An efficient multi-objective fuzzy logic based successive LP method for optimal reactive power planning

This paper presents a new and efficient method to solve the Optimal Reactiv e Power Planning (ORPP) problem, simultaneously minimising transmission los ses and improving voltage profile! using Multi-objective Fuzzy logic based LP (MFLP) model in Successive Linear Programming (SLP) framework. The ORPP problem is concerned with optimally siting and sizing new capacitors at pro spective sites in a planned grid with projected load demands such that, min imum number and quantum of new capacitors are allocated, the transmission l osses are minimised and a satisfactory voltage profile is obtained. In this paper, all the prospective capacitor locations are ranked using Voltage Pe rformance Index (VPI) based on sensitivity factors. From the top of the ran ked list, a minimum number of locations are chosen for siting new capacitor s. Reactive powers at these selected capacitor sites along with generator v oltage magnitudes, reactive powers of existing switchable VAR sources and o n-load tap changer (OLTC) settings of transformers are used as control vari ables in this method. In the MFLP framework, each of the objectives and eac h of the constraints are assigned a satisfaction parameter. The satisfactio n parameter corresponding to any objective quantifies the degree of closene ss of the current state of the objective to the optimum. The satisfaction p arameter corresponding to a constraint describes the degree of enforcement of that constraint. By maximising the minimum of these satisfaction paramet ers, the objectives are optimised and the constraint enforcements arc maxim ised. The MFLP based SLP method is found to be very effective while plannin g for systems which have severe under-voltage violations due to insufficien t reactive powers, as the method is capable of selectively relaxing certain load voltage constraints, while simultaneously enforcing all other constra ints strictly. The MFLP technique optimises all the objectives, i.e. minimi ses number and quantum of new capacitors, minimises transmission losses and maximises constraint enforcement of all violated constraints, while simult aneously imposing all other problem constraints strictly. This method uses compactly stored, factorised constant matrices in all MFLP steps, both for construction of the MFLP model as well as for the power flow solution. The method was tested on IEEE test systems and on a practical 191 bus electric utility system. The merits of the method, compared with SLP method using no n-fuzzy approach, are brought out. (C) 2001 Elsevier Science B.V. All right s reserved.