Mean-Square Error Minimization of Boiling Reactor Noise

An analytical approach is taken to develop a model of an optimum linear control system for a linearized approximation to a boiling water reactor. The optimization criterion used is the minimization of the mean-square error of the random fluctuation in the output variable due to boiling voids. In optimizing the system an acceleration constraint is placed on the control rod drive mechanism. The mean-square error of the output variable is calculated for various constraint levels. Results show that a considerable reduction in the mean-square error of the reactor output can be realized through external control.