Online infrared temperature monitoring system for rotary kiln

Abstract

The invention relates to the technical field of rotary kilns, and discloses an online infrared temperature monitoring system for a rotary kiln. The infrared temperature acquisition module is used for acquiring temperature data of the outer wall of the rotary kiln in real time and generating outer wall temperature distribution; the heat conduction positive problem modeling module is used for constructing a mathematical model for describing an evolution rule of a temperature field in the rotary kiln based on material parameters and operation conditions of the rotary kiln; and the heat conduction inverse problem solving module is used for reversely deducing three-dimensional temperature field distribution of the inner wall of the rotary kiln according to the outer wall temperature data. Through modular design and multi-source data fusion, a closed-loop system from outer wall temperature collection to internal temperature field inversion, anomaly recognition, trend prediction and linkage control is constructed. The real-time performance of the system comes from quick response of high-frequency acquisition and iterative calculation, and the stability depends on combined implementation of regularization processing, an optimization algorithm and real-time feedback control. And the reliability of monitoring the operation state of the rotary kiln is integrally improved.

Description

Technical Field

[0001] The present invention relates to the technical field of rotary kilns, and specifically to an online infrared temperature monitoring system for rotary kilns. Background Art

[0002] A rotary kiln is a key piece of equipment widely used in industries such as cement, chemical engineering, and metallurgy. Its main function is to conduct high-temperature calcination and treatment of raw materials to achieve the transformation of the physical and chemical properties of products. As an important high-temperature heat treatment equipment, the operating efficiency and stability of the rotary kiln directly affect the economy of production and the quality of products. However, due to the complex internal operating environment of the rotary kiln, with characteristics such as high temperature, high load, and rapid rotation, a series of problems are likely to occur during long-term operation.

[0003] In the prior art, the monitoring of the operating state of the rotary kiln mainly relies on the point measurement method of the outer wall temperature or traditional manual inspections. The outer wall temperature point measurement method usually uses an infrared thermometer or a thermocouple to measure the temperature at limited detection points, but this method has the following limitations.

[0004] The local acquisition of the outer wall temperature cannot comprehensively reflect the internal operating state of the rotary kiln. Especially when the abnormal area occurs at a location where no temperature measurement equipment is installed, it is extremely easy to cause a delay in discovering the fault or even misjudgment.

[0005] The temperature field distribution inside the rotary kiln is comprehensively affected by various factors such as material characteristics, combustion conditions, heat conduction, and convection. The prior art lacks effective means to inversely deduce the three-dimensional temperature field distribution inside the kiln body through external temperature data, making it difficult to provide accurate positioning and diagnosis of abnormal areas.

[0006] The existing monitoring methods for rotary kilns rely more on the direct judgment of abnormal outer wall temperatures and lack the ability to predict the development trend of abnormal areas, unable to achieve early intervention, and extremely easy to cause the expansion of faults or equipment damage. Summary of the Invention

[0007] In view of the deficiencies of the prior art, the present invention provides an online infrared temperature monitoring system for rotary kilns, which solves the problems of real-time monitoring of the three-dimensional temperature field inside the rotary kiln, accurate identification of abnormal faults, and linkage control.

[0008] To achieve the above objectives, the present invention is realized through the following technical solutions: An online infrared temperature monitoring system for rotary kilns, including; An infrared temperature acquisition module, used to collect the temperature data of the outer wall of the rotary kiln in real time and generate the outer wall temperature distribution; A forward heat conduction problem modeling module, which constructs a mathematical model describing the evolution law of the temperature field inside the rotary kiln based on the material parameters and operating conditions of the rotary kiln; The inverse heat conduction problem solving module inversely calculates the three-dimensional temperature field distribution of the inner wall of the rotary kiln based on the outer wall temperature data; The anomaly analysis and alarm module is used to identify the abnormal areas inside the rotary kiln according to the distribution characteristics of the inner wall temperature field and send out alarm signals; The trend prediction and linkage control module predicts potential faults based on historical temperature data and linkage-adjusts the operating parameters of the rotary kiln.

[0009] Preferably, the infrared temperature acquisition module includes; The infrared sensor acquisition unit is used to be arranged at multiple positions on the outer wall of the rotary kiln to collect the temperature data of different positions on the outer wall in real time; The data filtering unit is used to filter the collected temperature data to eliminate environmental noise; The interpolation and reconstruction unit is used to interpolate and spatially reconstruct the discontinuous temperature data to generate a complete outer wall temperature distribution.

