Pre-oxidation furnace temperature real-time monitoring system and method based on cloud platform

Abstract

The invention provides a pre-oxidation furnace temperature real-time monitoring system and method based on a cloud platform, and relates to the technical field of pre-oxidation furnace monitoring, and the pre-oxidation furnace temperature real-time monitoring system based on the cloud platform comprises a data acquisition unit, a temperature estimation unit, an abnormality analysis unit and a pre-oxidation furnace control unit. According to the invention, by combining the spatial correlation analysis result between the temperature data, the temperature interpolation model is constructed to estimate the temperature data of the non-measurement position in the pre-oxidation furnace, so that the accuracy and integrity of the temperature data are improved, the temperature distribution in the pre-oxidation furnace can be known more accurately, and the accuracy of the pre-oxidation furnace is improved. Abnormality analysis is carried out according to the preset space temperature threshold diagram of the pre-oxidation furnace, temperature abnormity can be found in time, the temperature in the pre-oxidation furnace is monitored in real time and accurately estimated, the system can optimize operation parameters of the pre-oxidation furnace, and the production efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pre-oxidation furnace monitoring, and in particular to a cloud platform-based pre-oxidation furnace temperature real-time monitoring system and method. Background Art

[0002] The pre-oxidation furnace plays a vital role in the carbon fiber preparation process, especially for the pre-oxidation treatment of PAN (polyacrylonitrile) based carbon fibers, it is an indispensable key equipment. In the pre-oxidation furnace, PAN fibers react violently with oxygen under the catalysis of high temperature environment, undergo a series of complex transformation processes such as cyclization, cross-linking and thermal cracking, and finally form a trapezoidal structure with high thermal stability. The formation of this structure lays a solid foundation for the subsequent carbonization and graphitization steps, ensuring the excellent performance of carbon fiber products.

[0003] In view of the importance of pre-oxidation treatment for the structural transformation of carbon fiber precursor, the pre-oxidation furnace must be strictly temperature-controlled during use. Accurate temperature control not only ensures that the carbon fiber precursor undergoes the necessary chemical reactions during the pre-oxidation process, thereby smoothly transforming into a heat-resistant ladder structure, but also plays a vital role in improving product quality, ensuring production efficiency, and extending the service life of equipment.

[0004] At present, in the existing technical system, the temperature monitoring system of the pre-oxidation furnace faces some significant challenges. Specifically, these systems may not be able to obtain the temperature data of all key locations in the furnace in real time and comprehensively, and the setting of monitoring points is relatively limited, making it difficult to achieve effective coverage of all key areas inside the pre-oxidation furnace. Especially in complex and changeable industrial environments, this lack of monitoring capability seriously limits the staff's immediate response to the conditions in the furnace, which not only increases the potential risks in the production process, but also may lead to a significant decrease in production efficiency.

[0005] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0006] In view of this, the present invention provides a pre-oxidation furnace temperature real-time monitoring system and method based on a cloud platform to solve the above-mentioned problems.

[0007] In order to solve the above problems, the specific technical solutions adopted by the present invention are as follows: According to one aspect of the present invention, a cloud platform-based real-time monitoring system for the temperature of a pre-oxidation furnace is provided. The cloud platform-based real-time monitoring system for the temperature of a pre-oxidation furnace comprises: a data acquisition unit, a temperature estimation unit, an abnormality analysis unit and a pre-oxidation furnace control unit; A data acquisition unit is used to acquire temperature data at different measurement locations inside the pre-oxidation furnace and establish a temperature monitoring database based on the cloud platform. The measurement locations include the heating element area, material inlet and outlet, material handling area and furnace wall; The temperature estimation unit is used to estimate the temperature data of non-measurement positions inside the pre-oxidation furnace based on the temperature monitoring database and in combination with the results of spatial correlation analysis between the temperature data, using the constructed temperature interpolation model; and establish a temperature estimation database based on the estimated temperature data; An abnormality analysis unit is used to construct a pre-oxidation furnace temperature database using a temperature monitoring database and a temperature estimation database, and to perform abnormality analysis according to a preset pre-oxidation furnace space temperature threshold diagram to obtain an abnormality analysis result; The pre-oxidation furnace control unit is used to execute corresponding pre-oxidation furnace control strategies based on abnormal analysis results and pre-oxidation furnace early warning rules.

[0008] Preferably, the data acquisition unit includes: a temperature data acquisition module, a temperature data transmission module, a temperature data preprocessing and temperature monitoring database establishment module; A temperature data acquisition module, used to collect temperature data using sensors pre-installed at different measuring positions inside the pre-oxidation furnace; A temperature data transmission module is used to transmit the collected temperature data to a data receiving terminal of the cloud platform by wireless means; Temperature data preprocessing is used to preprocess the received temperature data at the data receiving end of the cloud platform. The preprocessing includes data cleaning, data denoising and format conversion. The temperature monitoring database establishment module is used to store the pre-processed temperature data in the database of the cloud platform and establish a temperature monitoring database based on the measurement position of the sensor.

[0009] Preferably, the temperature estimation unit includes: a data extraction module, a correlation analysis module, a model construction module and an estimation database construction module; A data extraction module, used to extract the temperature data of all measured locations and their measurement locations from the temperature monitoring database; A correlation analysis module is used to analyze the spatial correlation between the heating element area and the material inlet and outlet, the material processing area and the furnace wall respectively using the correlation analysis method according to the temperature data of the measured position and its measurement position; A model building module is used to build a temperature interpolation model based on the spatial correlation analysis results using a Kriging interpolation algorithm based on a simulated annealing artificial bee colony hybrid algorithm; The estimation database building module is used to estimate the temperature data of non-measurement positions inside the pre-oxidation furnace by using a temperature interpolation model, and to establish a temperature estimation database based on the estimation results.

[0010] Preferably, according to the temperature data of the measured position and its measurement position, the spatial correlation between the heating element area and the material inlet and outlet, the material processing area and the furnace wall is analyzed by using the correlation analysis method, including: For the temperature data of the heating element area, the material inlet and outlet, the material handling area and the furnace wall, the Pearson correlation coefficient method is used to calculate the correlation coefficients between the heating element area and the material inlet and outlet, the material handling area and the furnace wall respectively; For the measurement position of the heating element area, the spatial distance between the measurement position of the heating element area and the temperature data measurement positions of the material inlet and outlet, the material processing area and the furnace wall is calculated using the Euclidean distance; Construct a spatial weight matrix based on the calculated spatial distance and correlation coefficient; According to the spatial weight matrix, the Moran index is used to evaluate the spatial correlation between the temperature data of the heating element area and the temperature data of the material inlet and outlet, material handling area and furnace wall.

