This paper reports on an evolutionary programming based method for solving the optimal power flow problem. The method incorporates an evolutionary programming based load flow solution. To demonstrate the global optimisation power of the new method it is applied to the IEEE30 bus test system with highly non-linear generator input/output cost curves and the results compared to those obtained using the method of steepest descent. The results demonstrate that the new method shows great promise for solving the optimal power flow problem when it contains highly non-linear devices.
|Title of host publication||Simulated Evolution and Learning|
|Editors||B. Yao McKay, X. Newton, C.S. Kim, J.H. Furuhashi, T.|
|Place of Publication||Berlin Heidelberg|
|Publication status||Published - 1999|
|Name||Lecture notes in Artificial Intelligence|