Determination of power system topological observability using the Boltzmann machine

A method for determining power system topological observability using a stochastic neural network is presented. The method is based on the Boltzmann machine that considers the stochastic characteristics of neurons. The Boltzmann machine is very useful for solving combinatorial problems, since it can avoid a local minimum in evaluating a global minimum of the cost function to be minimized. The problem of power system topological observability is formulated as an integer programming problem. The Boltzmann machine is then applied to the integer programming problem to obtain a global minimum. The method was successfully applied to a sample system