Coal machine equipment data analysis method and system based on big data platform

Abstract

The invention discloses a coal machine equipment data analysis method and system based on a big data platform, and the method comprises the steps: collecting multi-source data of coal machine equipment, and carrying out the preprocessing of the data; extracting features from the preprocessed data, and fusing the features of different data sources to form a unified equipment operation state feature set; constructing and training a deep learning model, and optimizing model parameters and structures through continuous learning to identify normal and abnormal states of the equipment; and performing a maintenance action according to the abnormal state. According to the invention, multi-source data are effectively integrated, and comprehensive equipment operation state evaluation is provided; the accuracy of anomaly detection is improved by using a deep learning model; dynamically adjusting model parameters to adapt to equipment state changes; potential faults are found in time and a maintenance strategy is optimized so that downtime is reduced and safety and economical efficiency of coal mine production are increased.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial equipment monitoring and safety protection, and specifically to a coal mining equipment data analysis method and system based on a big data platform. Background Art

[0002] With the development of industrial automation and informatization, the monitoring and maintenance of coal mining equipment has become an important part of ensuring the safety and efficiency of coal mine production. Traditional equipment monitoring mainly relies on regular manual inspections and monitoring of single sensor data. However, with the improvement of sensor technology and data acquisition capabilities, coal mining equipment can generate a large amount of multi-source data, including vibration, temperature, current and other parameters. These data can provide rich information and help to understand the operating status of the equipment more comprehensively. However, due to the lack of effective integration and analysis capabilities of traditional monitoring methods for multi-source data, it is often difficult to identify abnormal conditions of equipment in real time and accurately. This situation makes it difficult to detect and prevent equipment failures in a timely manner, increasing equipment downtime and maintenance costs.

[0003] There are many shortcomings in the existing data analysis technology for coal mining equipment. First, most methods are limited to processing data from a single sensor and fail to fully utilize the complementary information of multi-source data, thus failing to provide a comprehensive assessment of the equipment's operating status. Second, in the process of data processing and analysis, existing technologies usually rely on simple statistical analysis or rule-based methods, which have limited ability to recognize complex abnormal patterns and are difficult to cope with abnormal situations of equipment under variable working conditions. In addition, due to the lack of adaptive capabilities of traditional data analysis methods, it is impossible to dynamically adjust model parameters according to changes in the operating status of the equipment, resulting in low accuracy of anomaly detection. In addition, many existing methods lack the application of deep learning models for the operating status of equipment, which makes them powerless when faced with large amounts of unstructured data and unable to mine deep-level features in the data. Therefore, the deficiencies of existing technologies in equipment health status assessment and anomaly detection limit their effectiveness and wide application in practical applications.

[0004] The existing coal mining equipment data analysis technology mainly faces problems such as insufficient data integration, low anomaly detection accuracy, and poor model adaptability. The coal mining equipment data analysis method invented by us based on the big data platform collects and preprocesses multi-source data, extracts and integrates different features, forms a unified equipment operation status feature set, and uses deep learning models for continuous optimization and anomaly detection, which effectively solves the above problems and belongs to the field of industrial equipment monitoring and safety protection technology. This method not only improves the accuracy of equipment status assessment, but also can timely detect potential equipment failures and optimize maintenance strategies, thereby significantly improving the safety and economy of coal mine production. Summary of the invention

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is: how to make full use of the multi-source data of coal mining equipment, through effective data preprocessing and feature extraction, combined with deep learning models to comprehensively evaluate the equipment operation status, timely detect and identify abnormal conditions, and optimize the equipment maintenance strategy, thereby improving the accuracy of equipment fault detection and maintenance efficiency, reducing downtime, and ensuring the safety and economy of coal mine production.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: A method for analyzing coal mining equipment data based on a big data platform, comprising: collecting multi-source data of coal mining equipment and preprocessing the data;

[0008] Extract features from preprocessed data and fuse features from different data sources to form a unified feature set of equipment operation status;

[0009] Build and train deep learning models, and optimize model parameters and structures through continuous learning to identify normal and abnormal states of equipment;

[0010] Perform maintenance actions based on abnormal conditions.

[0011] As a preferred solution of the coal mining equipment data analysis method based on the big data platform described in the present invention, the multi-source data collected from the coal mining equipment includes vibration data, temperature data, acoustic data, current data and historical operation data;

[0012] The preprocessing includes data cleaning, denoising, standardization and time synchronization processing.

