CN-116464954-B - Automatic control method for main steam temperature of power generation boiler

Abstract

The invention discloses a main steam temperature control method of a power generation boiler, which comprises the steps of collecting step response data of the main steam temperature of the power generation boiler, carrying out data fitting by utilizing a least square method, establishing a related mathematical model of an inertia zone and a leading zone of a main steam control channel of the power generation boiler, introducing a pre-estimated compensation controller in a parallel feedback link, separating a hysteresis part by a constant separation method as a single item, optimizing PID parameters by utilizing an artificial fish swarm algorithm, and substituting an optimal solution into original parameters. Hysteresis influence in a system model can be effectively eliminated through smith estimated compensation control, system adjustment time is shortened, and then artificial fish swarm algorithm is utilized to optimize controller parameters, so that accuracy of the controller can be greatly improved, time for the system to reach a target value is effectively shortened, overshoot of the system is reduced, and influence of external interference is reduced.

Inventors

• TAN QIULIANG
• WU WEI
• YANG BUYUN
• XIE QIXIANG
• LIU YUN
• FENG XUGANG
• An shuo

Assignees

• 湖南华菱湘钢节能发电有限公司
• 安徽工业大学

Dates

Publication Date

20260505

Application Date

20230423

Claims (2)

1. 1. The automatic control method for the main steam temperature of the power generation boiler is characterized by comprising the following steps of: Step 1, establishing a related mathematical model of an inertia zone and a leading zone of a main steam control channel of a power generation boiler, and obtaining the model of the inertia zone and the leading zone of the main steam control channel of the power generation boiler by acquiring step response data of the main steam of the power generation boiler and performing data fitting by using a least square method: ; ; Wherein, the Indicating the temperature of the primary steam passing through the superheater, Representing the temperature of the steam at the outlet of the desuperheater, K 1 、K 2 is the gain coefficient of transfer functions of the inert zone and the leading zone, T 1 、T 2 is the time constant of transfer functions of the inert zone and the leading zone, and the initial PID parameters Kp0, ki0 and Kd0 are calculated by using a critical proportionality method; step2, introducing a pre-estimated compensation controller in a parallel feedback link, and then separating a hysteresis part by a constant separation method to be used as a single item; And 3, optimizing the PID parameters by using an artificial fish swarm algorithm, and substituting the optimal solution into the original parameters, wherein the method comprises the following specific steps: step 3.1, initializing parameters, and regarding the scale of the artificial fish school Degree of crowding of fish school Visual field range of artificial fish Step length of movement Number of iterations Setting parameters; Assuming that N artificial fishes exist in a certain water area, the state of each artificial fish is used as a vector Representation of wherein , Representing the solution vector of X, for the food concentration corresponding to each artificial fish The representation, wherein, Representing the objective function, the distance between two adjacent artificial fish Representing the density or crowding degree of the shoal of fish in the water area Representation of the visual field of the artificial fish Representing the step length of each movement Representing the number of heuristics or iterations in foraging A representation; step 3.2, calculating the food concentration Y of each artificial fish, namely calculating a fitness value according to an objective function, selecting an optimal state value through comparison of the fitness values, and recording the optimal state value on a bulletin board; Assume that the current position of the artificial fish is Randomly selecting a swimming position in the perceived water area Then the mathematical model of artificial fish swimming is expressed as: ; wherein: -random function, and Simultaneously assume positions The concentration of the food is , The concentration of the food is If the food concentration is Greater than Then the artificial fish will be from the position To the position Moving by one step, step length Meanwhile, the current value of the next search of the artificial fish is expressed as follows: ; Conversely, if the food concentration Less than Then the artificial fish will find new positions again in its field of view Then comparing the food concentrations at the two positions, if foraging, the number of iterations Reaching a maximum value, no higher than the current food concentration is found, then the artificial fish will randomly move one step within its field of view The mathematical expression is as follows: ; step 3.3, executing aggregation and rear-end collision behaviors, wherein the foraging behaviors are taken as default behaviors, and selecting the optimal behaviors to find food according to the fitness value calculated in the step 2; Assume that the current position of the artificial fish is For the number of fish shoals in the visual range Representation, position The food concentration at the location is Then the center position of the fish school The mathematical expression at this point is: ; Assuming the center position of the fish school The food concentration at the location is Then whether the artificial fish moves towards the center of the fish school is judged by the following formula; ; If the above equation is true, which indicates that the food in the center of the fish farm is sufficient and the fish farm is not crowded, the artificial fish will move toward the center of the fish farm to form a gathering behavior, and the mathematical expression is: ; in contrast, the artificial fish will perform step (2), assuming the current position of the artificial fish is The highest value of the food concentration which can be searched in the visual field perception range ) The corresponding positions are Then whether the artificial fish will be positioned The movement is determined by the following equation: ; if the above is true, the position is described The food is sufficient and the fish shoal is not crowded, then the artificial fish will be located The mathematical expression of the movement is as follows: ; Otherwise, the artificial fish will perform step (2).
2. 2. The method for automatically controlling the temperature of main steam of a power generation boiler according to claim 1, wherein in said step 2, a controller is designed , The equivalent transfer function is PID control transfer function: ; Wherein the method comprises the steps of For the Smith compensation function, from the above equation, it is required that ; The transfer function of the compensated system is ; After the Smith estimation compensation, the hysteresis part is separated as a single item and the unstable effect on the system is obviously weakened.

