A multivariable energy efficiency optimization control system for a furnace

Abstract

The application relates to the technical field of control, and particularly discloses a multivariable energy efficiency optimization control system for a heating furnace, which is used for solving the problems of local overheating, uneven temperature, and difficult accurate positioning and compensation of heat loss in the operation of an existing cracking heating furnace, and comprises a parameter detection module, a multivariable coupling modeling and simulation module, an optimization control module, and an execution and feedback module; the application constructs a dynamic digital twin through multimodal online sensing and data assimilation, generates a Pareto frontier solution and self-adaptively adjusts weights based on model predictive control and improved NSGA-II parallel optimization; when a hot spot / cold spot is triggered, a quadratic surface fitting compensation strategy is implemented and is issued for execution, so that high-precision simulation prediction, accurate closed-loop control, and real-time online energy efficiency optimization are realized.

Description

Technical Field

[0001] This invention relates to the field of control technology, and in particular to a multivariable energy efficiency optimization control system for heating furnaces. Background Technology

[0002] Chinese invention patent CN118500140A discloses a method and system for energy-saving monitoring and prediction of a heating furnace based on digital twin technology. Although it achieves the monitoring and evaluation of single indicators such as flue gas temperature, furnace surface temperature, flue gas oxygen content, excess air coefficient, and overall thermal efficiency by constructing a multi-zone digital twin simulation model and combining historical and real-time operating data to simulate, analyze, and predict the overall thermal efficiency and energy consumption of a pyrolysis heating furnace using petroleum coke or natural gas as fuel under different operating conditions, the pyrolysis heating furnace has several drawbacks. These include: the long rotary drum structure causing multiple circulation and retention of materials within the furnace; differences in the heat capacity and thermal inertia of the refractory lining in the zoned heating sections leading to localized temperature response lag; the time-varying heat transfer coefficient caused by the interaction between furnace rotation and gas-solid two-phase flow within the furnace; and the need for adjustments to fuel injection and combustion air volume. Considering the combustion characteristics of volatile and carbonaceous fuels, the large local flame temperature fluctuations caused by the burner being distributed in multiple combustion zones, the increased uncertainty of flue gas leakage and heat loss due to wear and tear of the furnace head and tail seals, and the nonlinear fluctuations in the heat transfer efficiency of the waste heat recovery heat exchanger due to dust content and scaling conditions, key variables such as fuel flow rate, combustion air flow rate, zoned temperature gradient, flue gas velocity, and oxygen content exhibit strong coupled dynamic changes. Even if the overall thermal efficiency meets the target, local overheating or cold spots, uneven heating of sidewalls, load fluctuations in the heat exchange section, and heat transfer losses are difficult to accurately locate and correct in a timely manner, resulting in the inability to further improve equipment life, operational stability, and energy-saving benefits. Therefore, it is urgent to propose a multivariate energy efficiency optimization control system for the pyrolysis heating furnace to address the above-mentioned multivariate coupled characteristics. Summary of the Invention

[0003] The technical problem to be solved by this invention is to provide a multivariable energy efficiency optimization control system for heating furnaces. By combining multi-modal online sensing and second-level data assimilation with multi-physics digital twins and model predictive control, along with NSGA-II multi-objective optimization and hot / cold point dynamic compensation closed loop, high-precision simulation prediction, accurate control and stable energy saving are achieved.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A multivariable energy efficiency optimization control system for a heating furnace includes a parameter detection module. This module comprises a flue gas parameter detection module, a furnace body temperature detection module, a fuel flow detection module, a combustion air flow detection module, a seal wear monitoring module, and a waste heat recovery monitoring module. These modules are used to collect real-time data on flue gas velocity and oxygen content, surface temperature of each zone of the furnace body, fuel injection rate, combustion air volume, furnace head and tail seal wear, and heat exchanger heat transfer efficiency. The parameter detection module is sequentially connected to a multivariable coupled modeling and simulation module, an optimization control module, and an execution and feedback module. The multivariable coupled modeling and simulation module constructs a system based on the acquired parameters, covering material circulation and retention, furnace body thermal inertia, and two-phase flow. A dynamic digital twin model of multi-physics coupling, including interaction, combustion characteristics, sealing, and heat exchange efficiency, is used. Combined with historical and real-time operating data, multi-condition simulation prediction is performed. The optimization control module solves the objective function based on the simulation results of the multivariable coupled modeling and simulation module and the set overall thermal efficiency, local temperature gradient, and heat loss indicators. It generates fuel and combustion air zoning distribution strategies, waste heat recovery reversing valve opening adjustment strategies, and layout hot and cold point dynamic and precise compensation control strategies. The execution and feedback module drives the burner group fuel distribution device, combustion air volume adjustment device, and waste heat recovery reversing valve respectively according to the optimization control decisions of the optimization control module, and updates the digital twin model in real time.

[0006] As a further aspect of the present invention, the flue gas parameter detection module employs multi-point ultrasonic anemometers arranged in different zones of the pyrolysis heating furnace. By measuring the time difference of sound wave propagation in the flue gas medium, the instantaneous flow velocity of each zone is calculated in real time. Combined with a multi-point static pressure difference compensation algorithm based on online sampling of dust concentration, the influence of dust disturbance is eliminated. The oxygen content detection adopts a Raman spectroscopy sensor set before and after the oxygen meter, supplemented by a dual-mode fusion calibration method based on an electrochemical oxygen sensor. The Raman signal drift and the temperature response compensation of the electrochemical sensor are calibrated in real time.

[0007] It should be noted that the multi-point ultrasonic anemometers arranged in different zones of the pyrolysis heating furnace used in this invention calculate the instantaneous flow velocity of each zone in real time by measuring the time difference of sound wave propagation in the flue gas medium. Combined with a multi-point static pressure difference compensation algorithm based on online sampling of dust concentration, the error caused by dust disturbance is eliminated. Compared with the traditional single-point detection method, this improves the response speed and measurement accuracy, providing flue gas flow parameters for the multivariate coupled model. At the same time, Raman spectroscopy sensors are set before and after the oxygen meter, and a dual-mode fusion calibration method of electrochemical oxygen sensor is used to improve the detection accuracy and anti-drift capability of oxygen content in high temperature and high dust environment by real-time calibration of Raman signal drift with temperature and temperature response deviation of electrochemical sensor, and shortens the response time. This provides effective confidence boundary conditions for simulation prediction, improves the robustness of system operation and fault early warning capability, and enhances the accuracy, stability and dynamic response performance of flue gas parameter detection. This provides a data foundation for the multivariate coupled simulation and optimization control of this invention.