[0013] As a preferred solution of the coal machinery equipment data analysis method based on the big data platform described in the present invention, wherein: the feature extraction from the preprocessed data includes obtaining the frequency domain feature X(f) of the vibration signal of the coal machinery equipment through frequency domain analysis, and its expression is:

[0014]

[0015] Among them, x[n] represents the time domain signal, N represents the signal length, f represents the frequency, j represents the imaginary unit, and X(f) is weighted as part of the feature set. The weight matrix is ​​W 1 ;

[0016] The trend feature T of temperature data is obtained through time series analysis, and its expression is:

[0017]

[0018] Among them, t 0 and tN They represent the starting point and end point of the time interval, t represents the time variable, φ i represents the coefficient of the autoregressive model, p represents the lag order of the autoregressive model, and x t-i Represents the temperature data at the lag time point, ∈ t represents white noise, α represents the attenuation coefficient of the time series feature; T is added to the feature set;

[0019] The Mel frequency cepstral coefficient feature MFCC(n) of acoustic data is obtained through voiceprint recognition, and its expression is:

[0020]

[0021] Where K represents the total number of Mel filters, E k Represents the energy of the kth filter; MFCC(n) is weighted and added to the feature set, and the weight matrix is ​​W 2 .

[0022] The final feature set F is expressed as:

[0023] F = {X(f), T, MFCC(n)}

[0024] Where β represents the characteristic attenuation coefficient.

[0025] As a preferred solution of the coal machinery equipment data analysis method based on the big data platform described in the present invention, the construction and training of the deep learning model includes performing forward propagation calculations, using the feature set F as input, and the formula is:

[0026] z (l) =g(W (l) F+b (l) )

[0027] Among them, z (l) represents the activation value of the lth layer, l is any layer number from 1 to L, W (l) represents the weight matrix of the lth layer, b (l) represents the bias vector of the lth layer, g(·) represents the activation function;

[0028] The loss function is calculated as follows:

[0029]

[0030] in, represents the loss value, m represents the number of samples, y i Indicates the actual value, represents the final predicted value;

[0031] Perform parameter optimization and feature fusion, the formula is:

[0032]

[0033] Among them, Z represents the optimized comprehensive features, λ and μ represent the regularization coefficients; represents the regularization term, which controls the change of the weight matrix over time.

[0034] As a preferred solution of the coal mining equipment data analysis method based on the big data platform described in the present invention, wherein: the normal and abnormal states of the equipment are identified, during the operation of the equipment, if the optimized comprehensive feature Z is within the normal state range, that is, Z∈[0, Z high ], the system uses a normal operating mode based on historical data to continuously monitor the equipment;

[0035] Normal state threshold Z high Through feature clustering and density estimation methods, the system performs real-time telemetry on all collected sensor data, and monitors and adjusts the status of key components of the equipment through remote control functions to keep the equipment in optimal operating state;

[0036] If the comprehensive feature Z is greater than the abnormal state threshold Z threshold , Z threshold <Z high , the system switches to early warning mode and increases the frequency of monitoring of key features;

[0037] When the comprehensive characteristic Z exceeds the normal range, that is, Z>Z high , the system enters the exception handling mode, starts the exception identification and isolation mechanism, and uses the multi-level anomaly detection algorithm to analyze and classify the anomalies to determine the type and severity of the anomaly.

[0038] As a preferred solution of the coal machinery equipment data analysis method based on the big data platform described in the present invention, the use of a multi-level anomaly detection algorithm to analyze and classify abnormal points includes performing preliminary anomaly detection on the comprehensive feature representation Z and screening out abnormal points that deviate from the normal range. The specific formula is:

[0039]

[0040] Among them, Z i represents the i-th eigenvalue, μ Z and σ Z Represent the mean and standard deviation of the feature set Z respectively;

[0041] If Anomaly i If it exceeds the preset threshold value θ, it is determined as a preliminary abnormal point;

[0042] For the outliers initially screened, a multi-level clustering algorithm is used to divide the forests and classify them;

[0043] The DBSCAN algorithm is used to perform density clustering on outliers to identify high-density outlier groups and noise points. The formula is:

[0044]

