Augmented Lagrangian Hopfield Network for Combined Heat and Power Economic Dispatch

This paper proposes an augmented Lagrangian Hopfield network (ALHN) for combined heat and power economic dispatch (CHPED) problem. The ALHN is the continuous Hopfield neural network based on augmented Lagrangian relaxation as its energy function. In the proposed ALHN, its energy function is augmented by Hopfield terms from Hopfield neural network and penalty factors from augmented Lagrangian relaxation which help to damp out oscillation of the neural network during transient process. Consequently, the proposed ALHN obtains fast convergence to optimal solution for the complicated CHPED problem. The proposed ALHN has been tested on various systems and results are compared to Lagrangian relaxation (LR), genetic algorithm (GA), improved ant colony search algorithm (IACSA), evolutionary programming (EP), and improved genetic algorithm with multiplier updating (IGA-MU). The results show that the proposed neural network has more advantages than other methods in terms of total costs and computational times, especially for large-scale systems of the CHPED problem. Keywords— Augmented Lagrangian Hopfield network, combined heat and power, economic dispatch.