[0010] Preferably, the forward heat conduction problem modeling module includes; The forward heat conduction equation establishment unit is used to construct a forward heat conduction equation based on the thermal conductivity, density, and specific heat capacity parameters of the materials inside the rotary kiln; The numerical discretization and calculation unit is used to discretize the forward heat conduction equation by using the finite element method and generate a numerical calculation model.

[0011] Preferably, the inverse heat conduction problem solving module includes; The optimization objective construction unit is used to construct an objective function based on minimizing the difference between the outer wall temperature measurement data and the calculated temperature; The regularization processing unit is used to introduce the smoothness constraint of the temperature field to avoid the problem of numerical instability in the inversion result; The optimization solution unit iteratively optimizes the objective function based on the Lagrange multiplier method to inversely calculate the internal temperature field.

[0012] Preferably, the optimization solution unit calculates the three-dimensional internal temperature field distribution according to the input outer wall temperature data and the forward heat conduction problem model by using the iterative optimization method, where the iteration termination condition is that the outer wall temperature measurement error is less than the preset threshold.

[0013] Preferably, the anomaly analysis and alarm module includes; The temperature gradient analysis unit is used to calculate the gradient distribution of the inner wall temperature field and identify the positions of gradient mutations; The anomaly feature extraction unit is used to extract the features of the abnormal areas according to the temperature gradient and distribution pattern; The alarm generation unit is used to generate alarm signals and output the position information and type of the abnormal areas.

[0014] Preferably, according to the extracted abnormal area features, the alarm generation unit determines that the refractory brick has fallen off when the local temperature gradient increases significantly, and determines that slag has adhered when the local temperature gradient decreases significantly.

[0015] Preferably, the trend prediction and linkage control module includes; A historical data modeling unit for storing and extracting historical temperature data and operating parameters; A fault prediction unit for predicting the temperature change trend and the potential fault occurrence time based on the time series analysis algorithm; A linkage control unit for adjusting the operating parameters of the rotary kiln according to the prediction results, including the kiln body rotation speed, the fuel supply amount, and the inclination angle.

[0016] Preferably, the fault prediction unit is based on a long short-term memory neural network model, and combines historical temperature data and the temperature field data collected in real time to dynamically update the fault prediction results.

[0017] Preferably, the linkage control unit adjusts the kiln body rotation speed in real time according to the fault prediction results to reduce the adhesion degree of the materials inside the kiln body.

[0018] The present invention provides an online infrared temperature monitoring system for a rotary kiln. It has the following beneficial effects: 1. Through modular design and multi-source data fusion, the present invention constructs a closed-loop system from outer wall temperature acquisition to internal temperature field inversion, anomaly identification, trend prediction, and linkage control. The real-time performance of the system comes from the fast response of high-frequency acquisition and iterative calculation, and the stability depends on the joint implementation of regularization processing, optimization algorithms, and real-time feedback control. Overall, the reliability of the rotary kiln operation state monitoring is improved.

[0019] 2. Through the infrared temperature acquisition module, using the multi-point infrared sensor unit arranged on the outer wall of the rotary kiln, combined with the Kalman filter algorithm to remove noise interference, and generating a two-dimensional continuous outer wall temperature distribution field through the interpolation and reconstruction unit, the high-precision and full-coverage monitoring of the outer wall temperature is realized, overcoming the limitations of the traditional point measurement method, and providing a complete and reliable boundary condition for the subsequent temperature field modeling.

[0020] 3. By adopting a long short-term memory (LSTM) neural network model to model and predict the expansion trend of the abnormal area, the model can process dynamic time series data, accurately predict the expansion speed and influence range of the abnormal area, so as to provide an early warning function.