[0008] As a further aspect of the present invention, the furnace body temperature detection module employs a combination of high-temperature short-wave infrared thermal imagers arranged in each heating zone on the outer surface of the pyrolysis heating furnace and fiber Bragg grating array temperature sensors embedded in the refractory lining for measurement. The infrared and FBG sensor data acquisition is synchronously triggered by a rotary position encoder, and the infrared radiation drift and FBG temperature drift are corrected online based on the established furnace body thermal inertia compensation model and multi-point data fusion algorithm, so as to continuously monitor the surface temperature of each zone in real time at a frequency of ≥20Hz.

[0009] It should be noted that the furnace body temperature detection module used in this invention uses high-temperature short-wave infrared thermal imagers arranged in each heating zone on the outer surface of the pyrolysis heating furnace, and fiber Bragg grating array temperature sensors embedded under the refractory lining for mutual supplementary measurement. It also combines a rotary position encoder for synchronous triggering and acquisition, and uses an online correction of infrared radiation drift and FBG temperature drift based on the furnace body thermal inertia compensation model and multi-point data fusion algorithm to achieve real-time continuous monitoring of the surface temperature of each zone. This effectively overcomes the temperature measurement blind zone and error drift caused by high-temperature dust obstruction, rapid furnace body rotation and local thermal inertia lag, improves the spatiotemporal consistency and accuracy of temperature measurement data, provides realistic boundary conditions with good temporal and spatial resolution for the multi-physics coupled simulation model, ensures the dynamic response capability and error convergence speed of simulation prediction, and provides data support for the optimization control module to compensate for local hot and cold points, thereby improving the refinement level and operational stability of the heating furnace energy efficiency optimization control.

[0010] As a further aspect of the present invention, the fuel flow detection module adopts a dual-mode measurement architecture of a mass flow sensor and an ultrasonic time-difference volumetric flow sensor arranged in series in the fuel pipeline at the front end of each burner. A microwave medium analyzer is set between the two sensors to collect fuel density and viscosity data in real time. The collected mass flow and volumetric flow signals are fused based on an improved gain adaptive Kalman filter algorithm and an online rheological model for dynamic calibration to compensate for measurement errors caused by changes in volatile matter and temperature and pressure fluctuations.

[0011] It should be noted that the fuel flow detection module used in this invention arranges a Coriolis mass flow sensor and an ultrasonic volumetric flow sensor in series at the front end of each burner, and introduces a microwave medium analyzer to acquire fuel density and viscosity data in real time. The mass and volumetric flow signals and physical property parameters are synchronously input into an improved gain adaptive Kalman filter, and the noise covariance used is dynamically calibrated based on an online rheological model. This enables the fusion and compensation of dual-mode signals under conditions of fuel volatile matter and temperature and pressure fluctuations. It can capture instantaneous signal distortion caused by sudden changes in fuel physical properties in real time, and maintain measurement consistency when switching between various operating conditions. It provides fuel flow boundary conditions for multivariate coupled digital twin simulation, improves the real-time performance, stability and fault early warning capability of optimized control, and provides data support for the dynamic energy efficiency adjustment of the combustion system.

[0012] As a further aspect of the present invention, the fuel flow detection module synchronously inputs the collected Coriolis mass flow rate signal and ultrasonic volumetric flow rate signal, along with the fuel density and viscosity parameters acquired in real time by the microwave medium analyzer, into a dual-channel extended Kalman filter. The process noise covariance of the first channel is dynamically updated based on an online regression reconstruction model of changes in fuel volatile content and historical operating deviations. The measurement noise covariance of the second channel is adjusted in real time through temperature-pressure hysteresis characteristic identification and an adaptive recursive least squares algorithm. After filtering and fusion, a secondary correction based on residual orthogonal projection is performed to compensate for the flow measurement error caused by changes in fuel volatile content and temperature and pressure fluctuations.

[0013] It should be noted that the fuel flow detection module used in this invention synchronously inputs the collected Coriolis mass flow rate signal and ultrasonic volumetric flow rate signal with the fuel density and viscosity parameters obtained in real time by the microwave medium analyzer into a dual-channel extended Kalman filter. The process noise covariance of the first channel is dynamically updated based on the online regression reconstruction model of changes in fuel volatile content and historical operating deviations. The measurement noise covariance of the second channel is adjusted in real time through temperature-pressure hysteresis characteristic identification and adaptive recursive least squares algorithm, and then corrected twice based on residual orthogonal projection after filtering and fusion. This enables online detection of mass and volumetric flow rate measurements under extreme conditions such as fluctuations in fuel volatile content, rapid temperature changes, and pressure pulsations. It provides fuel flow boundary conditions with excellent confidence for multivariate coupled digital twin simulation, improves the response speed, stability, and fault warning capability of the optimization control module for dynamic adjustment of the combustion system, and enhances the energy efficiency optimization control benefits of this invention under complex operating conditions.

[0014] As a further aspect of the present invention, the combustion air flow detection module is equipped with vortex flow meters and multi-point static pressure sensors at the inlet of the combustion air duct and in the branch pipes of each combustion zone. Temperature and humidity sensors are set up upstream and downstream of the flow meters to collect environmental parameters. Based on the collected static pressure difference, vortex pulse frequency and temperature and humidity data, a nonlinear adaptive observer algorithm combining CFD simulation pre-calibration curves is used to estimate the velocity profile distortion coefficient in real time and compensate for the vortex measurement values.

[0015] It should be noted that the combustion air flow detection module of the present invention arranges high-temperature resistant vortex flow meters and multi-point static pressure sensors at the inlet of the combustion air duct and in the branch pipes of each combustion zone, and adds temperature and humidity sensors upstream and downstream of the flow meters to synchronously collect environmental parameters. By inputting static pressure difference, vortex pulse frequency and temperature and humidity data and combining them with the nonlinear adaptive observer algorithm of the CFD simulation pre-calibration curve, the flow velocity profile distortion coefficient is identified and estimated in real time, and the vortex measurement value is dynamically corrected. Compared with the vortex measurement method in the prior art that relies on a single calibration coefficient and cannot respond to changes in operating conditions, this improves the flow measurement accuracy under high dust, high temperature and flow pulsation conditions, and eliminates the error caused by profile non-uniformity through online profile correction, providing consistent and confident airflow boundary conditions for subsequent multivariate coupled simulation models, thereby enhancing the reliability and response speed of optimization control decisions.

[0016] As a further embodiment of the present invention, the sealing wear monitoring module includes: an array of fiber Bragg grating strain sensors circumferentially distributed along the outer edge of the sealing rings at the furnace head and tail of the pyrolysis heating furnace, used to collect minute radial deformations of the sealing rings in real time; a high-frequency ultrasonic transducer pre-embedded at the contact surface of the sealing rings, used to detect ultrasonic signals generated by wear and gas leakage; an industrial infrared thermal imager installed on the outside of the sealing part, used to capture local temperature field abrupt changes caused by leaked flue gas; the three sensor signals are processed online by a multi-modal deep learning algorithm that fuses multi-scale wavelet packet decomposition and bidirectional long short-term memory network to detect the sealing wear and leakage velocity in real time, and the detection results are fed back to the optimization control decision module for dynamic compensation in real time.