[0045] Among them, ∈ represents the cluster radius parameter, Cluster i Represents point x i The clustering result centered on is . Points that cannot be classified into any density cluster are marked as noise points. x represents any data point in the data space. Represents a d-dimensional real number space, where d represents the dimension of the data;

[0046] According to the clustering results, GMM is used to evaluate the probability of the severity of the outliers. GMM fits each clustering result, calculates the probability that each outlier belongs to a different category, and determines the type of anomaly based on the maximum likelihood estimate. The specific formula is:

[0047]

[0048] Among them, P(x i ) represents point x i The probability density of k represents the mixing coefficient of the k-th Gaussian distribution, Indicates that the mean is μ k , the covariance is ∑ k Gaussian distribution; determine the abnormal type according to the category with the maximum probability density; when P(x i ) exceeds the set threshold value Pθ, the fault state of the current device is considered to be seriously abnormal.

[0049] As a preferred solution of the coal machine equipment data analysis method based on the big data platform described in the present invention, the maintenance action includes that when the early warning mode is triggered, the system automatically increases the remote control strength of the equipment to prevent the occurrence of abnormal conditions by adjusting the equipment operating parameters. At the same time, the early warning information is fed back to the management personnel in real time through the remote communication module;

[0050] When the equipment is seriously abnormal, the equipment protection program is automatically triggered to perform remote shutdown or isolation operations. During this process, the remote fault indicator and multimodal sensor fusion technology are used to accurately locate the fault location, and the abnormality report is sent to the maintenance team. The system records and analyzes the entire process of the abnormality. Through deep learning of data and continuous optimization of models, the system's fault detection and processing capabilities are continuously improved. When the abnormal state is resolved, the system automatically performs a self-healing process to restore the normal operation of the equipment.

[0051] A coal mining equipment data analysis system based on a big data platform, characterized by: including:

[0052] Data acquisition module: collects multi-source data of coal mining equipment and pre-processes the data;

[0053] Data feature fusion module: extracts features from preprocessed data, fuses features from different data sources, and forms a unified feature set of equipment operation status;

[0054] Model building module: Build and train deep learning models, optimize model parameters and structures through continuous learning to identify normal and abnormal states of equipment;

[0055] Abnormal alarm module: alarm and maintenance actions are taken according to abnormal conditions.

[0056] The beneficial effects of the present invention are as follows: the present invention effectively integrates multi-source data to provide a comprehensive equipment operating status assessment; utilizes a deep learning model to improve the accuracy of anomaly detection; dynamically adjusts model parameters to adapt to equipment status changes; timely detects potential faults and optimizes maintenance strategies, thereby reducing downtime and improving the safety and economy of coal mine production. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the figures required for use in the description of the embodiments will be briefly introduced below. Obviously, the figures described below are only some embodiments of the present invention. For ordinary technicians in this field, other figures can be obtained based on these figures without paying creative work.

[0058] Figure 1 A flow chart of a coal machinery equipment data analysis method based on a big data platform provided for the first embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the specification drawings. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0060] Example 1, reference Figure 1 , which is an embodiment of the present invention, provides a coal machinery equipment data analysis method based on a big data platform, including:

[0061] S1: Collect multi-source data of coal mining equipment and pre-process the data.

[0062] The multi-source data collected from coal mining equipment include vibration data, temperature data, acoustic data, current data and historical operation data.

[0063] Preprocessing includes data cleaning, denoising, standardization and time synchronization.

[0064] It should be noted that collecting multi-source data of coal mining equipment is the basis for accurate monitoring and analysis. These data sources include vibration data, temperature data, acoustic data, current data and historical operation data, and each data type provides different dimensions of information on the operating status of the equipment. Vibration data reflects the dynamic characteristics of mechanical components, temperature data indicates the thermal energy state, acoustic data provides sound characteristics during mechanical operation, current data reveals the operating conditions of the electrical system, and historical operation data provides a background basis for analyzing the changing trends of equipment status. By integrating these multi-source data, a comprehensive and three-dimensional picture of the equipment status can be constructed, which helps to deeply understand the operating status and potential problems of the equipment.

[0065] Since the sampling frequency and time point of multi-source data may be different, direct data fusion will lead to inconsistency of time series characteristics, which will affect the accuracy of analysis. Time synchronization processing uses precise time alignment technology to map data collected at different times to the same time axis, which not only ensures the accuracy of data fusion, but also provides a basis for dynamic feature analysis. For example, the relationship between mechanical vibration and current fluctuation can only be accurately reflected on the basis of time synchronization.

