Intelligence PID Control Research Of Boiler Main Steam Temperature In Thermal Power Unit

The automatic control system has become an important part of the large-scale thermal power plant, its reliability and performance play a critical role in safety and economy of the entire process. The super-heated steam temperature is the maximal temperature in the whole water/steam channels during of the thermal process in thermal power plant. If the steam temperature is too high or too low, it will bring on dangerous factors. We must control the steam temperature of the outlet of the super heaters to an expected range. The vapor object is a multi-volume link. It has pure lag and many disturbances. Its time constant is much bigger and its object model is not confirmable. It is the most difficult control system in the thermal process. The traditional control method such as PID controller is implemented widely in the thermal process now. But they work well only when the systemic load is steady and can not work well when the systemic load changes in a wide range. With deep study of the intelligent control theory, it provides a new control method for the automatic control of the thermal process in thermal power plant.Firstly, this paper analyzes many disturbance factors influencing vapor temperature on the basis of the produce technics in thermal power plant. The vapor object has complicated dynamic process and intricate object model. It has many disturbances. Such as vapor flux, burning medium, feed-water temperature, vapor enthalpy into the super heaters. These factors may affect on each other. On the basis of analyzing the three main disturbance factors in vapor temperature and main dynamic characteristics of the super-heated steam temperature control system in thermal power plant, this paper builds the vapor temperature mathematical model for simulation by use of mechanism analysis.Secondly, this paper discusses the fuzzy system and neural network theory, and researches the structure and algorithm of fuzzy neural network control (FNNC) . The design process for fuzzy controllers is based on the use of heuristic information from human experts, In order to design the fuzzy controller, the control engineer must gather information on how the artificial decision maker should act in the closed-loop system. Sometimes this information can come from a human decision maker who performs the control task, while at other times the control engineer can come to understand the object dynamic characteristics and write down a set of rules about how to control the system with outside help. These "rule" basically say, "If the object output and reference input are behaving in a certain manner, then the object input should be some value." A whole set of such "If-Then" rules is loaded into the rule-base, and an inference strategy is chosen, then the system is ready to be tested to see if the closed-loop specifications are met. Artificial neural networks are circuits, computer algorithms, or mathematical representations of the massively connected set of neurons that form biological neural networks. They have been shown to be useful as an alternative computing technology and have proven useful in a variety of pattern recognition, signal processing, estimation, and control problems. Their capabilities to learn from examples have been particularly useful.The neural network is a tunable nonlinearity can be tuned by changing the weights, biases, and parameters of the activation functions. While the fuzzy system is also a tunable nonlinearity whose shape can be changed by tuning, for example, the membership functions. Since both are tunable nonlinearities, we can use neural network as the identifier structures in fuzzy control schemes and use gradient or least squares method to update the parameters.The paper adopts fuzzy neural network control that is a kind of compound intelligence control based on the neural network theories and fuzzy logic, the study ability of neural network is used to optimize logical experience rule and adjust proportional gene, which can realize the valid control of the main steam temperature. And make use of the MATLAB proceeding simulation experiment, fuzzy neural network controller arrives the better simulation curve under the different loads, we can find out fuzzy neural network control system has very good robustness and control quality.At present, the application of intelligence control in main steam temperature system, including fuzzy control, neural network control etc, is placed mostly in the theory and simulation stage, this paper is also just placed in the theory and simulation stage, it still needs to do more work that applying the method to the engineering practice. Designing these PID adjusters usually follows a common thinking, that is, change information of the controlled object mathematical model is included in the error signal, error differential signal and error integral signal, after extracting this information, one will know how to change adjuster parameters and design adjusters with adaptability of object model. However, if changed information of the controlled object mathematical model is not included in above three signals, it is impossible for any method to get adjusters with adaptability of object model from the three signals.Based on this, finally, design method of PID adjuster with directive signal is proposed in this paper. The changes of object model manifest in three aspects including gain, time constant and rank. Gain has the biggest influence on control system performance quality, time constant has bigger influence on it and rank usually has little influence on it. If functions of gain and time constant with a directive signal can be determined in the mathematical model of controlled object, the functions can be used to determine parameters of PID adjuster so that realize the PID adjuster with directive signal. Thus, it is possible to maintain performance indexes of control system approximately invariable when working conditions change. The method can deal with nonlinear control problem by extending linear control theory, and convenient to use. Comparing it with fuzzy neural network control method, the method is easier to apply in the engineering practice.