Intelligent control method and system for co-production of gas and carbon based on positive pressure sealing and temperature coordination

Abstract

This invention discloses an intelligent control method and system for gas-coal cogeneration based on positive pressure sealing and temperature coordination, belonging to the field of intelligent control technology for gas-coal cogeneration. The method includes: collecting system operating parameters and calculating operating deviations; using a collaborative fuzzy neural network of a prediction layer-evaluation layer-execution layer to obtain the optimal weight vector and its estimated value, thus obtaining the weight deviation; constructing a Lyapunov stability functional by combining the operating deviation and the weight deviation, and deriving the adaptive adjustment law of the network weights. For the two core parameters of the pyrolysis furnace temperature and system pressure, optimal performance index functions and cost functions are constructed respectively, and the adaptive optimal control law is solved; the cooling gas recirculation, gasifier heat load, and nitrogen replenishment flow rate are dynamically adjusted to achieve duct self-cleaning and feed interlock protection. This invention can adapt to the strongly coupled, time-varying, and nonlinear operating conditions of gas-coal cogeneration systems, ensuring the safety of the pyrolysis furnace process temperature zone and micro-positive pressure sealing, and improving equipment automation and operational reliability.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for gas-carbon cogeneration, and more specifically to an intelligent control method and system for gas-carbon cogeneration based on positive pressure sealing and temperature coordination. Background Technology

[0002] Cogeneration of coal gas and charcoal, through deep coupling of gasification and pyrolysis units, produces clean fuel gas while simultaneously generating high-value-added charcoal materials, representing a core technological pathway for the graded and differentiated utilization of coal. The pyrolysis unit must stably maintain a reaction temperature range of 600–700°C to ensure the pyrolysis reaction rate and product quality. Simultaneously, the furnace must maintain a slightly positive pressure to prevent air infiltration and the formation of explosive mixtures, and to prevent high-temperature fuel gas leakage that could lead to safety and environmental risks. Fluctuations in the gasification side's inlet temperature can propagate along the process chain, causing a chain reaction of failures such as fluctuations in the gas composition, tar condensation and coking, and duct blockage.

[0003] Positive pressure gas-carbon cogeneration systems are characterized by strong coupling of multiple variables, high nonlinearity, large operating condition fluctuation range, and significant time-varying process parameters. Traditional distributed PID control cannot simultaneously meet the multiple requirements of temperature accuracy, pressure stability, and sealing safety. Independent operation of each loop cannot offset the coupling disturbances between variables. On-site manual control has the defects of response lag and adjustment accuracy depending on personnel experience, which can easily lead to operational accidents such as overheating and coking, pressure instability and leakage. Existing control schemes generally lack online adaptive approximation capabilities and cannot compensate for unknown system reaction mechanisms and external disturbances in real time, making it difficult to support long-term unattended stable operation.

[0004] Therefore, how to propose an intelligent control method and system for gas-carbon cogeneration based on positive pressure sealing and temperature coordination, and overcome the shortcomings of existing technologies, is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] In view of this, the present invention provides an intelligent control method and system for gas-coal cogeneration based on positive pressure sealing and temperature coordination, which solves the control bottlenecks of multi-variable coupling, strong nonlinearity, and poor adaptability to operating conditions in gas-coal cogeneration systems. It achieves online approximation of unknown dynamics through a three-layer collaborative fuzzy neural network, and combines Lyapunov stability theory and optimal control methods to achieve precise temperature control, positive pressure sealing safety assurance, and multi-parameter collaborative optimization, comprehensively improving the system's operational stability and automation level. To achieve the above objectives, the present invention adopts the following technical solution: A smart control method for gas-coke cogeneration based on positive pressure sealing and temperature coordination includes: Collect the operating parameters of the gas-coke cogeneration system and obtain the real-time detection values ​​and preset target values ​​of each parameter; The system operating deviation is calculated based on the real-time detection values ​​and the preset target values. A collaborative fuzzy neural network consisting of a prediction layer, an evaluation layer, and an execution layer is used to solve for the optimal weight vector and its online estimate corresponding to the running parameters. The weight deviation is then calculated based on the optimal weight vector and the online estimate. By combining the system operating deviation and the weight deviation, a Lyapunov stability functional for the control loop is constructed. The adaptive adjustment law of fuzzy neural network weights is derived based on the Lyapunov stability functional derivation. The output of each actuator is adjusted in real time according to the adaptive regulation law to achieve coordinated optimization control of the positive pressure sealing performance and temperature field distribution of the gas-carbon cogeneration system.

[0006] Optionally, the gas-coke cogeneration system includes a bubbling bed gasification unit and a pyrolysis reaction unit coupled in series, and the operating parameters include the furnace temperature of the pyrolysis furnace; The weighted adaptive adjustment law corresponding to the furnace temperature of the pyrolysis furnace is divided into three sets of weight update rules: prediction layer, evaluation layer, and execution layer. The weight update of the prediction layer combines the temperature detection value, temperature operation deviation and deviation barrier coefficient for dynamic correction, and is configured with a weight decay mechanism. The weight updates of both the evaluation layer and the execution layer are adjusted through Gaussian membership basis function mapping, and the weight updates of the execution layer are synchronously correlated with the weight deviations of the evaluation layer, thereby achieving coordinated adaptive adjustment of the three-layer network.

