Research On Economic Load Dispatch Methods In Thermal Power Plant

Economic load dispatch is an effective approach to enhance economic benefit. It dispatches the total power demand to all units satisfying the various constraints on the basis of operational character. Traditional optimization algorithm, however, tend to converge to local minima and have great workload to load dispatch of thermal power plant, which is a non-linear programming problem of high dimension and complex constrains.Genetic operator with good performance is designed and mutation-oriented genetic algorithm is put forward. Back-Propagation neural network has powerful non-linear fitting ability and self-learning ability, and its application to economic load dispatch in thermal power plants is discussed in detail. The new methods proposed overcome the disadvantages of traditional Genetic Algorithm such as low convergence speed and poor stability.Operation characteristics equations are fitted and mathematic model for load dispatch is established based on historical operating data. Then, the remarkable effectiveness of the proposed approaches applied to an example is demonstrated.Economic load dispatch optimal system of thermal power plant is developed. The effectiveness and stability is proved through tests. Consequently, the optimal system is able to enhance the economic benefit.