[0011] Preferably, based on the spatial correlation analysis results, constructing a temperature interpolation model using a Kriging interpolation algorithm based on a simulated annealing artificial bee colony hybrid algorithm includes: Based on the results of spatial correlation analysis, the temperature data of the measurement locations are grouped, and the inter-group distances between the groups are calculated; According to the inter-group distance and temperature data between the groups, the spatial relationship function of each group is calculated; The spatial relationship function is fitted based on the support vector machine algorithm to obtain the theoretical variation function model; The theoretical variation function model is mapped to the simulated annealing artificial bee colony hybrid algorithm, and the temperature interpolation model is obtained through iterative optimization.

[0012] Preferably, the spatial relationship function is fitted based on the support vector machine algorithm to obtain a theoretical variation function model including: The inter-group distances between groups and the spatial relationship functions of each group are combined into a set of data point pairs; The support vector machine regression model is constructed using the Gaussian radial basis kernel function; Configure the parameters of the support vector machine and train the support vector machine model using a set of data point pairs; The support vector machine regression function obtained after training is used as the theoretical variation function model.

[0013] Preferably, the theoretical variation function model is mapped to the simulated annealing artificial bee colony hybrid algorithm, and the temperature interpolation model is obtained through iterative optimization, including: The parameters to be optimized are extracted from the theoretical variogram model, and the extracted parameters are mapped to the simulated annealing artificial bee colony hybrid algorithm; Initialize the simulated annealing artificial bee colony hybrid algorithm; The hybrid strategy of simulated annealing algorithm and artificial bee colony algorithm is used to iteratively optimize the parameters of the mapped variogram model to obtain the optimal parameter combination. The theoretical variogram model is updated using the optimal parameter combination to obtain the temperature interpolation model.

[0014] Preferably, the abnormality analysis unit includes: a pre-oxidation furnace temperature database construction module, a temperature threshold map construction module and an abnormal state judgment module; A pre-oxidation furnace temperature database building module is used to integrate the temperature data in the temperature monitoring database and the temperature estimation database, and to establish a pre-oxidation furnace temperature database; A temperature threshold map building module is used to build a temperature threshold map of the pre-oxidation furnace space based on the properties of the processed materials of the pre-oxidation furnace and in combination with the design parameters and operation characteristics of the pre-oxidation furnace; The abnormal state judgment module is used to extract the temperature data of each position from the pre-oxidation furnace temperature database, and compare it with the pre-oxidation furnace space temperature threshold map, and based on the comparison result, make an abnormal state judgment on the pre-oxidation furnace.

[0015] Preferably, based on the properties of the processed materials of the pre-oxidation furnace and in combination with the design parameters and operation characteristics of the pre-oxidation furnace, establishing a pre-oxidation furnace space temperature threshold diagram includes: Based on the design parameters and operating characteristics of the pre-oxidation furnace, the temperature distribution inside the pre-oxidation furnace is simulated using thermal simulation tools; According to the simulation results and the properties of the materials to be processed in the pre-oxidation furnace, temperature thresholds are set for different areas inside the pre-oxidation furnace; Based on the set temperature threshold, draw the spatial temperature threshold map of the pre-oxidation furnace and mark the location of the temperature monitoring point.

[0016] According to another aspect of the present invention, a method for real-time monitoring of the temperature of a pre-oxidation furnace based on a cloud platform is provided. The method for real-time monitoring of the temperature of a pre-oxidation furnace based on a cloud platform comprises the following steps: S1. Obtain temperature data at different measurement locations inside the pre-oxidation furnace, and establish a temperature monitoring database based on the cloud platform. The measurement locations include the heating element area, material inlet and outlet, material handling area, and furnace wall; S2. Based on the temperature monitoring database and in combination with the results of spatial correlation analysis between temperature data, the temperature data of non-measurement positions inside the pre-oxidation furnace are estimated using the constructed temperature interpolation model; and a temperature estimation database is established based on the estimated temperature data; S3. Construct a pre-oxidation furnace temperature database using the temperature monitoring database and the temperature estimation database, and perform an abnormal analysis according to a preset pre-oxidation furnace space temperature threshold diagram to obtain an abnormal analysis result; S4. Based on the abnormal analysis results, the corresponding pre-oxidation furnace control strategy is executed according to the pre-oxidation furnace early warning rules.

[0017] The beneficial effects of the present invention are: 1. The present invention obtains temperature data at different measurement positions inside the pre-oxidation furnace and establishes a temperature monitoring database through a cloud platform, thereby realizing centralized management and storage of data, which is helpful for comprehensive and real-time monitoring of the temperature condition of the pre-oxidation furnace. By combining the spatial correlation analysis results between the temperature data, a temperature interpolation model is constructed to estimate the temperature data of non-measurement positions inside the pre-oxidation furnace, thereby improving the accuracy and completeness of the temperature data and helping to more accurately understand the temperature distribution inside the pre-oxidation furnace. A pre-oxidation furnace temperature database is constructed using a temperature monitoring database and a temperature estimation database, and anomaly analysis is performed according to a pre-set pre-oxidation furnace spatial temperature threshold map, so that temperature anomalies can be discovered in time. By real-time monitoring and accurate estimation of the temperature inside the pre-oxidation furnace, the system can optimize the operating parameters of the pre-oxidation furnace and improve production efficiency.

[0018] 2. The present invention utilizes the Pearson correlation coefficient method and Euclidean distance calculation, combined with the Moran index evaluation, to conduct in-depth spatial correlation analysis of temperature data at different measurement locations, which helps to reveal the intrinsic connections and laws between temperature data, and provides a scientific basis for constructing a temperature interpolation model. The Kriging interpolation algorithm based on the simulated annealing artificial bee colony hybrid algorithm is adopted, and combined with the spatial correlation analysis results, a high-precision temperature interpolation model is constructed. This hybrid algorithm can make full use of the advantages of the two algorithms, improve the convergence speed and optimization effect of the model, thereby more accurately estimating the temperature data at non-measurement locations inside the pre-oxidation furnace, and obtain the optimal parameter combination through iterative optimization. These parameters are used to update the theoretical variation function model to obtain the final temperature interpolation model, which can ensure the accuracy and rationality of the model parameters, thereby improving the accuracy and reliability of the temperature interpolation model.

[0019] 3. The present invention integrates the temperature data in the temperature monitoring database and the temperature estimation database to establish a unified pre-oxidation furnace temperature database, which helps to achieve comprehensive and unified management of the temperature data inside the pre-oxidation furnace and provide a reliable data basis for subsequent temperature anomaly analysis. By integrating temperature data, scientifically setting temperature thresholds and accurately judging abnormal states, the abnormal analysis unit can significantly improve the accuracy and efficiency of abnormal analysis, help to promptly discover and solve temperature problems in the operation of the pre-oxidation furnace, ensure the stable operation of the equipment and product quality, and its advantages in abnormal analysis provide a strong guarantee for the safe and efficient operation of the pre-oxidation furnace. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings: Figure 1 It is a principle block diagram of a real-time monitoring system for pre-oxidation furnace temperature based on a cloud platform according to an embodiment of the present invention; Figure 2 It is a flow chart of a method for real-time monitoring temperature of a pre-oxidation furnace based on a cloud platform according to an embodiment of the present invention.