[0021] 4. Through the anomaly analysis and alarm module, the present invention can quickly and accurately identify the abnormal areas inside the rotary kiln by using the calculation and feature extraction technology of temperature gradient. A significant increase or decrease in the gradient corresponds to the problems of refractory brick shedding and slag sticking respectively. The alarm signals generated by the module include the abnormal position, type, and severity, providing an intuitive decision-making basis for the operator and effectively improving the safety of the rotary kiln operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is the system framework diagram of the present invention; Figure 2 is the schematic diagram of the infrared temperature acquisition module of the present invention; Figure 3 is the schematic diagram of the forward heat conduction problem modeling module of the present invention; Figure 4 is the schematic diagram of the inverse heat conduction problem solving module of the present invention; Figure 5 is the schematic diagram of the anomaly analysis and alarm module of the present invention; Figure 6 is the schematic diagram of the trend prediction and linkage control module of the present invention; Figure 7 is the operation logic diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0024] Please refer to the attached Figure 1 - attached Figure 7 , the embodiment of the present invention provides an on-line infrared temperature monitoring system for a rotary kiln, including; An infrared temperature acquisition module for real-time acquisition of the temperature data on the outer wall of the rotary kiln to generate the outer wall temperature distribution; The infrared temperature acquisition module includes; An infrared sensor acquisition unit for being arranged at multiple positions on the outer wall of the rotary kiln to real-time acquire the temperature data at different positions on the outer wall; A data filtering unit for filtering the acquired temperature data to eliminate environmental noise; An interpolation and reconstruction unit is used to interpolate and spatially reconstruct discontinuous temperature data to generate a complete outer wall temperature distribution. Specifically, it is used to collect real-time outer wall temperature data of a rotary kiln and generate a complete outer wall temperature distribution field. The infrared temperature acquisition module collects the temperature data of the outer wall of the rotary kiln in real time through a plurality of infrared sensor units arranged on the outer wall of the rotary kiln. These temperature data are filtered to reduce external environmental interference, and at the same time, a two-dimensional continuous temperature field is generated through interpolation and reconstruction operations, so as to provide complete boundary conditions for subsequent mathematical modeling; In this embodiment, the infrared sensor acquisition unit is used to be arranged at multiple positions on the outer wall of the rotary kiln to form a grid-like sampling arrangement. The infrared sensors are arranged in a circumferential and axial uniform distribution manner according to the outer wall surface characteristics of the rotary kiln to ensure the uniformity of the sampling point distribution and the measurement coverage; Specifically, the temperature data collected by each sensor can be expressed as; T obs (x i ,t) = T true (x i ,t) + ε(x i ,t) Where: T obs (x i ,t) is the measured temperature of the sensor at position x i and time t; T true (x i ,t) is the true outer wall temperature; ε(x i ,t) is the measurement noise or interference; In this embodiment, the data filtering unit performs real-time denoising processing on the collected data through the Kalman filtering algorithm to reduce the interference of dust, radiation, and equipment vibration in the industrial field; The Kalman filtering process includes two main steps: prediction and update. The prediction equation is; The update equation is; Where: is the predicted temperature at the current moment; is the filtered temperature estimate; K t is the Kalman gain; T obs,t is the collected outer wall temperature; The Kalman gain K tDynamically calculated based on the prediction error covariance and the measurement noise covariance; Where: is the prediction error covariance; H t is the measurement matrix; R t is the measurement noise covariance; Since the temperature data collected by the infrared sensor is discrete in spatial distribution due to the limited number of sensors, in order to meet the requirements of the subsequent mathematical modeling for the continuity of the boundary conditions, in this embodiment, the interpolation and reconstruction unit performs spatial interpolation and two-dimensional reconstruction on the discontinuous temperature data; Specifically, in this embodiment, the bilinear interpolation method is used to process the discrete data. For any interpolation point (x, y), the interpolation temperature calculation formula is; T interp (x,y) = T(x 1 ,y 1 )·w 1 + T(x 2 ,y 2 )·w 2 + T(x 3 ,y 3 )·w 3 + T(x 4 ,y 4 )·w 4 Where: T interp (x,y) is the temperature of the interpolation point; T(x 1 ,y 1 ), T(x 2 ,y 2 ), T(x 3 ,y 3 ), T(x 4 ,y 4 ) are the measured temperatures of the four sensors around the interpolation point; w 1 ,w 2 ,w 3 ,w 4 are the interpolation weights, satisfying w 1 + w 2 + w 3 + w 4 = 1; The interpolation and reconstruction unit can also perform high-order fitting on the outer wall temperature field in combination with the surface fitting algorithm, so as to generate a smoother two-dimensional temperature distribution; The processed outer wall temperature distribution field Tobs (x, y, t) is input as the boundary condition of the forward heat conduction problem modeling module, which is used to construct the mathematical model of the temperature field inside the rotary kiln. Specifically, this data is used to define the outer wall temperature boundary condition of the heat conduction equation and provide an accurate fitting target for the inverse problem solving module. The forward heat conduction problem modeling module constructs a mathematical model that describes the evolution law of the temperature field inside the rotary kiln based on the material parameters and operating conditions of the rotary kiln; Specifically, based on the outer wall temperature distribution data of the rotary kiln, combined with the material parameters and operating conditions of the kiln body, a mathematical model that describes the evolution law of the temperature field inside the kiln body is established. By constructing the heat conduction equation and performing numerical discretization, a numerical calculation model that can be used for subsequent inverse problem solving is generated, thereby providing a basis for the inversion of the three-dimensional temperature field inside the kiln body; Specifically, the module receives the outer wall temperature distribution generated by the infrared temperature acquisition module as the boundary condition, and combines the geometric structure and thermophysical parameters of the rotary kiln to establish a three-dimensional unsteady heat conduction equation; In this embodiment, the forward heat conduction problem modeling module constructs a complete heat conduction mathematical model through the three-dimensional unsteady heat conduction equation that describes the internal temperature distribution of the rotary kiln, combined with the kiln body boundary conditions and material properties; Specifically, the basic form of the heat conduction equation is; ρ represents the density of the rotary kiln material, which is determined according to the material properties of the kiln body; c is the specific heat capacity of the material, which represents the heat absorption capacity of the material; k is the thermal conductivity, which represents the heat conduction ability of the material; T is the temperature, which is the main solution variable of the model; Q(x, y, z, t) is the volume heat source term, which represents the internal heat source distribution, such as the heat generated by fuel combustion; In this embodiment, the solution of the forward heat conduction problem model requires reasonable setting of the boundary conditions of the kiln body; the boundary conditions of the rotary kiln include the inner wall heat flux density condition and the outer wall temperature condition; For the inner wall, a heat flux density boundary condition is usually set to represent the heat transfer from the combustion-generated heat to the kiln body; Among them: n is the inner wall normal vector; q in is the heat flux density of the inner wall, which can be estimated according to the fuel combustion intensity and the combustion chamber temperature; For the outer wall, a temperature boundary condition is set to represent the outer wall temperature data provided by the infrared temperature acquisition module; T(x, y, z, t) = T obs (x, y, t) where; T obs (x, y, t) is the interpolated outer wall temperature distribution field; In this embodiment, after establishing the heat conduction equation, the finite element method is used to numerically discretize the equation for numerical calculation. The finite element method is achieved by discretizing the continuous temperature field into discrete node temperature values.