[0066] S2: Extract features from the preprocessed data and fuse the features of different data sources to form a unified feature set of equipment operation status.

[0067] The frequency domain characteristics X(f) of the vibration signal of coal mining equipment are obtained through frequency domain analysis, and its expression is:

[0068]

[0069] Among them, x[n] represents the time domain signal, N represents the signal length, f represents the frequency, j represents the imaginary unit, and X(f) is weighted as part of the feature set. The weight matrix is ​​W 1 .

[0070] The trend feature T of temperature data is obtained through time series analysis, and its expression is:

[0071]

[0072] Among them, t 0 and t N They represent the starting point and end point of the time interval, t represents the time variable, φi represents the coefficient of the autoregressive model, p represents the lag order of the autoregressive model, and x t-i Represents the temperature data at the lag time point, ∈ t represents white noise, α represents the attenuation coefficient of the time series feature; T is added to the feature set.

[0073] The Mel frequency cepstral coefficient feature MFCC(n) of acoustic data is obtained through voiceprint recognition, and its expression is:

[0074]

[0075] Where K represents the total number of Mel filters, E k Represents the energy of the kth filter; MFCC(n) is weighted and added to the feature set, and the weight matrix is ​​W 2 .

[0076] The final feature set F is expressed as:

[0077] F = {X(f), T, MFCC(n)}

[0078] Where β represents the characteristic attenuation coefficient.

[0079] It should be noted that the accuracy and comprehensiveness of equipment operation status assessment have been significantly improved through the extraction and fusion of multi-source data features. Frequency domain analysis, time series analysis, and voiceprint recognition technologies have all achieved good results when applied independently to equipment fault diagnosis, but through their fusion, the operation characteristics of the equipment can be captured more comprehensively, and the sensitivity and accuracy of fault identification can be improved. In addition, this multi-dimensional feature fusion method solves the problem of insufficient information from a single data source, and can more effectively support the preventive maintenance and fault warning of equipment, thereby improving the equipment's operational reliability and maintenance efficiency.

[0080] S3: Build and train deep learning models, and optimize model parameters and structures through continuous learning to identify normal and abnormal states of equipment.

[0081] Perform forward propagation calculations, using feature set F as input, and the formula is:

[0082] z (l) =g(W (l) F+b (l) )

[0083] Among them, z (l) represents the activation value of the lth layer, l is any layer number from 1 to L, W (l) represents the weight matrix of the lth layer, b (l) represents the bias vector of the lth layer, and g(·) represents the activation function.

[0084] It should be noted that the deep features of the equipment's operating status are captured through the calculation of a multi-layer neural network. Unlike traditional shallow feature extraction methods, deep learning models can automatically extract complex feature patterns in data, especially when processing nonlinear and high-dimensional data. This feature enables the present invention to more accurately identify minor abnormal states of equipment and improve the sensitivity of fault detection.

[0085] The loss function is calculated as follows:

[0086]

[0087] in, represents the loss value, m represents the number of samples, y i Indicates the actual value, Represents the final predicted value.

[0088] Perform parameter optimization and feature fusion, the formula is:

[0089]

[0090] Among them, Z represents the optimized comprehensive features, λ and μ represent the regularization coefficients; Represents the regularization term, which controls the change of the weight matrix over time. The regularization method effectively avoids overfitting of the model and ensures the generalization ability of the model. In particular, the time-varying control term introduced can dynamically adjust the model parameters to better adapt to the changes in the device status. This dynamic adaptability is particularly important in scenarios where the device operating status changes frequently, and can provide more accurate and timely anomaly detection.

[0091] During the operation of the equipment, if the optimized comprehensive characteristic Z is within the normal range, that is, Z∈[0, Z high ], the system adopts normal operation mode based on historical data to continuously monitor the equipment.

[0092] Normal state threshold Z high Through feature clustering and density estimation methods, the system performs real-time telemetry on all collected sensor data, and monitors and adjusts the status of key components of the equipment through remote control functions to keep the equipment in optimal operating condition.

[0093] If the comprehensive feature Z is greater than the abnormal state threshold Z threshold , Z threshold <Z high , the system switches to early warning mode and increases the frequency of monitoring of key features.