[0007] Optionally, the operating parameters include the pyrolysis furnace system pressure; The weighted adaptive adjustment law corresponding to the pressure of the pyrolysis furnace system is divided into three sets of weight update rules: prediction layer, evaluation layer, and execution layer. The weight update of the prediction layer combines the pressure detection value, pressure operation deviation and deviation barrier coefficient for dynamic correction, and is configured with a weight decay mechanism. The weight updates of both the evaluation layer and the execution layer are adjusted through Gaussian membership basis function mapping, and the weight updates of the execution layer are synchronously correlated with the weight deviations of the evaluation layer, thereby achieving coordinated adaptive adjustment of the three-layer network of the pressure loop.

[0008] Optionally, adjusting the output of each actuator in real time according to the adaptive regulation law includes: the process of regulating the furnace temperature of the pyrolysis furnace according to the adaptive regulation law. Based on the temperature operating deviation of the pyrolysis furnace, an optimal performance index function for the temperature control loop is constructed. By combining the optimal temperature control law to be solved with the optimal performance index function, the optimality equation for temperature control is established. The optimality equation is solved analytically by applying the extreme value condition, and the analytical initial value of the optimal temperature control law is obtained. A collaborative fuzzy neural network consisting of a prediction layer, an evaluation layer, and an execution layer is used to perform nonlinear approximation compensation on the analytical initial value, resulting in the final temperature adaptive optimal control law. Based on the temperature adaptive optimal control law, the proportion of cooling gas recirculation and the heat load of gasification furnace gas production are adjusted in real time to achieve closed-loop control of furnace temperature. When the gas inlet temperature of the gasifier is lower than the preset threshold, the reaction intensity of the gasifier is increased to improve the gas production temperature, and the self-cleaning program of the pyrolysis furnace airflow channel is triggered simultaneously.

[0009] Optionally, adjusting the output of each actuator in real time according to the adaptive control law includes: regulating the pressure of the pyrolysis furnace system according to the adaptive control law. The differential pressure signal and the pressure signal inside the furnace at each sealing node of the pyrolysis furnace are collected, and the pressure operation deviation is calculated. By combining the pressure optimal control law to be solved with the pressure operation deviation, a pressure control cost function is constructed. By introducing pyrolysis furnace volume parameters, pressure operation deviation, and internal pressure state variables, the pressure adaptive optimal control law is obtained by solving the problem. According to the pressure adaptive optimal control law, the sealing nitrogen supply flow rate and the feeding interlock device are adjusted in real time to maintain a slightly positive pressure condition inside the furnace. When the differential pressure at any sealing node exceeds the threshold, the nitrogen resealing procedure and the feed deceleration protection logic are automatically triggered.

[0010] Optionally, the pressure adaptive optimal control law consists of three parts: a deviation benchmark adjustment term, an execution layer fuzzy neural network approximation term, and a prediction layer fuzzy neural network approximation term. It dynamically outputs the optimal nitrogen replenishment flow rate based on the pressure deviation, the real-time pressure inside the furnace, and the volume of the pyrolysis furnace.

[0011] Optionally, the Lyapunov stability functional of the control loop includes: the Lyapunov stability functional of the temperature control loop, which consists of four parts: temperature deviation barrier term, prediction layer weight deviation term, execution layer weight deviation term and evaluation layer weight deviation term, used to characterize the energy level of the temperature control loop.

[0012] Optionally, the Lyapunov stability functional of the control loop includes: the Lyapunov stability functional of the pressure control loop, which consists of four parts: pressure deviation barrier term, prediction layer weight deviation term, execution layer weight deviation term and evaluation layer weight deviation term, used to characterize the energy level of the pressure control loop.

[0013] Optionally, the optimal performance index function for temperature control uses the cooling gas recirculation adjustment coefficient as the optimization variable, and calculates the minimum integral cost within the feasible region of temperature deviation, comprehensively balancing the temperature control accuracy and the energy consumption of the cooling system.

[0014] A smart control system for gas-coke cogeneration based on positive pressure sealing and temperature coordination includes: The condition sensing and target setting unit is used to collect the operating parameters of the gas-coke cogeneration system and determine the real-time detection value and preset target value of each parameter. A real-time deviation calculation unit is used to calculate the system operating deviation based on the real-time detection value and the preset target value; The fuzzy weight mapping unit is used to solve for the optimal weight vector and its online estimate value corresponding to the running parameters through a collaborative fuzzy neural network of prediction layer-evaluation layer-execution layer; The weight deviation calculation unit is used to calculate the weight deviation based on the optimal weight vector and the online estimated value. Stability functional building blocks are used to construct Lyapunov stability functionals for control loops by combining system operating deviations and weight deviations. An adaptive law generation unit is used to derive the adaptive adjustment law of the weights of the fuzzy neural network based on the Lyapunov stability functional derivation. The multi-loop coordinated control unit is used to adjust the output of each actuator in real time according to the adaptive regulation law, and to perform coordinated optimization control of positive pressure sealing performance and temperature field distribution.