[0021] In the figure: 1. Data acquisition unit; 2. Temperature estimation unit; 3. Abnormal analysis unit; 4. Pre-oxidation furnace control unit. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0023] According to an embodiment of the present invention, a system and method for real-time monitoring of the temperature of a pre-oxidation furnace based on a cloud platform are provided.

[0024] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to one embodiment of the present invention, a pre-oxidation furnace temperature real-time monitoring system based on a cloud platform is provided, and the pre-oxidation furnace temperature real-time monitoring system based on a cloud platform includes: a data acquisition unit 1, a temperature estimation unit 2, an abnormality analysis unit 3 and a pre-oxidation furnace control unit 4; The data acquisition unit 1 is used to acquire the temperature data of different measurement positions inside the pre-oxidation furnace and establish a temperature monitoring database based on the cloud platform. The measurement positions include the heating element area, the material inlet and outlet, the material processing area and the furnace wall; It should be noted that the heating element is one of the most critical parts of the pre-oxidation furnace and is responsible for providing heat. Placing temperature sensors near the heating elements can monitor the working status of the heating elements and the temperature distribution in the furnace in real time. The locations where materials enter and leave the pre-oxidation furnace are also important monitoring points. The temperature data at these locations helps to understand the state changes of the materials when entering and leaving the pre-oxidation furnace. Arranging temperature sensors in key areas of material handling ensures the temperature uniformity and stability of the materials during the pre-oxidation process. The temperature changes on the furnace wall can reflect the insulation effect of the pre-oxidation furnace and the health of the furnace body.

[0025] As a preferred embodiment, the data acquisition unit 1 includes: a temperature data acquisition module, a temperature data transmission module, a temperature data preprocessing and temperature monitoring database establishment module; A temperature data acquisition module, used to collect temperature data using sensors pre-installed at different measuring positions inside the pre-oxidation furnace; Specifically, before collecting temperature data, it is necessary to select a suitable temperature sensor according to the working environment and temperature range of the pre-oxidation furnace. Common types of temperature sensors include thermocouples, thermal resistors, etc. In the pre-oxidation furnace, the sensor should have the characteristics of high precision, high stability, and strong anti-interference ability to ensure that the collected temperature data is accurate and reliable. Sensors are installed at different locations inside the pre-oxidation furnace to fully cover the temperature distribution in the furnace. The installation location should take into account factors such as temperature gradient, heat flow direction, and material properties.

[0026] A temperature data transmission module is used to transmit the collected temperature data to a data receiving terminal of the cloud platform by wireless means; It should be noted that the cloud platform, also known as cloud computing platform, is a distributed computing platform composed of hardware, software and services. It provides on-demand computing resources, storage resources and network resources, and users can access these resources remotely through the Internet. The core of the cloud platform is to virtualize physical resources (such as servers, storage devices and networks) and provide flexible and scalable services to users through the Internet.

[0027] In the field of Internet of Things, there are various wireless data transmission methods, including Wi-Fi, Bluetooth, Zigbee, LoRa, NB-IoT, GPRS, etc. For the real-time temperature monitoring system of the pre-oxidation furnace, since the temperature data needs to be transmitted to the cloud platform in real time, and the pre-oxidation furnace may be located in a relatively harsh industrial environment, a wireless transmission method with a longer transmission distance, lower power consumption, higher stability and reliability is selected, such as Wi-Fi, GPRS or NB-IoT.

[0028] Temperature data preprocessing is used to preprocess the received temperature data at the data receiving end of the cloud platform. The preprocessing includes data cleaning, data denoising and format conversion. Specifically, data cleaning aims to remove invalid, incomplete or erroneous data and ensure the accuracy and completeness of the data, including removing invalid data, filling missing values ​​and consistency checks.

[0029] Data denoising is to remove random noise in the data, improve the signal-to-noise ratio of the data, and make the data smoother. It includes filtering processing and outlier detection.

[0030] Format conversion is to unify the data format so that it meets the requirements of subsequent processing, including data type conversion and unit unification.

[0031] For example, suppose some of the received temperature data is expressed in degrees Celsius (°C) and some is expressed in degrees Fahrenheit (°F). All data needs to be converted to the same unit, such as degrees Celsius. For example, a temperature data sequence [100, 212, 102, 103], where 212°F needs to be converted to degrees Celsius, that is, (212-32)*5 / 9=100°C, so the processed data sequence is [100, 100, 102, 103].

[0032] The temperature monitoring database establishment module is used to store the pre-processed temperature data in the database of the cloud platform and establish a temperature monitoring database based on the measurement position of the sensor.

[0033] It should be understood that cloud platforms usually provide a variety of database services, such as relational databases (such as MySQL, PostgreSQL) and non-relational databases (such as MongoDB, DynamoDB), and insert the pre-processed temperature data into the database table. This is usually achieved through SQL statements (for relational databases) or API calls (for non-relational databases).

[0034] The temperature estimation unit 2 is used to estimate the temperature data of the non-measurement position inside the pre-oxidation furnace based on the temperature monitoring database and in combination with the spatial correlation analysis results between the temperature data by using the constructed temperature interpolation model; and establish a temperature estimation database according to the estimated temperature data; As a preferred embodiment, the temperature estimation unit 2 includes: a data extraction module, a correlation analysis module, a model construction module and an estimation database construction module; A data extraction module, used to extract the temperature data of all measured locations and their measurement locations from the temperature monitoring database; A correlation analysis module is used to analyze the spatial correlation between the heating element area and the material inlet and outlet, the material processing area and the furnace wall respectively using the correlation analysis method according to the temperature data of the measured position and its measurement position; As a preferred embodiment, according to the temperature data of the measured position and its measurement position, the spatial correlation between the heating element area and the material inlet and outlet, the material processing area and the furnace wall is analyzed by using the correlation analysis method, including: For the temperature data of the heating element area, the material inlet and outlet, the material handling area and the furnace wall, the Pearson correlation coefficient method is used to calculate the correlation coefficients between the heating element area and the material inlet and outlet, the material handling area and the furnace wall respectively; Specifically, the Pearson correlation coefficient is used to measure the linear correlation between two variables. For each temperature data, the correlation coefficient between different measurement locations can be calculated.

[0035] Assume that n measurement locations and N The temperature data at each time point can form a n × m Then, for each pair of measurement locations ( i , j ), the Pearson correlation coefficient between them can be calculated.

[0036] The formula for the Pearson correlation coefficient is: ; In the formula, r jj Indicates the measurement position i and measurement location j The Pearson correlation coefficient of the temperature data, x ik and x jk Respectively indicate the measurement position i and measurement location j In the k Temperature data at a time point, and Respectively indicate the measurement position i and measurement location j Average temperature data, N Indicates the number of time points.

