A coal machine equipment operation and maintenance method and system based on a data integration platform

Abstract

The application discloses a coal machine equipment operation and maintenance method and system based on a data integration platform, relates to the technical field of coal machine equipment operation and maintenance, and comprises the following steps: collecting and fusing sensor data and external data of the coal machine equipment in real time through a data integration platform; performing abnormality detection and fault prediction on the coal machine equipment based on a fault prediction model; and generating a personalized operation and maintenance strategy and automatically updating a strategy library according to a fault prediction result. The coal machine equipment operation and maintenance method based on the data integration platform provided by the application reduces the false positive rate and the false negative rate, enhances the adaptability, reduces the equipment failure rate and the operation and maintenance cost through the intelligent abnormality detection and the fault prediction model, improves the foresight and effectiveness of the operation and maintenance, reduces the equipment failure risk and prolongs the service life of the equipment through the personalized operation and maintenance strategy generation and feedback closed loop, and the application achieves better effects in the false positive rate, the adaptability and the operation and maintenance cost.

Description

Technical Field

[0001] This invention relates to the field of coal mining equipment operation and maintenance technology, specifically a coal mining equipment operation and maintenance method based on a data integration platform. Background Technology

[0002] With the continuous improvement of the intelligence and automation level of coal mining equipment, equipment operation and maintenance technology has gradually become an important link in ensuring the continuity and safety of coal production. Traditional coal mining equipment operation and maintenance mainly relies on manual periodic inspections and routine maintenance. This method is not only inefficient and costly, but may also fail to detect potential equipment faults in a timely manner. In recent years, coal mining equipment operation and maintenance methods based on data integration platforms have been gradually researched and applied. By integrating multi-source sensor data and applying intelligent algorithms for real-time monitoring and predictive analysis, proactive operation and maintenance management of equipment can be achieved. In particular, with the support of technologies such as deep learning, intelligent sensing, and edge computing, the status monitoring and fault prediction capabilities of coal mining equipment have been significantly improved, enabling equipment maintenance to gradually shift from "passive response" to "proactive prediction." This technological development provides important technical support and data assurance for the efficient operation and maintenance of coal mining equipment.

[0003] However, existing operation and maintenance methods based on data integration platforms still have certain shortcomings, especially in terms of data real-time performance, prediction accuracy, and optimization of operation and maintenance strategies. First, data real-time performance is key to effective operation and maintenance. Traditional data acquisition and processing modes cannot ensure the accuracy and stability of real-time data and are difficult to adapt to the complex operating environment of coal mining equipment. Second, although deep learning models have been applied to fault prediction, due to the complexity of equipment status and the variability of environmental conditions, existing models often lack adaptive update capabilities, resulting in insufficient prediction accuracy and difficulty in meeting actual operation and maintenance needs. Finally, personalized operation and maintenance strategies have not been fully optimized in existing systems. Existing technologies lack a closed-loop feedback mechanism and cannot dynamically adjust operation and maintenance strategies according to changes in equipment status and maintenance effectiveness, thus limiting the level of intelligent operation and maintenance of coal mining equipment. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is that existing coal mining equipment operation and maintenance methods suffer from insufficient real-time data, low accuracy in fault prediction, lack of dynamic adjustment of operation and maintenance strategies, and the problem of how to optimize the efficient and intelligent operation and maintenance of coal mining equipment.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a coal mining equipment operation and maintenance method based on a data integration platform, comprising: real-time collection and fusion of sensor data and external data of the coal mining equipment through the data integration platform; anomaly detection and fault prediction of the coal mining equipment based on a fault prediction model; and generation of personalized operation and maintenance strategies and automatic updating of the strategy library based on the fault prediction results.

[0007] As a preferred embodiment of the coal mining equipment operation and maintenance method based on a data integration platform described in this invention, the real-time acquisition and fusion of sensor data and external data from the coal mining equipment includes the introduction of an intelligent filtering mechanism based on weighted average filtering. This mechanism evaluates the historical accuracy and real-time volatility of each sensor's data, assigning different weights accordingly. When processing multi-source data, it analyzes the noise levels and error rates of different sensors and sets initial weights based on relative stability. Simultaneously with data acquisition, the system automatically detects and records real-time data fluctuations for each sensor, reducing the weight of sensor data with error rates or noise exceeding preset values, and prioritizing the fusion of high-weight data. The intelligent filtering mechanism is expressed as follows:

[0008]

[0009] Among them, D f D represents the merged data value. i Let be the data from the i-th sensor, be the weight of the i-th sensor, and σ be the weight of the i-th sensor. i Let be the standard deviation of the data from the i-th sensor.

