Lime kiln combustion control method

Abstract

The invention relates to the technical field of lime kiln combustion control, and discloses a lime kiln combustion control method which comprises the following steps: acquiring lime kiln temperature data, and constructing a lime kiln temperature field; analyzing temperature deviation based on the temperature field, and judging whether a temperature abnormal region exists; when no temperature abnormal area exists, working condition parameters and environment parameters are collected, and the air-fuel ratio is calculated according to the working condition parameters; the environment parameters and the air-fuel ratio are input into a fuzzy control strategy, and adjustment control parameters are obtained through operation; smoke components generated in the combustion process are collected, heat efficiency is calculated according to component analysis, and whether the heat efficiency reaches the standard or not is judged; when the air-fuel ratio does not reach the standard, parameter self-learning is carried out on the fuzzy control strategy, the fuzzy control strategy is updated again, the air-fuel ratio is calculated again, and the control parameters are adjusted; combustion parameters can be adjusted in real time, heat efficiency is improved, uniformity of a temperature field in the kiln is improved, combustion control precision and energy utilization efficiency are improved, and the phenomenon of underburning or overburning is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of lime kiln combustion control, and in particular to a lime kiln combustion control method. Background Art

[0002] In the lime production process, lime kiln control is crucial for product quality, production efficiency, and energy consumption. Traditional lime kiln control methods often rely on manual experience to adjust fuel and air supply, resulting in uneven temperature fields, low control accuracy, and energy waste. For example, manual control makes it difficult to accurately adjust fuel supply and ventilation volume, resulting in premature or overburning of lime, affecting product quality. Furthermore, the inability to monitor and optimize the kiln temperature field in real time leads to low energy efficiency. Although existing technologies have introduced PID control or single-sensor monitoring, dynamic optimization of the temperature field and real-time tracking of thermal efficiency are still impossible in the face of complex operating conditions such as fuel calorific value fluctuations (such as gas calorific value fluctuations of ±10%) and kiln nodules. Furthermore, traditional thermal efficiency calculations rely on offline sampling and analysis, with a lag time of several hours, making it difficult to support timely adjustment of combustion parameters, leading to energy waste and excessive pollutant emissions (such as CO concentrations >1000ppm). Summary of the Invention

[0003] In view of the shortcomings of the prior art, the present invention aims to provide a lime kiln combustion control method.

[0004] In order to achieve the above object, the present invention provides the following technical solutions:

[0005] A lime kiln combustion control method comprising:

[0006] Collect lime kiln temperature data and construct lime kiln temperature field;

[0007] Analyze temperature deviation based on temperature field to determine whether there is an abnormal temperature area;

[0008] When there is no abnormal temperature area, collect working parameters and environmental parameters.

[0009] Calculate the air-fuel ratio based on operating parameters;

[0010] Input the environmental parameters and air-fuel ratio into the fuzzy control strategy and obtain the adjustment control parameters through calculation;

[0011] Collect the flue gas components generated during the combustion process, calculate the thermal efficiency based on the component analysis, and determine whether the thermal efficiency meets the standards;

[0012] When the standards are not met, the fuzzy control strategy performs parameter self-learning, re-updates the fuzzy control strategy, and recalculates the air-fuel ratio and adjusts the control parameters.

[0013] In the present invention, preferably, the construction of the lime kiln temperature field specifically includes:

[0014] Construct a multi-dimensional temperature monitoring network and collect a number of temperature data through the multi-dimensional temperature monitoring network;

[0015] Performing heterogeneous synchronous data fusion on the temperature data and calculating key temperature indicators;

[0016] Establish a three-dimensional heat conduction model of lime kiln;

[0017] Fusing the three-dimensional heat conduction model with a deep learning network to construct a three-dimensional temperature prediction model, and fusing the temperature predicted by the three-dimensional temperature prediction model into the three-dimensional heat conduction model;

[0018] Visualize a 3D heat conduction model.

[0019] In the present invention, preferably, the constructing of the three-dimensional temperature prediction model specifically includes:

[0020] A central temperature acquisition device is added to the multi-dimensional temperature monitoring network. The central temperature acquisition device is provided with several acquisition points. The central temperature acquisition device is placed in the center of the lime kiln and is fine-tuned according to the actual fuel position in the lime kiln.

[0021] The central temperature acquisition device collects temperature change data at the center of the lime kiln in real time, and the multi-dimensional temperature monitoring network collects temperature change data around the lime kiln. The temperature change data at the center and around the kiln collected at the same time are regarded as a set of temperature data. Several sets of temperature data from several combustion cycles are collected as training sets and test sets.

[0022] The three-dimensional temperature prediction model is trained using a training set, and the trained three-dimensional temperature prediction model is tested using a test set. After passing the test, the three-dimensional temperature prediction model is obtained.

[0023] In the present invention, preferably, the three-dimensional temperature prediction model input also includes the fuel valve and air valve openings. When the fuel valve and air valve openings change, the three-dimensional temperature prediction model predicts subsequent temperature change data based on the current fuel type and temperature data.

[0024] In the present invention, preferably, the determining whether there is a temperature abnormality area specifically includes:

[0025] Calculate the deviation between the measured temperature data and the corresponding grid temperature predicted by the three-dimensional temperature prediction model;

[0026] Determine whether the deviation value is greater than a set value, where the set value is the sum of the mean of the historical data of the grid and the standard deviation of the historical data;

[0027] When it is greater than the set value and lasts for more than 10 minutes, the corresponding area of ​​the grid is a temperature abnormality area.

[0028] In the present invention, preferably, when there is a temperature abnormality area, the linkage control strategy obtains the temperature value of the temperature abnormality area, and outputs adjustment control data corresponding to the ultra-high temperature or ultra-low temperature condition to control multiple valves in the abnormal area.

[0029] In the present invention, preferably, the calculating of the air-fuel ratio according to the operating condition parameters includes:

[0030] Identify fuel type based on operating parameters and correct calorific value and material level based on operating parameters;

[0031] Determine the theoretical air-fuel ratio based on the fuel type;

[0032] Calculate the excess air coefficient based on the theoretical air-fuel ratio;

[0033] The corrected calorific value, material level value and excess air coefficient are used to adjust the theoretical air-fuel ratio to obtain the final air-fuel ratio.