[0017] It should be noted that the method used in this invention, which employs a three-channel multimodal sensing system based on a fiber Bragg grating strain array, a high-frequency ultrasonic transducer, and an industrial infrared thermal imager, combined with a multi-scale wavelet packet decomposition and bidirectional LSTM deep learning fusion algorithm for real-time online monitoring of minute radial deformation of the sealing ring, ultrasonic leakage signals, and temperature field abrupt changes, provides boundary conditions for sealing wear and leakage velocity in the closed-loop adaptive execution stage of the multivariable coupled simulation model update and optimization control decision-making process. This not only enables rapid early warning of abnormal sealing conditions and timely issuance of compensation decisions, but also allows the digital twin model to dynamically self-calibrate, suppressing flue gas leakage and unpredictable heat loss caused by sealing wear. This improves the operational stability, response speed, and overall energy efficiency optimization level of the heating furnace system, and achieves long-term operational reliability improvement and extended downtime through continuous online monitoring.

[0018] As a further aspect of the present invention, the waste heat recovery monitoring module employs a multi-parameter coupled online detection method, specifically including: arranging K-type thermocouple arrays at the inlet and outlet flues of the heat exchanger for real-time measurement of the inlet and outlet flue gas temperatures; deploying multi-point thin-film heat flux meters within the heat exchanger tube bundle for real-time acquisition of heat flux per unit area of ​​the tube wall; and installing an acoustic imaging sensor array on the outside of the tube bundle for online identification of the location and thickness distribution of scaling. The waste heat recovery monitoring module dynamically calculates the heat transfer coefficient based on a bidirectional heat balance equation and an adaptive wall thermal resistance model, and combines this with a wall thermal resistance change rate threshold-triggered cleaning cycle prediction algorithm to locate and quantitatively assess the decrease in heat transfer efficiency caused by initial scaling and non-uniform distribution.

[0019] It should be noted that the waste heat recovery monitoring module of this invention, by arranging K-type thermocouple arrays, multi-point thin-film heat flow meters, and acoustic imaging sensor arrays in the inlet and outlet flues and tube bundles of the heat exchanger respectively, and combining the bidirectional heat balance equation and the adaptive wall thermal resistance model, dynamically calculates the heat transfer coefficient in real time, and triggers the cleaning cycle prediction algorithm with the wall thermal resistance change rate. This enables rapid location, accurate quantitative assessment, and proactive cleaning early warning of initial trace scaling and local non-uniform scaling inside the heat exchanger. Compared with the limitations of existing technologies that cannot achieve online high-sensitivity detection of initial scaling and are difficult to accurately predict the cleaning cycle, this invention achieves improved dynamic and accurate detection accuracy of heat exchange efficiency, improved accuracy of early scaling warning, improved heat exchanger operation stability and long-term heat transfer performance, effectively reduces maintenance downtime frequency, and improves the overall energy-saving effect of the system.

[0020] As a further aspect of this invention, the multivariable coupled modeling and simulation module employs a hierarchical dynamic adaptive modeling method, including: a multiphysics simulation engine based on the coupling of finite element heat conduction, CFD flow, discrete element material circulation, and combustion chemical reaction; the simulation engine integrates a real-time online data assimilation algorithm, injecting monitoring data into the simulation model at a second-level frequency, and performing self-calibration on thermal inertia, material residence time, and two-phase flow boundary conditions through a model error update mechanism driven by feedback residuals, gradually converging the simulation error to within ±0.5%; the simulation calculation runs on a GPU cluster parallel architecture.

[0021] It should be noted that the multivariable coupled modeling and simulation module of this invention adopts a hierarchical dynamic adaptive modeling method that couples multiple physics fields, including finite element heat conduction, CFD flow, discrete element material circulation, and combustion chemical reaction. It injects monitoring data in seconds using a real-time online data assimilation algorithm and achieves dynamic self-calibration of thermal inertia, material residence time, and two-phase flow boundary conditions through feedback residual driving. This enables the simulation error to converge to the set range in real time. Furthermore, the parallel computing of the GPU cluster significantly reduces the simulation time, overcoming the high-precision and high-real-time simulation requirements that cannot be achieved by existing single modeling methods. This improves the simulation prediction accuracy and real-time response performance, providing a model foundation and rapid optimization capability for dynamic and precise control.

[0022] As a further aspect of the present invention, the optimization control module includes: a multi-objective optimization engine based on model predictive control and an improved non-dominated sorting genetic algorithm, used to jointly solve the three objectives of maximizing overall thermal efficiency, homogenizing local temperature gradients, and minimizing heat loss; the engine automatically adjusts the weights of each objective through online sensitivity analysis and generates Pareto front solution sets in parallel within the feasible domain of each control variable; when a local hot spot or cold spot deviates from the threshold, the optimization control module automatically calls a local compensation sub-algorithm based on quadratic surface fitting to update the control strategy and sends the final optimal control command to the execution and feedback module in real time.

[0023] It should be noted that the optimization control module of this invention adopts a multi-objective optimization engine based on model predictive control and improved non-dominated sorting genetic algorithm. It automatically adjusts the objective weights online and generates Pareto front solution sets in parallel. When hot spots or cold spots appear, it calls the local compensation sub-algorithm of quadratic surface fitting in real time to accurately adjust the control strategy. Compared with the limitations of single-objective optimization and lack of adaptive local compensation in the existing technology, it realizes the synchronous optimization of overall and local energy efficiency indicators, and improves the accuracy of optimization decision, real-time response speed and energy efficiency stability.

[0024] The technical effects of a multivariable energy efficiency optimization control system for a heating furnace: This technical solution constructs a dynamic digital twin model coupled with multiple physics fields and collects multimodal monitoring data in real time, including flue gas velocity and oxygen content, furnace surface temperature, fuel and air flow, seal wear, and waste heat recovery efficiency. Adaptive data fusion and dynamic calibration algorithms are used for real-time simulation prediction and closed-loop self-calibration optimization control of the heating furnace's multiple variables. This enables simultaneous optimization of overall thermal efficiency and local temperature distribution under complex operating conditions of the pyrolysis heating furnace, eliminating strong coupling disturbances caused by fuel composition fluctuations, temperature and pressure disturbances, seal leakage, and heat exchange fouling. This improves the system's robustness, accuracy, and rapid response performance, achieving a coordinated improvement in energy efficiency indicators, equipment lifespan, and operational stability. Overall energy efficiency is improved, local temperature fluctuations are reduced, and maintenance cycles are effectively extended, reducing unplanned downtime. Attached Figure Description

[0025] Figure 1 This is a block diagram of a multivariable energy efficiency optimization control system for a heating furnace;

[0026] Figure 2 A flowchart for solving technical problems in a system;

[0027] Figure 3 Comparison of flow velocity simulation cloud maps before and after compensation for CFD simulation;

[0028] Figure 4 This is a flowchart of the signal processing for the seal wear monitoring module. Detailed Implementation

[0029] The technical solutions in 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.