[0010] As a preferred embodiment of the coal mining equipment operation and maintenance method based on a data integration platform described in this invention, the real-time acquisition and fusion of sensor data and external data from the coal mining equipment further includes setting a time window during the data fusion process. Only data within the time window is allowed to enter the fusion model. The length of the time window is determined comprehensively by the equipment's operating speed, data acquisition frequency, and maintenance requirements. The time window filters out old data exceeding the time range, periodically monitors data latency, and automatically adjusts the time window length based on the actual latency distribution. The time window model is represented as follows:

[0011] D rt ={D i |t c -t i ≤ΔT}

[0012] Among them, D rt For a valid dataset within a time window, t c t represents the current time. i D i The acquisition time is ΔT, where ΔT is the length of the time window.

[0013] As a preferred embodiment of the coal mining equipment operation and maintenance method based on a data integration platform described in this invention, the anomaly detection and fault prediction of the coal mining equipment includes a deep learning fault prediction model. Based on a deep neural network, this model learns the complex relationships of equipment status from multi-dimensional sensor data to predict potential faults. The deep learning fault prediction model has a multi-layer neuron structure. Each layer of neurons further analyzes the output of the previous layer, extracting deep-level features from the identification data layer by layer. Historical data is introduced during training, including sensor readings and actual fault conditions under different fault modes, enabling the model to learn to identify various types of anomalies. During operation, the model receives multi-dimensional sensor data as input in real time, and outputs the probability value of fault occurrence through the operation of layers of neurons and weight adjustment. The deep learning fault prediction model is expressed as follows:

[0014] P(Y=1|X=σ(W) n ·σ(W n-1 …σ(W1·X+b1)+b2)+…+b n )

[0015] Where σ is the activation function, X is the input layer sensor data vector, P is the fault probability, and W n and b n These are the weight matrix and bias for each layer.

[0016] As a preferred embodiment of the coal mining equipment operation and maintenance method based on a data integration platform described in this invention, the anomaly detection and fault prediction of the coal mining equipment further includes adaptive updating of anomaly data based on Kalman filtering. Kalman filtering optimizes the current equipment state estimate by fusing new data with previous state estimates. Whenever new data enters the system, Kalman filtering recalculates the optimal estimate of the current state, enabling the model to continuously adapt to the dynamically changing equipment environment. The adaptive updating of anomaly data is expressed as follows:

[0017] X t+1 =AX t +BU t +K(Z t -HX t )

[0018] Among them, X t Let A be the state estimate at time t, A be the state transition matrix, B be the control input matrix, and U be the state estimate at time t. t For control input, K is the Kalman gain, Z... t Here, H represents the measured value, and H is the observation matrix.

[0019] As a preferred embodiment of the coal mining equipment operation and maintenance method based on a data integration platform described in this invention, the generation of personalized operation and maintenance strategies and automatic updating of the strategy library includes analyzing the equipment's health status, estimated maintenance time, and spare parts requirements. By comparing various operation and maintenance options, the strategy with the lowest cost and highest efficiency is selected. When generating the operation and maintenance strategy, the execution cost of various operation and maintenance activities, the price of equipment spare parts, and the necessity of spare parts replacement are evaluated. Combining the current status of the equipment and historical data, it is determined whether spare parts need to be replaced or should be replaced later. The minimum cost of the operation and maintenance strategy is expressed as:

[0020]

[0021] Where C is the total operation and maintenance cost, c j M represents the replacement cost of the j-th spare part. j p represents the number of times the i-th spare part can be replaced. j Let T be the execution cost of the j-th maintenance project. j Let be the execution time of the j-th maintenance item.