[0015] As can be seen from the above technical solution, compared with the prior art, the present invention discloses an intelligent control method and system for gas-carbon co-production based on positive pressure sealing and temperature coordination, which has the following beneficial effects: This invention employs a three-layer collaborative fuzzy neural network architecture—prediction layer, evaluation layer, and execution layer—to approximate the unknown nonlinear reaction mechanism, time-varying parameter characteristics, and external disturbances of the gas-coal cogeneration process online in real time. It achieves high-precision closed-loop control without relying on a precise mechanistic model and exhibits strong adaptability to a wide range of operating condition fluctuations. Based on Lyapunov stability theory, a weighted adaptive adjustment law is designed, combined with a tangent barrier function to constrain the deviation range. Mathematically, it is proven that all state signals of the closed-loop system are consistent and ultimately bounded, with temperature and pressure deviations always converging within a preset range, fundamentally preventing faults such as overheating and coking, and pressure instability and leakage. Based on the HJB optimality equation, an optimal performance index functional is constructed. Under the premise of ensuring temperature control accuracy and positive pressure sealing performance, the cooling gas energy consumption, nitrogen consumption, and gasifier load are synergistically optimized to achieve the optimal balance between control quality and operating costs. It achieves multi-variable coupled and coordinated control of pyrolysis furnace temperature, system pressure, and inlet gas temperature, and automatically triggers protective actions such as duct self-cleaning, nitrogen resealing, and feed interlocking. It can cover all operating conditions without manual intervention, effectively improving the system's automation level and long-term operational reliability. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of a smart control method for gas-carbon cogeneration based on positive pressure sealing and temperature coordination, provided by the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] This invention discloses an intelligent control method for gas-carbon co-production based on positive pressure sealing and temperature coordination, such as... Figure 1 As shown, it includes: Collect the operating parameters of the gas-coke cogeneration system and obtain the real-time detection values ​​and preset target values ​​of each parameter; The system operating deviation is calculated based on the real-time detection value and the preset target value; A collaborative fuzzy neural network consisting of a prediction layer, an evaluation layer, and an execution layer is used to solve for the optimal weight vector and its online estimate corresponding to the running parameters; the weight deviation is then calculated based on the optimal weight vector and the online estimate. By combining the system operating deviation and weight deviation, a Lyapunov stability functional for the control loop is constructed. Based on the Lyapunov stability functional derivation, the adaptive adjustment law of the fuzzy neural network weights is derived; The output of each actuator is adjusted in real time according to the adaptive regulation law to converge the system operation deviation and realize the coordinated optimization control of positive pressure sealing performance and temperature field distribution of the gas-carbon cogeneration system.

[0020] Furthermore, the gas-coal cogeneration system includes a bubbling bed gasification unit and a pyrolysis reaction unit coupled in series. The operating parameters include the furnace temperature of the pyrolysis furnace. The weighted adaptive adjustment law corresponding to the furnace temperature of the pyrolysis furnace is divided into three sets of weight update rules: prediction layer, evaluation layer, and execution layer. The weight update of the prediction layer combines the temperature detection value, temperature operation deviation, and deviation barrier coefficient for dynamic correction, and is configured with a weight decay mechanism. The weight updates of the evaluation layer and the execution layer are both adjusted through Gaussian membership basis function mapping, and the weight update of the execution layer is synchronously associated with the weight deviation of the evaluation layer, realizing the coordinated adaptive adjustment of the three-layer network.

[0021] Specifically, the expression for the weighted adaptive adjustment law corresponding to the furnace temperature of the pyrolysis furnace is as follows: ; ; ; In the formula, Weight estimates for the fuzzy neural network in the temperature loop prediction layer The adaptive regulation law; This is the positive definite gain matrix of the temperature control loop; The second kind of Gaussian membership basis function vector of the temperature loop; This is the real-time temperature reading of the pyrolysis furnace chamber. This refers to the operating deviation of the pyrolysis furnace chamber temperature; This is the temperature deviation barrier coefficient, which is always positive. To estimate the layer weight decay coefficient, a constant positive value is used; Weight estimates for the fuzzy neural network in the temperature loop evaluation layer The adaptive regulation law; To evaluate the adjustment parameters of the layer network, satisfying >0; The first kind of Gaussian membership basis function vector of the temperature loop transpose; Weight estimation values ​​for the fuzzy neural network in the temperature loop execution layer. The adaptive regulation law; To adjust the parameters of the execution layer network to meet the requirements. >0.

[0022] Furthermore, the operating parameters include the pyrolysis furnace system pressure; the weighted adaptive adjustment law corresponding to the pyrolysis furnace system pressure is divided into three sets of weight update rules: prediction layer, evaluation layer, and execution layer. The weight update of the prediction layer combines the pressure detection value, pressure operation deviation, and deviation barrier coefficient for dynamic correction, and is configured with a weight decay mechanism; the weight updates of the evaluation layer and the execution layer are both adjusted through Gaussian membership basis function mapping, and the weight update of the execution layer is synchronously associated with the weight deviation of the evaluation layer, realizing the coordinated adaptive adjustment of the three-layer network of the pressure loop.