[0025] Specifically, after the finite element discretization of the heat conduction equation, the following discrete form is obtained; where: M is the mass matrix, used to describe the heat capacity effect; K is the stiffness matrix, used to describe the heat conduction effect; T is the node temperature vector; Q is the heat source term vector; The mass matrix M and the stiffness matrix K can be calculated by the following integral formula; M ij = ∫ Ω ρcφ i φ j dΩ where; φ i and φ j are finite element basis functions; (M + Δt·K)T n+1 = MT n + Δt·Q n where: T n and T n+1 are the node temperatures at the nth and (n + 1)th time steps respectively; Δt is the time step; The main outputs of the heat conduction forward problem modeling module are the numerical model of the temperature field inside the kiln body and the discretized matrix expression. These results provide theoretical support and basic input for the heat conduction inverse problem solving module. At the same time, the generated stiffness matrix and mass matrix can be directly used for the optimized solution of the inverse problem, further improving the calculation efficiency; The heat conduction inverse problem solving module inversely calculates the three-dimensional temperature field distribution of the inner wall of the rotary kiln based on the outer wall temperature data; Specifically, based on the mathematical model established by the heat conduction forward problem modeling module and the outer wall temperature data provided by the infrared temperature acquisition module, the three-dimensional temperature field distribution inside the rotary kiln is inversely calculated. The module constructs an optimization objective function, combines the constraints of the forward problem, and uses an iterative solution algorithm to achieve the stable solution of the inverse problem; In this embodiment, the inverse heat conduction problem constructs an optimization objective function to minimize the deviation between the measured outer wall temperature data and the calculation result of the forward problem, and at the same time introduces a regularization term to ensure the stability of the solution. The basic form of the objective function is: where: T int represents the temperature field distribution inside the rotary kiln, which is the variable to be solved; T obs is the outer wall temperature distribution data provided by the infrared temperature acquisition module; is the forward problem solver of the heat conduction forward problem modeling module, which is used to convert the internal temperature field T int into the outer wall temperature; R(T int ) is the regularization term, which is used to constrain the smoothness or physical rationality of the solution; λ is the regularization coefficient, which is used to balance the weights of the data fitting term and the regularization term; The regularization term R(T int ) can take the second norm of the temperature field gradient to limit the oscillation of the solution, and its expression is: where: Ω represents the internal area of the rotary kiln; In another possible implementation, the regularization term can adopt the Tikhonov regularization method, and its form is: R(T int ) = ∥T int ∥ 2 Through the construction of the above optimization objective function, this module can obtain a smooth and physically reasonable internal temperature field distribution while minimizing the fitting error of the outer wall temperature; To introduce the constraint conditions of the heat conduction forward problem into the optimization solution process, the Lagrange multiplier method is adopted in this embodiment to incorporate the partial differential equation constraints of the forward problem into the objective function. The expression of the extended Lagrangian function is: where: Q is the heat source distribution; μ is the Lagrange multiplier, which is used to represent the penalty for the forward problem constraint; Update the distribution of the internal temperature field during iteration until the convergence condition of the objective function is satisfied. In this embodiment, the gradient descent method is adopted as the optimization algorithm; Initial conditions: Initialize the internal temperature field and the heat source distribution Q 0 , for example, take a uniform distribution or according to the empirical data value; Gradient calculation; Calculate the gradient of the objective function with respect to T int The gradient expression is; where; represents the derivative of the forward problem solver with respect to the internal temperature field; Update formula; Update the distribution of the internal temperature field. The gradient descent update formula is; where; η is the learning rate, and k represents the current iteration step; Iteration termination condition; When the change in the objective function value is less than the set threshold, stop the iteration; or when the outer wall fitting error meets the requirements, stop the calculation; In this embodiment, the finite element method is used to spatially discretize the internal temperature field to obtain the discretized objective function and constraint conditions. Implicit difference method is used for time discretization to ensure numerical stability; The discretized objective function is expressed as; where: T int represents the discretized internal temperature field; T obs represents the discretized outer wall temperature field; F(T int ) is the discretized forward problem solver; R is the discretized regularization matrix; Through the above discretization operation, the optimization problem is transformed into a problem of solving a system of linear algebraic equations, which is convenient for numerical calculation; The inverse heat conduction problem solving module takes the outer wall temperature data provided by the infrared temperature acquisition module as input and combines the discretized matrix generated by the forward heat conduction problem modeling module to solve. The resulting internal temperature field will be used by the anomaly analysis and alarm module to identify the abnormal areas inside the kiln body.

