Cement grinding fault prediction method and system based on big data analysis

Abstract

The invention relates to the technical field of cement production equipment operation and maintenance, in particular to a cement grinding fault prediction method and system based on big data analysis, which can more accurately predict the fault of cement grinding equipment, improve the production efficiency, reduce the production cost and guarantee the production safety. The method comprises the following steps: continuously collecting a millstone temperature distribution diagram after multiple grinding operations; performing cell division on the millstone temperature distribution diagram to obtain a plurality of millstone temperature analysis unit blocks; performing image recognition processing on each millstone temperature analysis unit block to obtain a unit block temperature characteristic value; the unit block temperature characteristic value is used for representing the temperature performance of the millstone temperature analysis block; according to the shape of the millstone, the unit block temperature characteristic values are given to corresponding positions on the millstone, and a millstone temperature circular matrix is obtained; and based on the sequence of the acquisition time, carrying out central angle alignment on the plurality of grinding disc temperature circular matrixes to obtain a grinding disc temperature cylindrical matrix.

Description

Technical Field

[0001] The present invention relates to the technical field of cement production equipment operation and maintenance, and in particular to a cement grinding fault prediction method and system based on big data analysis. Background Art

[0002] In the cement production process, the grinding process is a vital link, and its operating conditions directly affect the output and quality of cement; however, due to the influence of various factors such as equipment aging, improper operation, and raw material changes, various faults often occur in the cement grinding process, such as grinding disc wear, blockage, and excessive temperature; these faults will not only lead to a decrease in production efficiency, but may also cause damage to the equipment and even cause safety accidents.

[0003] Early cement grinding fault prediction methods mainly relied on experience and regular maintenance; this method is not only inefficient, but also difficult to accurately detect potential faults; with the development and application of big data technology, more and more companies have begun to try to apply big data technology to cement grinding fault prediction; through the analysis and mining of a large amount of historical data, the rules and anomalies in the operation of the equipment can be discovered, so as to predict potential faults in advance; however, the existing cement grinding fault prediction methods based on big data analysis often only focus on the temperature changes and analysis of a certain point on the grinding disc, while ignoring the impact of the grinding disc temperature distribution on fault prediction. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a cement grinding fault prediction method and system based on big data analysis, which can more accurately predict the failure of cement grinding equipment, improve production efficiency, reduce production costs, and ensure production safety.

[0005] In a first aspect, the present invention provides a cement grinding fault prediction method based on big data analysis, the method comprising:

[0006] Continuously collect the temperature distribution diagram of the grinding disc after multiple grinding operations;

[0007] Dividing the grinding disc temperature distribution map into cells to obtain a plurality of grinding disc temperature analysis unit blocks;

[0008] Perform image recognition processing on each grinding disc temperature analysis unit block to obtain a unit block temperature characteristic value; the unit block temperature characteristic value is used to characterize the temperature performance of the grinding disc temperature analysis block;

[0009] According to the shape of the grinding disc, the temperature characteristic value of the unit block is assigned to the corresponding position on the grinding disc to obtain the grinding disc temperature circular matrix;

[0010] Based on the order of acquisition time, multiple grinding disc temperature circular matrices are aligned at their center angles to obtain a grinding disc temperature cylindrical matrix; the grinding disc temperature cylindrical matrix is ​​used to characterize the characteristic performance of the grinding disc temperature in the time series;

[0011] The grinding plate temperature cylindrical matrix is ​​input into the pre-built cement grinding fault analysis model to obtain the cement grinding fault type.

[0012] Furthermore, the method for dividing the grinding disc temperature distribution diagram into cells includes:

[0013] Segmenting the obtained grinding disc temperature distribution map;

[0014] Divide different areas of the grinding disc into cells of different sizes;

[0015] After the cell division, the data is balanced, including downsampling the data in the overcrowded area and interpolating the data in the sparse area;

[0016] Smoothing and special boundary unit division strategies are used to add a circle of buffer cells to reduce the interference caused by boundary effects;

[0017] Each cell is precisely labeled and verified against the physical structure diagram of the grinding disc.

[0018] Furthermore, the method for obtaining the unit block temperature characteristic value includes:

[0019] Convert the color temperature distribution map into a grayscale image;

[0020] Remove random noise from images;

[0021] Use edge detection algorithms to highlight boundaries where temperature changes are evident;

[0022] Extract temperature statistics, texture features, and hotspot detection for each unit block:

[0023] The extracted features are converted into numerical data to form the unit block temperature feature values.

[0024] Furthermore, the method for obtaining the grinding disc temperature circular matrix includes:

[0025] Create a digital model based on the actual grinding disc’s geometric dimensions and structure;

[0026] Convert the temperature analysis unit block to the corresponding position on the grinding disc three-dimensional model;

[0027] Map the temperature characteristic value of each unit block to the corresponding position of the grinding wheel model to form a virtual "temperature map";

[0028] Arrange the mapped temperature values ​​according to the actual structure of the grinding disc to construct a two-dimensional circular matrix, namely, the grinding disc temperature circular matrix;

[0029] Convert the grinding plate temperature matrix into an intuitive visualization.