[0034] In the present invention, preferably, in the fuzzy control strategy, the adjustment control parameters obtained by calculation include:

[0035] Preprocess the input environmental parameters and air-fuel ratio data;

[0036] Activate multiple rules and calculate the weights corresponding to each rule;

[0037] The output of each rule is summarized according to the corresponding weight of each rule to obtain the final adjustment control parameters.

[0038] In the present invention, preferably, the control parameters include a fuel valve opening adjustment amount and an air valve linkage adjustment amount, the fuel valve opening adjustment amount is calculated based on the environmental parameters, and the air valve linkage adjustment amount is calculated based on the air-fuel ratio.

[0039] In the present invention, preferably, the determining whether the thermal efficiency meets the standard specifically includes:

[0040] The flue gas and flue gas components are collected through the high temperature sampling probe.

[0041] Perform sliding average filtering on the collected smoke components and remove outliers;

[0042] Convert the measured smoke components into dry basis concentrations;

[0043] Thermal efficiency was calculated using a counter-balance strategy based on dry basis concentration;

[0044] Determine whether the thermal efficiency is greater than the set value. If it is, the thermal efficiency meets the standard.

[0045] Otherwise, the fuzzy control strategy is self-learned and the fuzzy control strategy is updated.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The present invention optimizes the air-fuel ratio by constructing a three-dimensional temperature field model, dynamically analyzing temperature anomaly areas, and combining fuzzy control strategies. It can adjust combustion parameters in real time and improve thermal efficiency, thereby improving the uniformity of the temperature field in the kiln, improving combustion control accuracy and energy utilization efficiency, and effectively reducing the phenomenon of raw or over-burning. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 The figure is a flow chart of a lime kiln combustion control method according to the present invention.

[0049] Figure 2 This is a schematic diagram of the process of constructing the temperature field of a lime kiln according to the present invention.

[0050] Figure 3 Schematic diagram of the flow chart for calculating the air-fuel ratio according to the present invention.

[0051] Figure 4 This is a schematic diagram of the process of determining whether the thermal efficiency meets the standard according to the present invention. DETAILED DESCRIPTION

[0052] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0053] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains. The terms used in this specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0054] In the existing technology, lime kiln control has long relied on manual experience to adjust the fuel and air supply, resulting in problems such as uneven temperature field distribution, control lag, and low energy utilization. The traditional method uses a single sensor to monitor local temperature, which is unable to obtain the overall thermal field distribution in the kiln, resulting in a mismatch between fuel supply and combustion demand. When the calorific value of the fuel fluctuates or nodules appear in the kiln, manual experience makes it difficult to adjust the air-fuel ratio in a timely manner, resulting in raw or overburning. Existing automatic control methods mostly use fixed parameter PID control, which has insufficient adjustment accuracy when facing complex working conditions, and the thermal efficiency calculation relies on offline sampling. Feedback delays lead to excessive pollutant emissions.

[0055] To address these issues, the inventors discovered that global temperature field monitoring is essential for precise control, but existing technologies lack the means to fuse multidimensional temperature data. By constructing a three-dimensional temperature prediction model, the dynamic changes in the kiln's thermal field can be reflected in real time. Fuzzy control strategies can handle nonlinear variables such as fluctuations in fuel calorific value, but they require a closed-loop system that combines dynamic calculations of the air-fuel ratio with real-time feedback on thermal efficiency. A parameter self-learning mechanism is key to overcoming the lag inherent in traditional control. Online analysis of flue gas composition triggers strategy updates, enabling the system to continuously optimize.

[0056] Therefore, the present application proposes a method comprising the following steps: collecting lime kiln temperature data, constructing a lime kiln temperature field, analyzing temperature deviations based on the temperature field, and determining whether there is an abnormal temperature area; when there is no abnormal temperature area, collecting operating parameters and environmental parameters, and calculating the air-fuel ratio based on the operating parameters; inputting the environmental parameters and the air-fuel ratio into the fuzzy control strategy, and obtaining the adjustment control parameters through calculation; collecting the flue gas components generated during the combustion process, calculating the thermal efficiency based on component analysis, and determining whether the thermal efficiency meets the standard; when it does not meet the standard, the fuzzy control strategy performs parameter self-learning, re-updates the fuzzy control strategy, and recalculates the air-fuel ratio and the adjustment control parameters.

[0057] Among them, constructing the lime kiln temperature field refers to obtaining spatial temperature distribution data through a multi-dimensional sensor network, which is achieved by combining a thermocouple array with an infrared scanner to provide a global data basis for temperature anomaly detection. Temperature deviation analysis refers to comparing the difference between the measured temperature and the grid temperature of the three-dimensional temperature prediction model. By setting a threshold to judge anomalies, local overheating or underburning areas can be identified. Air-fuel ratio calculation refers to dynamically determining the air-fuel ratio based on the fuel type and operating parameters, and requires combining calorific value correction with excess air coefficient adjustment. Fuzzy control strategy refers to a control algorithm that handles multivariable nonlinear relationships, mapping input parameters and output adjustment quantities through a rule base. Parameter self-learning refers to automatically adjusting the fuzzy rule weights based on thermal efficiency feedback, and a neural network algorithm can be used to optimize the control strategy.

[0058] Specifically, the temperature field construction phase establishes a three-dimensional thermal field model by fusing multi-source sensor data, and starts the main control process when no temperature anomaly is detected. The operating parameters include operating data such as fuel supply rate, material layer thickness, temperature, fuel type, etc., and the environmental parameters cover external factors such as atmospheric humidity and air pressure. The air-fuel ratio calculation module generates benchmark parameters based on the fuel calorific value and combustion requirements, and the fuzzy controller converts environmental variables and air-fuel ratio into fuel valve and air valve adjustment instructions. The flue gas composition analysis monitors the CO and O2 concentrations in real time through the smoke sensor, and calculates the instantaneous thermal efficiency through the counter-balance method. When the thermal efficiency is lower than the set threshold, the self-learning module adjusts the fuzzy rule base and regenerates the optimized control parameters.

[0059] Compared with existing technologies, traditional methods rely on single-point temperature monitoring and fixed control parameters, making them difficult to adapt to fluctuations in fuel calorific value and changes in kiln heat dissipation. This solution utilizes real-time monitoring and prediction of the three-dimensional temperature field, combined with dynamic air-fuel ratio calculation and fuzzy control, to achieve multivariable coordinated regulation of the combustion process. The introduction of online thermal efficiency feedback and a strategy self-learning mechanism transcends the static characteristics of traditional control systems, forming a closed-loop control system with adaptive capabilities.