[0023] Specifically, the expression for the weighted adaptive regulation law corresponding to the pressure of the pyrolysis furnace system is as follows: ; ; ; In the formula, Weight estimates for the fuzzy neural network in the pressure loop prediction layer The adaptive regulation law; This is the positive definite gain matrix for the pressure control loop; The second kind of Gaussian membership basis function vector of the pressure loop; This refers to the real-time pressure monitoring value of the pyrolysis furnace system. This refers to the operating deviation of the pyrolysis furnace system pressure; This is the pressure deviation barrier coefficient, which is always positive. To estimate the layer weight decay coefficient, a constant positive value is used; Weight estimates for the fuzzy neural network in the pressure loop evaluation layer The adaptive regulation law; Adjustment parameters for the pressure loop evaluation layer network; The first kind of Gaussian membership basis function vector of the pressure loop transpose; Weight estimates for the fuzzy neural network execution layer of the pressure loop. The adaptive regulation law; These are the adjustment parameters for the pressure loop execution layer network.

[0024] Furthermore, the process of regulating the furnace temperature of the pyrolysis furnace based on the adaptive regulation law includes: constructing the optimal performance index function of the temperature control loop based on the temperature operating deviation of the pyrolysis furnace; establishing the optimality equation of temperature control by combining the optimal temperature control law to be solved with the optimal performance index function; analytically solving the optimality equation through extreme value conditions to obtain the analytical initial value of the optimal temperature control law; using a predictor-evaluator-execution layer collaborative fuzzy neural network to perform nonlinear approximation compensation on the analytical initial value to obtain the final temperature adaptive optimal control law; adjusting the cooling gas recirculation ratio and the gasification furnace gas production heat load in real time according to the temperature adaptive optimal control law to achieve closed-loop control of the furnace temperature; when the gasification furnace inlet temperature is lower than the preset threshold, linking to increase the gasification furnace reaction intensity to increase the gas production temperature, and simultaneously triggering the pyrolysis furnace airflow channel self-cleaning program.

[0025] Furthermore, the optimal performance index function for temperature control uses the cooling gas recirculation adjustment coefficient as the optimization variable, and calculates the minimum integral cost within the feasible region of temperature deviation, comprehensively balancing the temperature control accuracy and the energy consumption of the cooling system.

[0026] For chicken legs, the optimal performance index functional expression of the temperature control loop is: ; In the formula, The optimal performance index functional for the temperature control loop; This is the cooling air recirculation adjustment coefficient; For a compact set of temperature deviations including the origin, Cooling gas recirculation adjustment coefficient The feasible control domain; The density function of the cost of temperature control; For integration time; The current moment; The optimal temperature control law to be solved is denoted as .

[0027] Furthermore, the optimality equation for temperature control integrates the temperature deviation barrier term, the control input cost term, and the temperature dynamic change term to describe the extreme conditions satisfied by optimal temperature control, providing a basis for solving the optimal control law.

[0028] Specifically, the HJB optimality equation for temperature control is expressed as follows: ; In the formula, The HJB optimality equation for temperature control; The optimal temperature control law to be solved; This is the temperature deviation barrier coefficient; The volumetric flow rate of the gas fed from the gasifier into the pyrolysis furnace; The density of the gas; The specific heat capacity of the gas at constant pressure; The outlet gas temperature of the gasifier; This refers to the temperature inside the pyrolysis furnace. This refers to the effective reaction volume of the pyrolysis furnace. The average density of the material inside the furnace; The average specific heat capacity of the materials inside the furnace; This refers to the endogenous heat power generated by the pyrolysis reaction; This refers to the temperature of the recirculating cooling gas. This is the derivative of the pyrolysis furnace temperature setpoint with respect to time.

[0029] Furthermore, the final temperature adaptive optimal control law consists of three parts: a deviation reference adjustment term, an execution layer fuzzy neural network approximation term, and a prediction layer fuzzy neural network approximation term. It can dynamically output the optimal cooling gas reflux adjustment coefficient based on the temperature deviation, the real-time furnace temperature, and the cooling gas temperature.

[0030] Specifically, the final expression of the temperature adaptive optimal control law is: ; In the formula, This is the final temperature-adaptive optimal control law; The gain is designed for temperature control and is always positive. This is the temperature deviation barrier coefficient; This refers to the operating deviation of the pyrolysis furnace chamber temperature; This is the real-time temperature of the pyrolysis furnace chamber. This refers to the temperature of the recirculating cooling gas. The weight estimates for the fuzzy neural network in the temperature loop execution layer; The vector of the first kind of Gaussian membership basis functions of the temperature loop; For the temperature loop prediction layer, a fuzzy neural network is used to determine the optimal weight vector. The estimated value; This is the second kind of Gaussian membership basis function vector of the temperature loop.

[0031] Furthermore, the process of regulating the pressure of the pyrolysis furnace system based on the adaptive control law includes: collecting the pressure difference signal and the furnace pressure signal at each sealing node of the pyrolysis furnace, and calculating the pressure operation deviation; constructing a pressure control cost function by combining the pressure optimal control law to be solved and the pressure operation deviation; introducing the pyrolysis furnace volume parameters, pressure operation deviation, and furnace pressure state variables to solve for the pressure adaptive optimal control law; adjusting the sealing nitrogen supply flow rate and the feeding interlock device in real time according to the pressure adaptive optimal control law to maintain a slightly positive pressure condition in the furnace; and automatically triggering the nitrogen supply sealing program and the feeding deceleration protection logic when the pressure difference at any sealing node exceeds the threshold.