[0030] Furthermore, the method for constructing the cement grinding fault analysis model includes:

[0031] Collect historical data, including grinding disc temperature data and corresponding fault conditions;

[0032] Based on the collected historical data, a fault analysis model is constructed using a deep learning model; the deep learning model includes a convolutional neural network, a recurrent neural network, and a long short-term memory network;

[0033] Extract features from historical data, including average temperature, temperature change rate, and temperature volatility;

[0034] Use the extracted features as the input of the model and train the selected model;

[0035] After building the model, evaluate and tune the model;

[0036] Deploy the trained model to an actual cement production facility.

[0037] Furthermore, the design structure of the cement grinding fault analysis model includes an input layer, a feature extraction layer, a fusion layer, a hidden layer and a classification layer.

[0038] Furthermore, the cement grinding fault types include overheating fault, wear fault, imbalance fault, blockage fault and bearing fault.

[0039] On the other hand, the present application also provides a cement grinding fault prediction system based on big data analysis, the system comprising:

[0040] A data acquisition module, used to continuously collect the temperature distribution diagram of the grinding disc after multiple grinding operations;

[0041] A unit division module is used to divide the grinding disc temperature distribution map into cells to obtain a plurality of grinding disc temperature analysis unit blocks;

[0042] An image recognition module is used to perform image recognition processing on each grinding disc temperature analysis unit block to obtain a unit block temperature characteristic value; the unit block temperature characteristic value is used to characterize the temperature performance of the grinding disc temperature analysis block;

[0043] The temperature characteristic value mapping module is used to assign the temperature characteristic value of the unit block to the corresponding position on the grinding disc according to the shape of the grinding disc, and obtain the grinding disc temperature circular matrix;

[0044] The cylindrical matrix construction module aligns the central angles of multiple grinding disc temperature circular matrices based on the order of acquisition time to obtain a grinding disc temperature cylindrical matrix; the grinding disc temperature cylindrical matrix is ​​used to characterize the characteristic performance of the grinding disc temperature in the time series;

[0045] The fault analysis module is used to input the grinding disc temperature cylindrical matrix into the pre-built cement grinding fault analysis model to obtain the cement grinding fault type; the cement grinding fault type includes overheating fault, wear fault, imbalance fault, blockage fault and bearing fault.

[0046] In a third aspect, the present application provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and the computer program, when executed by the processor, implements the steps of any one of the above methods.

[0047] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps in any one of the above-mentioned methods when executed by a processor.

[0048] Compared with the prior art, the present invention has the following beneficial effects: the method can more accurately predict equipment failures through big data analysis, based on the mining and analysis of historical data; the method can fully understand the temperature distribution of the grinding disc by continuously collecting the temperature distribution diagram of the grinding disc after multiple grinding operations and performing temperature analysis on the entire grinding disc, so as to more effectively find faults; through image recognition and extraction of unit block temperature characteristic values, detailed temperature characteristic analysis can be performed on each grinding disc temperature analysis unit block, thereby more accurately describing the temperature performance of the grinding disc, which is helpful to find potential faults;

[0049] By constructing a grinding disc temperature cylindrical matrix, the changing trend of the grinding disc temperature can be analyzed in the time series, which can not only detect the current fault situation, but also predict possible future faults and take measures to intervene and repair in advance; it can identify various types of faults, including overheating faults, wear faults, imbalance faults, blockage faults and bearing faults; it enables operation and maintenance personnel to take targeted measures in time to avoid equipment damage and safety accidents;

[0050] In summary, this method can more accurately predict the failure of cement grinding equipment, improve production efficiency, reduce production costs and ensure production safety by making full use of big data analysis technology and combining image recognition and time series analysis methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a flow chart of the present invention;

[0052] Figure 2 Structure diagram of the circular matrix of grinding disc temperature;

[0053] Figure 3 This is the structural diagram of the cement grinding fault prediction system based on big data analysis. DETAILED DESCRIPTION

[0054] In the description of the present application, those skilled in the art should know that the present application can be implemented as a method, an apparatus, an electronic device, and a computer-readable storage medium. Therefore, the present application can be specifically implemented in the following forms: complete hardware, complete software (including firmware, resident software, microcode, etc.), a combination of hardware and software. In addition, in some embodiments, the present application can also be implemented in the form of a computer program product in one or more computer-readable storage media, and the computer-readable storage medium contains computer program code.

