The Resistance Furnace Decoupling Control System Based On Neural Network

With the rapid development of modern industry, Producing of all kinds of precise products, We have higher requirements to all kinds of metal materials, and heat treatment technology would develop to higher quality, higher efficiency, energy-saving, environment-friendly direction. Heat resistance furnace is one of the most widely used in the production of heating equipment, It is a large inertia of the system with pure time delay, and many factors will affect stability of the system,such as switching door, heating, temperature environmental, volatility of grid voltage and so on. So it would become the key technology that how control of heating temperature well. First, the high precision of temperature control; Second, when the production environment changes and affects the precision of temperature control, Appropriate methods must be implied to satisfy the requirements of precision system, and in order to facilitate the process of research, we need to preserve temperature data; Third, in actual production, control equipments be required to operation convenience,easy maintenance, low cost;Finally, for the modern large-scale industrial resistance furnace, they often are multiple input and multiple output of multivariable systems, each temperature between general always exist large coupling, We would try to use the control algorithm to eliminate the coupling between variables to achieve better control.This paper first introduces current situation,development trend and the existing problems about the resistance furnace temperature decoupling control system. Secondly, according to the resistance furnace temperature control system with time delay, large inertia and strong coupling characteristics, we get many variables of PID neural network (MPIDNN) decoupling control algorithm to resolve successfully the problem of the resistance furnace temperature area with the coupling, so that the system has good decoupling control performance. According to the heat treatment process requirement, we have completed second development on kingview6.53 using kingview6.53 well software and hardware interface and rich screen design platform to the system have many functions such as monitoring, alarming, real-time curves showing, data reports etc.The resistance furnace of actual operation shows that the system has higher control precision, stability and reliability of work, it can satisfy completely the requirement of resistance furnace heating treatment.