Dry quenching boiler inlet temperature control system based on multivariable PID

Abstract

The invention relates to the field of intelligent coke discharge control, and in particular relates to a dry quenching boiler inlet temperature control system based on multivariable PID (Proportion Integration Differentiation), which is used for solving the problems of adjustment lag, control overshoot and the like in the face of multivariable collaborative change due to the fact that a univariate PID control method cannot comprehensively consider the coupling relationship between variables. Comprising a parameter end control module, a data decoupling calculation module, an adaptive parameter setting module, a control output module and a closed-loop feedback correction module. According to the method, coupling interference among variables is eliminated and temperature control deviation is greatly reduced through a multivariable coordinated regulation and decoupling algorithm, interference such as red coke amount fluctuation and gas component change can be quickly responded and the temperature recovery time is shortened by combining adaptive PID parameter setting, and the temperature recovery efficiency is improved through three-variable coordinated regulation. System oscillation caused by single variable violent fluctuation is avoided, the service life of the dry quenching furnace and boiler equipment is prolonged, and the maintenance cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of intelligent coke discharge control, specifically to a multivariable PID-based inlet temperature control system for dry quenching boilers. Background Technology

[0002] Dry quenching technology, or dry coke quenching for short, is a new type of coke quenching method relative to wet quenching. The dry quenching and waste heat power generation system uses inert gas nitrogen as a carrier, which is introduced into the dry quenching furnace to cool the hot coke. The coke is cooled from 1000°C to below 180°C and then discharged. The heat from the circulating inert gas is recovered and used to generate steam for power generation.

[0003] In this process, the boiler inlet temperature, as a core process parameter, directly determines the steam output, quality, and operational stability of the entire dry quenching system. Excessively high boiler inlet temperature can lead to overheating of the boiler heating surfaces, shortening the lifespan of refractory bricks and metal pipes, and even causing tube rupture accidents. Conversely, excessively low boiler inlet temperature will reduce steam output and quality, decrease power generation, and reduce the economic efficiency of the dry quenching system. Stable and precise boiler inlet temperature control not only ensures efficient operation of the dry quenching unit and improves energy recovery efficiency but also reduces thermal stress loss and extends equipment lifespan.

[0004] Currently, the industry mostly relies on traditional single-variable PID control technology to control the inlet temperature of dry quenching coke boilers. This control mode is based on the feedback regulation principle and only performs closed-loop control on a single dominant variable. Temperature stability is achieved through proportional, integral, and derivative regulation. However, the dry quenching coke system is essentially a complex dynamic system with strong coupling of multiple variables, nonlinearity, and large time lag.

[0005] For example, many factors, such as circulating air volume, coke production, boiler evaporation rate, pre-storage section pressure, initial coke temperature, and inert gas composition, are interconnected and influence each other. For instance, fluctuations in coke production directly change the heat release in the dry quenching furnace and cause changes in gas flow rate, affecting heat exchange efficiency. Adjustments to boiler evaporation rate change the system's thermal balance, which in turn affects the boiler inlet temperature. Fluctuations in pre-storage section pressure can interfere with gas flow patterns, leading to uneven temperature distribution.

[0006] Therefore, single-variable PID control methods, due to their inability to fully consider the coupling relationship between variables, often exhibit problems such as regulation lag and control overshoot when facing multi-variable coordinated changes. This can easily lead to drastic fluctuations, even exceeding the safe operating range, resulting in decreased steam quality and increased thermal fatigue of equipment. Thus, how to overcome the control bottleneck caused by multi-variable coupling and achieve high-precision, adaptive control of the inlet temperature of dry quenching boilers has become a key issue that urgently needs to be addressed in the current dry quenching technology field. Summary of the Invention

[0007] This invention proposes a multivariable PID-based inlet temperature control system for dry quenching boilers. By using multivariable collaborative regulation and decoupling algorithms to eliminate coupling interference between variables, temperature control deviation is significantly reduced. Combined with adaptive PID parameter tuning, it can quickly respond to disturbances such as fluctuations in red coke quantity and changes in gas composition, shortening temperature recovery time. Furthermore, through three-variable collaborative regulation, it avoids system oscillations caused by drastic fluctuations in a single variable, extending the lifespan of the dry quenching furnace and boiler equipment, and reducing maintenance costs.