[0026] Anomaly analysis and alarm module, used to identify the abnormal areas inside the rotary kiln based on the distribution characteristics of the inner wall temperature field and send out alarm signals; Use the three-dimensional temperature field distribution output by the inverse heat conduction problem solving module to identify and analyze the abnormal areas inside the rotary kiln, and generate alarm signals according to the anomaly type and severity; The anomaly analysis and alarm module analyzes the distribution characteristics and gradient changes of the inner wall temperature field to extract possible abnormal characteristics; In this embodiment, the temperature gradient analysis unit is the basic module for anomaly recognition, which is used to calculate the gradient distribution of the temperature field inside the kiln body, so as to identify the possible gradient anomaly regions. The calculation formula of the temperature gradient is as follows; Where: T is the three-dimensional temperature field distribution output by the inverse heat conduction problem solving module; x, y, and z are spatial coordinates; To more clearly capture the change trend of the gradient, the gradient modulus is calculated as the key index for gradient analysis; Specifically, the anomaly region usually shows a significant mutation in the gradient modulus. When the refractory brick falls off, the local heat flux density increases, resulting in a significant increase in the gradient; while when the slag adheres, the local temperature is homogenized, causing the gradient to decrease; In this embodiment, the anomaly feature extraction unit is used to further analyze and extract the features of the gradient anomaly region to distinguish different anomaly types and perform pattern recognition; Region size feature: Measure the scale of the anomaly region by calculating the volume V abn of the anomaly region. The calculation formula of the volume is; Where; I is the indicator function, indicating whether the gradient modulus exceeds the threshold; Temperature deviation feature: Calculate the average temperature deviation ΔT avg ; Where; N abn is the number of grid points in the anomaly region; T ref is the reference temperature value, usually taking the average temperature of the normal region of the kiln body; Distribution pattern feature: Use principal component analysis (PCA) to extract the main distribution direction of the gradient anomaly region and calculate its aspect ratio L / W to judge whether the anomaly region is a local hot spot or a widespread adhesion phenomenon; In this embodiment, the alarm generation unit generates an alarm signal according to the extracted anomaly features and outputs the location information and type of the anomaly; The generation of the alarm signal can be based on the following rules; Significant increase in local temperature gradient: Determine the area where the local gradient modulus exceeds the threshold as the falling off of the refractory brick; Significant decrease in local temperature gradient: Determine the area where the local gradient modulus is lower than the average level as the adhesion of the slag; Specifically, the format of the alarm information can include: Abnormal type: such as refractory brick shedding or slag sticking; Abnormal location: Locate the center of the abnormal area through three-dimensional coordinates; Abnormal volume: such as the calculated V abn ; Severity: Evaluate according to the abnormal temperature deviation ΔT avg or gradient abnormal degree; Through temperature gradient analysis, abnormal feature extraction, and alarm generation, it can effectively identify and locate the abnormal area inside the rotary kiln and generate accurate alarm signals according to the abnormal type. Through reasonable feature extraction methods and classification rules, this module can not only distinguish different abnormal types but also provide important input data for subsequent trend prediction and linkage control, thus realizing the intelligent monitoring of the operating state of the rotary kiln.