[0037] For the measurement position of the heating element area, the spatial distance between the measurement position of the heating element area and the temperature data measurement positions of the material inlet and outlet, the material processing area and the furnace wall is calculated using the Euclidean distance; Specifically, Euclidean distance is a method for calculating the straight-line distance between two points, which is suitable for distance calculation in multidimensional space.

[0038] Construct a spatial weight matrix based on the calculated spatial distance and correlation coefficient; Specifically, constructing a spatial weight matrix is ​​an important step in spatial statistical analysis, which is used to describe the degree of interaction or influence between different spatial locations. The Pearson correlation coefficient between the heating element area and the material inlet and outlet, material handling area and furnace wall, as well as the spatial distance between these areas based on Euclidean distance, are calculated. Next, this information will be used to construct a spatial weight matrix.

[0039] For example, the spatial weight matrix is ​​usually a symmetric matrix with elements w ij Indicates the measurement position i and measurement location j The weight or influence between them. The weight can be a function based on the distance, a function based on the correlation coefficient, or both.

[0040] A common way to construct a spatial weight matrix is ​​to use the inverse of the distance as the weight, that is: w ij =1 / d ij ; in, d ij Is the measurement position i and measurement location j This method assumes that closer locations have a greater influence on each other.

[0041] However, since the correlation coefficient is taken into account, the weight can be defined as a combination of distance and correlation coefficient. For example, the following formula can be used: w ij= r ij / d ij ; in, r ij Is the measurement position i and measurement location j The Pearson correlation coefficient between d ij is the Euclidean distance between them. This weight definition takes into account the spatial distance between locations.

[0042] According to the spatial weight matrix, the Moran index is used to evaluate the spatial correlation between the temperature data of the heating element area and the temperature data of the material inlet and outlet, material handling area and furnace wall.

[0043] Moran's I is a statistical indicator used to assess the global correlation of spatial data. It takes into account the similarity between the spatial weight matrix and the temperature data to reveal spatial clustering patterns. The value of Moran's I ranges from -1 to 1, where: Values ​​close to 1 indicate strong positive spatial correlation, that is, places with high temperatures tend to cluster together, and places with low temperatures also tend to cluster together. Values ​​close to -1 indicate strong negative spatial correlation, that is, places with high temperatures tend to be adjacent to places with low temperatures. Values ​​close to 0 indicate random spatial distribution, with no obvious clustering pattern.

[0044] A model building module is used to build a temperature interpolation model based on the spatial correlation analysis results using a Kriging interpolation algorithm based on a simulated annealing artificial bee colony hybrid algorithm; As a preferred implementation, based on the results of spatial correlation analysis, a temperature interpolation model is constructed using a Kriging interpolation algorithm based on a simulated annealing artificial bee colony hybrid algorithm, including: Based on the results of spatial correlation analysis, the temperature data of the measurement locations are grouped, and the inter-group distances between the groups are calculated; Specifically, the present invention groups the temperature data of the measurement positions by a clustering algorithm, which specifically includes the following steps: The correlation coefficient and spatial distance are standardized to the interval [0,1] respectively, and the standardized correlation coefficient and spatial distance are combined to form a comprehensive weight matrix. This matrix comprehensively considers the influence of correlation and spatial distance.

[0045] Based on the comprehensive weight matrix W, the following grouping strategies can be adopted: Use a clustering algorithm (such as spectral clustering) to divide the measurement locations into several groups. The specific steps are as follows: Construct similarity matrix: Use the comprehensive weight matrix W as the similarity matrix; Construct the Laplace matrix: Calculate the Laplace matrix L, where L=D W, D are degree matrices (the diagonal is the degree of the node, and the rest are 0); Eigendecomposition: Perform eigendecomposition on the Laplace matrix L to obtain the eigenvector; Select eigenvectors: Select the first k eigenvectors (k is the number of clusters) to form the matrix U; Clustering: Cluster the rows of matrix U. You can use the K-means algorithm or other clustering methods.

[0046] In addition, the inter-group distance between groups can be expressed as the distance between the centroids.

[0047] According to the inter-group distance and temperature data between the groups, the spatial relationship function of each group is calculated; Specifically, the spatial relationship function, also known as the semivariogram, is a statistical tool used to describe the variability of spatial data, which can help understand the spatial correlation of data. The spatial relationship function reflects how the differences between observations change as the distance increases.

[0048] The calculation formula of the spatial relationship function is: ; In the formula, G represents the spatial relationship function, T ( h ) represents the distance between two sets of data. Z ( m i ) indicates the measurement position m i Temperature data, Z( m i + h ) indicates the measurement position m i + h Temperature data, i The index representing the measurement location.

[0049] The spatial relationship function is fitted based on the support vector machine algorithm to obtain the theoretical variation function model; As a preferred implementation, the spatial relationship function is fitted based on the support vector machine algorithm to obtain a theoretical variation function model including: The inter-group distances between groups and the spatial relationship functions of each group are combined into a set of data point pairs; The support vector machine regression model is constructed using the Gaussian radial basis kernel function; Configure the parameters of the support vector machine and train the support vector machine model using a set of data point pairs; The support vector machine regression function obtained after training is used as the theoretical variation function model.

[0050] Specifically, Gaussian Radial Basis Function (RBF) is a kernel function widely used in support vector machines (SVM) and other machine learning models. This kernel function can map input data to a high-dimensional space, making it easier to solve problems that are difficult to linearly segment in low-dimensional space.

[0051] Among them, the expression of the support vector machine regression model is: f(e)=ω T (e)+b; Where f(e) represents the predicted spatial relationship function, ω represents the weight vector, (e) represents the feature mapping function, which maps the input e to a high-dimensional space, b represents the bias term, and T represents the transpose of the matrix.

[0052] In addition, configure the parameters of the support vector machine. The parameters of the support vector machine mainly include: Penalty coefficient C: This is a parameter used to control the complexity of the model. A larger C value means a lower tolerance for training errors, which may lead to overfitting; a smaller C value means a looser error tolerance, which helps avoid overfitting.

[0053] Kernel function parameter γ: This is an important parameter when using the Gaussian radial basis kernel function (RBF). The size of γ determines the width of the kernel function. A larger γ value will make the model more sensitive and may introduce overfitting; a smaller γ value will make the model smoother and help improve generalization ability.

[0054] Training the support vector machine model using the configured parameters usually requires model selection and tuning. After the training is completed, the obtained support vector machine regression function is used as the theoretical variation function model.

[0055] The theoretical variation function model is mapped to the simulated annealing artificial bee colony hybrid algorithm, and the temperature interpolation model is obtained through iterative optimization.