[0060] Through the above technical solutions, this application effectively solves the problem of raw and overburning caused by uneven temperature distribution within the kiln, improving the quality and stability of lime calcination. Dynamic air-fuel ratio adjustment combined with environmental parameter compensation improves fuel utilization and reduces energy consumption. Real-time thermal efficiency monitoring and strategy self-learning ensure that the combustion control system continuously adapts to changing operating conditions, reducing pollutant emission fluctuations. Three-dimensional temperature field construction technology supports the rapid location of abnormal areas, shortening fault response time.

[0061] The present application further proposes a specific method for constructing a lime kiln temperature field, including constructing a multidimensional temperature monitoring network to collect temperature data, collecting a plurality of temperature data through the multidimensional temperature monitoring network, performing heterogeneous synchronous data fusion on the temperature data and calculating key temperature indicators, establishing a three-dimensional heat conduction model based on COMSOL Multiphysics, fusing the three-dimensional heat conduction model with a deep learning network to construct a three-dimensional temperature prediction model, fusing the temperature predicted by the three-dimensional temperature prediction model into the three-dimensional heat conduction model, and visualizing the model.

[0062] Among them, the multi-dimensional temperature monitoring network refers to the arrangement of differentiated sensor arrays in different temperature zones of the kiln body. Specifically, it can be achieved by arranging 2 groups of 4 circumferential thermocouples in the cooling zone and 3 groups of 6 circumferential thermocouples in the calcining zone in combination with infrared scanners. The temperature collection coverage rate is improved by densely arranging sensors in different areas.

[0063] Among them, heterogeneous synchronous data fusion refers to the integration of multi-source data of thermocouples and infrared images, specifically including: spatiotemporal alignment, mapping the thermocouple data in the temperature data to the infrared image coordinate system, eliminating the spatial misalignment of multi-source data to ensure data consistency; Kalman filtering to filter all temperature data; outlier removal, removing abnormal temperature data based on the 3σ principle.

[0064] Among them, the key indicator for calculating temperature is the temperature uniformity index: Where σ is the standard deviation and μ is the mean, with an ideal value of >0.9. Temperature zone proportion: The area proportion of the calcination zone with a temperature >1250°C should be <10%.

[0065] Among them, the three-dimensional heat conduction model refers to a physical simulation model built based on the actual size of the kiln body. Specifically, it can be achieved by setting the grid unit size to less than 0.5 meters using COMSOL Multiphysics software, and improving the matching degree between the model and the actual structure through refined grid division. Then, the temperature data after data fusion is mapped to the model grid nodes, and the missing data is filled using the Kriging interpolation method. The Kriging interpolation method refers to a spatial interpolation algorithm. Specifically, it can be achieved by using the variation function to analyze the spatial correlation of temperature data, and the temperature data of the missing areas is supplemented by considering the structural characteristics of the kiln body. Then, the fuel combustion heat flux density, material flow heat transfer coefficient, and kiln wall heat dissipation coefficient are used as boundary conditions. The boundary conditions are corrected in real time, that is, the model parameters are dynamically updated. Specifically, an iterative algorithm can be used to feed back real-time sensor data to the material thermal conductivity calculation module, and the model accuracy is maintained through online parameter calibration.

[0066] Specifically, thermocouple arrays and infrared scanners were placed in the cooling zone, calcination zone, and preheating zone, respectively. After collecting multi-source temperature data, spatiotemporal registration was performed to eliminate coordinate discrepancies. Kalman filtering was then used to reduce noise and eliminate outliers using the 3σ criterion. The processed data was then fed into a three-dimensional heat conduction model, where missing node data was supplemented using Kriging interpolation. After setting boundary conditions such as the fuel combustion heat flux density, real-time sensor data was fed back into the model for iterative parameter updates. The three-dimensional heat conduction model was then integrated with a 3D CNN-LSTM deep learning network. The 3D CNN extracted spatial features of the temperature field, while the LSTM captured temporal patterns. Ultimately, a three-dimensional temperature prediction result was generated that integrated physical laws and data features, and displayed in real time as a cloud map.

[0067] Compared with existing technologies, traditional methods use only a single type of sensor, resulting in temperature monitoring blind spots, and rely on static thermodynamic models that are unable to adapt to changing operating conditions. This solution achieves full kiln coverage monitoring through a multidimensional sensor network. Combining a dynamically corrected three-dimensional model with a deep learning network, this approach preserves the interpretability of the physical model while enhancing adaptability to complex operating conditions. Existing interpolation methods fail to consider the structural characteristics of the kiln. This solution employs kriging interpolation to improve the accuracy of missing data filling through spatial correlation analysis.

[0068] Through the above-mentioned technical solution, this application effectively solves the problem of incomplete three-dimensional temperature distribution monitoring within the kiln, improves the real-time and accuracy of temperature field construction, and provides reliable three-dimensional temperature data support for combustion control. A dynamic correction mechanism ensures the model's prediction accuracy under conditions such as fuel fluctuations, and a visual cloud map helps operators quickly identify areas of abnormal temperature. The multi-source data fusion method overcomes the measurement limitations of a single sensor and lays the data foundation for subsequent air-fuel ratio optimization control.

[0069] This application further proposes a specific method for constructing a three-dimensional temperature prediction model, including: adding a central temperature acquisition device on the basis of a multi-dimensional temperature monitoring network, wherein the central temperature acquisition device is provided with a number of acquisition points, that is, each acquisition point corresponds to a temperature sensor, and the central temperature acquisition device is placed at the center of the lime kiln and fine-tuned according to the actual fuel position in the lime kiln; real-time acquisition of temperature change data at the center position and the surrounding areas forms a training set and a test set; the training set is used to train a three-dimensional temperature prediction model including 3D CNN and LSTM, and the effectiveness of the model is verified through the test set.

