Analysis of Stability Degree for Combustion of Waste Plants with Fuzzy Relational Map of Sensors

Abstract

In waste plants, we need to predict unstable combustion and control it appropriately for keeping combustion state stable. Waste plants equip many sensors but a PID controller use only its local few sensors to analyze the stability degree of combustion. In this paper, we propose a method for computing a stability degree based on fuzzy relational maps between many sensors. Fuzzy relational maps between many sensors are constructed using sensor values in the stable state. Stability degrees for unknown data are derived from the consistency with the fuzzy relational map. We applied this method to data of certain plant to detect several kinds of unstable combustion states. Before at least several minutes of unstable states, we detected them for CO moving average and deviation rate of steam amount. We also propose several extensions for improving detection accuracies of our method.