[0022] As a preferred embodiment of the coal mining equipment operation and maintenance method based on a data integration platform described in this invention, the generation of personalized operation and maintenance strategies and automatic updating of the strategy library further includes automatically recording the maintenance effect after equipment maintenance is completed, updating the operation and maintenance strategy library based on the data, adjusting the strategy library using actual operation and maintenance data, so that the system continuously optimizes predictions and operation and maintenance plans, and identifies the current optimal maintenance plan by analyzing historical data and the latest feedback. Operation and maintenance optimization is expressed as follows:

[0023]

[0024] Where f(x) represents the effect of the policy in the current policy library, x + The current optimal strategy This is the expectation operator.

[0025] Another objective of this invention is to provide a coal mining equipment operation and maintenance system based on a data integration platform, which can perform anomaly detection and fault prediction on coal mining equipment through a fault prediction module, thus solving the problems of current systems being unable to adapt to dynamic environmental changes and having low prediction accuracy.

[0026] As a preferred embodiment of the coal mining equipment operation and maintenance system based on a data integration platform according to the present invention, it includes: a data acquisition and fusion module, a fault prediction module, and an operation and maintenance strategy module; the data acquisition and fusion module is used to collect and fuse sensor data and external data of the coal mining equipment in real time through the data integration platform; the fault prediction module is used to perform anomaly detection and fault prediction on the coal mining equipment based on a fault prediction model; the operation and maintenance strategy module is used to generate personalized operation and maintenance strategies and automatically update the strategy library based on the fault prediction results.

[0027] A computer device includes a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program as a step in implementing a coal mining equipment operation and maintenance method based on a data integration platform.

[0028] A computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of a coal mining equipment operation and maintenance method based on a data integration platform.

[0029] The beneficial effects of this invention are as follows: The coal mining equipment operation and maintenance method based on a data integration platform provided by this invention, through the design of intelligent filtering mechanisms and time window screening mechanisms, performs real-time collection and fusion of multi-source data from different sensors, realizing dynamic data integration of coal mining equipment, reducing false alarm rate and false negative rate, and enhancing adaptability. Through intelligent anomaly detection and fault prediction models, it achieves accurate analysis of equipment status and anomaly early warning, reducing equipment failure rate and operation and maintenance costs, and improving the foresight and effectiveness of operation and maintenance. Through personalized operation and maintenance strategy generation and feedback closed loop, it reduces equipment failure risk and extends equipment service life. This invention achieves better results in terms of false alarm rate, adaptability, and operation and maintenance cost. Attached Figure Description

[0030] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort. Wherein:

[0031] Figure 1 The first embodiment of the present invention provides an overall flowchart of a coal mining equipment operation and maintenance method based on a data integration platform.

[0032] Figure 2 The third embodiment of the present invention provides an overall module diagram of a coal mining equipment operation and maintenance system based on a data integration platform. Detailed Implementation

[0033] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0034] Example 1

[0035] Reference Figure 1 As an embodiment of the present invention, a method for the operation and maintenance of coal mining equipment based on a data integration platform is provided, comprising:

[0036] S1: Through the data integration platform, sensor data and external data from coal mining equipment are collected and integrated in real time.

[0037] Furthermore, the real-time acquisition and fusion of sensor data and external data from coal mining equipment includes the introduction of an intelligent filtering mechanism based on weighted average filtering.

[0038] It should be noted that by evaluating the historical accuracy and real-time volatility of each sensor's data and assigning different weights, when processing multi-source data, the noise levels and error rates of different sensors are analyzed, and initial weights are set based on relative stability. During data acquisition, the system automatically detects and records the real-time data fluctuations of each sensor, reducing the weight of sensor data with error rates or noise exceeding preset values, and prioritizing the fusion of high-weight data. This intelligent filtering mechanism is represented as follows:

[0039]

[0040] Among them, D f D represents the merged data value. i Let be the data from the i-th sensor, be the weight of the i-th sensor, and σ be the weight of the i-th sensor. i Let be the standard deviation of the data from the i-th sensor.

[0041] Furthermore, the real-time acquisition and fusion of sensor data and external data from coal mining equipment is also included in the data fusion process.