[0032] Furthermore, the pressure adaptive optimal control law consists of three parts: a deviation benchmark adjustment term, an execution layer fuzzy neural network approximation term, and a prediction layer fuzzy neural network approximation term. It can dynamically output the optimal nitrogen replenishment flow rate based on the pressure deviation, the real-time pressure inside the furnace, and the volume of the pyrolysis furnace.

[0033] Specifically, the expression for the pressure adaptive optimal control law is: ; In the formula, The optimal control law for nitrogen replenishment flow rate; The gain is designed for pressure control and is always positive. This is the pressure deviation barrier coefficient; This refers to the effective reaction volume of the pyrolysis furnace. This refers to the operating deviation of the pyrolysis furnace system pressure; This refers to the real-time pressure value of the pyrolysis furnace system. The weight estimates for the fuzzy neural network in the pressure loop execution layer; The vector of the first kind of Gaussian membership basis functions of the pressure loop; For the pressure loop prediction layer, a fuzzy neural network is used to determine the optimal weight vector. The estimated value; This is the second-class Gaussian membership basis function vector of the pressure loop.

[0034] Furthermore, the Lyapunov stability functional of the temperature control loop consists of four parts: a temperature deviation barrier term, a prediction layer weight deviation term, an execution layer weight deviation term, and a judgment layer weight deviation term. These components characterize the energy level of the temperature control loop and provide a stability basis for the design of the adaptive regulation law.

[0035] Specifically, the Lyapunov stability functional expression of the temperature control loop is: ; In the formula, For the Lyapunov stability functional of the temperature control loop; This is the temperature deviation barrier coefficient; This refers to the operating deviation of the pyrolysis furnace chamber temperature; The weight bias of the fuzzy neural network for the temperature loop prediction layer; T is the vector transpose sign; This is the positive definite gain matrix of the temperature control loop; The weight bias of the fuzzy neural network for the temperature loop execution layer; The weight bias of the fuzzy neural network for the temperature loop evaluation layer.

[0036] Furthermore, the Lyapunov stability functional of the pressure control loop consists of four parts: a pressure deviation barrier term, a prediction layer weight deviation term, an execution layer weight deviation term, and a judgment layer weight deviation term. These components are used to characterize the energy level of the pressure control loop and ensure the stable convergence of the pressure closed-loop system.

[0037] Specifically, the Lyapunov stability functional expression of the pressure control loop is: ; In the formula, For the Lyapunov stability functional of the pressure control loop; This is the pressure deviation barrier coefficient; This refers to the operating deviation of the pyrolysis furnace system pressure; The weight bias of the fuzzy neural network for the pressure loop prediction layer; T is the vector transpose sign; This is the positive definite gain matrix for the pressure control loop; The weight bias of the fuzzy neural network for the pressure loop execution layer; The weight bias of the fuzzy neural network for evaluating the pressure loop is used.

[0038] Furthermore, the kinetic model of the pyrolysis furnace temperature covers three energy exchange processes: heat introduced by the fuel gas, heat generated by the pyrolysis reaction, and heat introduced by the cooling gas, which is used to describe the dynamic change law of the pyrolysis furnace temperature.

[0039] Specifically, the kinetic model expression for the furnace temperature of the pyrolysis furnace is as follows: ; In the formula, t represents the furnace temperature of the pyrolysis furnace; t represents time. The volumetric flow rate of the gas fed from the gasifier into the pyrolysis furnace; The density of the gas; The specific heat capacity of the gas at constant pressure; The outlet gas temperature of the gasifier; This refers to the effective reaction volume of the pyrolysis furnace. The average density of the material inside the furnace; The average specific heat capacity of the materials inside the furnace; This refers to the endogenous heat power generated by the pyrolysis reaction; This is the cooling air recirculation adjustment coefficient; This refers to the temperature of the recirculating cooling gas.

[0040] Furthermore, the dynamic model of the pyrolysis furnace system pressure is constructed based on the gas state equation, reflecting the dynamic influence of gas furnace inlet gas, nitrogen replenishment, outlet exhaust gas and seal leakage on the furnace pressure.

[0041] Specifically, the kinetic model expression for the pressure in the pyrolysis furnace system is as follows: ; In the formula, t represents the pressure of the pyrolysis furnace system; t represents time. This is the universal gas constant; The thermodynamic temperature of the pyrolysis furnace chamber; This refers to the effective reaction volume of the pyrolysis furnace. The average molar mass of the mixed gas in the furnace; The volumetric flow rate of the fuel gas supplied to the gasifier; For the nitrogen supply flow rate to be sealed; This refers to the gas flow rate at the outlet of the pyrolysis furnace. This represents the leakage flow rate at the sealing point.

[0042] In a specific implementation, a smart control system for gas-coke cogeneration based on positive pressure sealing and temperature coordination includes: The condition sensing and target setting unit is used to collect the operating parameters of the gas-coke cogeneration system and determine the real-time detection value and preset target value of each parameter. A real-time deviation calculation unit is used to calculate the system operating deviation based on the real-time detection value and the preset target value; The fuzzy weight mapping unit is used to solve for the optimal weight vector and its online estimate value corresponding to the running parameters through a collaborative fuzzy neural network of prediction layer-evaluation layer-execution layer; The weight deviation calculation unit is used to calculate the weight deviation based on the optimal weight vector and the online estimated value. Stability functional building blocks are used to construct Lyapunov stability functionals for control loops by combining system operating deviations and weight deviations. An adaptive law generation unit is used to derive the adaptive adjustment law of the weights of the fuzzy neural network based on the Lyapunov stability functional derivation. The multi-loop coordinated control unit is used to adjust the output of each actuator in real time according to the adaptive regulation law, so as to achieve coordinated optimization control of positive pressure sealing performance and temperature field distribution.