[0094] When the comprehensive characteristic Z exceeds the normal range, that is, Z>Z high, the system enters the exception handling mode, starts the exception identification and isolation mechanism, and uses the multi-level anomaly detection algorithm to analyze and classify the anomalies to determine the type and severity of the anomaly.

[0095] Perform preliminary anomaly detection on the comprehensive feature representation Z to filter out abnormal points that deviate from the normal range. The specific formula is:

[0096]

[0097] Among them, Z i represents the i-th eigenvalue, μ Z and σ Z They represent the mean and standard deviation of the feature set Z respectively.

[0098] If Anomaly i If it exceeds the preset threshold θ, it is determined as a preliminary abnormal point.

[0099] For the outliers initially screened out, a multi-level clustering algorithm is used to divide the forests and classify them.

[0100] The DBSCAN algorithm is used to perform density clustering on outliers to identify high-density outlier groups and noise points. The formula is:

[0101]

[0102] Among them, ∈ represents the cluster radius parameter, Cluster i Represents point x i The clustering result centered on is . Points that cannot be classified into any density cluster are marked as noise points. x represents any data point in the data space. Represents a d-dimensional real number space, where d represents the dimension of the data.

[0103] According to the clustering results, GMM is used to evaluate the probability of the severity of the outliers. GMM fits each clustering result, calculates the probability that each outlier belongs to a different category, and determines the type of anomaly based on the maximum likelihood estimate. The specific formula is:

[0104]

[0105] Among them, P(x i ) represents point x i The probability density of k represents the mixing coefficient of the k-th Gaussian distribution, Indicates that the mean is μ k , the covariance is ∑ k Gaussian distribution; determine the abnormal type according to the category with the maximum probability density; when P(x i) exceeds the set threshold value Pθ, the fault state of the current device is considered to be seriously abnormal.

[0106] It should be noted that the combined application of preliminary anomaly detection, density clustering and GMM can effectively solve the problems of unclear identification of anomaly types and difficulty in quantifying severity in traditional methods. First, through standardized deviation calculation, the degree of abnormality of the feature value is effectively quantified, and preliminary anomalies worthy of attention are screened out. Secondly, the DBSCAN algorithm accurately identifies the group characteristics of anomalies by clustering the density distribution of data. Different from traditional methods such as K-means, this algorithm does not require a preset number of clusters and can naturally handle noise points. Finally, GMM further refines the classification and severity assessment of anomalies, and provides an intuitive quantitative evaluation standard through probability density functions. The combination of this series of steps not only enhances the early warning capability of equipment failures, but also provides a clear classification basis and quantitative evaluation for subsequent fault handling, providing important support for equipment maintenance and management.

[0107] S4: Perform maintenance actions according to the abnormal status.

[0108] When the early warning mode is triggered, the system automatically increases the remote control of the equipment and prevents the occurrence of abnormal conditions by adjusting the equipment operating parameters. At the same time, the early warning information is fed back to the management personnel in real time through the remote communication module.

[0109] When the equipment is seriously abnormal, the equipment protection program is automatically triggered to perform remote shutdown or isolation operations. During this process, the remote fault indicator and multimodal sensor fusion technology are used to accurately locate the fault location, and the abnormality report is sent to the maintenance team. The system records and analyzes the entire process of the abnormality. Through deep learning of data and continuous optimization of models, the system's fault detection and processing capabilities are continuously improved. When the abnormal state is resolved, the system automatically performs a self-healing process to restore the normal operation of the equipment.

[0110] It should be noted that according to the abnormal status of the equipment, the system performs corresponding maintenance actions to ensure the safe operation and efficient maintenance of the equipment. When the system triggers the early warning mode, the early warning mechanism is activated and the system automatically increases the remote control of the equipment. This includes preventing the further development of potential abnormalities by adjusting key operating parameters (such as power output, voltage, temperature, etc.). At this time, the early warning information is fed back to the management personnel in real time through the remote communication module, so that the operator can understand the status of the equipment in time and make necessary manual intervention. This remote control and real-time feedback mechanism not only helps to prevent the deterioration of faults, but also identifies and solves problems at an early stage, reduces equipment downtime, and improves the reliability and stability of the system.