[0055] The above-mentioned computer-readable storage medium may adopt any combination of one or more computer-readable storage media. Computer-readable storage media include: electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or devices, or any combination of the above. More specific examples of computer-readable storage media include: portable computer disks, hard disks, random access memories, read-only memories, erasable programmable read-only memories, flash memories, optical fibers, optical disc read-only memories, optical storage devices, magnetic storage devices, or any combination of the above. In the present application, computer-readable storage media can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or device.

[0056] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws.

[0057] The present application describes the provided methods, devices, and electronic devices through flowcharts and / or block diagrams.

[0058] It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer-readable program instructions. These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine, and these computer-readable program instructions are executed by a computer or other programmable data processing device to produce a device that implements the functions / operations specified by the boxes in the flowchart and / or block diagram.

[0059] These computer-readable program instructions may also be stored in a computer-readable storage medium that enables a computer or other programmable data processing device to work in a specific manner. In this way, the instructions stored in the computer-readable storage medium produce an instruction device product including functions / operations specified in the blocks in the flowchart and / or block diagram.

[0060] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device, so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby enabling the instructions executed on the computer or other programmable data processing apparatus to provide a process for implementing the functions / operations specified in the blocks in the flowchart and / or block diagram.

[0061] The present application is described below in conjunction with the drawings in the present application.

[0062] Embodiment 1: Figure 1 to Figure 2 As shown, the cement grinding fault prediction method based on big data analysis of the present invention specifically includes the following steps:

[0063] S1, continuously collecting the temperature distribution diagram of the grinding disc after multiple grinding operations;

[0064] In the field of cement production equipment operation and maintenance, continuous collection of the grinding disc temperature distribution map after grinding operation is a basic and critical step. In order to achieve this goal, the technical solution involves the following aspects:

[0065] High-precision temperature sensor deployment: First, install high-sensitivity temperature sensors in key areas of the grinding disc; the sensors should be able to cover the entire surface of the grinding disc to ensure no blind spot monitoring; select sensors suitable for high temperature and dusty environments to ensure long-term stable operation and data accuracy;

[0066] Data transmission and integration: Each sensor is connected to the central data collection system through wired or wireless means; the data transmission protocol should ensure the real-time and integrity of the data and avoid information loss caused by network delay or packet loss;

[0067] Real-time monitoring software platform: Establish a central monitoring system that receives temperature data from various sensors and draws a real-time temperature distribution map of the grinding disc. The software platform needs to have data preprocessing capabilities. In addition, the platform should have a visualization function to intuitively display the temperature changes on the surface of the grinding disc, so that operation and maintenance personnel can quickly identify abnormal areas.

[0068] Timing and triggering mechanism: set the automatic collection frequency, such as once every minute or according to the production cycle, to ensure the continuity of data;

[0069] Data storage and management: The collected temperature distribution map data needs to be properly stored, usually using cloud storage or local servers to ensure data security and accessibility; the data should be saved in order according to time series to facilitate subsequent analysis and input into fault prediction models.

[0070] By deploying high-precision temperature sensors and establishing a real-time monitoring software platform, operation and maintenance personnel can monitor the temperature distribution of the grinding disc at any time; once an abnormal temperature area is found, measures can be taken immediately to make adjustments to prevent the occurrence or expansion of faults; the continuously collected grinding disc temperature distribution map data can be used to achieve preventive maintenance; by analyzing historical data and monitoring trends, operation and maintenance personnel can identify potential signs of failure and take preventive maintenance measures in time to extend the service life of equipment and reduce downtime; real-time monitoring of grinding disc temperature distribution helps optimize the production process; timely discovery and resolution of temperature unevenness or abnormal areas can reduce energy consumption and improve the efficiency and quality of cement production; by continuously collecting and storing grinding disc temperature distribution map data, a historical database can be established to provide support for subsequent data analysis and fault prediction; data-based decisions can help managers formulate more scientific production plans and equipment maintenance strategies; implementing preventive maintenance and responding to abnormal situations in a timely manner can reduce maintenance costs; by continuously monitoring the grinding disc temperature distribution, downtime and equipment damage caused by failures can be avoided, and the cost of repairs and replacement parts can be reduced;

[0071] In summary, the technical solution of continuously collecting the temperature distribution map of the grinding disc after multiple grinding operations can not only improve the reliability and production efficiency of the equipment, but also reduce maintenance costs, creating greater economic benefits and social value for cement production enterprises.

[0072] S2, dividing the grinding disc temperature distribution map into cells to obtain a plurality of grinding disc temperature analysis unit blocks;

[0073] In the field of cement production equipment operation and maintenance, cell division of the grinding disc temperature distribution map is a core preprocessing step, which aims to convert complex temperature distribution data into structured information that is easy to analyze; the following is a detailed technical implementation plan for this step:

[0074] Determine the segmentation strategy: Use image processing technology to segment the acquired grinding wheel temperature distribution map; Considering the circular characteristics of the grinding wheel, select a segmentation algorithm based on a circular coordinate system, such as the polar coordinate grid division method, which can maintain the high efficiency of data processing and better adapt to the geometric shape of the grinding wheel;

[0075] Adaptive grid size determination: Different areas of the grinding disc have different temperature sensitivities and failure probabilities. Therefore, the adaptive grid size needs to be considered when dividing cells. By analyzing historical data, high-prone areas of failure and areas sensitive to temperature changes can be identified. A smaller grid size can be used for these areas to improve analysis accuracy. For other areas, the grid size can be appropriately increased to balance the analysis accuracy and computing resource requirements.

