Combustion system of the hottest industrial boiler

Industrial boiler combustion system

different types and brands Abstract: in view of the model mismatch in general predictive function control, a solution to compensate the control quantity with fuzzy reasoning is proposed. The predictive function control based on fuzzy compensation is applied to the chain boiler combustion control system. The system simulation results show that this controller has strong robustness, adaptability and high control accuracy

Keywords: fuzzy reasoning; Predictive function control; Boiler control; Continuous system simulation

0 preface

industrial boiler is a power equipment widely used in industrial production and one of the important equipment for energy conversion. The combustion control of industrial boilers is directly related to energy saving, so it is the key of boiler control system and is of great significance to safe and economic operation

predictive functional control (PFC) method is a new predictive control algorithm developed in the control research based on the predictive control principle [1]. Simply using the prediction model to predict the process output value is only an ideal way, because there are nonlinear, time-varying, interference and other factors in the actual process, the output prediction value obtained based on the prediction model cannot be completely consistent with the reality, and when the model mismatch is serious, it may even fundamentally destroy the stability of the control. In order to better improve the robustness of predictive function control, this paper attempts to use fuzzy reasoning to compensate the error, that is, the control quantity is the sum of predictive function control quantity and fuzzy compensation quantity, and its control structure is shown in Figure 1. The algorithm is applied to the design of boiler combustion control system, and the simulation shows that the algorithm is effective. 1 industrial boiler combustion system and its control

the combustion process of industrial boiler is a nonlinear, time-varying multivariable process with strong interference, which can be divided into three circuits: steam pressure, flue gas oxygen content and furnace negative pressure

the steam pressure is usually required to be stable around a set value, so the steam pressure is selected as the controlled quantity, and the control quantity is the slip motor speed that controls the forging bar speed. In order to overcome the influence of load change on the steam pressure, the average steam flow is introduced as the feedforward signal. Due to the influence of coal quality in the combustion process, the time delay is large and sometimes variable. Therefore, the control interval and other items are appropriately selected

air supply and induced draft are very important for boiler combustion. Only when the air supply, induced draft and coal feeding are well coordinated, can the best combustion be achieved and the thermal efficiency of the boiler be improved. The oxygen content of flue gas is used as the controlled quantity in the air supply circuit, and the opening of the air supply baffle is used as the control quantity. In fact, as early as 2010, the Fire Department of the Ministry of public security has expressly prohibited such materials. The negative pressure of the furnace is used as the controlled quantity, and the opening of the induced draft baffle is used as the controlled quantity. In order to overcome the influence of the change of coal feeding amount on the oxygen content of flue gas, the equivalent amount of ball disk friction pair of steam pressure control is introduced as the feedforward signal of air supply channel. In order to ensure the safe combustion of coal, the oxygen content and negative pressure of flue gas are required to be kept near an appropriate value. Generally, the oxygen content is required to be low, and the negative pressure is micro negative pressure

the chain furnace combustion system is a complex 3-input and 3-output object. The output includes: boiler outlet steam pressure PM, flue gas oxygen content O2%, furnace negative pressure st; The input quantities include: coal feeding quantity m, air supply volume V, and induced air volume s, and there is a correlation between these quantities. After in-depth study of the actual object, it is not difficult to find that the furnace negative pressure is mainly affected by air supply and induced draft, while other quantities have little influence on it. Therefore, the furnace negative pressure can be treated as a single loop control system with air supply feedforward. In this way, the chain furnace combustion system is simplified to 2 × 2 object, as shown in Figure 2. 2 predictive function control

predictive function control (PFC) has three characteristics of general predictive control: predictive model, rolling optimization and feedback correction. The biggest difference between PFC and other predictive control algorithms is that it pays attention to the structural form of control quantity, and believes that the control quantity is related to a group of functions corresponding to process characteristics and tracking set values. Therefore, the control quantity calculated at each time is equal to the linear combination of a set of pre selected functions, which are called basis functions. Using the known process responses of these basis functions, the weight coefficients of each basis function are obtained by optimizing the objective function, and the corresponding control quantities are obtained

for the convenience of algorithm implementation, the prediction process model of PFC adopts the parameter model in the form of discrete state equation

the control quantity is regarded as a linear combination of basis functions, and the selection of basis functions is related to the characteristics of the process and the tracking set value. Where u (K + I) is the control quantity at K + I time

fn (I) is the value of the basis function in the ith sampling period

n is the number of basis functions

p is the prediction optimization time domain length

μ N is the linear combination coefficient, which needs to be optimized

in order to prevent violent changes and overshoot of control quantity, an exponential curve in the prediction time domain should be introduced as the reference track

the optimization objective function of predictive function control is to minimize the sum of squares of the difference between the output of the prediction process and the reference trajectory value on the fitting point of the selected prediction time domain. 3 fuzzy compensation

let e (k) and EC (k) be the deviation and deviation change rate of the process output at k sampling time, and there is a d-step pure time delay in the process, then the fuzzy compensation control quantity is determined by E (K + D) and EC (K + D). E (K + D) and EC (K + D) can be calculated from the prediction model and reference trajectory

from imitation to self renovation, there are always two input fuzzy variables and output control compensation Δ The membership functions of u are all triangular in shape. Using uneven distribution helps to improve the control accuracy of the system. There are seven fuzzy subsets: Pb, PM, PS, Ze, NS, nm, Nb. By solving the ambiguity with the barycenter method, the Δ u. The process input control quantity is the sum of the calculated fuzzy compensation control quantity and the predictive function control quantity. 4 realization of predictive function controller based on fuzzy compensation