[0111] When a serious abnormality is detected in the equipment, the system automatically triggers the equipment protection program and performs remote shutdown or isolation operations. This operation is achieved through remote fault indicators and multimodal sensor fusion technology, which can accurately locate the specific location of the equipment failure. This positioning technology combines data from multiple sensors, including vibration, temperature and acoustic data, to provide a multi-dimensional fault view to ensure the accuracy and reliability of positioning. In the case of serious abnormalities, immediate shutdown or isolation of faulty equipment can effectively prevent the expansion and spread of the fault and protect the safety of other equipment and systems. At the same time, the system generates a detailed abnormality report and sends it to the maintenance team through remote communication, so that maintenance personnel can quickly understand the details of the fault and formulate corresponding maintenance plans.

[0112] During the entire process of recording and analyzing abnormalities, the system uses deep learning and continuous model optimization technology to continuously improve fault detection and processing capabilities. This adaptive learning capability enables the system to adjust the detection model according to new fault modes, improving the accuracy and response speed of fault prediction. During the fault handling process, the system continuously accumulates and learns new data, continuously optimizes and updates model parameters, which not only improves the ability to identify unknown faults, but also better adapts to changes in the equipment operating environment.

[0113] When the abnormal state is resolved, the system automatically performs a self-healing process to restore the normal operation of the equipment. The self-healing process includes readjusting the operating parameters of the equipment, recalibrating sensors, and restoring normal monitoring and control modes. The automation of the self-healing process enables the equipment to quickly return to normal state, reduce downtime, reduce maintenance costs, and ensure the continued stable operation of the equipment.

[0114] The above embodiments also include a coal mining equipment data analysis system based on a big data platform, specifically:

[0115] Data acquisition module: collects multi-source data of coal mining equipment and pre-processes the data;

[0116] Data feature fusion module: extracts features from preprocessed data, fuses features from different data sources, and forms a unified feature set of equipment operation status;

[0117] Model building module: Build and train deep learning models, optimize model parameters and structures through continuous learning to identify normal and abnormal states of equipment;

[0118] Abnormal alarm module: alarm and maintenance actions are taken according to abnormal conditions.

[0119] The computer device may be a server. The computer device includes a processor, a memory, an input / output interface (I / O for short) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data cluster data of the power monitoring system. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method for analyzing coal machinery equipment data based on a big data platform is implemented.

[0120] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0121] Example 2 is an embodiment of the present invention, which provides a coal machinery equipment data analysis method based on a big data platform. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation / comparative experiments.

[0122] This experiment aims to verify the effectiveness of a data analysis method for coal mining equipment based on a big data platform in equipment fault detection and maintenance. The experimental objects include six coal mining equipment, numbered M1 to M6. Each device is equipped with a variety of sensors to collect multi-source data, covering vibration, temperature, acoustics, current and historical operation information. Specifically, vibration data is recorded by accelerometers, temperature data is obtained by thermocouples, acoustic data is collected by acoustic sensors, and current data is monitored by current transformers. In addition, historical operation data of the equipment is also collected, including cumulative working hours and past fault records.

[0123] The data preprocessing step first cleans the collected data to remove outliers and noise signals. Then, all data are standardized to eliminate differences caused by different units and magnitudes. Time synchronization ensures the time consistency of different data sources and unifies the timestamps of each data stream.

[0124] In the feature extraction and fusion stage, the main frequency components were extracted from the vibration signal through frequency domain analysis, the time series characteristics of the temperature data were analyzed by using the autoregressive model, and the Mel-frequency cepstral coefficients (MFCC) of the acoustic data were extracted through voiceprint recognition. These features were fused into a comprehensive feature set to represent the overall operating status of the device.

[0125] Next, a deep learning model was built and trained based on the extracted feature set. The model uses a multi-layer neural network structure and is optimized using the cross-entropy loss function. During the training process, the model parameters are continuously adjusted through forward propagation and back-propagation algorithms to minimize the prediction error. To prevent overfitting, early stopping and cross-validation techniques were used in the experiment.

[0126] In actual applications, the system performs different maintenance measures based on the abnormal conditions detected. For equipment that enters the early warning mode, the system automatically adjusts its operating parameters and sends early warning information to management personnel in real time through remote communication. If a serious abnormality is detected, the system immediately executes a protection mechanism, including remote shutdown and equipment isolation, while using multimodal sensor technology to accurately locate the fault point and generate a detailed abnormality report. The specific experimental data are shown in Table 1.