[0076] Data balancing: After the cell division, in order to avoid analysis deviation caused by abnormal temperature data volume in some areas, the data needs to be balanced. This includes downsampling the data in overcrowded areas and interpolating the data in sparse areas to ensure that the amount of information contained in each cell block is roughly equal, thereby improving the fairness and accuracy of subsequent analysis.

[0077] Boundary processing: The temperature data processing of the grinding disc edge is easily affected by the external environment; smoothing processing and special boundary unit division strategies are adopted, such as adding a circle of buffer cells, to reduce the interference caused by the boundary effect and ensure the consistency of the temperature characteristic analysis of the entire grinding disc;

[0078] Cell labeling and verification: To ensure that the divided cells correspond one-to-one to the actual physical location, each cell needs to be accurately labeled and verified against the physical structure diagram of the grinding wheel; by comparing with historical fault records, check whether the cell division can effectively identify known fault modes and further optimize the division strategy.

[0079] In this step, the segmentation algorithm based on the circular coordinate system is adopted, which can better adapt to the geometric shape of the grinding disc, thereby improving the efficiency and accuracy of data processing; through adaptive grid size determination, the temperature sensitivity and fault probability of different areas of the grinding disc are distinguished, making the fault prediction analysis more targeted and accurate; through data balancing processing, the amount of information contained in each unit block is ensured to be roughly equal, avoiding the analysis deviation caused by abnormal data volume in some areas, and further improving the accuracy of the fault prediction model; the boundary processing strategy can reduce the interference of the grinding disc edge temperature data affected by the external environment, and ensure the consistency of the entire grinding disc temperature characteristic analysis; through precise labeling and verification of cells, and comparison with historical fault records, the effectiveness of the division strategy can be verified, and the cell division strategy can be further optimized, thereby improving the reliability and stability of the fault prediction model; cell division of the grinding disc temperature distribution map can effectively improve the accuracy and reliability of fault prediction, and provide important technical support for the operation and maintenance management of cement production equipment.

[0080] S3, performing image recognition processing on each grinding disc temperature analysis unit block to obtain a unit block temperature characteristic value; the unit block temperature characteristic value is used to characterize the temperature performance of the grinding disc temperature analysis block;

[0081] Perform image recognition processing on each grinding disc temperature analysis unit block to obtain a unit block temperature characteristic value; the unit block temperature characteristic value is used to characterize the temperature performance of the grinding disc temperature analysis block; this step is implemented by the following technical solution:

[0082] Image preprocessing:

[0083] Grayscale: First, convert the color temperature distribution map into a grayscale image to simplify the processing complexity and focus on the temperature intensity information;

[0084] Noise removal: Apply median filtering or Gaussian filtering algorithms to remove random noise in the image and improve the accuracy of subsequent analysis;

[0085] Edge enhancement: Use edge detection algorithms to highlight boundaries where temperature changes are significant, helping to identify areas of temperature anomalies;

[0086] Feature extraction:

[0087] Temperature statistical characteristics: calculate the average temperature, maximum temperature, minimum temperature, and temperature range of each unit block, directly reflecting the basic characteristics of temperature distribution;

[0088] Texture features: The temperature distribution pattern of the unit block is extracted through texture analysis method, which helps to identify the temperature distribution characteristics caused by specific faults;

[0089] Hotspot detection: Apply hotspot detection algorithms to identify areas with abnormally high temperatures;

[0090] The extracted features are converted into numerical data to form the unit block temperature characteristic value, which integrates various properties of the temperature in the area to facilitate subsequent analysis and model input.

[0091] In this step, grayscale processing is used to simplify image information, and the temperature intensity is analyzed intensively, which reduces the processing complexity. At the same time, noise removal improves data quality and ensures the accuracy of subsequent analysis. Edge enhancement and hotspot detection techniques help to quickly identify abnormal temperature areas, improving the sensitivity and timeliness of fault prediction. By extracting temperature statistical features and texture features, not only the basic statistical information of temperature is captured, but also the complex pattern of temperature distribution is considered, which can more comprehensively reflect the temperature state of the grinding wheel. The extracted features are converted into numerical data to form unit block temperature eigenvalues. This process standardizes the data format, so that the eigenvalues ​​can be directly used as model input, which is convenient for algorithm processing and comparison and optimization between models, and accelerates the development and application of fault prediction models.