[0030] Example 1

[0031] This invention proposes a multivariable energy efficiency optimization control system for a heating furnace, comprising a parameter detection module. This module includes a flue gas parameter detection module, a furnace body temperature detection module, a fuel flow detection module, a combustion air flow detection module, a seal wear monitoring module, and a waste heat recovery monitoring module. These modules are used to collect real-time data on flue gas velocity and oxygen content, surface temperature of each zone of the furnace body, fuel injection rate, combustion air volume, furnace head and tail seal wear, and heat exchanger heat transfer efficiency. The parameter detection module is sequentially connected to a multivariable coupled modeling and simulation module, an optimization control module, and an execution and feedback module. The multivariable coupled modeling and simulation module constructs a system based on the acquired parameters, covering material circulation and retention, furnace body thermal inertia, etc. A dynamic digital twin model of multi-physics coupling, considering two-phase flow interaction, combustion characteristics, sealing, and heat exchange efficiency, is used. This model, combined with historical and real-time operating data, performs multi-condition simulations and predictions. The optimization control module solves for objective functions based on the simulation results from the multivariate coupled modeling and simulation module, along with set overall thermal efficiency, local temperature gradient, and heat loss indices. This generates fuel and combustion air zoning strategies, waste heat recovery reversing valve opening adjustment strategies, and dynamic precise compensation control strategies for hot and cold spots. The execution and feedback module drives the burner fuel distribution device, combustion air volume adjustment device, and waste heat recovery reversing valve according to the optimization control module's decisions, updating the digital twin model in real time. Figure 1 In this process, raw materials are conveyed by a feeder, and after the burner is ignited, high-temperature flue gas enters the pyrolysis heating furnace and recovers heat energy through a waste heat recovery heat exchanger. The system monitors flue gas velocity and oxygen content through composite sensors, monitors zone temperature using infrared and FBG sensors, obtains flow rate through dual fuel sensors, corrects combustion air flow rate using vortex shedding, static pressure, and CFD curve algorithms, and uses a multi-mode fusion algorithm to detect seal wear and leakage. Thermocouple, heat flow meter, and acoustic imaging data are collected from the heat exchanger and tube bundle, and heat exchange efficiency and pigging timing are evaluated based on thermal balance and adaptive wall resistance models. The parameter detection module filters and corrects the signals. The input is a multiphysics coupled digital twin module. This module generates thermal efficiency, zone temperature and heat loss predictions through multiphysics simulation and online assimilation of data. The optimized control module combines the improved NSGA-II with the MPC framework to generate Pareto front solutions in parallel and adaptively adjust the weights. When the predicted temperature exceeds the limit, the quadratic surface fitting compensation strategy is called to generate fuel, combustion air volume and reversing valve control schemes and send them to the execution and feedback module for closed-loop implementation by DCS / PLC. This achieves high precision, high efficiency and stable energy saving of the pyrolysis heating furnace under dynamic operating conditions and improves the overall energy efficiency level.

[0032] like Figure 2 As shown, to solve the technical problem raised in the background section of this invention, the process for solving the technical problem using the system proposed in this invention is as follows:

[0033] Step 1, Real-time parameter detection: The flue gas parameter detection module, furnace body temperature detection module, fuel flow detection module, combustion air flow detection module, furnace head and tail seal wear monitoring module, and waste heat recovery efficiency monitoring module collect online data on exhaust gas velocity and oxygen content, surface temperature of each zone of the furnace body, gas injection volume, combustion air flow, wear degree of seals, and heat transfer efficiency of the waste heat recovery heat exchanger.

[0034] Step 2, Multivariable Coupled Modeling and Simulation: Based on the real-time and historical data collected in Step 1, the multivariable coupled modeling and simulation module constructs a multi-physics digital twin simulation module, which includes multiple material circulation and retention, differences in thermal inertia of furnace sections, gas-solid two-phase flow coupling in the rotating furnace, combustion dynamics, nonlinearity of sealing leakage, and fluctuations in heat exchange efficiency. Dynamic simulation is performed under different fuel ratios, combustion air volume, and load conditions to predict the overall thermal efficiency, local temperature difference, and heat loss distribution.

[0035] Step 3, Optimal Control Decision Generation: Based on simulation results and preset objectives (maximizing overall thermal efficiency, homogenizing local temperature gradient, and minimizing heat loss), the optimization control module constructs a multi-objective optimization problem to solve the optimal strategies for fuel and combustion air distribution ratio in each combustion zone, preheating recovery reversing valve opening, and hot / cold point compensation control. The model predictive control (MPC) framework is adopted, combined with an improved non-dominated sorting genetic algorithm (NSGA-II) to generate the Pareto optimal solution set in parallel. Online sensitivity analysis is used to dynamically adjust the weights of each objective, and a quadratic surface fitting local sub-algorithm is used for rapid compensation under sudden operating conditions.

[0036] Step 4, Strategy Execution and Closed-Loop Feedback: The execution and feedback module drives the burner group fuel distribution device, and issues control commands through proportional / servo valves, variable frequency fans and PLC-RTU or DCS systems to achieve fine adjustment of fuel injection quantity in each zone, control combustion fans or bypass valves to adjust combustion air flow in each zone, operate preheat recovery reversing valves to achieve dynamic switching between waste heat recovery and direct exhaust of exhaust gas, implement local temperature compensation control in hot / cold areas, and feed back execution results and real-time response data to the digital twin model to complete adaptive iterative updates.

[0037] To clearly illustrate the implementation of the technical solution of this invention, the following test cases are used for explanation.

[0038] First, the pyrolysis furnace was preheated and run continuously for 4 hours until the temperature fluctuation in each zone was within ±2℃ before testing began. The ambient temperature was 25℃, the relative humidity was 45%, and three-phase AC power of 380V and 50Hz was used. Historical operating data from the past two weeks (including load, fuel / air volume, temperature, sealing status, and heat exchange efficiency) was loaded into the system proposed in this invention and processed through a complete cycle. In the tests, test group 1 used natural gas as fuel with a standard calorific value of 38MJ / m³, test group 2 used petroleum coke as fuel with a calorific value of 29MJ / m³, a sulfur content of 0.8%, and a volatile matter content of 12%. Test group 3 used a 60:40 co-firing mode of natural gas and petroleum coke. The test results are shown in Table 1.