[0070] The central temperature acquisition device is a temperature measurement device installed in the core area of ​​the kiln based on the characteristics of fuel combustion. It consists of a stainless steel bracket secured to the kiln's internal flange. Its central axis aligns with the kiln's axis. A threaded slider is built into the bracket for mounting a high-temperature sensor, and the slider's position can be fine-tuned based on actual conditions. The bracket and slider are encased in a ceramic protective layer, with a silicon carbide layer and then a thermal insulation layer inside. The sensor chip is encapsulated in a molybdenum alloy (Mo-41Ti) housing filled with magnesium oxide powder for insulation and improved high-temperature resistance. During installation, the sensor is tilted downward at a 60° angle to the horizontal, allowing gravity to naturally drain dust and prevent accumulation. A tungsten carbide-coated conical shroud is also installed at the front of the casing. The shroud has a 120° angle. During combustion, the high-speed airflow (>30 m / s) in the kiln creates vortices as it passes through the shroud, dispersing accumulated dust and preventing inaccurate temperature measurements caused by dust accumulation.

[0071] The temperature sensor uses an offline acquisition method, and the collected data is stored in the internal chip. No external wiring is required during acquisition. It is only placed in the center area of ​​the lime kiln when collecting training set and test set data. After collecting the complete combustion cycle data, it is taken out and its internal data is transmitted and then used for training and testing of the three-dimensional temperature prediction model. This can make the collected temperature data more accurate, thereby improving the accuracy of the predicted data of the trained three-dimensional temperature prediction model.

[0072] The training set and test set refer to data sets consisting of central and surrounding temperature data collected synchronously during multiple combustion cycles. This can be achieved by continuously collecting temperature data for 30 combustion cycles to enhance the model's adaptability to different working conditions.

[0073] The 3D CNN network is a three-dimensional convolutional neural network, implemented using a deep network structure consisting of convolutional and pooling layers. It is used to extract the spatial distribution characteristics of the temperature field. LSTM, a long short-term memory neural network, is added to the 3D convolutional neural network. This is achieved by adding a recurrent neural network structure to the original 3D convolutional neural network to capture the dynamic characteristics of temperature changes over time.

[0074] Specifically, the location of the central temperature acquisition device is dynamically adjusted based on the fuel type. For example, when using gas fuel, it is set at the midpoint of the kiln axis, and when using pulverized coal fuel, it is offset to the burner nozzle area. Central and surrounding temperature data are synchronously collected once per second, forming a data set containing spatial location and timestamps. During training, a 3D CNN extracts features from the three-dimensional spatial distribution of the temperature field, and an LSTM network models the temperature variation patterns in continuous time series. The feature vectors output by both are fused through a fully connected layer to ultimately predict the temperature field distribution for future time steps. During the testing phase, independent data is used to verify the model's prediction accuracy, and network parameters are re-optimized when the prediction error exceeds a threshold.

[0075] Compared with existing technologies, traditional methods rely solely on temperature monitoring data around the kiln body, failing to reflect temperature variations in the core combustion area and requiring model recalibration when switching fuel types. This solution, by adding fuel-type-adaptive central monitoring points and combining a hybrid 3D CNN and LSTM network structure, achieves comprehensive modeling of the spatiotemporal evolution of the three-dimensional temperature field within the kiln, addressing the issues of single-sensor monitoring blind spots and the inability of static models to adapt to dynamic operating conditions.

[0076] Through the above technical solution, this application can accurately predict the temperature distribution in the central area of ​​the lime kiln under different fuel types, avoiding temperature field prediction errors caused by the lack of core area monitoring. By integrating spatial features with time series analysis, the model can adapt to changes in combustion characteristics caused by fuel switching, improving the generalization ability of three-dimensional temperature prediction. The model trained based on multi-cycle data effectively covers the range of operating condition fluctuations, enhancing the robustness of prediction results in complex production environments.

[0077] This application further proposes that the input of the three-dimensional temperature prediction model also includes the opening of the fuel valve and the air valve. When the opening of the fuel valve and the air valve changes, the three-dimensional temperature prediction model predicts the subsequent temperature change data based on the current fuel type and temperature data.

[0078] The fuel valve and air valve openings refer to the position parameters of the actuators that regulate fuel supply and air flow. These parameters can be implemented using position feedback signals from electric actuators or data from valve opening sensors, representing the real-time regulation status of the combustion system. Fuel type refers to the type of fuel currently in use. This can be achieved through online fuel composition analyzers or by matching and identifying pre-set fuel databases, determining the thermodynamic characteristics of fuel combustion. Temperature data refers to real-time temperature information collected via a multi-dimensional temperature monitoring network. This can be achieved using a combination of thermocouple arrays and infrared scanners, reflecting the spatial distribution of the temperature field within the kiln.

[0079] Specifically, when the fuel valve and air valve openings are adjusted, the change in their openings is input into the three-dimensional temperature prediction model in real time. The model combines the combustion characteristic parameters corresponding to the current fuel type, such as the high calorific value rapid release characteristics of natural gas or the slow combustion and diffusion characteristics of coal powder, to establish a correlation between the valve opening change and the combustion intensity and airflow distribution. By integrating real-time temperature field data, the model can simulate the expansion or contraction process of the combustion area caused by changes in fuel supply, as well as the changes in air flow velocity and mixing efficiency caused by air flow adjustment. Based on the time series prediction algorithm, the model outputs the temperature evolution trend of each grid node in the future period, providing advanced prediction data for the control strategy, so that the temperature field adjustment and valve action can be synchronized in the time dimension.

[0080] Compared to existing technologies, traditional temperature prediction models rely solely on extrapolated predictions based on historical temperature data, failing to consider the direct impact of dynamic valve opening adjustments on the combustion process. This can lead to prediction lags when fuel supply changes suddenly or the ventilation system responds quickly. By using valve opening as a model input variable, this solution establishes a real-time coupling relationship between actuator movement and temperature field changes, addressing the problem of prediction inaccuracy caused by the time lag between control command execution and temperature response.

[0081] Through the above technical solution, the present application realizes real-time quantitative evaluation of the impact of fuel supply and air flow regulation operations on the temperature field in the kiln, so that the temperature prediction results can dynamically track the valve adjustment action, effectively eliminating the temperature control deviation caused by the response delay of the combustion system, and ensuring that the air-fuel ratio optimization strategy is synchronized with the temperature field changes, thereby improving the thermal efficiency stability of the combustion process.

[0082] The present application further proposes a method for determining whether there is a temperature anomaly area, specifically including: calculating the deviation value between the measured temperature data and the corresponding grid temperature predicted by the three-dimensional temperature prediction model; determining whether the deviation value is greater than a set value, where the set value is the sum of the mean of the historical data of the grid and the standard deviation of the historical data; when it is greater than the set value and the duration is greater than 10 minutes, the corresponding area of ​​the grid is a temperature anomaly area.

