Role of adaptive heuristic criticism in cascade temperature control of an industrial tubular furnace

The purpose of this research is to improve and apply the multivariable control structure of an industrial furnace on the basis of the adaptive heuristic criticism (AHC). This algorithm is a three-layer feed-forward artificial neural network (ANN) that uses supervised learning with reinforcement in a unique topology. It shows how a system consisting of two neurone-like adaptive elements can solve a difficult learning control problem, i.e. the learning system consists of a single associative search element (ASE) and a single adaptive critic element (ACE). The task is to balance a pole that hinges on the manipulated variable by applying disturbance forces to the furnace. This approach to solve control problems of furnaces using AHC is discussed and compared with the results from the fuzzy temperature control of the system in this work.