2015). The Teaching Learning Based Optimization (TLBO) algorithm and Chaotic Global Best ABC (CGBABC) algorithm have been applied for the economic solution considering tie line flows and MFS (Basu, 2014; Secui, 2015).