[0043] In a specific embodiment, a smart control method for gas-coke co-production based on positive pressure sealing and temperature coordination includes the following steps: Raw coal is fed into the bubbling bed gasifier via a metering feeder. The high-temperature crude gas generated by the gasification reaction is then fed into the pyrolysis furnace as a heat source after large particles of dust are removed by a cyclone dust collector. The gas produced by the pyrolysis furnace is purified sequentially by an electrostatic precipitator and a scrubbing tower. Part of the purified gas is sent to a steam boiler to recover waste heat, while the other part is transported to the user's pipeline network via a gas compressor. The solid semi-coke generated by pyrolysis is sent to a cold coke bin for storage via a cooling and coke discharge device. The pyrolysis furnace is equipped with a cooling gas return pipeline, a nitrogen sealing interface, and a self-cleaning mechanism for the airflow channel. Differential pressure sensors are installed at sealing nodes such as flanges and feed inlets. Multiple temperature and pressure detection elements are arranged inside the furnace. The normal operating temperature range of the pyrolysis furnace is 600~700℃, and the furnace is maintained at a slight positive pressure of 50~100Pa.

[0044] For the above-mentioned process system, this application proposes a multivariate intelligent optimization control method based on a collaborative fuzzy neural network of prediction layer-evaluation layer-execution layer, which simultaneously regulates the temperature field of the pyrolysis furnace, system pressure and gas intake conditions to achieve positive pressure sealing and temperature coordinated control.

[0045] The overall control process of this application is achieved through steps S1 to S7: Step S1: Acquisition of operating parameters and setting of control targets. Determine the core control parameters of the gas-coke cogeneration system, including the pyrolysis furnace temperature, pyrolysis furnace system pressure, and gasifier inlet temperature; the pyrolysis furnace temperature setpoint is dynamically adjusted within the range of 600~700℃ according to the charcoal product process requirements, the system pressure is set to a slightly positive pressure condition, and the inlet temperature threshold is determined according to the design conditions.

[0046] Step S2: Real-time operating deviation calculation. The operating deviation is calculated based on the real-time measured values ​​of each parameter and the preset target values; for the temperature loop, a tangent barrier function is introduced. The temperature deviation range is constrained; similarly, a tangent barrier function is introduced to constrain the pressure deviation for the pressure loop. Through barrier function design, it is ensured that the system operating deviation remains within the preset safety boundary, preventing operational failures caused by deviation exceeding limits.

[0047] Step S3: Solving for the optimal weights and estimates of the fuzzy neural network. A collaborative fuzzy neural network architecture of prediction layer-evaluation layer-execution layer is adopted to approximate the unknown nonlinear terms of the temperature and pressure loops, respectively, and to obtain the optimal weight vectors and their online estimates for each loop.

[0048] Step S4: Calculation of weight bias. The online estimated value of the fuzzy neural network is subtracted from the theoretically optimal weight vector to obtain the weight bias of each layer of the network, which is used for stability analysis and adaptive law design.

[0049] Step S5: Construction of Lyapunov stability functionals. Combining the system operating deviation and the weight deviations of each layer, Lyapunov stability functionals for the temperature control loop and pressure control loop are constructed respectively, serving as the basis for stability analysis and adaptive law derivation.

[0050] Step S6: Derivation of the Adaptive Adjustment Law for Weights. By differentiating the Lyapunov stability functional over time and employing inequality scaling and negative qualitative design, the adaptive adjustment law for the weights of the prediction, evaluation, and execution layers is derived. This enables real-time online updates of the weights, adaptively tracking the dynamic changes of the system.

[0051] Step S7: Multivariable Collaborative Optimization Control. Based on the derived adaptive regulation law, the actuator outputs of the temperature and pressure loops are adjusted in real time to converge system operating deviations and simultaneously realize the working condition linkage protection logic.

[0052] Furthermore, the sub-process for optimizing the temperature control of the pyrolysis furnace is achieved through steps S71 to S75: Step S71: Construction of the optimal performance index functional for temperature control. Using the pyrolysis furnace temperature operating deviation and control input as optimization variables, an integral-type optimal performance index functional is constructed to comprehensively balance control accuracy and energy consumption cost.

[0053] Step S72: Establishment of the HJB optimality equation for temperature control. Combining the optimal temperature control law and the optimal performance index functional to be solved, the HJB optimality equation for temperature control is derived and established, describing the necessary conditions for optimal control.

[0054] Step S73: Solving for the analytical initial value of the optimal control law. The HJB equation is solved analytically using the control input extremum conditions to obtain the analytical initial value form of the optimal control law.

[0055] Step S74: Fuzzy Neural Network Approximation Compensation. A three-layer collaborative fuzzy neural network is used to approximate and compensate for the unknown nonlinear terms in the analytical initial values, resulting in the final temperature adaptive optimal control law.