[0083] The deviation value refers to the temperature difference between the measured temperature data and the three-dimensional temperature prediction model at the same grid node. This can be achieved by calculating the difference between the measured value of the temperature sensor at the grid node and the model prediction value, and is used to quantify the degree to which the local temperature deviates from the predicted value. The set value refers to the abnormal judgment threshold that is dynamically adjusted based on historical data. It can be calculated by adding the mean and standard deviation of the temperature data of the grid node over several cycles in the past. It is used to adapt to the temperature fluctuation characteristics of different regions. The duration refers to the continuous length of time that the temperature deviation exceeds the set value. This can be achieved by using a timer to record the difference between the start time and the end time of the deviation limit, and is used to eliminate misjudgments caused by transient interference.

[0084] Specifically, in the gridded temperature field generated by the three-dimensional temperature prediction model, the measured temperature of each grid node is compared with the model-predicted temperature in real time. When the temperature deviation of a node exceeds the sum of its historical mean and standard deviation, the abnormality monitoring mechanism is triggered. If the deviation state lasts for more than a preset time, such as 10 minutes, it is determined that there is a temperature anomaly in the physical area corresponding to the grid. By combining the dual verification mechanism of dynamic threshold and time persistence, it can effectively distinguish between real anomalies and transient interference, such as short-term fluctuations in fuel supply or sensor noise. This mechanism adaptively adjusts the threshold range through historical data to avoid missed or misjudgment problems caused by fixed thresholds.

[0085] Compared to existing technologies, traditional methods typically use fixed temperature thresholds or single time points to identify anomalies, which are unable to adapt to the dynamic changes in the kiln's temperature field. For example, fixed threshold methods are prone to misjudgment when the fuel's calorific value fluctuates or when materials are unevenly distributed, while single-time over-limit judgments are susceptible to transient interference. This solution dynamically sets thresholds based on historical data and combines them with time-based verification to make anomaly judgments more consistent with actual operating conditions and reduce the risk of misjudgment.

[0086] Through the above technical solution, this application can accurately identify localized temperature anomalies within the lime kiln caused by uneven fuel distribution, nodules, or equipment failures. This solution addresses the lack of sensitivity and reliability of traditional methods under complex operating conditions, providing accurate anomaly location data for subsequent valve linkage control, thereby avoiding combustion efficiency degradation and product quality fluctuations caused by localized temperature deviations.

[0087] The present application further proposes that when there is a temperature abnormality area, a linkage control strategy obtains the temperature value of the temperature abnormality area, and outputs adjustment control data corresponding to the ultra-high temperature or ultra-low temperature condition to control multiple valves in the abnormal area.

[0088] The linkage control strategy refers to the control logic that coordinates the actions of multiple valves. This strategy can be implemented using a preset valve opening combination table or a dynamic optimization algorithm. Its purpose is to generate a coordinated multi-valve adjustment plan based on the actual temperature conditions in the abnormal area. The temperature value in the abnormal area refers to the local temperature deviation value determined by comparing a three-dimensional temperature prediction model with measured data. This strategy can be implemented using a grid node temperature comparison module. This strategy accurately locates the spatial location of the abnormal area and the magnitude of the temperature fluctuation. Ultra-high or ultra-low temperature conditions refer to conditions where the temperature exceeds a preset threshold. This strategy can be implemented using a temperature interval classifier. This strategy distinguishes the abnormality type and matches the corresponding control mode. Adjustment control data refers to a parameter set containing multiple valve opening adjustment values. This strategy can be implemented in a matrix format to store the coordinated adjustment values ​​for fuel and air valves. This strategy synchronizes the operating parameters of multiple actuators. The multiple valves refer to the control valve groups distributed across the fuel and air supply branches corresponding to the abnormal area. This strategy can be implemented using a combination of electric control valves and pneumatic butterfly valves. This strategy rapidly compensates for the local temperature field through multi-channel regulation.

[0089] Specifically, when the temperature value in the temperature abnormality area is identified as ultra-high temperature, the linkage control strategy combines and outputs a fuel valve opening reduction instruction with a corresponding air valve opening increase instruction, achieving temperature reduction by reducing fuel supply and increasing cooling air flow. When the temperature value in the temperature abnormality area is identified as ultra-low temperature, the linkage control strategy combines and outputs a fuel valve opening increase instruction with a corresponding air valve opening reduction instruction, achieving temperature increase by increasing fuel supply and reducing excess air. In this process, the opening adjustment amount of multiple valves is dynamically calculated based on the temperature deviation amplitude, so that the fuel-air mixture ratio can be precisely controlled in the local area.

[0090] Compared to existing technologies, traditional methods typically only adjust a single fuel or air valve, failing to address the temperature field recovery delay caused by multi-variable coupling. This solution establishes a multi-valve linkage mechanism to synchronously adjust fuel supply and air flow when a temperature anomaly is detected. This overcomes the thermal inertia lag caused by single-parameter adjustment and allows the local temperature field to return to the set range in a shorter time.

[0091] Through the above technical solution, the present application can synchronously generate multi-valve collaborative control instructions when an abnormal temperature area is detected, eliminate local temperature deviations through the linkage adjustment of fuel and air supply, avoid the problem of low temperature field recovery efficiency caused by insufficient action of a single valve, and effectively improve the uniformity of temperature distribution in the kiln.

[0092] The present application further proposes that the calculation of the air-fuel ratio based on the operating parameters includes: identifying the fuel type based on the operating parameters, and correcting the calorific value and the material level value based on the operating parameters; determining the theoretical air-fuel ratio based on the fuel type; calculating the excess air coefficient based on the theoretical air-fuel ratio; and adjusting the theoretical air-fuel ratio using the corrected calorific value, material level value and excess air coefficient to obtain the final air-fuel ratio.

[0093] Among them, fuel type identification refers to determining the type of fuel currently in use through fuel composition analysis or matching with a preset database. This can be achieved by using an online gas analyzer to detect fuel components or manually entering the fuel type code through the operation interface. This is used to provide an accurate fuel characteristic benchmark for subsequent calculations. Calorific value correction refers to the dynamic adjustment of the theoretical calorific value based on the real-time collected fuel calorific value data. Specifically:

[0094]

[0095] Where Q corr It is obtained by modifying the operating parameters (such as fuel inlet temperature and pressure) through thermodynamic formulas, for example:

[0096] Q corr =Q std ×(1+β Q ΔT),

[0097] β Q is the temperature correction coefficient, and ΔT is the difference between the actual temperature and the standard temperature.