[0042] It should be noted that a time window is set, allowing only data within that time window to enter the fusion model. The length of the time window is determined by a combination of factors, including the device's operating speed, data acquisition frequency, and maintenance requirements. The time window filters out old data that exceeds the specified time range. Data latency is monitored periodically, and the time window length is automatically adjusted based on the actual latency distribution. The time window model is represented as follows:

[0043] D rt ={D i |t c -t i ≤ΔT}

[0044] Among them, D rt For a valid dataset within a time window, t c t represents the current time. i D i The acquisition time is ΔT, where ΔT is the length of the time window.

[0045] S2: Based on the fault prediction model, perform anomaly detection and fault prediction on coal mining equipment.

[0046] Furthermore, anomaly detection and fault prediction for coal mining equipment include deep learning fault prediction models.

[0047] It should be noted that, based on deep neural networks, the deep learning fault prediction model learns the complex relationships of equipment states from multi-dimensional sensor data to predict potential faults. The model has a multi-layered neuron structure, with each layer's neurons further analyzing the output of the previous layer, extracting deeper features from the identification data layer by layer. Historical data, including sensor readings and actual fault conditions under different fault modes, is incorporated during training, enabling the model to learn to identify various types of anomalies. During runtime, the model receives multi-dimensional sensor data as input in real time, and through the computation and weight adjustment of each layer of neurons, outputs the probability of fault occurrence. The deep learning fault prediction model is represented as follows:

[0048] P(Y=1|X=σ(W) n ·σ(W n-1 …σ(W1·X+b1)+b2)+…+b n )

[0049] Where σ is the activation function, X is the input layer sensor data vector, P is the fault probability, and W n and b n These are the weight matrix and bias for each layer.

[0050] Furthermore, anomaly detection and fault prediction for coal mining equipment also include adaptive updating of anomaly data based on Kalman filtering.

[0051] It should be noted that Kalman filtering optimizes the current device state estimate by fusing new data with previous state estimates. Whenever new data enters the system, Kalman filtering recalculates the optimal estimate of the current state, allowing the model to continuously adapt to the dynamically changing device environment. Adaptive updates for abnormal data are represented as follows:

[0052] X t+1 =AX t +BU t +K(Z t -HX t )

[0053] Among them, X t Let A be the state estimate at time t, A be the state transition matrix, B be the control input matrix, and U be the state estimate at time t. t For control input, K is the Kalman gain, Z... t Here, H represents the measured value, and H is the observation matrix.

[0054] S3: Based on the fault prediction results, generate personalized operation and maintenance strategies and automatically update the strategy library.

[0055] Furthermore, it generates personalized operation and maintenance strategies and automatically updates the strategy library, including analyzing the health status of equipment, estimated maintenance time, and spare parts requirements.

[0056] It should be noted that by comparing various maintenance options, the strategy with the lowest cost and highest efficiency is selected. When generating the maintenance strategy, the execution cost of various maintenance activities, the price of equipment spare parts, and the necessity of spare parts replacement are evaluated. Combined with the current status of the equipment and historical data, it is determined whether spare parts need to be replaced or whether replacement should be postponed. The minimum cost of the maintenance strategy is expressed as:

[0057]

[0058] Where C is the total operation and maintenance cost, c j M represents the replacement cost of the j-th spare part. j p represents the number of times the i-th spare part can be replaced. j Let T be the execution cost of the j-th maintenance project. j Let be the execution time of the j-th maintenance item.

[0059] Furthermore, generating personalized operation and maintenance strategies and automatically updating the strategy library also includes automatically recording the maintenance results after equipment maintenance is completed.

[0060] It should be noted that the operation and maintenance strategy library is updated based on data, and adjustments are made to the strategy library using actual operation and maintenance data. This allows the system to continuously optimize predictions and operation and maintenance plans. By analyzing historical data and the latest feedback, the optimal maintenance plan is identified. Operation and maintenance optimization is represented as follows:

[0061]

[0062] Where f(x) represents the effect of the policy in the current policy library, x + The current optimal strategy This is the expectation operator.

