Augmented Lagrange Hopfield Network for economic dispatch

This paper proposes an augmented Lagrange Hopfield Network (ALHN) for the economic dispatch (ED) problem. The ALHN is a combination of continuous Hopfield neural network and augmented Lagrange function. In the ALHN, the energy function of the Hopfield neural network is based on Lagrange function augmented by penalty factor. An augmented Lagrange function consists of quadratic cost function and power balance equation. The ALHN optimally dispatches online generator units with a minimum total generation cost while satisfying power balance equations and network operating constraints. Test results on 13 to 120 generator units with various load demand show that ALHN can obtain lower total cost but faster computing times than Lambda-Iteration method, Genetic Algorithm (GA), and Sequential Quadratic Programming (SQP) methods. That leads to generator fuel cost savings.