[0127] Table 1 Implementation data table

[0128]

[0129] It can be observed from the experimental data that there are significant differences in the multi-source data of each device, especially in terms of vibration, temperature and acoustic data. These data provide an important basis for the status assessment of the equipment. For example, equipment M5 and M6 showed higher vibration and temperature data, reaching 0.70g, 0.68g and 85℃, 83℃ respectively, and the acoustic data also reached 95dB and 93dB, exceeding the normal operating levels of other equipment. The increase in current data further indicates that these devices may be in a high-load operation state. Historical operation data also shows that M5 and M6 have the longest cumulative operating time, 6500 hours and 6400 hours respectively, and the most failures, 5 times and 4 times respectively.

[0130] Through the training and application of deep learning models, the system can accurately identify the status of each device. M5 and M6 are detected as abnormal, which is consistent with their historical fault records and current monitoring data. In contrast, M1 and M3 are judged to be normal, and the actual data also shows that their parameters are within the normal range. M2 and M4 are in the early warning state, which is consistent with their slightly higher vibration and temperature data, and the relatively high number of historical faults.

[0131] Compared with traditional single parameter monitoring, the present invention provides a more comprehensive equipment status assessment capability through deep fusion and analysis of multi-source data. In particular, the accuracy and sensitivity of fault detection are significantly improved in the extraction of data features and the application of deep learning models. The combined use of DBSCAN algorithm and GMM not only effectively identifies abnormal groups, but also quantifies the severity of abnormalities, providing a quantitative basis for maintenance decisions.

[0132] In addition, the experimental results show that the system can effectively reduce the impact of equipment failures through early warning mode and automatic maintenance measures under abnormal conditions. For example, in the early warning state, timely parameter adjustment and early warning information feedback reduce the risk of possible equipment damage. In abnormal conditions, automatic shutdown and fault location ensure rapid isolation and processing of faults. This fully automatic monitoring and maintenance mechanism not only improves the operational safety of equipment, but also reduces maintenance costs and downtime, providing strong protection for industrial production.

[0133] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A coal mining equipment data analysis method based on a big data platform, characterized in that: include: Collect multi-source data of coal mining equipment and pre-process the data; Extract features from preprocessed data and fuse features from different data sources to form a unified feature set of equipment operation status; Build and train deep learning models, and optimize model parameters and structures through continuous learning to identify normal and abnormal states of equipment; Perform maintenance actions based on abnormal conditions.

2. The method for analyzing coal mining equipment data based on a big data platform according to claim 1, characterized in that: The multi-source data collected from the coal mining equipment includes vibration data, temperature data, acoustic data, current data and historical operation data; The preprocessing includes data cleaning, denoising, standardization and time synchronization processing.

3. The method for analyzing coal mining equipment data based on a big data platform according to claim 2, characterized in that: The feature extraction from the preprocessed data includes obtaining the frequency domain feature X(f) of the vibration signal of the coal mining equipment through frequency domain analysis, and the expression thereof is: Where x[n] represents the time domain signal, N represents the signal length, f represents the frequency, j represents the imaginary unit, and X(f) is weighted as part of the feature set, and the weight matrix is ​​W1; The trend feature T of temperature data is obtained through time series analysis, and its expression is: Among them, t0 and t N They represent the starting point and end point of the time interval, t represents the time variable, φ i represents the coefficient of the autoregressive model, p represents the lag order of the autoregressive model, and x t-i Represents the temperature data at the lag time point, ∈ t represents white noise, α represents the attenuation coefficient of the time series feature; T is added to the feature set; The Mel frequency cepstral coefficient feature MFCC(n) of acoustic data is obtained through voiceprint recognition, and its expression is: Where K represents the total number of Mel filters, E k Represents the energy of the kth filter; MFCC(n) is weighted and added to the feature set, and the weight matrix is ​​W2. The final feature set F is expressed as: F = {X(f), T, MFCC(n)} Where β represents the characteristic attenuation coefficient.

4. The method for analyzing coal mining equipment data based on a big data platform according to claim 3, characterized in that: The construction and training of the deep learning model includes performing forward propagation calculations, using the feature set F as input, and the formula is: z (l) =g(W (l) F+b (l) ) Among them, z (l) represents the activation value of the lth layer, l is any layer number from 1 to L, W (l) represents the weight matrix of the lth layer, b (l) represents the bias vector of the lth layer, g(·) represents the activation function; The loss function is calculated as follows: in, represents the loss value, m represents the number of samples, y i Indicates the actual value, represents the final predicted value; Perform parameter optimization and feature fusion, the formula is: Among them, Z represents the optimized comprehensive features, λ and μ represent the regularization coefficients; represents the regularization term, which controls the change of the weight matrix over time.