[0056] Step S75: Temperature closed-loop control and operating condition linkage. The cooling gas recirculation ratio and gasifier gas production heat load are adjusted in real time according to the optimal control law to achieve precise control of furnace temperature; when the gasifier inlet gas temperature is lower than the preset threshold, the gasifier blast intensity is increased to improve the gas production temperature, and the pyrolysis furnace airflow channel self-cleaning program is triggered simultaneously to avoid tar condensation and blockage of the channel due to low temperature.

[0057] Furthermore, the sub-process for optimizing the pressure control of the pyrolysis furnace system is implemented through steps S76 to S79: Step S76: Pressure Operation Deviation and Sealing Pressure Difference Acquisition. Real-time acquisition of furnace pressure signals and differential pressure signals at each sealing node is performed to calculate the pressure operation deviation and determine the sealing status.

[0058] Step S77: Construction of Pressure Control Cost Function. Using pressure operating deviation and nitrogen refill flow rate as variables, a pressure control cost function is constructed to balance sealing performance and nitrogen consumption.

[0059] Step S78: Solving the pressure adaptive optimal control law. Introducing the pyrolysis furnace volume parameters, pressure state variables, and deviation variables, and combining them with fuzzy neural network approximation terms, the pressure adaptive optimal control law is obtained.

[0060] Step S79: Pressure closed-loop control and sealing protection. The nitrogen replenishment flow rate and feed interlock device are adjusted in real time according to the optimal control law; when the pressure difference at any sealing node exceeds the safety threshold, the nitrogen replenishment flow rate is automatically increased, and the feed deceleration or cut-off protection is triggered at the same time to maintain the positive pressure condition in the furnace and prevent gas leakage.

[0061] Furthermore, this application uses Gaussian membership functions as the basis functions of the fuzzy neural network, which can achieve universal approximation of any continuous nonlinear function. The output expression of the fuzzy neural network is: ; In the formula, For the dimensions of the input variables, The total number of fuzzy rules; This is the output of the fuzzy neural network that approximates the unknown function. The network input vector; This is the connection weight vector from the rule layer to the output layer; The output vector of the normalized basis functions; The activation output for the l-th fuzzy rule; and These are the membership function width and center parameter corresponding to the i-th input for the j-th rule, respectively.

[0062] To verify the effectiveness of this method, a simulation experiment was conducted, selecting three typical operating conditions: full load, 70% load, and 50% load. The pyrolysis furnace temperature setpoint was 650℃, and the pressure setpoint was 80Pa. Temperature control loop parameter settings: design gain... Adaptive law parameters Gain matrix Deviation barrier coefficient Pressure control loop parameter settings: design gain Adaptive law parameters , , Gain matrix Deviation barrier coefficient Simulation results verify the control effectiveness and robustness of the proposed method under different load conditions.

[0063] The methods and systems provided in this application have been described in detail above. This document has illustrated the principles and implementation methods of this application through specific embodiments. The descriptions of the embodiments above are only for the purpose of assisting in understanding the methods and core ideas of this application. At the same time, for those skilled in the art, there will be variations in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

[0064] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0065] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A smart control method for gas-coke co-production based on positive pressure sealing and temperature coordination, characterized in that, include: Collect the operating parameters of the gas-coke cogeneration system and obtain the real-time detection values ​​and preset target values ​​of each parameter; The system operating deviation is calculated based on the real-time detection values ​​and the preset target values. A collaborative fuzzy neural network consisting of a prediction layer, an evaluation layer, and an execution layer is used to solve for the optimal weight vector and its online estimate corresponding to the running parameters. The weight deviation is then calculated based on the optimal weight vector and the online estimate. By combining the system operating deviation and the weight deviation, a Lyapunov stability functional for the control loop is constructed. The adaptive adjustment law of fuzzy neural network weights is derived based on the Lyapunov stability functional derivation. The output of each actuator is adjusted in real time according to the adaptive regulation law to achieve coordinated optimization control of the positive pressure sealing performance and temperature field distribution of the gas-carbon cogeneration system.

2. The intelligent control method for gas-coal co-production based on positive pressure sealing and temperature coordination according to claim 1, characterized in that, The gas-coke cogeneration system includes a bubble bed gasification unit and a pyrolysis reaction unit coupled in series, and the operating parameters include the furnace temperature of the pyrolysis furnace. The weighted adaptive adjustment law corresponding to the furnace temperature of the pyrolysis furnace is divided into three sets of weight update rules: prediction layer, evaluation layer, and execution layer. The weight update of the prediction layer combines the temperature detection value, temperature operation deviation and deviation barrier coefficient for dynamic correction, and is configured with a weight decay mechanism. The weight updates of both the evaluation layer and the execution layer are adjusted through Gaussian membership basis function mapping, and the weight updates of the execution layer are synchronously correlated with the weight deviations of the evaluation layer, thereby achieving coordinated adaptive adjustment of the three-layer network.

3. The intelligent control method for gas-coal co-production based on positive pressure sealing and temperature coordination according to claim 1, characterized in that, The operating parameters include the pyrolysis furnace system pressure; The weighted adaptive adjustment law corresponding to the pressure of the pyrolysis furnace system is divided into three sets of weight update rules: prediction layer, evaluation layer, and execution layer. The weight update of the prediction layer combines the pressure detection value, pressure operation deviation and deviation barrier coefficient for dynamic correction, and is configured with a weight decay mechanism. The weight updates of both the evaluation layer and the execution layer are adjusted through Gaussian membership basis function mapping, and the weight updates of the execution layer are synchronously correlated with the weight deviations of the evaluation layer, thereby achieving coordinated adaptive adjustment of the three-layer network of the pressure loop.

