Generation allocation problem using a Hopfield-bisection approach including transmission losses

The objective of the paper is to solve generation allocation problem by minimizing total production cost, including transmission losses using a Hopfield neural network (HNN) algorithm. The generation allocation problem is commonly known as economic dispatch (ED). The computation procedure of the proposed HNN method is direct and do not need training and has been developed and mapped to solve the generation allocation problem of thermal generators. The procedure employs a linear input–output model for the neurons instead of the sigmoidal function. Formulations for solving the ED problem are explored. Through the application of these formulations, direct computation instead of iterations for solving the problem becomes possible. Not like the usual Hopfield methods, which select the weighting factors of the energy function by trials, the proposed method determines the corresponding factors by calculations. To include transmission losses in ED solution, we propose a dichotomy solution combined to the HNN. The effectiveness of the developed method is identified through its application to the 15-unit system. Computational results manifest that the method has a lot of excellent performances.

Research highlights
► We solve generation allocation problem with losses using a Hopfield neural network.
► The procedure is direct and do not need training.
► The procedure employs a linear input–output model for the neurons.
► To include losses in ED solution, we propose a dichotomy solution combined to the HNN.