[0039] Table 1 Comparison of Test Indicators

[0040]

[0041] The above data shows that under 50%, 75%, and 100% load and fuel conditions, the use of a multi-physics coupled digital twin model combined with MPC+NSGA-II multi-objective optimization and hot / cold point dynamic compensation control improves the overall thermal efficiency by 68 percentage points compared to before optimization, reduces the maximum local temperature gradient by 15-30℃, reduces heat loss per unit area by 4-10kW / m², narrows the model simulation error to ±2.5%, shortens the response time to 3 seconds, and improves the anti-disturbance capability by 40%. This demonstrates that the technical solution of this invention achieves synergistic effects in three major innovations: accurate prediction, optimal decision-making, and rapid compensation. This improves the overall thermal efficiency, balances the temperature distribution, reduces heat loss, lowers the energy consumption of the pyrolysis heating furnace, enhances equipment stability and extends equipment life, and strengthens the equipment's anti-disturbance capability.

[0042] Example 2

[0043] Unlike Example 1, this example provides a detailed description of the flue gas parameter detection module and the furnace body temperature detection module.

[0044] The flue gas parameter detection module uses multi-point ultrasonic anemometers arranged in different sections of the pyrolysis heating furnace. It calculates the instantaneous flow velocity of each section in real time by measuring the time difference of sound wave propagation in the flue gas medium, and combines it with a multi-point static pressure difference compensation algorithm based on online sampling of dust concentration to eliminate the influence of dust disturbance. The oxygen content detection uses a Raman spectroscopy sensor set before and after the oxygen meter, supplemented by a dual-mode fusion calibration method based on electrochemical oxygen sensor, and compensates for Raman signal drift and electrochemical sensor temperature response in real time.

[0045] For example, at 9:00 AM on a certain day, ultrasonic anemometers were installed in the preheating zone (flue section A), middle section (flue section B), and tail section (flue section C) of a ceramic pyrolysis heating furnace. Each group consisted of a transmitting probe and a receiving probe. The instantaneous flow velocities v1=ΔL1 / Δt1 and v2=ΔL2 / Δt2 were calculated by measuring the time differences Δt1 and Δt2 of the sound waves along the paths A→B and B→C, respectively. Combined with online dust sampling, the dust concentration C_powder was determined to be 15 mg / m³. 3 The original wind speed was corrected by inputting the C-powder and the static pressure differences ΔP1, ΔP2, and ΔP3 at each point into a multi-point static pressure difference compensation algorithm based on Kalman filtering to eliminate dust interference. At the same time, a Raman spectral sensor was installed before and after the oxygen meter to collect the original spectra R1 and R2 respectively. The two Raman signals and the output voltage E of the electrochemical oxygen sensor were fused into a dual-mode calibration model constructed by multiple linear regression and temperature response compensation algorithm. The measurement errors caused by Raman signal drift and electrochemical sensor temperature drift were corrected in real time, and the instantaneous oxygen content O2=20.8% was finally output. All collected and corrected data were transmitted to the host computer via industrial Ethernet at a frequency of 1 second / time for edge computing and stored in a time-series database for real-time visualization by the system. This realized the online detection process of flue gas parameters across the entire chain of ultrasonic-static pressure difference coupling noise reduction and Raman-electrochemical dual-mode fusion calibration.

[0046] The furnace body temperature detection module uses a combination of high-temperature short-wave infrared thermal imagers arranged in each heating zone on the outer surface of the pyrolysis heating furnace body and fiber Bragg grating array temperature sensors embedded in the refractory lining for measurement. The infrared and FBG sensor data acquisition is triggered synchronously by a rotary position encoder, and the infrared radiation drift and FBG temperature drift are corrected online based on the established furnace body thermal inertia compensation model and multi-point data fusion algorithm. The surface temperature of each zone is monitored in real time at a frequency of ≥20Hz.

[0047] For example, at 10:00 AM on a certain day, the following measurement procedure was implemented in three heating zones (Zone A, Zone B, and Zone C) of a ceramic pyrolysis furnace:

[0048] (1) Sensor placement and synchronous triggering:

[0049] A high-temperature short-wave infrared thermal imager (model IR-SW1024, resolution 1024×768, temperature measurement range 300~1200℃) was installed in each of the three outer surface areas of the furnace (A, B, and C). A 16-point fiber Bragg grating (FBG) temperature sensor array (center wavelength 1550nm, sensitivity 10pm / ℃, spacing 50mm) was also embedded under the refractory lining of each area.

[0050] A rotary position encoder (10,000 pulses / revolution) is installed on the furnace body support shaft. It generates a trigger signal every 0.036° of rotation, which simultaneously triggers the thermal imager to acquire one frame of infrared image and the FBG array to read all 16 wavelength data once, with a sampling frequency of 50Hz.

[0051] (2) Online correction and compensation:

[0052] Infrared radiation drift correction: Infrared calibration points are collected every 10 minutes by a 500℃ standard blackbody reference target fixedly mounted on the furnace body to correct the radiation drift of the thermal imager caused by changes in ambient temperature.

[0053] FBG temperature drift correction: Set up a set of room temperature reference FBG (constant temperature 25±0.1℃) outside the furnace room, monitor its wavelength drift in real time, and use the obtained drift amount as a baseline to eliminate the temperature drift error of each array reading;

[0054] Thermal inertia compensation model: A first-order RC thermal inertia model is established for each heating zone, wherein the time constant in the model is set to 120s, 150s and 180s respectively based on the refractory lining and thermal conductivity of zone A, zone B and zone C.

[0055] (3) Multi-point data fusion algorithm:

[0056] For the nth frame of data, calculate the corrected infrared temperature;

[0057] The fusion weights are adjusted based on the average temperature of the FBG and their respective historical residuals, and the fusion temperature is output.

[0058] (4) Real-time continuous monitoring:

[0059] After the above correction and fusion, the system outputs the real-time surface temperature curves of areas A / B / C at a frequency of ≥20Hz, and sends the results to the DCS via industrial Ethernet for real-time viewing by the optimization control module and maintenance personnel.

[0060] The furnace body temperature detection module coordinates short-wave infrared thermal imaging with an FBG array sensor under the refractory lining to overcome the limitations of a single sensing method. It uses a rotary position encoder to synchronously trigger infrared images and FBG readings in real time, eliminating the timing differences caused by rotation. It uses a standard blackbody inside the furnace and a room temperature reference FBG to achieve online compensation for infrared radiation drift and FBG temperature drift without the need for furnace shutdown and recalibration. It performs time-domain correction on the measurement results based on a segment thermal inertia model to overcome the temperature hysteresis effect of high-temperature, large-volume furnace bodies. It adopts a weighted adaptive infrared-FBG data fusion algorithm to improve the temperature measurement accuracy to ±1℃ and maintain high-frequency monitoring of ≥20Hz.

