Abstract :
Recently, power-supply failures have caused major social losses. Therefore, power-supply systems need to be discussed from various points of view. The objective of this study is to present a concept of serviceability in investment to a power system. In this study, the serviceability is interpreted from the reliability and risks of units, which are evaluated with a variance-covariance matrix, and the effects and expenses of replacement are analyzed. The mean-variance analysis is formulated as a mathematical program with the following two objectives: (1) to maximize the serviceability, that is, minimize the risk and (2) to maximize the expected return. Finally, a structural learning model of a mutual connection neural network is proposed to solve these problems defined by mixed-integer quadratic programming, and employed in the mean-variance analysis after proving its convergence. Our method is applied to a power system network in a cetain urban area. This method enables us to select results more effectively and enhance decision making. In other words, decision-makers can select the investment rate and serviceability of each ward within a given total budget.
Keywords :
covariance matrices; electricity supply industry; integer programming; investment; learning (artificial intelligence); neural nets; power engineering computing; quadratic programming; decision making; decision-makers; investment rate; mathematical program; mean-variance analysis; mixed-integer quadratic programming; mutual connection neural network; power system network; power-supply failures; power-supply systems; reliability; serviceability based investment; social losses; structural learning model; variance-covariance matrix; Biological neural networks; Equations; Hopfield neural networks; Investments; Mathematical model; Neurons; Portfolios;
