VOLTAGE CONTROL USING A COMBINED INTEGER LINEAR-PROGRAMMING AND RULE-BASED APPROACH

Voltage control of Taiwan Power Company is investigated using a nile-based expert system. The developed expert system, which is implemented on a personal computer, is written using the artificial intelligence language PROLOG. To benefit from human experts' past experience, the heuristic rules currently followed by the operators at Taiwan Power Company are compiled and coded in the rule base of the expert system. To determine the required adjustments to the various VAR control devices, an integer linear programming (LP) routine is developed using the branch and bound method. With the developed integer LP routine, integer solutions to the control variables such as switchable inductors and capacitors and transformer tap settings can be achieved by the expert system without rounding off the results. The objective function employed in this study is to minimize the number of required adjustments, making the approach suitable for practical applications. The designed expert system has been applied to correct the abnormal voltages in Taiwan Power Company during emergency conditions. It is found from the study results that the proposed expert system is a valuable aid to system operators for voltage control.