Quadratic load flow calculation in electric power systems using a Hopfield model

This paper presents a neural net based alternative method for the load flow calculation in electric power systems. Unlike the conventional method, the proposed method makes use of the fact that the load flow equation may be expressed as the exact quadratic form. In other words, the proposed method solves the quadratic programming problem rather than the linearized model. As a result, the computational effort is highly reduced with consideration of the nonlinearity.

[1]  John C. Platt,et al.  Constrained Differential Optimization , 1987, NIPS.

[2]  Y. Akimoto,et al.  The representation of large numbers in neural networks and its application to economical load dispatching of electric power , 1989, International 1989 Joint Conference on Neural Networks.

[3]  M. R. Irving,et al.  Power System State Estimation Using Linear Programming , 1978 .

[4]  Yasuhiro Hayashi,et al.  Neural network application to state estimation computation , 1991, Proceedings of the First International Forum on Applications of Neural Networks to Power Systems.

[5]  S. Iwamoto,et al.  A Fast Load Flow Method Retaining Nonlinearity , 1978, IEEE Transactions on Power Apparatus and Systems.

[6]  M. Shanblatt,et al.  A constrained optimization neural net technique for economic power dispatch , 1990, IEEE International Symposium on Circuits and Systems.