Description

Automatic control method for main steam temperature of power generation boiler Technical Field The invention belongs to the technical field of automatic control of power generation boilers, and particularly relates to an automatic control method for optimizing the main steam temperature of a power generation boiler by Smith-PID based on an artificial fish swarm algorithm. Background The temperature of the main steam of the boiler is one of main parameters of boiler control, and the quality of the control effect of the main steam is directly related to the quality of the steam, so that the economic operation of a unit is affected. The main steam temperature of the boiler is a complex controlled object, and has larger hysteresis, so that the main steam temperature is influenced by combustion change, smoke change, manual intervention temperature reduction water quantity change, external main steam load change and primary/secondary air change. Aiming at the factors, the control of the temperature of the main steam not only requires a relatively high response speed and a relatively high rebound rate to the outside, but also requires a relatively high restraining force to the change of the temperature reduction pressure, so that the system has good stability and ensures high-quality steam. For the conventional DCS software algorithm at present, the control loop lap joint is mainly traditional PID control or cascade PID control, and the feedforward of signals such as load, fuel quantity and the like is introduced. However, for the main steam temperature control with large hysteresis, the dynamic effect of actual operation still has large deviation, long transition time and dynamic swing, and the requirement of high-precision control cannot be met. While the intelligent control methods such as fuzzy control, neural network control, sliding mode control, no-identification pre-estimation control and the like realize better effects in the study of main steam temperature control, the intelligent control methods cannot be realized in DCS due to the complexity of operation and the restriction of a DCS platform, and the intelligent control methods are realized by a computer controlled by an externally hung APC (automatic control) so as to increase great investment cost. The traditional main steam temperature control of the power generation boiler adopts PID control (proportional, integral and differential control), and the proportion (P), integral (I) and differential (D) of deviation form control quantity through linear combination, so the control quantity is called PID controller. The traditional PID control can obtain better control effect for a clear system, but has poor control effect for a system which is difficult to describe accurately by mathematics. The inability to automatically adjust parameters for changes in the controlled object provides better control. However, along with the progress of technology, people meet the fields that the traditional PID control system cannot be well made or even cannot be applied, wherein the hysteresis link caused by mechanical actions of each link and the power generation boiler per se in the power generation boiler process cannot be well adjusted, and the PID control precision is poor and the adjusting time is long. Disclosure of Invention Aiming at the defects that the main steam temperature control of the industrial power generation boiler has a plurality of uncertain factors, so that the system is nonlinear and the main steam temperature control of the power generation boiler has large hysteresis, the invention provides a power generation boiler main steam temperature control method based on the optimization of Smith-PID by using an artificial fish swarm algorithm, and the control rate is improved by using the method of optimizing the estimated compensation PID control of Smith by using the artificial fish swarm algorithm, so as to solve the technical problem of accurate temperature control. The technical scheme of the invention is as follows: an automatic control method for the temperature of main steam of a power generation boiler comprises the following steps: step 1, establishing a related mathematical model of an inert zone and a leading zone of a main steam control channel of a power generation boiler, wherein the specific steps are as follows: Acquiring step response data of main steam of the power generation boiler, and performing data fitting by using a least square method to obtain a model of an inert zone and a leading zone of a main steam control channel of the power generation boiler Wherein, the Indicating the temperature of the primary steam passing through the superheater,Representing the temperature of the steam at the outlet of the desuperheater,The flow of the cooling water is represented, K 1、K2 is the gain coefficient of the transfer function of the inert zone and the leading zone, T 1、T2 is the time constant of the transfer function of the inert zone a