[0063] Example 2

[0064] One embodiment of the present invention provides a coal mining equipment operation and maintenance method based on a data integration platform. To verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0065] The test subjects were six coal mining machines. Their performance on key parameters such as multi-source data fusion accuracy, fault prediction accuracy, real-time data response time, reduced operation and maintenance costs, increased maintenance cycle, and increased equipment uptime was recorded. The test preparation work included building a data integration platform and installing various sensor devices to collect temperature, vibration, pressure, and environmental information of each coal mining machine in real time. The platform was configured with intelligent filtering and time window screening mechanisms to ensure the real-time performance and accuracy of data fusion. In addition, deep neural network models and adaptive Kalman filters were used to realize equipment fault prediction and dynamic updates. After the equipment status prediction was completed, the platform would generate personalized maintenance plans. Through a closed-loop feedback mechanism, the results of each maintenance were fed back to the strategy library for continuous optimization.

[0066] During implementation, real-time data from the equipment is first collected and processed through the platform. After filtering, the data from each device is monitored in real time for fluctuations in multi-source data to remove noise interference. Subsequently, the data integration platform detects and evaluates the status of the equipment based on a fault prediction model, issues early warnings for potential faults, and generates personalized operation and maintenance strategies that cover the maintenance cycle, spare parts requirements, and optimal maintenance methods for each device. After each maintenance, the platform feeds back the latest status of the equipment and maintenance results to the strategy library, automatically adjusting subsequent maintenance strategies to achieve continuous optimization of the equipment status. The entire process strictly follows the predetermined experimental procedures to ensure the scientific validity and accuracy of the results.

[0067] As shown in Table 1, the multi-source data fusion accuracy of each device is above 95%, with the highest reaching 98.7%. This indicates that the intelligent filtering and time window screening mechanisms effectively improve the data integration accuracy. Compared with the false alarm problem caused by data noise in the traditional operation and maintenance mode, the fusion accuracy of this invention is significantly enhanced, ensuring the reliability of monitoring data and providing a solid data foundation for subsequent analysis. The fault prediction accuracy ranges from 93.0% to 95.9% among the test objects, which is a significant improvement compared with the error caused by the lack of adaptive adjustment in traditional prediction models. Especially under dynamic environmental conditions, the real-time update of the prediction model is achieved through adaptive Kalman filtering, ensuring the high reliability of the prediction results. This predictive advantage provides early warning. It provides higher accuracy and further reduces the equipment failure rate. The real-time data response time is between 150 and 162 milliseconds, indicating that the data integration platform can complete data fusion processing within a millisecond response time. Compared with the delayed feedback problem in traditional equipment operation and maintenance, this invention achieves real-time management of equipment status through a time window filtering mechanism, which helps to improve the system response speed. The average operation and maintenance cost of each device is reduced by about 19.6%, with a maximum reduction of up to 21.2%. Due to the personalized operation and maintenance strategy and closed-loop feedback mechanism of this invention, while extending the maintenance cycle and reducing unnecessary spare parts replacement, the operation and maintenance cost is optimized. The average maintenance cycle of the equipment reaches more than 45 days, which significantly extends the maintenance interval and reduces the operation and maintenance frequency and labor costs.

[0068] Table 1 Experimental Data

[0069]

[0070]

[0071] Example 3

[0072] Reference Figure 2 As an embodiment of the present invention, a coal mining equipment operation and maintenance system based on a data integration platform is provided, including: a data acquisition and fusion module, a fault prediction module, and an operation and maintenance strategy module.

[0073] The data acquisition and fusion module is used to collect and fuse sensor data and external data of coal mining equipment in real time through the data integration platform; the fault prediction module is used to perform anomaly detection and fault prediction of coal mining equipment based on the fault prediction model; and the operation and maintenance strategy module is used to generate personalized operation and maintenance strategies and automatically update the strategy library based on the fault prediction results.