[0056] As a preferred implementation, the theoretical variation function model is mapped to the simulated annealing artificial bee colony hybrid algorithm, and the temperature interpolation model is obtained through iterative optimization, including: The parameters to be optimized are extracted from the theoretical variogram model, and the extracted parameters are mapped to the simulated annealing artificial bee colony hybrid algorithm; Initialize the simulated annealing artificial bee colony hybrid algorithm; The hybrid strategy of simulated annealing algorithm and artificial bee colony algorithm is used to iteratively optimize the parameters of the mapped variogram model to obtain the optimal parameter combination. The theoretical variogram model is updated using the optimal parameter combination to obtain the temperature interpolation model.

[0057] Specifically, once the parameters to be optimized are determined, the next task is to map these parameters into the simulated annealing artificial bee colony hybrid algorithm. The specific steps are as follows: Step 1: Define the search space Define the search range for C and γ. Typically, C can range from a small positive number (such as 0.01) to a large number (such as 1000), while γ usually ranges from a small number close to zero (such as 0.001) to a relatively large number (such as 1).

[0058] Step 2: Initialize the simulated annealing artificial bee colony hybrid algorithm The initialization algorithm requires defining several key parameters: Initial temperature: The initial temperature of the simulated annealing algorithm determines the probability of accepting new solutions in the initial stage.

[0059] Cooling rate: The ratio of temperature reduction after each iteration.

[0060] Iteration number: The number of iterations in the entire optimization process.

[0061] Colony size: the number of hired bees, scout bees and waiting bees in a colony.

[0062] Step 3: Set the evaluation function Define an evaluation function to measure the quality of the model. In this case, this function may be the mean squared error (MSE), or any other indicator that reflects the prediction accuracy of the model.

[0063] Step 4: Run the simulated annealing artificial bee colony hybrid algorithm Simulated annealing part: In each iteration, a new combination of C and γ is randomly generated according to the current temperature, and the model performance under the new combination is calculated. If the performance of the new combination is better or meets a certain acceptance probability, the new combination is accepted and the current solution is updated.

[0064] Artificial bee colony part: In the bee colony, employed bees and scout bees perform local searches based on the current best solution to find a better combination of C and γ. Waiting bees decide whether to abandon the current search area and explore a new area based on a certain probability.

[0065] Step 5: Iterative Optimization The steps of simulated annealing and artificial bee colony algorithm are repeatedly executed until a predetermined number of iterations or convergence condition is reached.

[0066] Step 6: Get the optimal parameter combination After multiple rounds of iterations, a set of C and γ parameter combinations that give the best model performance is found.

[0067] The estimation database building module is used to estimate the temperature data of non-measurement positions inside the pre-oxidation furnace by using a temperature interpolation model, and to establish a temperature estimation database based on the estimation results.

[0068] It should be noted that by analyzing the coordinate data of all non-measurement positions in the pre-oxidation furnace where the temperature needs to be estimated, the temperature of each non-measurement position is estimated using the temperature interpolation model. The specific steps are as follows: Step 1: Standardize coordinate data Since the coordinate data are normalized when training the SVM model, the coordinate data of non-measured locations also need to be normalized before estimation.

[0069] Step 2: Apply the model to make predictions The standardized coordinate data are input into the support vector machine model to obtain the predicted temperature value.

[0070] Store the estimated temperature data in a structured database for easy query and analysis. The database should contain at least the following information: Position coordinates: X, Y, and Z coordinates of non-measured locations.

[0071] Estimated Temperature: The estimated temperature value at this location.

[0072] Estimation time: Record the timestamp of the estimate to facilitate tracking of data changes.

[0073] The abnormality analysis unit 3 is used to construct a pre-oxidation furnace temperature database using the temperature monitoring database and the temperature estimation database, and to perform abnormality analysis according to a preset pre-oxidation furnace space temperature threshold diagram to obtain an abnormality analysis result; As a preferred embodiment, the abnormality analysis unit 3 includes: a pre-oxidation furnace temperature database construction module, a temperature threshold map construction module and an abnormal state judgment module; A pre-oxidation furnace temperature database building module is used to integrate the temperature data in the temperature monitoring database and the temperature estimation database, and to establish a pre-oxidation furnace temperature database; A temperature threshold map building module is used to build a temperature threshold map of the pre-oxidation furnace space based on the properties of the processed materials of the pre-oxidation furnace and in combination with the design parameters and operation characteristics of the pre-oxidation furnace; As a preferred implementation, based on the properties of the processed materials of the pre-oxidation furnace and in combination with the design parameters and operation characteristics of the pre-oxidation furnace, a pre-oxidation furnace space temperature threshold diagram is established, including: Based on the design parameters and operating characteristics of the pre-oxidation furnace, the temperature distribution inside the pre-oxidation furnace is simulated using thermal simulation tools; It should be understood that in order to establish the space temperature threshold map of the pre-oxidation furnace, a series of orderly steps need to be followed, which combine the design characteristics of the pre-oxidation furnace, operating conditions and the nature of the processed materials. The following is a detailed process: Collect the design parameters of the pre-oxidation furnace, including: Size and shape: The length, width, height and any special geometry of the furnace.

[0074] Material properties: thermal conductivity, specific heat capacity, etc. of furnace walls, heating elements and insulation materials.

[0075] Heating and cooling systems: heating methods (such as electric heating, gas heating, etc.), cooling mechanisms (such as natural cooling, forced air cooling, etc.).

[0076] Airflow and pressure: flow pattern, inlet and outlet locations and pressure distribution of gases in the furnace.

[0077] Determine operational characteristics, including: Working temperature range: the temperature range required for the pre-oxidation process.

[0078] Heating and Cooling Rates: The rate at which temperature changes.

[0079] Processing time: The time the material stays in the furnace.

[0080] Understand the nature of the materials being handled, including: Thermal conductivity and thermal diffusivity: the ability of a material to conduct and diffuse heat.

[0081] Thermal stability and reaction characteristics: the stability of the material at high temperatures and possible chemical reactions Specifically, by selecting a thermal simulation tool such as ANSYS, a geometric model is constructed according to the actual size and shape of the pre-oxidation furnace. Then material properties are assigned to the furnace body, heating elements, and insulation materials. Boundary conditions and initial conditions are set, such as heating power, cooling rate, ambient temperature, etc. Airflow and pressure conditions are defined. Finally, a thermal simulation is performed to simulate the temperature distribution inside the pre-oxidation furnace.

[0082] According to the simulation results and the properties of the processed materials in the pre-oxidation furnace, temperature thresholds are set for different areas inside the pre-oxidation furnace; Specifically, setting temperature thresholds for different areas inside the pre-oxidation furnace is a comprehensive decision-making process based on thermal simulation results, the properties of the processed materials, and the pre-oxidation process requirements. The following are the general steps and considerations for setting temperature thresholds for the pre-oxidation furnace: 1. Analyze thermal simulation results Temperature distribution: First, carefully analyze the temperature distribution diagram in the thermal simulation results to identify high temperature areas, low temperature areas, and areas with large temperature gradients.

[0083] Hot and Cold Spots: Identify the hot spots (highest temperature points) and cold spots (lowest temperature points) within the furnace, as well as their location and number.

