Fuzzy modeling of power flow solution

Over the past few decades many power system (viz state estimation, stability analysis contingency and security analysis, VAR management and optimization etc) have well recognized the usefulness of power flow analytical tools. The basic requirement of a power flow program is to organize a set of input data, which describe the system parameters, line parameters, bus injection and initial state variables. Conventionally these inputs are assigned values that are crisp in nature. However, in many systems this may prove to be unrealistic as these parameter show some imprecision and uncertainty. In such cases, the information need be qualitatively soft and natural. The proposed fuzzy load flow (FLF) method requires the repeated solution of load flow equations set via fuzzy logic control .The computational iterations will not terminate until the maximum real and reactive power mismatches are within an acceptable limit. The power mismatches per voltage magnitude at each node of the network are the crisp input values of the designed fuzzy load flow controller (FLFC) . Crisp output values are the correctives of the state variables. The components of the fuzzy logic controller (FLC), the number of fuzzy-membership functions and their shapes are selected from computational experience to minimize the computing time and the number of iterations required for convergence of the solution. Only 2 m calculations per iteration are required in the proposed FLF method, where m is the number of nodes of the system, in contrast to the large number of calculations executed at every iteration of the FDLF method. Hence, the benefit of the FLF method is the short computing time. Test results on various size networks also indicate that the proposed FLF is more efficient than the existing fast decoupled load flow (FDLF) and full Newton methods.