[0092] In summary, this step greatly improves the depth and breadth of the analysis of grinding disc temperature data, provides a more accurate and comprehensive basis for the early prediction of cement grinding failures, and is conducive to taking measures in advance to prevent failures and ensure production safety and efficiency.

[0093] S4. According to the shape of the grinding disc, the temperature characteristic value of the unit block is assigned to the corresponding position on the grinding disc to obtain a grinding disc temperature circular matrix;

[0094] In the field of cement production equipment operation and maintenance, the S4 step is a key step to remap the temperature characteristic values ​​extracted in the early stage to the physical space layout of the grinding disc, forming an intuitive and structured temperature distribution matrix. The specific implementation details are as follows:

[0095] Constructing a digital model of the grinding disc: First, using modeling tools, an accurate digital model is created based on the actual geometric dimensions and structure of the grinding disc, ensuring the accuracy and intuitiveness of subsequent temperature characteristic value mapping;

[0096] Coordinate system conversion: Since the temperature analysis unit block is obtained in the two-dimensional image processing stage, it needs to be converted to the corresponding position on the three-dimensional model of the grinding wheel; it involves the conversion of the coordinate system, from pixel coordinates to polar coordinates or directly mapped to the Cartesian coordinate system of the three-dimensional model;

[0097] Eigenvalue mapping: accurately map the temperature eigenvalue of each unit block to the corresponding position of the grinding wheel model to form a virtual "temperature map". This step ensures the perfect correspondence between the physical space and the temperature data, providing a basis for subsequent analysis.

[0098] Matrix construction: Arrange the mapped temperature values ​​according to the actual structure of the grinding disc to construct a two-dimensional circular matrix, namely the grinding disc temperature circular matrix; each element in the matrix contains not only the position information but also the corresponding temperature characteristic value, forming an intuitive grinding disc temperature distribution matrix, which intuitively shows the temperature status of each part of the grinding disc;

[0099] Visual presentation: Finally, data visualization technology is used to convert the grinding disc temperature matrix into an intuitive visual image, which helps operation and maintenance personnel to quickly identify abnormal temperature areas and provide direction for troubleshooting.

[0100] In this step, by accurately mapping the temperature characteristic values ​​to the physical position of the grinding disc, the circular temperature matrix formed can more accurately reflect the actual temperature distribution; this precise temperature distribution map helps operation and maintenance personnel identify potential abnormal areas, so that they can intervene in time before the problem develops into a serious failure; through real-time temperature monitoring and analysis, it is possible to more accurately determine the parts that need to be maintained or replaced, thereby achieving more targeted maintenance activities; this not only extends the service life of the equipment, but also helps to reduce unnecessary maintenance costs; the grinding disc temperature circular matrix provides a data-rich decision-making basis, so that operation and maintenance decisions can rely more on real-time data and trend analysis, rather than just on the experience and intuition of operators; it can improve the efficiency and effectiveness of decision-making and reduce human errors; by continuously monitoring and analyzing the grinding disc temperature matrix, early warning of equipment status can be achieved; it helps to implement preventive maintenance strategies, thereby avoiding production interruptions and safety accidents caused by equipment failures; using advanced visualization tools to present temperature data as graphics can make problem identification more intuitive and easy to understand; it not only helps technicians to quickly diagnose problems, but also facilitates non-professionals to understand the operating status of the equipment;

[0101] In summary, step S4 provides a higher level of monitoring and maintenance capabilities for grinding equipment in the cement production process through highly structured and visual data presentation, significantly improving operation and maintenance efficiency and equipment reliability.

[0102] S5. Based on the order of the acquisition time, align the central angles of multiple grinding disc temperature circular matrices to obtain a grinding disc temperature cylindrical matrix; the grinding disc temperature cylindrical matrix is ​​used to characterize the characteristic performance of the grinding disc temperature in the time series;

[0103] In the field of cement production equipment operation and maintenance, the monitoring and analysis of grinding disc temperature is an important means to predict grinding failures. Step S5 aligns the central angles of multiple grinding disc temperature circular matrices based on the order of acquisition time to obtain the grinding disc temperature cylindrical matrix, which is a key step in constructing time series temperature characteristics. The following is a detailed introduction to this step:

[0104] The principle of center angle alignment:

[0105] Since the grinding disc is circular and the temperature distribution may change with time and operating conditions, it is necessary to align the grinding disc temperature circular matrix at different time points in order to analyze the changing rules of temperature distribution in the time series; the purpose of center angle alignment is to ensure that the position correspondence of each grinding disc temperature analysis unit block at different time points is consistent, so as to accurately compare and analyze the temperature change trend;

[0106] Alignment process:

[0107] First, determine a unified circle center as the reference point for alignment; this circle center can be the actual geometric center of the grinding disc or a virtual center set during data processing;