[0008] The objective of this invention can be achieved through the following technical solution: a multivariable PID-based inlet temperature control system for dry quenching boilers, comprising a parameter control module, a data decoupling calculation module, an adaptive parameter tuning module, a control output module, and a closed-loop feedback correction module; The parameter control module is used to collect multiple dynamic variables of the dry quenching boiler and use a multivariate model to predict the future impact of each adjustment variable on the temperature under the current operating conditions. The data decoupling calculation module can acquire real-time deviations and generate decoupling matrices. Based on the decoupling matrix, it eliminates the coupling interference of variables and then calculates the initial control increment of each regulation variable in combination with the temperature deviation. The adaptive parameter tuning module can update PID parameters based on fuzzy inference and correct the control increment. The control output module is used to send the corrected control increment to multiple devices in the dry quenching boiler for coordinated adjustment; The closed-loop feedback correction module is used to collect parameters such as temperature in real time, evaluate the control effect, and update the model parameters.

[0009] In a preferred embodiment of the present invention, the multiple dynamic variables collected by the parameter control module include: inlet temperature, coke discharge rate, circulating fan speed, and air intake rate. Using a multivariate model, based on the current coke discharge rate, pressure, and air intake rate, the temperature deviation over a certain period of time can be predicted.

[0010] In a preferred embodiment of the present invention, the real-time deviations obtained by the data decoupling calculation module include real-time temperature deviation e(k), coke discharge deviation eM(k), pressure deviation eP(k), and air volume deviation eA(k), where k is the time period; The decoupling matrix generated by the data decoupling calculation module is Kd: Where uM(k), uP(k), and uA(k) are the control increments of each regulating variable.

[0011] In a preferred embodiment of the present invention, when the data decoupling calculation module adjusts one adjustment variable to eliminate real-time deviation, it simultaneously compensates and adjusts other adjustment variables through the decoupling matrix to avoid disturbance to other adjustment variables.

[0012] In a preferred embodiment of the present invention, the PID parameters generated by the adaptive parameter tuning module include: Proportional gain Kp, integral gain Ki, and derivative gain Kd; The adaptive parameter tuning module obtains the temperature deviation e(k) and calculates the deviation change rate ec(k) based on the temperature deviation, where ec(k) = Δe(k). The adaptive parameter tuning module dynamically adjusts the PID parameters based on the temperature deviation and the rate of change of the deviation. Specifically, Kp = fp(e(k), ec(k), ΔP); Ki = fi(e(k), ec(k), ΔP); Kd=fd(e(k),ec(k),ΔP); Where ΔP is the root mean square of the pressure deviation, and fp, fi, and fd are fuzzy rule bases obtained through experience or data training.

[0013] In a preferred embodiment of the present invention, the method by which the adaptive parameter tuning module performs adaptive parameter adjustment to correct the control increment is as follows: uM′=Kp×e(k)+Ki×∑e(k)+Kd×Δe(k), and uP′ and uA′ are calculated similarly.

[0014] In a preferred embodiment of the present invention, after the control output module obtains the corrected control increment, the target position sent includes the coke discharge device, the air inlet valve and the circulating fan, which respectively correspond to the coke discharge volume adjustment uM′, the air inlet volume adjustment uA′ and the gas pressure adjustment uP′.

[0015] In a preferred embodiment of the present invention, the method for real-time effect evaluation of the closed-loop feedback correction module is as follows: The temperature T(k+1) at the next time step (k+1) is collected, and the actual deviation e(k+1) = Tset - T(k+1) is calculated to evaluate the control effect, where Tset is the setpoint of the temperature.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention establishes a multivariate prediction model through a parameter-side control module and combines it with a data decoupling calculation module to decouple and compensate for the coupling interference between multiple variables such as coke discharge rate, circulating gas pressure, and air introduction rate in real time. This fundamentally overcomes the regulation lag and overshoot problems caused by coupling in single-variable PID control. The system can stabilize the control deviation of boiler inlet temperature within a very small range, significantly improving the stability of steam output and quality compared to traditional methods, and extending the service life of boiler and dry quenching furnace refractory materials.

[0017] In this invention, the adaptive parameter tuning module dynamically adjusts the proportional, integral, and derivative gains based on the real-time temperature deviation and its rate of change through fuzzy inference. This enables the PID parameters to adapt to complex operating conditions such as sudden changes in coke production and fluctuations in steam load, significantly improving the system's ability to suppress strong disturbances and its response speed. After encountering interference, the system can quickly restore the temperature to the set value, effectively avoiding equipment over-temperature alarms or sudden drops in steam production caused by drastic temperature fluctuations, and ensuring the continuous, efficient, and safe operation of the entire system.