[0098] Material level correction refers to the compensation calculation based on the impact of material height changes in the kiln on the combustion space, specifically:

[0099]

[0100] Where, L actual is the material level height measured in real time (unit: m), L designis the standard material level height under design conditions (unit: m), β L is the material level correction sensitivity coefficient (dimensionless, usually 0.05 to 0.2).

[0101] The theoretical air-fuel ratio refers to the oxygen requirement derived from the chemical reaction equation for complete fuel combustion. It can be calculated using the formula for the fuel's carbon-hydrogen content and the oxidant ratio, and is used to establish the basic combustion ratio for different fuel types. Excess air coefficient adjustment refers to scaling the theoretical air volume based on actual combustion conditions, specifically:

[0102]

[0103] Where, It is the oxygen content in dry flue gas (%). The excess air coefficient is used to compensate for the uneven airflow distribution in the kiln or the fluctuation of fuel atomization effect.

[0104] Specifically, the fuel type identification module first determines the fuel type through online detection or manual input. For example, when the methane concentration is detected to be over 95%, it is determined to be natural gas fuel. The calorific value correction module receives real-time measurement data from the fuel calorific value sensor. For example, when it is detected that the current fuel calorific value is 5% lower than the standard value, the calorific value correction coefficient is automatically increased. The material level correction module reduces the air supply according to the preset algorithm based on the material height sensor data. For example, when the material layer height increases by 0.3 meters, the theoretical air-fuel ratio calculation unit calls the corresponding stoichiometric formula according to the fuel type. For example, natural gas calculates the basic air demand according to the reaction formula CH4+2O2→CO2+2H2O. The excess air coefficient adjustment module combines the flue gas oxygen content data. For example, when the oxygen content is detected to be lower than 2%, the air supply ratio is automatically increased. The final air-fuel ratio is formed by superimposing the calorific value correction coefficient, the material level correction factor and the excess air coefficient to form a combustion control parameter that dynamically adapts to the current operating conditions. The final air-fuel ratio is:

[0105] A / F final =A / F th ×α×K Q ×K L ,

[0106] A / F th is the theoretical air-fuel ratio.

[0107] Compared to existing technologies, traditional methods rely on fixed air-fuel ratio tables or single parameter adjustments, which are unable to cope with fluctuations in fuel calorific value and changes in material distribution. For example, existing technologies only use a table lookup to determine a fixed air-fuel ratio based on fuel type. Even when the fuel calorific value fluctuates by ±10%, the original parameters are still used for control, resulting in incomplete combustion. This solution uses dynamic multi-parameter correction to automatically compensate for air supply when the fuel calorific value decreases, adjust the spatial distribution of the combustion area when the material level increases, and optimize the excess air coefficient based on real-time combustion performance feedback, achieving coordinated adjustment in three dimensions.

[0108] Through the above-mentioned technical solution, the present application can compensate in real time for the impact of fuel composition fluctuations on combustion efficiency. For example, when the calorific value of coal gas drops by 8% due to gas source switching, the system automatically increases the gas supply and adjusts the air ratio accordingly. It can dynamically adapt to changes in the combustion space caused by material accumulation. For example, when the preheating material layer thickens, causing increased airflow resistance, the air valve opening distribution is automatically optimized. This effectively solves the air-fuel ratio inaccuracy problem caused by traditional control methods that ignore calorific value fluctuations, material level changes, and differences in operating conditions, ensuring that the combustion process in the kiln always maintains the optimal air-fuel ratio.

[0109] This application further proposes a method for obtaining adjustment control parameters through calculation in a fuzzy control strategy, including preprocessing the input environmental parameters and air-fuel ratio data, activating multiple rules and calculating the weights corresponding to each rule, and summarizing the output of each rule according to the corresponding weights of each rule to obtain the final adjustment control parameters.

[0110] Among them, preprocessing refers to the standardization and noise elimination of the original data. Specifically, the normalization algorithm can be used to convert parameters of different dimensions into dimensionless values. At the same time, the sliding average filtering algorithm can be used to eliminate high-frequency interference signals to ensure the validity and consistency of the input data.

[0111] Among them, activating multiple rules means triggering the corresponding fuzzy control rules according to the membership of the input parameters. The fuzzy control rule table is shown in Table 1 below:

[0112] Table 1

[0113] ΔT ΔAFR Fuel valve adjustment Air valve adjustment PB(>50℃) PS (+2%) -5％ +3％ PM (20-50℃) ZO (0%) -2％ +1％ ZO(±20℃) NS (-3%) +3％ -2％

[0114] Among them, ΔT is the temperature deviation, ΔAFR is the air-fuel ratio deviation, PB, PM and ZO represent the degree of temperature deviation from large to small, and PS, ZO and NS represent the deviation of air-fuel ratio deviation from large to small.

[0115] Calculating weights refers to determining the contribution of each rule to the current operating conditions. Specifically, the Euclidean distance method is used to calculate the degree of match between input parameters and rule antecedents, and then the weighted average method is used to distribute the activation strength of each rule. Aggregating outputs refers to the control instructions that fuse multiple rules. Specifically, the center of gravity method is used for defuzzification. The membership functions output by each rule are superimposed, and their center of mass coordinates are calculated to convert them into precise fuel and air valve adjustment values.

[0116] Specifically, when the fuel calorific value fluctuates or the kiln pressure becomes abnormal, the preprocessing module first normalizes the ambient temperature, air pressure, and air-fuel ratio to eliminate errors caused by sensor range variations. Subsequently, through membership function matching, multiple control rules relevant to the current operating conditions are simultaneously activated. For example, a "low calorific value compensation rule" is triggered when the fuel calorific value decreases, and a "negative pressure adjustment rule" is triggered when the kiln pressure increases. Each rule is dynamically weighted based on its degree of match with real-time data. Ultimately, a defuzzification operation generates a comprehensive control command that takes multiple factors into account. This process enables control parameters to adapt to changing fuel characteristics and complex operating conditions, avoiding the lag or oscillation caused by single-rule control.