[0084] Temperature uniformity: It is particularly important to evaluate the uniformity of the temperature in the furnace, especially for the pre-oxidation process which requires strict control of temperature uniformity.

[0085] 2. Consider the nature of the material being processed Thermal stability: Understand the thermal stability of the material being processed at high temperatures and determine the maximum temperature it can withstand to avoid overheating of the material resulting in performance degradation or damage.

[0086] Reaction characteristics: Consider the temperature requirements of the pre-oxidation reaction and determine the most suitable reaction temperature range.

[0087] Temperature sensitivity: For temperature-sensitive materials, a stricter temperature control range needs to be set to ensure product quality.

[0088] 3. Combined with the pre-oxidation process requirements Process temperature: Determine the target temperature of each area in the furnace according to the requirements of the pre-oxidation process.

[0089] Heating and cooling rates: Consider the impact of heating and cooling rates on the pre-oxidation process and set a reasonable temperature change rate.

[0090] Processing time: Combined with the processing time and temperature requirements, determine the temperature thresholds of different areas to ensure sufficient pre-oxidation reaction.

[0091] 4. Set temperature threshold High temperature zone: For the high temperature zone, an upper temperature threshold should be set to prevent the material from overheating. This threshold should be lower than the thermal stability limit of the material and leave a certain safety margin.

[0092] Low temperature zone: For the low temperature zone, a lower temperature threshold should be set to ensure that the pre-oxidation reaction can proceed sufficiently. This threshold should be higher than the minimum temperature required for the reaction.

[0093] Transition zone: For transition zones with large temperature gradients, a stricter temperature control range should be set to reduce the impact of thermal stress on materials and furnace body.

[0094] Safety margin: When setting the temperature threshold, a certain safety margin should be considered to cope with possible temperature fluctuations and measurement errors.

[0095] For example, assume that the interior of the pre-oxidation furnace is divided into three zones: heating zone, reaction zone, and cooling zone. Based on the thermal simulation results and the properties of the processed materials, the following temperature thresholds can be set for these three zones: Heating zone: The upper temperature threshold is set at 400°C (below the material thermal stability limit of 450°C), and the lower temperature threshold is set at 300°C (to ensure a moderate heating rate).

[0096] Reaction zone: The temperature threshold is set at 350°C ± 10°C (to ensure that the pre-oxidation reaction proceeds within the optimal temperature range).

[0097] Cooling zone: The upper temperature threshold is set to 200°C (to prevent material cracking due to excessive cooling), and the lower temperature threshold is set to ambient temperature (to ensure cooling effect).

[0098] Based on the set temperature threshold, draw the spatial temperature threshold map of the pre-oxidation furnace and mark the location of the temperature monitoring point.

[0099] It should be noted that, according to the actual size and shape of the pre-oxidation furnace, a two-dimensional or three-dimensional contour map of the furnace body is drawn, and according to the setting of the temperature threshold, the space inside the furnace is divided into different temperature zones. Different colors or patterns can be used to represent these zones.

[0100] In each temperature zone, label the temperature threshold range for that zone. For example, use "350°C ± 10°C" to represent the temperature range of the reaction zone. Add text descriptions next to or below the graphic to explain the basis and purpose of setting the temperature threshold.

[0101] The abnormal state judgment module is used to extract the temperature data of each position from the pre-oxidation furnace temperature database, and compare it with the pre-oxidation furnace space temperature threshold map, and based on the comparison result, make an abnormal state judgment on the pre-oxidation furnace.

[0102] Specifically, to extract the temperature data of each position from the pre-oxidation furnace temperature database and compare it with the pre-oxidation furnace space temperature threshold map to determine the abnormal state of the pre-oxidation furnace, the following steps can be followed: 1. Extracting temperature data Connecting to the database: First, you need to connect to the database that stores the pre-oxidation furnace temperature data. This usually requires the use of database management software or programming languages ​​(such as SQL, Python, etc.) to achieve this.

[0103] Query data: Query the temperature data of each position of the pre-oxidation furnace in the database. Make sure the extracted data includes the number, location, timestamp and temperature value of the temperature monitoring point.

[0104] Data organization: Organize the extracted data into a format that is easy to analyze, such as tables or charts.

[0105] 2. Comparison of temperature data and threshold graph Loading threshold graph: Load the pre-oxidation furnace space temperature threshold graph into the analysis tool. This can be a graphic file in electronic form (such as PDF, PNG, etc.) or a graphical interface integrated directly in the database management software.

[0106] Data mapping: Map the extracted temperature data to the threshold map. According to the number and location of the temperature monitoring point, mark the corresponding temperature value at the corresponding position of the threshold map.

[0107] Comparative Analysis: Compare temperature data with the temperature range on the threshold graph. Check if the temperature at any location exceeds the set threshold range.

[0108] 3. Abnormal status judgment Identify anomalies: If the temperature at a location is outside the threshold range, the location is marked as an anomaly. Anomalies may indicate that the pre-oxidation furnace is overheating, insufficient cooling, or uneven temperature distribution.

[0109] Assess severity: Assess the severity of the anomaly based on the abnormal temperature value and duration. Severe anomalies may require immediate action, such as stopping the pre-oxidation process, adjusting the heating or cooling system, etc.

[0110] Record and report: Record relevant information of abnormal status, including abnormal location, time, temperature value and possible cause. Report this information to relevant personnel so that they can take appropriate actions.

[0111] The pre-oxidation furnace control unit 4 is used to execute a corresponding pre-oxidation furnace control strategy based on the abnormal analysis result and the pre-oxidation furnace early warning rule.

[0112] Specifically, based on the abnormal analysis results, when the pre-oxidation furnace is in an abnormal state, the corresponding control strategy needs to be executed according to the predefined early warning rules. The following is a general process for executing the pre-oxidation furnace control strategy based on the abnormal analysis results: 1. Review of Abnormal Analysis First, review the results of the anomaly analysis to determine which temperature monitoring points have anomalies, as well as the type and severity of the anomalies. This typically involves reviewing temperature data, comparing temperature threshold graphs, and evaluating the impact of the anomalies on the pre-oxidation process.

[0113] 2. Warning rule matching According to the predefined warning rules, the abnormal analysis results are matched with the rules. Warning rules usually include conditions such as temperature threshold, abnormal duration, abnormal frequency, and corresponding warning levels and response measures.

[0114] 3. Execution Control Strategy According to the matching early warning rules, the corresponding pre-oxidation furnace control strategy is executed. This may include the following measures: Adjust heating or cooling power: If the abnormality is caused by excessively high or low temperature, the temperature can be controlled within the normal range by adjusting the power of the heating or cooling system.

[0115] Change the material feed speed: If the abnormality is caused by uneven temperature due to the material feed speed being too fast or too slow, the feed speed can be adjusted appropriately to improve the temperature distribution.