[0108] Next, according to the center position of each grinding disc temperature circular matrix, it is rotated to coincide with the center of the reference circle; this involves the rotation transformation algorithm in image processing, ensuring that the temperature analysis unit blocks in each circular matrix can be correctly aligned to a unified position;

[0109] During the alignment process, the relative position relationship between the unit blocks in each circular matrix needs to be kept unchanged, that is, only the overall rotation operation is performed without changing the temperature data inside the unit block;

[0110] Construct the grinding disc temperature cylindrical matrix:

[0111] After the center angle alignment is completed, multiple aligned grinding disc temperature circular matrices are stacked in the order of collection time to form a three-dimensional grinding disc temperature cylindrical matrix;

[0112] The grinding disc temperature cylindrical matrix not only contains the temperature distribution information at each time point, but also reflects the changing law of temperature distribution in the time series.

[0113] In this step, the center angle alignment ensures that the temperature distribution at different time points can be aligned in space, so that the temperature change trend can be accurately compared and analyzed; it helps engineers and operators to better understand the evolution law of the grinding disc temperature in the time series and detect abnormalities in time; after stacking multiple aligned grinding disc temperature circular matrices to form a grinding disc temperature cylindrical matrix, the change of temperature distribution in the time series can be more intuitively observed; abnormal temperature change trends indicate potential equipment failures or poor operations. By timely detecting and analyzing these anomalies, the occurrence of failures can be prevented and the reliability and stability of the equipment can be improved; based on the grinding disc temperature cylindrical matrix The time series data of the column matrix can be used to build a prediction model using machine learning or deep learning methods to predict the fault type and probability of cement grinding equipment. By analyzing the temperature change characteristics and combining historical fault data, possible faults can be discovered in advance and corresponding maintenance measures can be taken to avoid production interruptions and equipment damage. The three-dimensional data structure formed by the grinding plate temperature column matrix can be displayed to operation and maintenance personnel and managers through visualization tools, so that they can intuitively understand the changes in the equipment temperature distribution. Through the intuitive visualization interface, personnel can quickly discover abnormal situations and take timely measures to improve production efficiency and equipment utilization.

[0114] In summary, the implementation of step S5 can effectively help cement production equipment operation and maintenance personnel monitor and analyze the changing trend of grinding disc temperature, discover equipment failures in advance, and provide a reliable data basis for equipment maintenance and fault prediction.

[0115] S6. Inputting the grinding disc temperature cylindrical matrix into a pre-built cement grinding fault analysis model to obtain cement grinding fault types; the cement grinding fault types include overheating fault, wear fault, imbalance fault, blockage fault and bearing fault;

[0116] Through the S6 step, a closed loop from data processing to fault prediction is achieved. Advanced algorithms are used to automatically analyze the temperature changes of the grinding disc and efficiently identify a variety of potential faults. Compared with traditional empirical judgment and regular maintenance, the efficiency and accuracy of fault prediction are greatly improved, providing a scientific basis for cement production safety and maintenance strategies.

[0117] The method for constructing the cement grinding fault analysis model comprises:

[0118] Collect historical data, including grinding disc temperature data and corresponding fault conditions; these data come from the real-time monitoring system and historical records of the cement production site;

[0119] Based on the collected historical data, a fault analysis model is constructed using a deep learning model; the deep learning model includes a convolutional neural network CNN, a recurrent neural network RNN ​​and a long short-term memory network LSTM;

[0120] Extract features from historical data, including average temperature, temperature change rate, and temperature volatility;

[0121] Use the extracted features as the input of the model and train the selected model;

[0122] After building the model, evaluate and tune the model; evaluate the performance of the model through cross-validation, ROC curve, and confusion matrix methods, and adjust and improve the model as needed;

[0123] Deploy the trained model to actual cement production equipment; in actual application, the model will continuously monitor the grinding disc temperature data and predict the type of possible failure in real time;

[0124] The design structure of the cement grinding fault analysis model includes:

[0125] Input layer: the entrance of the model, receiving the processed grinding disc temperature cylindrical matrix as input;

[0126] Feature extraction layer: Use convolutional neural network to capture the spatial characteristics of grinding disc temperature distribution; use recurrent neural network (RNN) and long short-term memory network (LSTM) to analyze the temperature change trend in time series;

[0127] Fusion layer: The fusion layer integrates spatial features and time series features to form a unified feature representation;

[0128] Hidden layers: used to further process and abstract features obtained from the feature extraction layer; these layers help the model learn more complex failure modes through nonlinear transformations;

[0129] Classification layer: The last layer of the model, used to output the probability distribution of each fault type; each probability value represents the confidence that the current grinding disc state meets the specific fault type;

[0130] Loss function and optimizer: During model training, the cross entropy loss function is used to measure the difference between the predicted fault type and the actual label, and the model parameters are adjusted through optimization algorithms such as gradient descent to minimize the loss.