[0027] Trend prediction and linkage control module, predict potential faults based on historical temperature data and linkage-adjust the operating parameters of the rotary kiln; Specifically, based on the abnormal features and historical data provided by the abnormal analysis and alarm module, predict the development trend of potential faults, and dynamically adjust the operating parameters of the rotary kiln according to the prediction results. The trend prediction and linkage control module establish a fault development model of the rotary kiln by combining multi-source data and generate control instructions; In this embodiment, the trend prediction unit first needs to extract features from historical data and establish a prediction model for the fault development trend. The historical data includes the temperature field distribution, abnormal features, and operating parameters collected during the operation of the rotary kiln; X t =[T int (t), T obs (t), R(t), u(t)] Where: T int (t) is the temperature field distribution inside the kiln body; T obs (t) is the temperature field distribution on the outer wall of the kiln; R(t) is the feature of the abnormal area; u(t) is the operating parameter vector, including rotational speed, fuel supply, and tilt angle; In this embodiment, the trend prediction unit uses a long short-term memory neural network to model and predict the fault development trend. The LSTM model can effectively handle the long-term dependence relationship of time series data and is suitable for industrial processes with dynamic and nonlinear characteristics such as rotary kilns; The input of the LSTM model is the feature vector {X t-N , X t-N+1 … X t} of the past N time steps, and the output is the fault development trend of the next M time steps where R(t) is the abnormal feature; h t = f(W h ·X t + U h ·h t-1 + b h ) where: h t is the hidden state vector; f(·) and g(·) are the activation function and the output function respectively; W h ,U h ,b h ,W o ,b o are the model parameters; In this embodiment, the linkage control unit adjusts the operating parameters of the rotary kiln according to the trend prediction result, including the kiln body rotation speed, the fuel supply amount, and the inclination angle, so as to prevent the failure from further expanding; The goal of the linkage control is to optimize the operating conditions of the kiln body and minimize the expansion speed of the abnormal area. The objective function of the optimization problem can be expressed as; where; u is the control parameter vector; R safe is the safety threshold; α is the weight coefficient, which is used to balance the failure control and the operating cost; u 0 is the initial operating parameter.

[0028] Specifically: When the predicted expansion speed of the abnormal area is relatively fast, the rotation speed of the kiln body can be reduced to reduce the internal thermal stress; When the predicted temperature deviation of the abnormal area is relatively large, the fuel supply amount can be increased to improve the uniformity of the temperature distribution; When the abnormal area is concentrated at a specific position of the kiln body, the inclination angle can be adjusted to optimize the flow path trend prediction of the material. The input of the linkage control module includes the abnormal feature provided by the abnormal analysis and alarm module and the result of the historical data modeling. The optimized parameters generated by the linkage control module will directly act on the operating control system of the rotary kiln and dynamically adjust the control strategy through the real-time collected data; Specifically, the control effect of this module will be fed back to the infrared temperature acquisition module, so as to close-loop adjust the monitoring and analysis process of the system.

