The Resistance Furnace Decoupling Control System Based On Nerve Cell Network

In industrial production, resistance furnace is one of the most widely used heating equipments. As the production process is becoming more complex, the control variables are often more than one variable, the problem is not a single variable, but many variables, and these variables often associates with each other by a certain form. Therefore, we have to decouple them. For a control system with the characteristic of non-linear, pure time delay, large inertia, and strong coupling, such as resistance furnace, only by eliminating the coupling between each temperature range we can achieve the technological requirements well.First, this paper introduces decoupling control system development status and existing problems. And the traditional decoupling methods are discussed in detail, I use traditional feed-forward decoupling algorithm to do Matlab simulation, the detect of this method is relying on the model of the control object largely. Next, the paper gets P, I, D parameters by using relay feedback method and uses incremental PID decoupling method to decouple, and then the paper proposes nerve cell network decoupling which doesn't depend on the model. Last, the two decoupling methods are compared .In addition, according to the system's requirement of process, this paper designs a set of resistance furnace decoupling control system by using King view platform, this system has data monitoring, alarm detection, real-time curve display and so on.The debugging result shows that it has small overshoot, high strong resistance to interference, fast response and operates simply, it realizes the decoupling between two channels of the resistance furnace. It has a certain theoretical significance and using value.