[0117] Compared to existing technologies, traditional methods use fixed thresholds to trigger a single control rule, making them unable to handle complex operating conditions where both fuel calorific value and kiln pressure fluctuate simultaneously. This solution, however, achieves coordinated handling of intersecting factors through the parallel activation of multiple rules and dynamic weight allocation. For example, when the calorific value of coal gas decreases and the negative pressure in the kiln increases, the system can simultaneously adjust the fuel supply and ventilation volume. Traditional methods require a step-by-step approach, resulting in delayed response.

[0118] Through the above technical solution, this application can improve the regulation accuracy of combustion parameters within a ±15% range of fuel calorific value fluctuations, controlling the dynamic deviation of the air-fuel ratio within a ±0.5 range. It also shortens the control response time under complex operating conditions to less than 30 seconds, effectively avoiding temperature field imbalances caused by insufficient manual experience. The CO concentration in the flue gas can be stably maintained below 800 ppm, significantly reducing the risk of pollutant exceeding the standard due to control lag.

[0119] The present application further proposes that the control parameters include a fuel valve opening adjustment amount and an air valve linkage adjustment amount. The fuel valve opening adjustment amount is calculated based on environmental parameters, and the air valve linkage adjustment amount is calculated based on the air-fuel ratio.

[0120] The fuel valve opening adjustment refers to the control variable that dynamically adjusts the fuel supply based on environmental parameters. This is achieved by using ambient temperature, humidity, and air pressure sensor data as input parameters and establishing a mapping model between these environmental parameters and the required fuel calorific value. This parameter is used to compensate for the impact of external environmental changes on combustion stability. For example, when the ambient temperature drops sharply, the fuel supply is automatically increased to maintain thermal balance within the kiln.

[0121] The air valve linkage adjustment value is a control variable that dynamically matches the air supply based on the air-fuel ratio. This is achieved by using real-time fuel flow and oxygen concentration data, combined with the combustion chemical reaction equation to calculate the theoretical oxygen demand. This parameter is used to ensure a precise fuel-air ratio, for example, by automatically adjusting the air damper opening to maintain optimal combustion efficiency when the fuel's calorific value fluctuates.

[0122] Specifically, the fuel valve opening adjustment variable incorporates external interference factors such as kiln heat loss and intake air temperature changes into the control closed loop through an independent environmental parameter response mechanism. When it detects that rising ambient humidity is causing incomplete fuel combustion, the control system prioritizes adjusting the fuel valve opening to compensate for latent heat loss from evaporation. The air valve linkage adjustment variable, through an air-fuel ratio feedback mechanism, synchronously adjusts the air supply in response to changes in the fuel supply. For example, when the fuel flow increases, the blower speed is automatically increased according to the preset excess air coefficient. The decoupled control of the two enables the fuel supply and air ratio to form two orthogonal adjustment dimensions, avoiding the adjustment oscillation caused by the cross-influence of parameters in traditional coupled control.

[0123] Compared to existing technologies, conventional combustion control systems typically utilize linked fuel-air control valves, which are prone to regulation lag and overshoot when environmental parameters change suddenly. For example, when humidity surges during a rainstorm, conventional systems must simultaneously recalculate both fuel and air control levels, potentially causing temporary mismatches. This solution, however, utilizes a separate, independent control mechanism, ensuring that environmental parameter changes trigger only a rapid response from the fuel valve while maintaining steady-state regulation of the air-fuel ratio. This effectively addresses control mismatches under dynamic operating conditions.

[0124] Through the above-mentioned technical solution, this application can achieve dynamic adaptation of fuel supply to environmental conditions, while ensuring precise matching of air ratio with combustion requirements. When sudden heat dissipation from the kiln increases, rapid compensation by the fuel valve prevents temperature fluctuations; when the fuel calorific value fluctuates, precise adjustment of the air valve maintains combustion efficiency. The coordinated control of these two factors improves the stability of the temperature field within the kiln while keeping the concentration of unburned carbon oxides in the combustion products within the process requirements.

[0125] The present application further proposes specific steps for determining whether the thermal efficiency meets the standard, including collecting flue gas and flue gas components through a high-temperature sampling probe, performing sliding average filtering and outlier removal on the collected flue gas components, converting the measured flue gas components into dry basis concentrations, calculating the thermal efficiency using a counter-balance strategy based on the dry basis concentration, and determining whether the thermal efficiency is greater than a set value. If it is, the thermal efficiency meets the standard; otherwise, the fuzzy control strategy is self-learned and the fuzzy control strategy is re-updated.

[0126] The high-temperature sampling probe is a flue gas collection device capable of withstanding the high temperatures within the kiln. Specifically, it can be implemented using a probe structure with a water-cooled sleeve and ceramic coating. Its purpose is to directly obtain real-time flue gas samples during the combustion process, avoiding the delays associated with traditional offline sampling. A sliding average filter applies a moving window mean to continuously collected flue gas composition data. Specifically, a window of 5-10 sampling points can be used for data smoothing. Its purpose is to eliminate the interference of instantaneous fluctuations in flue gas composition on thermal efficiency calculations. Dry-basis concentration conversion converts wet-basis flue gas composition data, which is affected by moisture, into concentration values ​​in a water-free state. This can be achieved by measuring the flue gas moisture content with a humidity sensor and then performing a mathematical conversion. It aims to eliminate the impact of moisture fluctuations on the accuracy of thermal efficiency calculations. A counterbalance strategy is a method for calculating the combustion heat utilization rate by inversely inferring it from flue gas composition. Specifically, a heat loss model is established using the functional relationship between oxygen content and carbon dioxide concentration. Its purpose is to adapt to the dynamic thermal efficiency calculation requirements under conditions with fluctuating fuel calorific values.

[0127] Specifically, during the combustion process, a high-temperature sampling probe continuously collects flue gas samples and transmits them to the analyzer. The flue gas composition data is filtered through a sliding average to eliminate instantaneous fluctuation interference, and outliers are eliminated using the 3σ principle to ensure data reliability. The humidity-compensated dry basis concentration data is input into the counter-balance calculation model, and combined with the preset fuel calorific value parameters to generate a real-time thermal efficiency value. When the thermal efficiency falls below the set threshold, the self-learning mechanism of the fuzzy control strategy is triggered, and the air-fuel ratio calculation logic is optimized by adjusting the membership function or rule weights, thereby forming a closed-loop control circuit. This process replaces traditional offline analysis with real-time data collection and dynamic calculation, reducing the lag time for combustion parameter adjustment from hours to minutes.

