Deaerator Water Level Control Based on Neuron Intelligent Control by Fieldbus Intelligent Control Network

Difficulties in water level control of deaerator lie in the strong couples between water level of deaerator and pressure in outlet of condenser pump, which causes unavailability of automatic control with traditional PID controllers. General problems in traditional decoupling control and PID control schemes are analyzed. Control scheme of combining neuron intelligent regulator with static decoupling, in which adaptive and self-organizing functions are realized by tuning weight factors in neuron adaptive control with gradient descent algorithm and neuron intelligent regulators act in control while decoupling network acts in decoupling, is proposed. Algorithms of online optimization for Kp based on linear reinforcement and self-learning rate based on gradual reduction are proposed and analyzed. Features, characteristics, advantages and simulation results of the proposed scheme are analyzed and presented. Control implementation based on fieldbus intelligent control network is presented. Operational practice shows that satisfactory control effects are achieved with the proposed control scheme.