[0116] Stop the pre-oxidation process: If the abnormality is so serious that it may affect product quality or equipment safety, the pre-oxidation process should be stopped immediately and the equipment should be inspected and repaired.

[0117] Trigger the alarm system: According to the early warning rules, trigger the corresponding alarm system to notify the operator or technical support team to handle the abnormality in time.

[0118] 4. Monitoring and Feedback After executing the control strategy, the temperature and other relevant parameters of the pre-oxidation furnace need to be continuously monitored to ensure that the abnormality is effectively handled and the equipment returns to normal operation. At the same time, according to the monitoring results and feedback information, the control strategy is adjusted in time to optimize the pre-oxidation process.

[0119] For example, assume that the warning rules of the pre-oxidation furnace include: when the temperature in the reaction zone exceeds 360°C for more than 5 minutes, a high-level warning is triggered and the heating power is automatically reduced to 50%; when the temperature drops to the normal range, the heating power is gradually restored.

[0120] Abnormal analysis review: It was found that the temperature of the reaction zone temperature monitoring point 3 has been higher than 360°C in the past 10 minutes.

[0121] Warning rule matching: According to the warning rules, the anomaly triggered a high-level warning.

[0122] Execution of control strategy: The automatic control system immediately reduces the heating power to 50% and continuously monitors temperature changes.

[0123] Monitoring and feedback: After reducing the heating power, the temperature gradually drops and returns to the normal range. The automatic control system gradually restores the heating power to normal levels based on the feedback information and continues to monitor the pre-oxidation process.

[0124] like Figure 2 According to another embodiment of the present invention, a method for real-time monitoring of the temperature of a pre-oxidation furnace based on a cloud platform is provided. The method for real-time monitoring of the temperature of a pre-oxidation furnace based on a cloud platform comprises the following steps: S1. Obtain temperature data at different measurement positions inside the pre-oxidation furnace and establish a temperature monitoring database based on the cloud platform; S2. Based on the temperature monitoring database and in combination with the results of spatial correlation analysis between temperature data, the temperature data of non-measurement positions inside the pre-oxidation furnace are estimated using the constructed temperature interpolation model; and a temperature estimation database is established based on the estimated temperature data; S3. Construct a pre-oxidation furnace temperature database using the temperature monitoring database and the temperature estimation database, and perform an abnormal analysis according to a preset pre-oxidation furnace space temperature threshold diagram to obtain an abnormal analysis result; S4. Based on the abnormal analysis results, the corresponding pre-oxidation furnace control strategy is executed according to the pre-oxidation furnace early warning rules.

[0125] In summary, with the aid of the above-mentioned technical scheme of the present invention, the present invention realizes centralized management and storage of data by acquiring temperature data at different measuring positions inside the pre-oxidation furnace and establishing a temperature monitoring database through a cloud platform, which is helpful for comprehensive and real-time monitoring of the temperature condition of the pre-oxidation furnace. By combining the spatial correlation analysis results between the temperature data, a temperature interpolation model is constructed to estimate the temperature data of non-measurement positions inside the pre-oxidation furnace, thereby improving the accuracy and completeness of the temperature data and helping to more accurately understand the temperature distribution inside the pre-oxidation furnace. A pre-oxidation furnace temperature database is constructed using the temperature monitoring database and the temperature estimation database, and anomaly analysis is performed according to a pre-set pre-oxidation furnace spatial temperature threshold map, so that temperature anomalies can be discovered in time. By real-time monitoring and accurate estimation of the temperature inside the pre-oxidation furnace, the system can optimize the operating parameters of the pre-oxidation furnace and improve production efficiency. The present invention utilizes the Pearson correlation coefficient method and Euclidean distance calculation, combined with Moran's index evaluation, to conduct in-depth spatial correlation analysis on temperature data at different measurement locations, which is helpful to reveal the intrinsic connection and law between temperature data, and provides a scientific basis for constructing a temperature interpolation model. The Kriging interpolation algorithm based on the simulated annealing artificial bee colony hybrid algorithm is adopted, and combined with the spatial correlation analysis result, a high-precision temperature interpolation model is constructed. The hybrid algorithm can make full use of the advantages of the two algorithms, improve the convergence speed and optimization effect of the model, so as to more accurately estimate the temperature data of non-measurement positions inside the pre-oxidation furnace, obtain the optimal parameter combination through iterative optimization, and use these parameters to update the theoretical variation function model to obtain the final temperature interpolation model, which can ensure the accuracy and rationality of the model parameters, thereby improving the accuracy and reliability of the temperature interpolation model. The present invention integrates the temperature data in the temperature monitoring database and the temperature estimation database to establish a unified pre-oxidation furnace temperature database, which helps to achieve comprehensive and unified management of the temperature data inside the pre-oxidation furnace and provide a reliable data basis for subsequent temperature anomaly analysis. By integrating temperature data, scientifically setting temperature thresholds and accurately judging abnormal states, the abnormal analysis unit can significantly improve the accuracy and efficiency of abnormal analysis, help to promptly discover and solve temperature problems in the operation of the pre-oxidation furnace, ensure the stable operation of the equipment and product quality, and its advantages in abnormal analysis provide a strong guarantee for the safe and efficient operation of the pre-oxidation furnace.

[0126] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) containing computer-usable program code.

[0127] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A real-time monitoring system for pre-oxidation furnace temperature based on a cloud platform, characterized in that: The cloud platform-based pre-oxidation furnace temperature real-time monitoring system includes: a data acquisition unit, a temperature estimation unit, an abnormality analysis unit and a pre-oxidation furnace control unit; The data acquisition unit is used to acquire temperature data at different measurement positions inside the pre-oxidation furnace and establish a temperature monitoring database based on the cloud platform, wherein the measurement positions include the heating element area, the material inlet and outlet, the material processing area and the furnace wall; The temperature estimation unit is used to estimate the temperature data of non-measurement positions inside the pre-oxidation furnace based on the temperature monitoring database and in combination with the results of spatial correlation analysis between the temperature data using the constructed temperature interpolation model; and establish a temperature estimation database based on the estimated temperature data; The abnormality analysis unit is used to construct a pre-oxidation furnace temperature database using a temperature monitoring database and a temperature estimation database, and to perform abnormality analysis according to a preset pre-oxidation furnace space temperature threshold diagram to obtain an abnormality analysis result; The pre-oxidation furnace control unit is used to execute a corresponding pre-oxidation furnace control strategy according to the pre-oxidation furnace early warning rule based on the abnormal analysis result; The temperature estimation unit includes: a data extraction module, a correlation analysis module, a model construction module and an estimation database construction module; The data extraction module is used to extract the temperature data of all measured locations and their measurement locations from the temperature monitoring database; The correlation analysis module is used to analyze the spatial correlation between the heating element area and the material inlet and outlet, the material processing area and the furnace wall respectively using the correlation analysis method according to the temperature data of the measured position and its measurement position; The model building module is used to build a temperature interpolation model based on the spatial correlation analysis results by using a Kriging interpolation algorithm based on a simulated annealing artificial bee colony hybrid algorithm; The estimation database building module is used to estimate the temperature data of non-measurement positions inside the pre-oxidation furnace by using a temperature interpolation model, and to establish a temperature estimation database according to the estimation result.

