Implementing soft computing techniques to solve economic dispatch problem in power systems
Introduction
The economic dispatch (ED) is a constrained optimization problem and the nature of the problem is to find the most economical schedule of the generating units while satisfying load demand and operational constraints. The problem has been tackled by many researchers in the past. The economic dispatch problem using conventional methods are surveyed by Chowdhury and Rahman, 1990, Talaq et al., 1994. Due to the network connection of power systems and the further innovation in the electricity market, the power systems become large scale non-linear dynamic systems (Bai & Zhao, 2006). In consequence, conventional techniques become very complicated when dealing with such increasingly complex dynamic system to solve economic dispatch problems, and are further limited by their lack of robustness and efficiency in a number of practical applications. Thus developing a reliable, fast and efficient algorithm is still an active area in power systems. In the last decay, the success of artificial intelligence techniques in broad area of optimization problems and promising research direction in literature pave the way to employ artificial intelligence and soft computing techniques to solve long standing power system problems, such as economic dispatch (Hong and Li, 2002, Kulkarni et al., 2000, Roa-Sepulveda and Herrera, 2000, Song and Chou, 1997, Yalcinoz and Altun, 2005, Yalcinoz and Altun, 2000, Yalcinoz and Short, 1997, Zhu and Tomsovic, 2007), environmental/economical dispatch (Chiang, 2007, Jayabarathi, 2003, Song et al., 1997), unit commitment (Dieu and Ongsakul, 2006, Rajan et al., 2002, Rajan and Mohan, 2003).
In this paper, a brief survey covering recent implementation of soft computing techniques in ED problem is presented. Then performance evaluation of a number of soft computing techniques to solve the problem is elaborated. Some point to improve the performance of the soft computing techniques is explained. Genetic algorithm, Hopfield neural network, multi-layered perceptron neural network (MLP NN) and tabu search algorithm are employed to solve the economic dispatch problem for power plants consisting of 6 and 20 units.
The organization of the paper is as follows: Section 2 covers a review of the soft computing techniques employed. In Section 3, the economic dispatch problem is briefly defined. The implementation issues related to the soft computing techniques in ED problem are discussed in Section 4 and the simulation results are given in Section 5. Finally, we end the paper with the conclusions and discussion to point out further direction in the implementation of soft computing techniques in ED problem.
Section snippets
Soft computing techniques
Soft computing is the state-of-the-art approach to artificial intelligence and its role in effect is to model the human mind. In this respect, the soft computing techniques differ from the respective conventional computing techniques in that they are tolerant of imprecision, uncertainty, partial truth, and approximation. The soft computing techniques comprises of fuzzy logic, artificial neural networks, probabilistic reasoning and meta-heuristic techniques such as genetic algorithm, tabu
Economic dispatch problem
The economic dispatch problem aims to supply the required quantity of power at the lowest possible cost (Wood & Wollenberg, 1996). The dispatch problem can be described mathematically as an objective function with two constraints.
The total fuel cost at thermal plants should be minimized aswhere Fi is cost function for unit i and ai,bi and ci are cost coefficients of unit i. Pi is the power output of the ith generator and n is the number of generators
Implementation of soft computing techniques in ED problem
For a successful application of soft computing techniques in ED problem, some issues related to the implementation of the soft computing techniques should be taken into account. As in the case of MLP neural network, the determination of the network structural parameters as well as the presentation of the training data are highly important for a successful application of the technique. It is also true for Hopfield neural network in that; network performance is highly relied on the success of
Simulation results
The results of the economic dispatch problem illustrated in this study are obtained by the techniques given in the literature i.e. an improved Hopfield NN approach (IHN) (Yalcinoz & Short, 1997), a fuzzy logic controlled genetic algorithm (FLCGA) (Song et al., 1997), an advance engineered-conditioning genetic approach (AECGA) (Song & Chou, 1997) and an advance Hopfield NN approach (AHNN) (Yalcinoz & Altun, 2000), genetic algorithm with arithmetic crossover (GAwAC) (Yalcinoz & Altun, 2005). Also
Discussion and conclusion
In this paper, solution to the economic dispatch problem as a constrained optimization problem has been obtained using various soft computing techniques. The implementation of soft computing techniques such as tabu search, genetic algorithm, Hopfield NN and MLP NN for solving economic dispatch problem has been presented. Then, the performance of soft computing techniques for economic dispatch problem has been evaluated. The soft computing techniques were tested on power systems consisting of 6
Acknowledgement
This work was supported by the Research Fund of Nigde University under the project number of FBE 2001/4.
References (68)
- et al.
Treatment of skewed multi-dimensional training data to facilitate the task of engineering neural models
Expert Systems with Applications
(2007) - et al.
Generalized regression neural network in modeling river sediment yield
Advances in Engineering Software
(2006) - et al.
Tabu search algorithm for maintenance scheduling of generating units
Electric Power Systems Research
(2000) - et al.
An alternative approach for neural network evolution with a genetic algorithm: Crossover by combinatorial optimization
Neural Networks
(2006) - et al.
An analytical framework for optimizing neural networks
Neural Networks
(1993) - et al.
A new genetic-based tabu search algorithm for unit commitment problem
Electric Power Systems Research
(1999) - et al.
Evolutionary computation in power systems
Electric Power Systems Research
(1998) - et al.
Solution to the economic dispatch problem using decision trees
Electric Power Systems Research
(2000) - et al.
A dynamic genetic algorithm based on continuous neural networks for a kind of non-convex optimation problems
Applied Mathematics and Computation
(2004) - et al.
Economic dispatch solutions with piecewise quadratic cost functions using improved genetic algorithm
International Journal of Electrical Power and Energy Systems
(2003)