5. The method for analyzing coal mining equipment data based on a big data platform according to claim 4, characterized in that: The identification of the normal and abnormal state of the device includes, during the operation of the device, if the optimized comprehensive feature Z is within the normal state range, that is, Z∈[0,Z high ], the system uses a normal operating mode based on historical data to continuously monitor the equipment; Normal state threshold Z high Through feature clustering and density estimation methods, the system performs real-time telemetry on all collected sensor data, and monitors and adjusts the status of key components of the equipment through remote control functions to keep the equipment in optimal operating state; If the comprehensive feature Z is greater than the abnormal state threshold Z threshold , Z threshold <Z high , the system switches to early warning mode and increases the frequency of monitoring of key features; When the comprehensive characteristic Z exceeds the normal range, that is, Z>Z high , the system enters the exception handling mode, starts the exception identification and isolation mechanism, and uses the multi-level anomaly detection algorithm to analyze and classify the anomalies to determine the type and severity of the anomaly.

6. The method for analyzing coal mining equipment data based on a big data platform according to claim 5, characterized in that: The use of a multi-level anomaly detection algorithm to analyze and classify anomalies includes performing preliminary anomaly detection on the comprehensive feature representation Z and screening out anomalies that deviate from the normal range. The specific formula is: Among them, Z i represents the i-th eigenvalue, μ Z and σ Z Represent the mean and standard deviation of the feature set Z respectively; If Anomaly i If it exceeds the preset threshold value θ, it is determined as a preliminary abnormal point; For the outliers initially screened, a multi-level clustering algorithm is used to divide the forests and classify them; The DBSCAN algorithm is used to perform density clustering on outliers to identify high-density outlier groups and noise points. The formula is: Among them, ∈ represents the cluster radius parameter, Cluster i Represents point x i The clustering result centered on is . Points that cannot be classified into any density cluster are marked as noise points. x represents any data point in the data space. Represents a d-dimensional real number space, where d represents the dimension of the data; According to the clustering results, GMM is used to evaluate the probability of the severity of the outliers. GMM fits each clustering result, calculates the probability that each outlier belongs to a different category, and determines the type of anomaly based on the maximum likelihood estimate. The specific formula is: Among them, P(x i ) represents point x i The probability density of k represents the mixing coefficient of the k-th Gaussian distribution, Indicates that the mean is μ k , the covariance is Σ k Gaussian distribution; determine the abnormal type according to the category with the maximum probability density; when P(x i ) exceeds the set threshold value Pθ, the fault state of the current device is considered to be seriously abnormal.

7. The method for analyzing coal mining equipment data based on a big data platform according to claim 6, characterized in that: The maintenance action includes that when the early warning mode is triggered, the system automatically increases the remote control strength of the equipment to prevent the occurrence of abnormal conditions by adjusting the equipment operation parameters. At the same time, the early warning information is fed back to the management personnel in real time through the remote communication module; When the equipment is seriously abnormal, the equipment protection program is automatically triggered to perform remote shutdown or isolation operations. During this process, the remote fault indicator and multimodal sensor fusion technology are used to accurately locate the fault location, and the abnormality report is sent to the maintenance team. The system records and analyzes the entire process of the abnormality, and continuously improves the system's fault detection and processing capabilities through deep learning of data and continuous optimization of models. When the abnormal state is resolved, the system automatically performs a self-healing process to restore the normal operation of the equipment.

8. A coal mining equipment data analysis system based on a big data platform using the method according to any one of claims 1 to 7, characterized in that: Data acquisition module: collects multi-source data of coal mining equipment and pre-processes the data; Data feature fusion module: extracts features from preprocessed data, fuses features from different data sources, and forms a unified feature set of equipment operation status; Model building module: Build and train deep learning models, optimize model parameters and structures through continuous learning to identify normal and abnormal states of equipment; Abnormal alarm module: alarm and maintenance actions are taken according to abnormal conditions.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of 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 of the method according to any one of claims 1 to 7 are implemented.

Images