[0131] In this step, by using the deep learning model to analyze the grinding disc temperature data, the model can more accurately identify various potential fault types. Compared with traditional empirical judgment and periodic maintenance methods, this data-driven fault prediction method is more accurate and timely, and can help prevent equipment failures and production interruptions, and improve production efficiency; by timely discovering potential faults and taking preventive maintenance measures, it can reduce downtime and production losses caused by equipment failures, thereby reducing maintenance costs; in addition, predicting faults can also help reduce the occurrence of sudden failures, reduce the risk of safety accidents, and ensure the safety of production equipment and personnel; through the analysis of fault data, production management personnel can better understand the operating status and failure mode of the equipment, thereby optimizing production plans and resource allocation; the construction and deployment of fault analysis models make the operation and maintenance management of cement production equipment more intelligent and data-driven; through the continuous accumulation and analysis of historical data, the model can be continuously optimized and improved, gradually improving the accuracy and reliability of fault prediction, and providing enterprises with continuous and stable production guarantees and economic benefits.

[0132] Embodiment 2: Figure 3 As shown, the cement grinding fault prediction system based on big data analysis of the present invention specifically includes the following modules:

[0133] A data acquisition module, used to continuously collect the temperature distribution diagram of the grinding disc after multiple grinding operations;

[0134] A unit division module is used to divide the grinding disc temperature distribution map into cells to obtain a plurality of grinding disc temperature analysis unit blocks;

[0135] An image recognition module is used to perform image recognition processing on each grinding disc temperature analysis unit block to obtain a unit block temperature characteristic value; the unit block temperature characteristic value is used to characterize the temperature performance of the grinding disc temperature analysis block;

[0136] The temperature characteristic value mapping module is used to assign the temperature characteristic value of the unit block to the corresponding position on the grinding disc according to the shape of the grinding disc, and obtain the grinding disc temperature circular matrix;

[0137] The cylindrical matrix construction module aligns the central angles of multiple grinding disc temperature circular matrices based on the order of acquisition time to obtain a grinding disc temperature cylindrical matrix; the grinding disc temperature cylindrical matrix is ​​used to characterize the characteristic performance of the grinding disc temperature in the time series;

[0138] The fault analysis module is used to input the grinding disc temperature cylindrical matrix into the pre-built cement grinding fault analysis model to obtain the cement grinding fault type; the cement grinding fault type includes overheating fault, wear fault, imbalance fault, blockage fault and bearing fault.

[0139] Through the unit division module and image recognition module, the system is no longer limited to monitoring the temperature of a certain point on the grinding disc, but performs grid analysis on the entire grinding disc to capture the temperature characteristics of each small area; it greatly improves the sensitivity and accuracy of fault detection, can detect local anomalies earlier, and avoid faults caused by ignoring subtle changes; the combined use of the temperature eigenvalue mapping module and the cylindrical matrix construction module not only considers the spatial distribution characteristics of the grinding disc temperature, but also integrates the data at different time points to form a temperature change pattern in the time series, which can more effectively reveal the dynamic trends and potential problems of equipment operation; compared with traditional experience judgment and regular maintenance, the system can monitor and predict faults in real time, greatly shortening the fault detection cycle and facilitating timely action. The fault analysis module provides the possibility for taking maintenance measures; the prediction results based on the big data model enable the operation and maintenance team to take action before the fault actually occurs, avoiding production interruptions and safety accidents, and effectively improving production efficiency and safety; the whole process from data collection to fault prediction is automated, which reduces manual intervention, reduces the misjudgment rate, and improves operation and maintenance efficiency; the intelligence level of the system helps enterprises optimize resource allocation and reduce unnecessary maintenance costs. At the same time, through continuous learning and optimization of models, the prediction accuracy is expected to be further improved; through comprehensive and detailed data collection and analysis, combined with advanced image processing and machine learning technology, the system can achieve accurate prediction of cement grinding faults, which can improve cement production efficiency, ensure production safety and promote industrial upgrading.

[0140] The various variations and specific embodiments of the cement grinding fault prediction method based on big data analysis in the aforementioned embodiment 1 are also applicable to the cement grinding fault prediction system based on big data analysis in this embodiment. Through the aforementioned detailed description of the cement grinding fault prediction method based on big data analysis, those skilled in the art can clearly know the implementation method of the cement grinding fault prediction system based on big data analysis in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0141] In addition, the present application also provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are respectively connected via a bus. When the computer program is executed by the processor, each process of the above-mentioned method for controlling output data is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0142] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A cement grinding fault prediction method based on big data analysis, characterized in that: The method comprises: Continuously collect the temperature distribution diagram of the grinding disc after multiple grinding operations; Dividing the grinding disc temperature distribution map into cells to obtain a plurality of grinding disc temperature analysis unit blocks; Perform image recognition processing on each grinding disc temperature analysis unit block to obtain a unit block temperature characteristic value; the unit block temperature characteristic value is used to characterize the temperature performance of the grinding disc temperature analysis block; According to the shape of the grinding disc, the temperature characteristic value of the unit block is assigned to the corresponding position on the grinding disc to obtain the grinding disc temperature circular matrix; Based on the order of acquisition time, multiple grinding disc temperature circular matrices are aligned at their center angles to obtain a grinding disc temperature cylindrical matrix; the grinding disc temperature cylindrical matrix is ​​used to characterize the characteristic performance of the grinding disc temperature in the time series; The grinding plate temperature cylindrical matrix is ​​input into the pre-built cement grinding fault analysis model to obtain the cement grinding fault type.