[0061] Example 3

[0062] The difference between this embodiment and embodiments 1 and 2 is that this embodiment is a detailed introduction and explanation of the fuel flow detection module.

[0063] The fuel flow detection module adopts a dual-mode measurement architecture with a mass flow sensor and an ultrasonic time-difference volumetric flow sensor arranged in series in the fuel line at the front end of each burner. A microwave medium analyzer is set between the two sensors to collect fuel density and viscosity data in real time. The collected mass flow and volumetric flow signals are fused based on an improved gain adaptive Kalman filter algorithm and an online rheological model to compensate for measurement errors caused by changes in volatile matter and temperature and pressure fluctuations.

[0064] The fuel flow detection module synchronously inputs the collected Coriolis mass flow rate signal and ultrasonic volumetric flow rate signal, along with the fuel density and viscosity parameters acquired in real time by the microwave medium analyzer, into a dual-channel extended Kalman filter. The process noise covariance of the first channel is dynamically updated based on an online regression reconstruction model of changes in fuel volatile content and historical operating deviations. The measurement noise covariance of the second channel is adjusted in real time through temperature-pressure hysteresis characteristic identification and an adaptive recursive least squares algorithm. After filtering and fusion, a second correction based on residual orthogonal projection is performed to accurately compensate for the flow measurement error caused by changes in fuel volatile content and temperature and pressure fluctuations.

[0065] Specifically, the fuel flow detection module uses a Coriolis mass flow sensor and an ultrasonic time-of-flight volumetric flow sensor connected in series in the fuel line at the front end of each burner, and adds a microwave medium analyzer between them to acquire fuel density and viscosity data in real time. The collected mass flow, volumetric flow, and medium parameters are synchronously input into a dual-channel extended Kalman filter. The process noise covariance of the first channel is dynamically updated based on an online regression reconstruction model of changes in fuel volatile content and historical operating deviations. The measurement noise covariance of the second channel is adjusted in real time through temperature-pressure hysteresis characteristic identification and adaptive recursive least squares algorithm. After filtering and fusion, the results are corrected twice by residual orthogonal projection. With an improved gain adaptive Kalman filter algorithm and an online rheological model, real-time dynamic calibration and error compensation of fuel composition, viscoelasticity, and operating condition fluctuations are achieved.

[0066] Example 4

[0067] Unlike Examples 1, 2, and 3, this example provides a detailed description of the combustion air flow detection module and the seal wear monitoring module.

[0068] The combustion air flow detection module is equipped with vortex flow meters and multi-point static pressure sensors at the inlet of the combustion air duct and in the branch pipes of each combustion zone. Temperature and humidity sensors are set up upstream and downstream of the flow meters to collect environmental parameters. Based on the collected static pressure difference, vortex pulse frequency and temperature and humidity data, a nonlinear adaptive observer algorithm combined with CFD simulation pre-calibration curve is used to estimate the velocity profile distortion coefficient in real time and compensate for the vortex measurement value.

[0069] The combustion air flow detection module is equipped with vortex flow meters and multi-point static pressure sensors at the inlet of the combustion air duct and in the branch pipes of each combustion zone. Temperature and humidity sensors are set up upstream and downstream of the vortex flow meters to collect environmental parameters. Based on the non-uniform velocity profile curve pre-calibrated by CFD simulation, combined with static pressure difference, vortex pulse frequency and temperature and humidity data, a nonlinear adaptive observer algorithm is used to estimate the velocity profile distortion coefficient online and compensate for the vortex measurement value. The correction result is updated in real time at a frequency of 1Hz to eliminate the measurement error caused by secondary flow, geometric distortion of pipe branches and fluctuations in ambient temperature and humidity. This solves the problems of low accuracy, slow response and poor stability of traditional single vortex method in complex flow fields, improves the accuracy of combustion air flow measurement, response time and system anti-disturbance capability, provides input for combustion control optimization and improves combustion efficiency and reduces emissions.

[0070] To illustrate the improved measurement effect of the combustion air flow detection module at the combustion air duct inlet, a fluid dynamics CFD simulation was performed on typical geometric sections of the combustion air duct and its branch pipes in the pyrolysis heating furnace, including the main duct and several branches. Assuming air is incompressible and a Newtonian fluid, the steady-state Reynolds-averaged Navier-Stokes (RANS) equations were used, combined with the standard k–ε turbulence model to describe the turbulent flow. In ANSYS Fluent, a structured / unstructured mesh of approximately 500,000 to 1,000,000 elements was generated for the cross-sectional area, ensuring mesh refinement at the shear layer and critical branch points to guarantee convergence. The nominal average inlet velocity was set to 15 m / s, the outlet static pressure to 1 atm, and the pipe wall to be non-slip. Roughness was set according to the actual pipe material, and an isothermal assumption was used. A steady-state solver and the SIMPLE algorithm with pressure-velocity coupling were employed, resulting in residual convergence to 1 × 10⁻⁶. -6 The output velocity distribution contour map is obtained by directly simulating the original inlet profile before compensation. After compensation, the velocity profile distortion coefficient calibrated by the nonlinear adaptive observer algorithm is applied to the inlet profile, and CFD calculations are performed again with the same settings. Figure 3 As shown, by comparing the velocity contour maps obtained under the same grid, boundary conditions and solution parameters, it can be clearly seen that the velocity profile recovers from distortion (high at the center, low at the edge, asymmetric) to a more symmetrical and uniform distribution. This is an intuitive verification of the real-time correction of vortex shedding measurement error, the improvement of measurement accuracy and the consistency of the flow field in this technical solution.

[0071] like Figure 4 As shown, the sealing wear monitoring module includes: an array of fiber Bragg grating strain sensors circumferentially distributed along the outer edge of the sealing rings at the furnace head and tail of the pyrolysis heating furnace, used to collect minute radial deformations of the sealing rings in real time; a high-frequency ultrasonic transducer pre-embedded at the contact surface of the sealing rings, used to detect ultrasonic signals generated by wear and gas leakage; and an industrial infrared thermal imager installed on the outside of the sealing area, used to capture local temperature field abrupt changes caused by leaking flue gas; the three sensor signals are processed online by a multi-modal deep learning algorithm that fuses multi-scale wavelet packet decomposition and bidirectional long short-term memory network to detect the sealing wear and leakage velocity in real time, and the detection results are fed back to the optimization control decision module for dynamic compensation in real time.

[0072] By fusing signals from three sensors and performing multi-scale wavelet packet decomposition and bidirectional LSTM online processing, the accuracy of sealing wear measurement, leakage velocity estimation error, and response speed are improved. This dynamically compensates for heat loss caused by sealing leakage, thereby enhancing the operating stability and lifespan of the heating furnace.