[0029] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The rotary kiln online infrared temperature monitoring system is characterized by: include; Infrared temperature acquisition module, used to collect temperature data of the outer wall of the rotary kiln in real time and generate the outer wall temperature distribution; Heat conduction forward problem modeling module, based on the material parameters and operating conditions of the rotary kiln, constructs a mathematical model to describe the evolution law of the temperature field in the rotary kiln; The heat conduction inverse problem solving module infers the three-dimensional temperature field distribution of the inner wall of the rotary kiln based on the outer wall temperature data; The abnormal analysis and alarm module is used to identify abnormal areas inside the rotary kiln according to the distribution characteristics of the inner wall temperature field and issue an alarm signal; The trend prediction and linkage control module predicts potential faults based on historical temperature data and adjusts the operating parameters of the rotary kiln in a linked manner.

2. The rotary kiln online infrared temperature monitoring system according to claim 1, characterized in that: The infrared temperature acquisition module comprises: Infrared sensor acquisition units are arranged at multiple locations on the outer wall of the rotary kiln to collect temperature data at different locations on the outer wall in real time; A data filtering unit is used to filter the collected temperature data and eliminate environmental noise; The interpolation and reconstruction unit is used to interpolate and spatially reconstruct discontinuous temperature data to generate a complete outer wall temperature distribution.

3. The rotary kiln online infrared temperature monitoring system according to claim 1, characterized in that: The heat conduction forward problem modeling module includes: Heat conduction equation building unit, used to build heat conduction equation based on thermal conductivity, density and specific heat capacity parameters of the internal materials of the rotary kiln; The numerical discretization and calculation unit is used to discretize the heat conduction equation using the finite element method and generate a numerical calculation model.

4. The rotary kiln online infrared temperature monitoring system according to claim 1, characterized in that: The heat conduction inverse problem solving module includes: An optimization target construction unit is used to construct an objective function based on minimizing the difference between the outer wall temperature measurement data and the calculated temperature; Regularization processing unit, used to introduce smoothness constraints on the temperature field to avoid numerical instability in the inversion results; The optimization solution unit iteratively optimizes the objective function based on the Lagrange multiplier method and infers the internal temperature field.

5. The rotary kiln online infrared temperature monitoring system according to claim 4, characterized in that: The optimization solving unit calculates the three-dimensional internal temperature field distribution using an iterative optimization method based on the input outer wall temperature data and the heat conduction forward problem model, wherein the iteration termination condition is that the outer wall temperature measurement error is less than a preset threshold.

6. The rotary kiln online infrared temperature monitoring system according to claim 1, characterized in that: The abnormal analysis and alarm module includes: Temperature gradient analysis unit, used to calculate the gradient distribution of the inner wall temperature field and identify the location of gradient mutation; An abnormal feature extraction unit, used to extract the features of the abnormal area according to the temperature gradient and distribution pattern; The alarm generating unit is used to generate an alarm signal and output the location information and type of the abnormal area.

7. The rotary kiln online infrared temperature monitoring system according to claim 6, characterized in that: The alarm generating unit determines that a significant increase in the local temperature gradient is a refractory brick falling off, and determines that a significant decrease in the local temperature gradient is a slag sticking, based on the extracted abnormal area features.

8. The rotary kiln online infrared temperature monitoring system according to claim 1, characterized in that: The trend prediction and linkage control module includes: A historical data modeling unit for storing and extracting historical temperature data and operating parameters; Fault prediction unit, which predicts temperature change trends and potential fault occurrence times based on time series analysis algorithms; The linkage control unit is used to adjust the operating parameters of the rotary kiln in a linkage manner according to the prediction results, including the kiln body rotation speed, fuel supply and inclination angle.

9. The rotary kiln online infrared temperature monitoring system according to claim 8, characterized in that: The fault prediction unit dynamically updates the fault prediction result based on the long short-term memory neural network model and in combination with the historical temperature data and the temperature field data collected in real time.

10. The rotary kiln online infrared temperature monitoring system according to claim 8, characterized in that: The linkage control unit adjusts the kiln body rotation speed in real time according to the fault prediction result to reduce the sticking degree of the material inside the kiln body.

Images