2. The cement grinding fault prediction method based on big data analysis according to claim 1, characterized in that: The method for dividing the grinding plate temperature distribution diagram into cells includes: Segmenting the obtained grinding disc temperature distribution map; Divide different areas of the grinding disc into cells of different sizes; After the cell division, the data is balanced, including downsampling the data in the overcrowded area and interpolating the data in the sparse area; Smoothing and special boundary unit division strategies are used to add a circle of buffer cells to reduce the interference caused by boundary effects; Each cell is precisely labeled and verified against the physical structure diagram of the grinding disc.

3. The cement grinding fault prediction method based on big data analysis according to claim 1, characterized in that: The method for obtaining the unit block temperature characteristic value includes: Convert the color temperature distribution map into a grayscale image; Remove random noise from images; Use edge detection algorithms to highlight boundaries where temperature changes are evident; Extract temperature statistics, texture features, and hotspot detection for each unit block: The extracted features are converted into numerical data to form the unit block temperature feature values.

4. The cement grinding fault prediction method based on big data analysis according to claim 1, characterized in that: The method for obtaining the grinding disc temperature circular matrix includes: Create a digital model based on the actual grinding disc’s geometric dimensions and structure; Convert the temperature analysis unit block to the corresponding position on the grinding disc three-dimensional model; Map the temperature characteristic value of each unit block to the corresponding position of the grinding wheel model to form a virtual "temperature map"; Arrange the mapped temperature values ​​according to the actual structure of the grinding disc to construct a two-dimensional circular matrix, namely, the grinding disc temperature circular matrix; Convert the grinding plate temperature matrix into an intuitive visualization.

5. The cement grinding fault prediction method based on big data analysis according to claim 1, characterized in that: The method for constructing the cement grinding fault analysis model comprises: Collect historical data, including grinding disc temperature data and corresponding fault conditions; Based on the collected historical data, a fault analysis model is constructed using a deep learning model; the deep learning model includes a convolutional neural network, a recurrent neural network, and a long short-term memory network; Extract features from historical data, including average temperature, temperature change rate, and temperature volatility; Use the extracted features as the input of the model and train the selected model; After building the model, evaluate and tune the model; Deploy the trained model to an actual cement production facility.

6. The cement grinding fault prediction method based on big data analysis according to claim 5, characterized in that: The design structure of the cement grinding fault analysis model includes an input layer, a feature extraction layer, a fusion layer, a hidden layer and a classification layer.

7. The cement grinding fault prediction method based on big data analysis according to claim 1, characterized in that: The cement grinding fault types include overheating fault, wear fault, imbalance fault, blockage fault and bearing fault.

8. A cement grinding fault prediction system based on big data analysis, characterized in that: The system comprises: A data acquisition module, used to continuously collect the temperature distribution diagram of the grinding disc after multiple grinding operations; A unit division module is used to divide the grinding disc temperature distribution map into cells to obtain a plurality of grinding disc temperature analysis unit blocks; An image recognition module is used to perform image recognition processing on each grinding disc temperature analysis unit block to obtain a unit block temperature characteristic value; the unit block temperature characteristic value is used to characterize the temperature performance of the grinding disc temperature analysis block; The temperature characteristic value mapping module is used to assign the temperature characteristic value of the unit block to the corresponding position on the grinding disc according to the shape of the grinding disc, and obtain the grinding disc temperature circular matrix; The cylindrical matrix construction module aligns the central angles of multiple grinding disc temperature circular matrices based on the order of acquisition time to obtain a grinding disc temperature cylindrical matrix; the grinding disc temperature cylindrical matrix is ​​used to characterize the characteristic performance of the grinding disc temperature in the time series; The fault analysis module is used to input the grinding disc temperature cylindrical matrix into the pre-built cement grinding fault analysis model to obtain the cement grinding fault type; the cement grinding fault type includes overheating fault, wear fault, imbalance fault, blockage fault and bearing fault.

9. An electronic device for predicting cement grinding faults based on big data analysis, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and characterized in that: When the computer program is executed by the processor, the steps in the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method according to any one of claims 1 to 7 are implemented.

Images