2. A cloud platform-based pre-oxidation furnace temperature real-time monitoring system according to claim 1, characterized in that: The data acquisition unit includes: a temperature data acquisition module, a temperature data transmission module, a temperature data preprocessing module and a temperature monitoring database establishment module; The temperature data acquisition module is used to collect temperature data using sensors pre-installed at different measurement positions inside the pre-oxidation furnace; The temperature data transmission module is used to transmit the collected temperature data to the data receiving end of the cloud platform by wireless means; The temperature data preprocessing is used to preprocess the received temperature data at the data receiving end of the cloud platform, and the preprocessing includes data cleaning, data denoising and format conversion; The temperature monitoring database establishment module is used to store the pre-processed temperature data in the database of the cloud platform and establish a temperature monitoring database based on the measurement position of the sensor.

3. The real-time monitoring system for pre-oxidation furnace temperature based on a cloud platform according to claim 1 is characterized in that: The method of analyzing the spatial correlation between the heating element area and the material inlet and outlet, the material processing area and the furnace wall respectively using the correlation analysis method according to the temperature data of the measured position and the measurement position thereof includes: For the temperature data of the heating element area, the material inlet and outlet, the material handling area and the furnace wall, the Pearson correlation coefficient method is used to calculate the correlation coefficients between the heating element area and the material inlet and outlet, the material handling area and the furnace wall respectively; For the measurement position of the heating element area, the spatial distance between the measurement position of the heating element area and the temperature data measurement positions of the material inlet and outlet, the material processing area and the furnace wall is calculated using the Euclidean distance; Construct a spatial weight matrix based on the calculated spatial distance and correlation coefficient; According to the spatial weight matrix, the Moran index is used to evaluate the spatial correlation between the temperature data of the heating element area and the temperature data of the material inlet and outlet, material handling area and furnace wall.

4. A cloud platform-based pre-oxidation furnace temperature real-time monitoring system according to claim 3, characterized in that: The method of constructing a temperature interpolation model based on the spatial correlation analysis result and using a Kriging interpolation algorithm based on a simulated annealing artificial bee colony hybrid algorithm includes: Based on the results of spatial correlation analysis, the temperature data of the measurement locations are grouped, and the inter-group distances between the groups are calculated; According to the inter-group distance and temperature data between the groups, the spatial relationship function of each group is calculated; The spatial relationship function is fitted based on the support vector machine algorithm to obtain the theoretical variation function model; The theoretical variation function model is mapped to the simulated annealing artificial bee colony hybrid algorithm, and the temperature interpolation model is obtained through iterative optimization.

5. A cloud platform-based pre-oxidation furnace temperature real-time monitoring system according to claim 4, characterized in that: The spatial relationship function is fitted based on the support vector machine algorithm to obtain a theoretical variation function model including: The inter-group distances between groups and the spatial relationship functions of each group are combined into a set of data point pairs; The support vector machine regression model is constructed using the Gaussian radial basis kernel function; Configure the parameters of the support vector machine and train the support vector machine model using a set of data point pairs; The support vector machine regression function obtained after training is used as the theoretical variation function model.

6. A cloud platform-based pre-oxidation furnace temperature real-time monitoring system according to claim 4, characterized in that: The method of mapping the theoretical variation function model to the simulated annealing artificial bee colony hybrid algorithm and obtaining the temperature interpolation model through iterative optimization includes: The parameters to be optimized are extracted from the theoretical variogram model, and the extracted parameters are mapped to the simulated annealing artificial bee colony hybrid algorithm; Initialize the simulated annealing artificial bee colony hybrid algorithm; The hybrid strategy of simulated annealing algorithm and artificial bee colony algorithm is used to iteratively optimize the parameters of the mapped variogram model to obtain the optimal parameter combination. The theoretical variogram model is updated using the optimal parameter combination to obtain the temperature interpolation model.

7. The real-time monitoring system for pre-oxidation furnace temperature based on a cloud platform according to claim 1, characterized in that: The abnormality analysis unit includes: a pre-oxidation furnace temperature database construction module, a temperature threshold map construction module and an abnormal state judgment module; The pre-oxidation furnace temperature database construction module is used to integrate the temperature data in the temperature monitoring database and the temperature estimation database, and establish a pre-oxidation furnace temperature database; The temperature threshold map building module is used to establish a temperature threshold map of the pre-oxidation furnace space based on the properties of the processed material of the pre-oxidation furnace and in combination with the design parameters and operation characteristics of the pre-oxidation furnace; The abnormal state judgment module is used to extract temperature data of each position from the pre-oxidation furnace temperature database, compare it with the pre-oxidation furnace space temperature threshold map, and judge the abnormal state of the pre-oxidation furnace based on the comparison result.

8. A cloud platform-based pre-oxidation furnace temperature real-time monitoring system according to claim 7, characterized in that: The method of establishing a temperature threshold diagram of the pre-oxidation furnace space based on the material properties of the pre-oxidation furnace and in combination with the design parameters and operation characteristics of the pre-oxidation furnace includes: Based on the design parameters and operating characteristics of the pre-oxidation furnace, the temperature distribution inside the pre-oxidation furnace is simulated using thermal simulation tools; According to the simulation results and the properties of the materials to be processed in the pre-oxidation furnace, temperature thresholds are set for different areas inside the pre-oxidation furnace; Based on the set temperature threshold, draw the spatial temperature threshold map of the pre-oxidation furnace and mark the location of the temperature monitoring point.

9. A method for real-time monitoring temperature of a pre-oxidation furnace based on a cloud platform, used to implement the operation of the real-time monitoring temperature of a pre-oxidation furnace based on a cloud platform according to any one of claims 1 to 8, characterized in that: The real-time monitoring method of the pre-oxidation furnace temperature based on the cloud platform comprises the following steps: S1. Obtain temperature data at different measurement locations inside the pre-oxidation furnace, and establish a temperature monitoring database based on the cloud platform. The measurement locations include the heating element area, material inlet and outlet, material processing area, and furnace wall; S2. Based on the temperature monitoring database and in combination with the results of spatial correlation analysis between temperature data, the temperature data of non-measurement positions inside the pre-oxidation furnace are estimated using the constructed temperature interpolation model; and a temperature estimation database is established based on the estimated temperature data; S3. Construct a pre-oxidation furnace temperature database using the temperature monitoring database and the temperature estimation database, and perform an abnormal analysis according to a preset pre-oxidation furnace space temperature threshold diagram to obtain an abnormal analysis result; S4. Based on the abnormal analysis results, the corresponding pre-oxidation furnace control strategy is executed according to the pre-oxidation furnace early warning rules.

Images