[0018] This invention also achieves fully automated closed-loop control of the entire process from data acquisition, model prediction, decoupling calculation, parameter self-tuning to actuator coordinated adjustment through a control output module and a closed-loop feedback correction module. It does not require frequent manual intervention, and the core algorithm has the ability to learn and correct itself through historical data. It can adapt to the characteristics of different dry quenching devices and long-term operating condition drift, and realize the intelligent upgrade of the dry quenching process. Attached Figure Description

[0019] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0020] Figure 1 This is a system block diagram of the present invention; Figure 2 This is a system flowchart of the present invention. Detailed Implementation

[0021] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1: Please refer to Figures 1-2 As shown, the inlet temperature control system for a dry quenching coke boiler based on multivariable PID includes a parameter-side control module, a data decoupling calculation module, an adaptive parameter tuning module, a control output module, and a closed-loop feedback correction module. The parameter-side control module collects multiple dynamic variables of the dry quenching boiler through the DCS system. These dynamic variables include: inlet temperature, coke discharge rate, circulating fan speed, and air intake rate. Among these, the coke discharge rate, circulating fan speed, and air intake rate are all parameters that affect the inlet temperature. The DCS system (Distributed Control System) is equipped with communication interfaces for temperature sensors, pressure transmitters, coke discharge weighing sensors, and air flow meters to ensure that real-time data such as temperature T, coke discharge M, pressure P, and air intake A are collected to the controller at a set sampling period. Specifically, the coke discharge rate is adjusted by changing the amount of red coke discharged to regulate the heat carried by the circulating gas. Air intake regulation is achieved by controlling the amount of air injected into the circulating gas, thereby regulating the temperature through the exothermic or endothermic combustion reaction. The speed regulation of the circulating fan indirectly affects the temperature by changing the flow rate of the circulating gas. However, there is a strong coupling relationship between the coke discharge rate, air introduction rate, and inlet temperature in the dry quenching system: First, changes in the amount of coke discharged will directly affect the flow rate and temperature of the circulating gas. When the flow rate increases, the heat absorbed per unit volume of gas decreases, and the inlet temperature decreases. However, at the same time, more high-temperature red coke enters the dry quenching furnace, which will push up the overall temperature in the long run. Secondly, changes in the amount of air introduced will alter the oxygen content in the circulating gas, triggering a localized combustion reaction and thus affecting the temperature. However, excessive air volume will lead to an increase in the coke discharge temperature, which in turn affects the inlet temperature. Furthermore, fluctuations in circulating gas pressure caused by factors such as coke discharge rate and fan speed can alter gas density and flow rate, indirectly affecting heat transfer efficiency.

[0023] Traditional single-variable PID control only adjusts a single variable (such as temperature deviation), which cannot compensate for the coupling interference between multiple variables. This results in poor control performance, lag in temperature regulation, and overshoot of ±15℃ or more. In severe cases, it can trigger boiler over-temperature alarms or fluctuations in steam production, reducing system operating efficiency. Therefore, based on the physical model of the dry quenching system, a multivariable coupled model of inlet temperature and coke discharge rate (M), circulating gas pressure (P), and air introduction rate (A) is established; a multivariable PID controller is designed, which separates the independent influence of each variable through a decoupling compensator, and adjusts the PID parameters according to the real-time operating conditions to achieve coordinated control of the inlet temperature.

[0024] Based on the law of conservation of energy, a dynamic relationship was established between the inlet temperature T(k) and the coke discharge rate M(k-d1), pressure P(k-d2), and air introduction rate A(k-d3) (d1, d2, and d3 are lag times, determined through step response experiments; typical values ​​are d1=2min, d2=1min, and d3=5min): T(k)=αM(k-d1)+βP(k-d2)+γA(k-d3)+δT(k-1)+ε(k) Where α, β, γ, δ are model coefficients, calculated from the design parameters of the dry quenching system, such as gas specific heat capacity cp = 1.05 kJ / (kg·℃), boiler heat exchange area S = 500 m2, etc.), and ε(k) is a random disturbance term; Data-driven correction: Collect historical data from the site, extract the correspondence between temperature deviation e(k)=Tset-T(k) and the deviations of each variable eM(k)=Mset-M(k), eP(k)=Pset-P(k), eA(k)=Aset-A(k), and correct the coefficients α, β, and γ of the mechanism model through the least squares method or neural network (NARX model) to ensure that the model prediction error is ≤±2℃.