[0073] Example 5

[0074] Unlike Examples 1, 2, 3 and 4, this example is a specific description of the waste heat recovery monitoring module.

[0075] The waste heat recovery monitoring module employs a multi-parameter coupled online detection method, specifically including: arranging K-type thermocouple arrays at the inlet and outlet flues of the heat exchanger for real-time measurement of inlet and outlet flue gas temperatures; deploying multi-point thin-film heat flux meters within the heat exchanger tube bundle for real-time acquisition of heat flux per unit area of ​​the tube wall; and installing an acoustic imaging sensor array on the outside of the tube bundle for online identification of scale location and thickness distribution. The waste heat recovery monitoring module dynamically calculates the heat transfer coefficient based on a two-way heat balance equation and an adaptive wall thermal resistance model, and combines this with a wall thermal resistance change rate threshold-triggered cleaning cycle prediction algorithm to locate and quantitatively assess the decrease in heat transfer efficiency caused by initial scaling and non-uniform distribution.

[0076] Specifically, the methods for multi-parameter online monitoring by the waste heat recovery monitoring module include:

[0077] Step 1, Sensor Arrangement and Calibration: Arrange K-type thermocouple arrays at equal intervals in the inlet and outlet flues of the heat exchanger, with at least 8 points on each side, and the calibration accuracy is ±0.5℃. Install thin-film heat flow meters uniformly on the outer wall of key pipelines inside the tube bundle. Set an acoustic imaging sensor array around the outside of the tube bundle to identify the thickness of the internal structure in real time through ultrasonic reflection intensity profile.

[0078] Step 2, Online data acquisition: After sampling, the signals from thermocouples, heat flow meters and acoustic arrays are processed for edge computing, and signal preprocessing is used to remove noise and environmental interference;

[0079] Step 3, Two-way thermal balance calculation: Perform two-way thermal balance calculation for each tube bundle unit according to the two-way thermal balance equation;

[0080] Step 4, Adaptive wall thermal resistance model: Establish a first-order adaptive model of the wall thermal resistance of the tube bundle, wherein the learning rate is automatically adjusted according to the historical error and the wall resistance increase caused by fouling is tracked in real time.

[0081] Step 5, scale location and thickness assessment: The acoustic imaging profile is mapped onto the tube bundle mesh to identify the local last scale area and thickness distribution. The structural area and thickness distribution are used as input for the wall resistance model, and the local tube bundle wall thermal resistance at the corresponding location is spatially weighted and corrected.

[0082] Step 6, Pipe cleaning cycle prediction and triggering: Monitor and visualize the real-time calculated pipe wall heat transfer coefficient, tube bundle wall resistance, local scale thickness and pipe cleaning countdown time, and output to the optimization control module for dynamic compensation measures such as adjusting preheated air distribution or increasing flushing pipeline flow.

[0083] Example 6

[0084] Unlike Examples 1, 2, 3, 4 and 5, this example describes the multivariable coupling modeling and simulation module and the optimization control module.

[0085] The multivariable coupled modeling and simulation module adopts a hierarchical dynamic adaptive modeling method, including: a multiphysics simulation engine based on the coupling of finite element heat conduction, CFD flow, discrete element material circulation, and combustion chemical reaction; the simulation engine integrates a real-time online data assimilation algorithm, injects monitoring data into the simulation model at a frequency of seconds, and performs self-calibration on thermal inertia, material residence time, and two-phase flow boundary conditions through a model error update mechanism driven by feedback residuals, gradually converging the simulation error to within ±0.5%; the simulation calculation runs on a GPU cluster parallel architecture.

[0086] The optimization control module includes a multi-objective optimization engine based on model predictive control and an improved non-dominated sorting genetic algorithm, used to jointly solve the three objectives of maximizing overall thermal efficiency, homogenizing local temperature gradients, and minimizing heat loss. The engine automatically adjusts the weights of each objective through online sensitivity analysis and generates Pareto front solution sets in parallel within the feasible domain of each control variable. When a local hot spot or cold spot deviates from the threshold, the optimization control module automatically calls a local compensation sub-algorithm based on quadratic surface fitting to update the control strategy and sends the final optimal control command to the execution and feedback module in real time.

[0087] The multivariable coupled modeling and simulation module first runs finite element heat conduction, CFD flow, and DEM material circulation and chemical reaction sub-models in parallel within a unified framework. Intermediate quantities such as temperature, flow rate, particle distribution, and reaction rate are exchanged via the MPI interface. Every 1 second, the system automatically collects measured data such as flue gas temperature, heat flow, and material concentration from the parameter detection module. A feedback residual-driven correction process is used—comparing the simulation output with the monitored values ​​and iteratively updating the furnace thermal inertia coefficient, material residence time, and gas-solid boundary conditions online—until the simulation results highly match the actual operating conditions. All calculations are deployed on a GPU cluster, ensuring that multi-condition simulations are completed within a limited time window, achieving a dynamic balance between simulation depth and online speed, and maintaining high reliability of the model over long periods without manual intervention. The optimized control module organically integrates Model Predictive Control (MPC) with an improved Non-Dominated Sort Genetic Algorithm (NSGA-II): In each control cycle, the module first calls the multi-condition results predicted by the digital twin to construct objective functions for three indicators: thermal efficiency, local temperature difference, and heat loss, and automatically assigns weights through online sensitivity analysis. Then, it searches the Pareto front solution set in parallel within the feasible region of the given fuel flow rate, combustion air volume, and directional valve opening. When the predicted temperature in any segment exceeds a preset threshold, the system immediately activates a local compensation sub-algorithm based on quadratic surface fitting to refine the control path near the current optimal solution, forming a new fuel allocation and airflow regulation scheme, which is then sent to the execution module via the OPC UA / DCS interface. This process is completed in a closed loop within the same control time slot, ensuring real-time integration between simulation decisions and on-site execution, achieving precise energy efficiency optimization under high-frequency, multi-variable coupled environments.

