Sensitivity and functionally oriented models for power system planning, operation, and control

This paper presents results of research into the use of factorial experimental design for constructing sensitivity as well as functionally oriented models for power system planning, operation, and control. The rationality of applying factorial experimental design for this goal is associated with the possibility to "build" secants but not tangents (how it is done in the use of existing techniques) to improve the adequateness of constructed models. Considering that the experiments with mathematical models of systems (but not with real systems) create difficulties in all stages of statistical evaluating the results of experiments (the impossibility to estimate reproducibility dispersions without the results of so-called parallel experiments), questions of their overcoming are discussed. The sensitivity models obtained with the use of experimental design techniques are applicable to planning, operation, and control technologies related to applying knowledge based approaches (in particular, a fuzzy logic based approach) as well as to utilizing optimization approaches. Taking this into account, the results of the paper are illustrated by their applications to solving the following power engineering problems: (a) fuzzy logic based coordinated voltage and reactive power control in regulated and deregulated environments and (b) power system operation, based on the application of procedures of sequential multicriteria optimization to realize dispatch on several objectives. The paper also describes procedures for constructing functionally oriented models (models with given or predefined structures, which can serve, in particular, as flexible equivalents for diverse functionalities in power engineering) on the basis of applying experimental design techniques. Their application is illustrated by the construction of functionally oriented equivalents to evaluate power system reaction in solving problems of monocriteria or multicriteria optimizing network configuration in distribution systems. (c) 2012 Elsevier Ltd. All rights reserved.