[0025] The parameter-side control module uses a multivariate model to predict the future impact of each adjustment variable on temperature under the current operating conditions. Using the multivariate model, based on the current coke discharge rate, pressure, and air intake rate, it predicts the temperature deviation over a certain period of time, such as the temperature dynamics within the next 5 minutes. The data decoupling calculation module can acquire real-time deviations and generate decoupling matrices. Based on the decoupling matrix, the coupling interference of variables is eliminated, and the initial control increment of each regulation variable is calculated in combination with the temperature deviation. For example, when adjusting the coke discharge rate, its effect on pressure is offset by a compensator; when adjusting the air volume, its hysteresis effect on temperature is offset. The decoupling matrix can be obtained through the inverse matrix of the system model or by identification. The data decoupling calculation module sets the inlet temperature setpoint Tset and initializes the coke discharge setpoint Mset, pressure setpoint Pset, and air introduction setpoint Aset. The real-time deviations obtained by the data decoupling calculation module include real-time temperature deviation e(k), coke discharge deviation eM(k), pressure deviation eP(k), and air volume deviation eA(k), where k is the time period; The decoupling matrix generated by the data decoupling calculation module is Kd: Where uM(k), uP(k), and uA(k) are the control increments of each regulating variable; When adjusting one control variable to eliminate real-time deviation, the data decoupling calculation module uses a decoupling matrix to synchronously compensate and adjust other control variables, thus avoiding disturbance to other control variables. The adaptive parameter tuning module can update PID parameters based on fuzzy inference and correct control increments; The PID parameters generated by the adaptive parameter tuning module include: proportional gain Kp, integral gain Ki, and derivative gain Kd; The adaptive parameter tuning module obtains the temperature deviation e(k) and calculates the deviation change rate ec(k) based on the temperature deviation, where ec(k) = Δe(k). The adaptive parameter tuning module dynamically adjusts the PID parameters based on the temperature deviation and the rate of change of the deviation. Specifically, Kp = fp(e(k), ec(k), ΔP); Ki = fi(e(k), ec(k), ΔP); Kd=fd(e(k),ec(k),ΔP); Where ΔP is the root mean square of the pressure deviation, and fp, fi, and fd are fuzzy rule bases obtained through experience or data training. The adaptive parameter tuning module corrects the control increment by adaptively adjusting the parameters as follows: uM′=Kp×e(k)+Ki×∑e(k)+Kd×Δe(k), and uP′ and uA′ are calculated similarly.

[0026] The control output module is used to send the corrected control increment to multiple devices in the dry quenching boiler for coordinated adjustment; After the control output module obtains the corrected control increment, the target positions sent include the coke discharge device, the air inlet valve, and the circulating fan, which correspond to the coke discharge volume adjustment uM′, the air inlet volume adjustment uA′, and the gas pressure adjustment uP′, respectively. Specifically, the coke discharge rate adjustment converts the control increment into a frequency command for the variable frequency motor of the coke discharge device, and adjusts the motor speed through the frequency converter; Air intake regulation: Converts the control increment into an opening command for the electric regulating valve to control the valve opening; Pressure-assisted regulation adjusts the speed of the circulating fan via a frequency converter based on the control increment; The closed-loop feedback correction module is used to collect parameters such as temperature in real time, evaluate the control effect, and update the model parameters. The method for real-time effect evaluation of the closed-loop feedback correction module is as follows: The temperature T(k+1) at the next time step (k+1) is collected, and the actual deviation e(k+1) = Tset - T(k+1) is calculated to evaluate the control effect, where Tset is the setpoint of the temperature.

[0027] In summary, by using multi-variable collaborative regulation and decoupling algorithms to eliminate coupling interference between variables, the temperature control deviation is significantly reduced. Combined with adaptive PID parameter tuning, it can quickly respond to disturbances such as fluctuations in red coke quantity and changes in gas composition, shortening the temperature recovery time. Furthermore, through three-variable collaborative regulation, it avoids system oscillations caused by drastic fluctuations in a single variable, extends the life of the dry quenching furnace and boiler equipment, reduces maintenance costs, and achieves full-condition adaptive control without manual intervention, reducing operational intensity and adapting to dynamic changes in coke oven load.