[0128] Compared with existing technologies, traditional methods rely on manual, periodic collection of flue gas samples for laboratory analysis, resulting in thermal efficiency calculations lagging behind actual operating conditions and failing to support timely optimization of combustion parameters. This solution, through online flue gas composition monitoring and real-time calculation, increases the frequency of thermal efficiency data updates to coincide with the combustion control cycle, effectively addressing the problem of accumulated control deviations. Furthermore, the application of a counter-balance strategy overcomes the calculation errors of traditional positive-balance methods when the fuel calorific value fluctuates, enhancing adaptability under complex operating conditions.

[0129] Through the above technical solution, this application achieves real-time monitoring and closed-loop control of thermal efficiency during the combustion process, avoiding the lag in air-fuel ratio adjustment caused by offline sampling delays, reducing energy waste and the risk of excessive pollutant emissions. By combining dry-base concentration conversion with counter-balance calculations, the accuracy of thermal efficiency calculations is improved, ensuring that the fuzzy control strategy can optimize parameters based on actual operating conditions, thereby maintaining efficient and stable operation of the combustion process.

[0130] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, various modifications and variations of the present application are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0131] In some other preferred embodiments of the present invention, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the method described in the above embodiment.

[0132] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0133] The above description is a detailed description of the preferred embodiments of the present invention, but the embodiments are not intended to limit the scope of the patent application of the present invention. Any equivalent changes or modifications completed under the technical spirit suggested by the present invention should fall within the patent scope covered by the present invention.

Claims

1. A lime kiln combustion control method, characterized in that: include: Collect lime kiln temperature data and construct lime kiln temperature field; Analyze temperature deviation based on temperature field to determine whether there is an abnormal temperature area; When there is no abnormal temperature area, collect working parameters and environmental parameters. Calculate the air-fuel ratio based on operating parameters; Input the environmental parameters and air-fuel ratio into the fuzzy control strategy and obtain the adjustment control parameters through calculation; Collect the flue gas components generated during the combustion process, calculate the thermal efficiency based on the component analysis, and determine whether the thermal efficiency meets the standards; When the standards are not met, the fuzzy control strategy performs parameter self-learning, re-updates the fuzzy control strategy, and recalculates the air-fuel ratio and adjusts the control parameters.

2. A lime kiln combustion control method according to claim 1, characterized in that: The lime kiln temperature field construction specifically includes: Construct a multi-dimensional temperature monitoring network and collect a number of temperature data through the multi-dimensional temperature monitoring network; Performing heterogeneous synchronous data fusion on the temperature data and calculating key temperature indicators; Establish a three-dimensional heat conduction model of lime kiln; Fusing the three-dimensional heat conduction model with a deep learning network to construct a three-dimensional temperature prediction model, and fusing the temperature predicted by the three-dimensional temperature prediction model into the three-dimensional heat conduction model; Visualize a 3D heat conduction model.

3. A lime kiln combustion control method according to claim 2, characterized in that: The construction of the three-dimensional temperature prediction model specifically includes: A central temperature acquisition device is added to the multi-dimensional temperature monitoring network. The central temperature acquisition device is provided with several acquisition points. The central temperature acquisition device is placed in the center of the lime kiln and is fine-tuned according to the actual fuel position in the lime kiln. The central temperature acquisition device collects combustion temperature change data at the center of the lime kiln in real time. The multi-dimensional temperature monitoring network collects temperature change data around the lime kiln. The temperature change data at the center and around the kiln collected at the same time are regarded as a set of temperature data. Several sets of temperature data from several combustion cycles are collected as training sets and test sets. The three-dimensional temperature prediction model is trained using a training set, and the trained three-dimensional temperature prediction model is tested using a test set. After passing the test, the three-dimensional temperature prediction model is obtained.

4. A lime kiln combustion control method according to claim 3, characterized in that: The three-dimensional temperature prediction model input also includes the fuel valve and air valve openings. When the fuel valve and air valve openings change, the three-dimensional temperature prediction model predicts subsequent temperature change data based on the current fuel type and temperature data.

5. A lime kiln combustion control method according to claim 1, characterized in that: Determining whether there is an abnormal temperature area specifically includes: Calculate the deviation between the measured temperature data and the corresponding grid temperature predicted by the three-dimensional temperature prediction model; Determine whether the deviation value is greater than a set value, where the set value is the sum of the mean of the historical data of the grid and the standard deviation of the historical data; When it is greater than the set value and lasts for more than 10 minutes, the corresponding area of ​​the grid is a temperature abnormality area.

6. A lime kiln combustion control method according to claim 4, characterized in that: When there is an abnormal temperature area, the linkage control strategy obtains the temperature value of the abnormal temperature area, and outputs adjustment control data according to the ultra-high temperature or ultra-low temperature conditions to control multiple valves in the abnormal area.

7. A lime kiln combustion control method according to claim 1, characterized in that: Calculating the air-fuel ratio based on the operating condition parameters includes: Identify fuel type based on operating parameters and correct calorific value and material level based on operating parameters; Determine the theoretical air-fuel ratio based on the fuel type; Calculate the excess air coefficient based on the theoretical air-fuel ratio; The corrected calorific value, material level value and excess air coefficient are used to adjust the theoretical air-fuel ratio to obtain the final air-fuel ratio.

8. A lime kiln combustion control method according to claim 1, characterized in that: In the fuzzy control strategy, the adjustment control parameters obtained by calculation include: Preprocess the input environmental parameters and air-fuel ratio data; Activate multiple rules and calculate the weights corresponding to each rule; The output of each rule is summarized according to the corresponding weight of each rule to obtain the final adjustment control parameters.

9. A lime kiln combustion control method according to claim 8, characterized in that: The control parameters include a fuel valve opening adjustment amount and an air valve linkage adjustment amount. The fuel valve opening adjustment amount is calculated based on the environmental parameters, and the air valve linkage adjustment amount is calculated based on the air-fuel ratio.

10. A lime kiln combustion control method according to claim 8, characterized in that: The determination of whether the thermal efficiency meets the standard specifically includes: The flue gas and flue gas components are collected through the high temperature sampling probe. Perform sliding average filtering on the collected smoke components and remove outliers; Convert the measured smoke components into dry basis concentrations; Thermal efficiency was calculated using a counter-balance strategy based on dry basis concentration; Determine whether the thermal efficiency is greater than the set value. If it is, the thermal efficiency meets the standard. Otherwise, the fuzzy control strategy is self-learned and the fuzzy control strategy is updated.

Images