[0088] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0089] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multivariable energy efficiency optimization control system for a heating furnace, characterized in that, The parameter detection module includes a flue gas parameter detection module, a furnace body temperature detection module, a fuel flow detection module, a combustion air flow detection module, a sealing wear monitoring module, and a waste heat recovery monitoring module, which are respectively used to collect the exhaust gas flow rate and oxygen content, the surface temperature of each partition of the furnace body, the fuel injection amount, the combustion air amount, the sealing wear condition of the furnace head and tail, and the heat transfer efficiency of the heat exchanger in real time. The parameter detection module is sequentially connected with a multivariable coupling modeling and simulation module, an optimization control module, and an execution and feedback module. The multivariable coupling modeling and simulation module constructs a multi-physical field coupling dynamic digital twin model based on the obtained parameters, which covers material circulation retention, furnace thermal inertia, two-phase flow interaction, combustion characteristics, sealing, and heat transfer efficiency. The model is combined with historical and real-time operation data to perform multi-working condition simulation prediction. The optimization control module combines the simulation prediction results and the set overall thermal efficiency, local temperature gradient, and heat loss index to solve the objective function, generate fuel and combustion air partition distribution strategies and waste heat recovery reversing valve opening regulation strategies, and generate hot spot and cold spot dynamic compensation control strategies. The execution and feedback module drives the burner group fuel distribution device, combustion air regulation device, and waste heat recovery reversing valve according to the optimization control decision, and updates the digital twin model in real time. The sealing wear monitoring module includes an optical fiber Bragg grating strain sensor array arranged in a circumferential direction on the outer edge of the sealing ring of the cracking furnace head and tail, which is used to collect the small radial deformation of the sealing ring in real time. A high-frequency ultrasonic transducer is embedded at the contact surface of the sealing ring, which is used to detect ultrasonic signals generated by wear and gas leakage. An industrial infrared thermal imager is arranged outside the sealing part, which is used to capture the local temperature field mutation signal caused by leakage flue gas. Three-way sensor signals are processed online by a multi-modal deep learning algorithm based on multi-scale wavelet packet decomposition and bidirectional long short-term memory network fusion, which can detect the sealing wear and leakage flow rate in real time, and feed back the detection results to the optimization control decision module for dynamic compensation. The optimization control module includes a multi-objective optimization engine based on model predictive control and improved non-dominated sorting genetic algorithm, which is used to jointly solve the three objectives of maximizing overall thermal efficiency, uniformizing local temperature gradient, and minimizing heat loss. The engine automatically adjusts the weight of each target through online sensitivity analysis, and generates a Pareto frontier solution set in parallel within the feasible region of each control variable. When the local hot spot or cold spot deviates from the threshold, the optimization control module automatically calls a local compensation sub-algorithm based on quadratic surface fitting to update the control strategy, and sends the final optimal control instruction to the execution and feedback module in real time.

2. A multi-variable energy efficiency optimization control system for a furnace as recited in claim 1, wherein, The flue gas parameter detection module adopts multi-point ultrasonic wind speed sensors arranged in different partitions of the cracking furnace flue. The instantaneous flow rate of each partition is calculated in real time by measuring the time difference of sound wave propagation in the flue gas medium, and the influence of dust disturbance is eliminated by combining a multi-point static pressure difference compensation algorithm based on on-line sampling of soot concentration. The oxygen content detection adopts one Raman spectrum sensor before and after the oxygen meter, and a double-mode fusion calibration method based on an electrochemical oxygen sensor is used to compensate for the real-time calibration of Raman signal drift and electrochemical sensor temperature response.

3. The multivariable energy efficiency optimization control system for a furnace of claim 1, wherein, The furnace body temperature detection module adopts a composite measurement of high-temperature short-wave infrared thermal imager and fiber Bragg grating array temperature sensor buried in the refractory lining of each heating partition of the cracking furnace furnace body. The infrared and FBG sensor data acquisition is triggered synchronously by the rotary position encoder, and the infrared radiation drift and FBG temperature drift are corrected online based on the established furnace body thermal inertia compensation model and multi-point data fusion algorithm, so as to realize real-time continuous monitoring of the surface temperature of each partition at a frequency of ≥20Hz.

4. The multi-variable energy efficiency optimization control system for a furnace of claim 1, wherein, The fuel flow detection module adopts a double-mode measurement architecture of mass flow sensor and ultrasonic time difference volume flow sensor arranged in series in the fuel pipeline in front of each burner, and a microwave medium analyzer is arranged between the two sensors to collect fuel density and viscosity data in real time. The collected mass flow and volume flow signals are fused based on an improved gain adaptive Kalman filter algorithm and an online rheological model dynamic calibration to compensate for the measurement errors caused by the change of volatile matter and the fluctuation of temperature and pressure.

5. A multi-variable energy efficiency optimization control system for a furnace as recited in claim 4, wherein, The fuel flow detection module synchronously inputs the collected Coriolis mass flow signal and ultrasonic volume flow signal and the real-time fuel density and viscosity parameters obtained by the microwave medium analyzer into a double-channel extended Kalman filter. The process noise covariance of the first channel is dynamically updated based on the online regression reconstruction model of fuel volatile content change and historical running deviation, and the measurement noise covariance of the second channel is adjusted in real time by the temperature-pressure lag characteristic identification and adaptive recursive least squares algorithm, and then the fused signal is corrected based on the residual orthogonal projection quadratic correction to accurately compensate for the flow measurement error caused by the change of fuel volatile matter and the fluctuation of temperature and pressure.

6. The multi-variable energy efficiency optimization control system for a furnace of claim 1, wherein, The combustion air flow detection module arranges vortex flow meters and multi-point static pressure sensors at the inlet of the combustion air duct and each combustion zone branch pipeline, respectively, and sets temperature and humidity sensors upstream and downstream of the flow meters to collect environmental parameters. Based on the collected static pressure difference, vortex pulse frequency and temperature and humidity data, a nonlinear adaptive observer algorithm combined with CFD simulation pre-calibration curve is used to estimate the flow rate profile distortion coefficient in real time and compensate for the vortex measurement value.

7. The multi-variable energy efficiency optimization control system for a furnace of claim 1, wherein, The waste heat recovery monitoring module adopts a multi-parameter coupling online detection method, specifically including: arranging a K-type thermocouple array at the inlet and outlet flues of the heat exchanger for real-time measurement of the inlet and outlet flue gas temperatures; arranging a multi-point thin-film heat flow meter inside the heat exchanger tube bundle for real-time collection of the heat flux per unit area of the tube wall; installing an acoustic imaging sensor array outside the tube bundle for online identification of the fouling location and thickness distribution; the waste heat recovery monitoring module dynamically calculates the heat transfer coefficient based on the two-way heat balance equation and the adaptive wall heat resistance model, and combines the wall heat resistance change rate threshold to trigger the pigging cycle prediction algorithm, to position and quantitatively evaluate the heat transfer efficiency decline caused by the initial stage of fouling and non-uniform distribution.

8. The multivariable energy efficiency optimization control system for a furnace of claim 1, wherein, The multivariate coupling modeling and simulation module adopts a hierarchical dynamic adaptive modeling method, including: a multi-physical field simulation engine based on the coupling of finite element heat conduction, CFD flow, discrete element material circulation and combustion chemical reaction; the simulation engine integrates a real-time online data assimilation algorithm, injects the monitoring data into the simulation model at a frequency of seconds, and through a model error updating mechanism based on feedback residual driving, self-calibrates the thermal inertia, material residence time and two-phase flow boundary conditions, and gradually converges the simulation error to within ±0.5%; the simulation calculation runs on a GPU cluster parallel architecture.

Images