[0028] Thresholds, preset values, preset ranges, etc. are set for result comparison and analysis to determine whether they are good or bad. The value of these thresholds is determined by a combination of large-scale model analysis of sample data and human experience. They can also be adjusted appropriately based on seasonal or common-sense influences. Furthermore, the settings for weighting ratios, influence factors, etc., are based on the magnitude of each parameter's influence on the results. The specific values ​​are allocated to ultimately reflect the impact on the results. The settings for input and storage are also determined by a combination of large-scale model analysis of sample data and human experience. Appropriate adjustments can also be made based on seasonal or rational influence conditions.

[0029] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A multivariable PID-based inlet temperature control system for dry quenching coke boilers, characterized in that, It includes a parameter control module, a data decoupling calculation module, an adaptive parameter tuning module, a control output module, and a closed-loop feedback correction module; The parameter control module is used to collect multiple dynamic variables of the dry quenching boiler and use a multivariate model to predict the future impact of each adjustment variable on the temperature under the current operating conditions. The data decoupling calculation module can acquire real-time deviations and generate a decoupling matrix. Based on the decoupling matrix, the coupling interference of variables is eliminated, and the initial control increment of each regulation variable is calculated in combination with the temperature deviation. The adaptive parameter tuning module can update PID parameters based on fuzzy inference and correct the control increment. The control output module is used to send the corrected control increment to multiple devices in the dry quenching boiler for coordinated adjustment; The closed-loop feedback correction module is used to collect parameters such as temperature in real time, evaluate the control effect, and update the model parameters.

2. The inlet temperature control system for a dry quenching coke boiler based on multivariable PID according to claim 1, characterized in that, The parameter control module collects multiple dynamic variables, including: inlet temperature, coke discharge rate, circulating fan speed, and air intake rate. Using a multivariate model, based on the current coke discharge rate, pressure, and air intake rate, the temperature deviation over a certain period of time can be predicted.

3. The inlet temperature control system for a dry quenching coke boiler based on multivariable PID according to claim 1, characterized in that, The real-time deviations obtained by the data decoupling calculation module include real-time temperature deviation e(k), coke discharge deviation eM(k), pressure deviation eP(k), and air volume deviation eA(k), where k is the time period; The decoupling matrix generated by the data decoupling calculation module is Kd: Where uM(k), uP(k), and uA(k) are the control increments of each regulating variable.

4. The inlet temperature control system for a dry quenching coke boiler based on multivariable PID according to claim 3, characterized in that, When the data decoupling calculation module adjusts one type of regulation variable to eliminate real-time deviation, it simultaneously compensates and adjusts other regulation variables through the decoupling matrix to avoid disturbance to other regulation variables.

5. The inlet temperature control system for a dry quenching coke boiler based on multivariable PID according to claim 1, characterized in that, The PID parameters generated by the adaptive parameter tuning module include: Proportional gain Kp, integral gain Ki, and derivative gain Kd; The adaptive parameter tuning module obtains the temperature deviation e(k) and calculates the deviation change rate ec(k) based on the temperature deviation, where ec(k) = Δe(k). The adaptive parameter tuning module dynamically adjusts the PID parameters based on the temperature deviation and the rate of change of the deviation. Specifically, Kp = fp(e(k), ec(k), ΔP); Ki=fi(e(k),ec(k),ΔP); Kd=fd(e(k),ec(k),ΔP); Where ΔP is the root mean square of the pressure deviation, and fp, fi, and fd are fuzzy rule bases obtained through experience or data training.

6. The inlet temperature control system for a dry quenching coke boiler based on multivariable PID according to claim 1, characterized in that, The adaptive parameter tuning module performs adaptive parameter adjustment to correct the control increment as follows: uM′=Kp×e(k)+Ki×∑e(k)+Kd×Δe(k), and uP′ and uA′ are calculated similarly.

7. The inlet temperature control system for a dry quenching coke boiler based on multivariable PID according to claim 1, characterized in that, After obtaining the corrected control increment, the control output module sends target locations including the coke discharge device, the air inlet valve, and the circulating fan, which correspond to the coke discharge volume adjustment uM′, the air inlet volume adjustment uA′, and the gas pressure adjustment uP′, respectively.

8. The inlet temperature control system for a dry quenching coke boiler based on multivariable PID according to claim 1, characterized in that, The method for real-time effect evaluation of the closed-loop feedback correction module is as follows: The temperature T(k+1) at the next time step (k+1) is collected, and the actual deviation e(k+1) = Tset - T(k+1) is calculated to evaluate the control effect, where Tset is the setpoint of the temperature.

Images