4. The intelligent control method for gas-coal co-production based on positive pressure sealing and temperature coordination according to claim 2, characterized in that, The process of adjusting the output of each actuator in real time according to the adaptive regulation law includes: regulating the furnace temperature of the pyrolysis furnace according to the adaptive regulation law. Based on the temperature operating deviation of the pyrolysis furnace, an optimal performance index function for the temperature control loop is constructed. By combining the optimal temperature control law to be solved with the optimal performance index function, the optimality equation for temperature control is established. The optimality equation is solved analytically by applying the extreme value condition, and the analytical initial value of the optimal temperature control law is obtained. A collaborative fuzzy neural network consisting of a prediction layer, an evaluation layer, and an execution layer is used to perform nonlinear approximation compensation on the analytical initial value, resulting in the final temperature adaptive optimal control law. Based on the temperature adaptive optimal control law, the proportion of cooling gas recirculation and the heat load of gasification furnace gas production are adjusted in real time to achieve closed-loop control of furnace temperature. When the gas inlet temperature of the gasifier is lower than the preset threshold, the reaction intensity of the gasifier is increased to improve the gas production temperature, and the self-cleaning program of the pyrolysis furnace airflow channel is triggered simultaneously.

5. The intelligent control method for gas-coal co-production based on positive pressure sealing and temperature coordination according to claim 4, characterized in that, Real-time adjustment of the output of each actuator based on the adaptive regulation law includes: regulating the pressure of the pyrolysis furnace system based on the adaptive regulation law. The differential pressure signal and the pressure signal inside the furnace at each sealing node of the pyrolysis furnace are collected, and the pressure operation deviation is calculated. By combining the pressure optimal control law to be solved with the pressure operation deviation, a pressure control cost function is constructed. By introducing pyrolysis furnace volume parameters, pressure operation deviation, and internal pressure state variables, the pressure adaptive optimal control law is obtained by solving the problem. According to the pressure adaptive optimal control law, the sealing nitrogen supply flow rate and the feeding interlock device are adjusted in real time to maintain a slightly positive pressure condition inside the furnace. When the differential pressure at any sealing node exceeds the threshold, the nitrogen resealing procedure and the feed deceleration protection logic are automatically triggered.

6. The intelligent control method for gas-coal co-production based on positive pressure sealing and temperature coordination according to claim 5, characterized in that, The pressure adaptive optimal control law consists of three parts: a deviation benchmark adjustment term, an execution layer fuzzy neural network approximation term, and a prediction layer fuzzy neural network approximation term. It dynamically outputs the optimal nitrogen replenishment flow rate based on the pressure deviation, the real-time pressure inside the furnace, and the volume of the pyrolysis furnace.

7. The intelligent control method for gas-coal co-production based on positive pressure sealing and temperature coordination according to claim 1, characterized in that, The Lyapunov stability functional of the control loop includes: the Lyapunov stability functional of the temperature control loop, which consists of four parts: temperature deviation barrier term, prediction layer weight deviation term, execution layer weight deviation term, and evaluation layer weight deviation term, and is used to characterize the energy level of the temperature control loop.

8. The intelligent control method for gas-coal co-production based on positive pressure sealing and temperature coordination according to claim 1, characterized in that, The Lyapunov stability functional of the control loop includes: the Lyapunov stability functional of the pressure control loop, which consists of four parts: pressure deviation barrier term, prediction layer weight deviation term, execution layer weight deviation term, and evaluation layer weight deviation term, and is used to characterize the energy level of the pressure control loop.

9. The intelligent control method for gas-coal co-production based on positive pressure sealing and temperature coordination according to claim 4, characterized in that, The optimal performance index function for temperature control uses the cooling gas recirculation adjustment coefficient as the optimization variable, and finds the minimum integral cost within the feasible region of temperature deviation, comprehensively balancing the temperature control accuracy and the energy consumption of the cooling system.

10. An intelligent control system for gas-coal cogeneration based on positive pressure sealing and temperature coordination, characterized in that, include: The condition sensing and target setting unit is used to collect the operating parameters of the gas-coke cogeneration system and determine the real-time detection value and preset target value of each parameter. A real-time deviation calculation unit is used to calculate the system operating deviation based on the real-time detection value and the preset target value; The fuzzy weight mapping unit is used to solve for the optimal weight vector and its online estimate value corresponding to the running parameters through a collaborative fuzzy neural network of prediction layer-evaluation layer-execution layer; The weight deviation calculation unit is used to calculate the weight deviation based on the optimal weight vector and the online estimated value. Stability functional building blocks are used to construct Lyapunov stability functionals for control loops by combining system operating deviations and weight deviations. An adaptive law generation unit is used to derive the adaptive adjustment law of the weights of the fuzzy neural network based on the Lyapunov stability functional derivation. The multi-loop coordinated control unit is used to adjust the output of each actuator in real time according to the adaptive regulation law, and to perform coordinated optimization control of positive pressure sealing performance and temperature field distribution.

Images