[0074] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0075] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0076] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0077] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc. It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0078] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for the operation and maintenance of coal mining equipment based on a data integration platform, characterized in that, include: Through the data integration platform, sensor data and external data from coal mining equipment are collected and integrated in real time; Based on the fault prediction model, anomaly detection and fault prediction are performed on coal mining equipment. Based on the fault prediction results, personalized operation and maintenance strategies are generated and the strategy library is automatically updated; The real-time acquisition and fusion of sensor data and external data from coal mining equipment includes the introduction of an intelligent filtering mechanism based on weighted average filtering. This intelligent filtering mechanism is expressed as follows: Among them, D f D represents the merged data value. i For the data from the i-th sensor, w i Let σ be the weight of the i-th sensor. i Let be the standard deviation of the data from the i-th sensor. During data fusion, a time window is set, allowing only data within that time window to enter the fusion model. The time window model is represented as follows: D rt ={D i |t c -t i ≤ΔT} Among them, D rt For a valid dataset within a time window, t c t represents the current time. i D i The acquisition time, ΔT is the length of the time window; The anomaly detection and fault prediction of the coal mining equipment includes a deep learning fault prediction model, which is expressed as follows: P(Y=1|Y)=σ(W n ·σ(W n-1 …σ(W1·X+b1)+b2)+…+b n ) Where σ is the activation function, X is the input layer sensor data vector, P is the fault probability, and W n and b n These are the weight matrix and bias for each layer, respectively; Adaptive update of outlier data based on Kalman filtering, which is expressed as: X t+1 =AX t +BU t +K(Z t -HX t ) Among them, X t Let A be the state estimate at time t, A be the state transition matrix, B be the control input matrix, and U be the state estimate at time t. t For control input, K is the Kalman gain, Z... t Here, H represents the measured values, and H is the observation matrix. The process of generating personalized operation and maintenance strategies and automatically updating the strategy library includes analyzing the health status of equipment, estimated maintenance time, and spare parts requirements. By comparing various operation and maintenance options, the strategy with the lowest cost and optimal efficiency is selected. The minimum cost of the operation and maintenance strategy is expressed as: Where C is the total operation and maintenance cost, c j M represents the replacement cost of the j-th spare part. j p represents the number of times the i-th spare part can be replaced. j Let T be the execution cost of the j-th maintenance project. j Let j be the execution time of the j-th maintenance item; After equipment maintenance is completed, the system automatically records the maintenance results and updates the operation and maintenance strategy library based on the data. It then adjusts the strategy library using actual operation and maintenance data, continuously optimizing predictions and operation and maintenance plans. By analyzing historical data and the latest feedback, it identifies the optimal maintenance plan. Operation and maintenance optimization is represented as follows: Where f(x) represents the effect of the policy in the current policy library, x + The current optimal strategy This is the expectation operator.

2. The coal mining equipment operation and maintenance method based on a data integration platform as described in claim 1, characterized in that: The real-time acquisition and fusion of sensor data and external data from coal mining equipment includes the introduction of an intelligent filtering mechanism based on weighted average filtering. By evaluating the historical accuracy and real-time fluctuations of each sensor's data, different weights are assigned. When processing multi-source data, the noise level and error rate of different sensors are analyzed, and initial weights are set based on relative stability. While acquiring data, the system automatically detects and records the real-time data fluctuations of each sensor, reduces the weight of sensor data with error rates or noise exceeding preset values, and prioritizes the fusion of high-weight data.

3. The coal mining equipment operation and maintenance method based on a data integration platform as described in claim 2, characterized in that: The real-time acquisition and fusion of sensor data and external data from coal mining equipment also includes setting a time window during the data fusion process. Only data within the time window is allowed to enter the fusion model. The length of the time window is determined by a combination of the equipment's operating speed, data acquisition frequency, and maintenance requirements. The time window filters out old data that exceeds the time range, regularly monitors data latency, and automatically adjusts the time window length based on the actual latency distribution.

4. The coal mining equipment operation and maintenance method based on a data integration platform as described in claim 3, characterized in that: The anomaly detection and fault prediction of coal mining equipment includes a deep learning fault prediction model. Based on a deep neural network, it learns the complex relationships of equipment status from multi-dimensional sensor data to predict potential faults. The deep learning fault prediction model has a multi-layer neuron structure. Each layer of neurons further analyzes the output of the previous layer, extracting deep features from the identification data layer by layer. Historical data is introduced during training, including sensor readings and actual fault conditions under different fault modes, enabling the model to learn to identify various types of anomalies. During operation, the model receives multi-dimensional sensor data as input in real time, and outputs the probability value of fault occurrence through the operation of layers of neurons and weight adjustment.

5. The coal mining equipment operation and maintenance method based on a data integration platform as described in claim 4, characterized in that: The anomaly detection and fault prediction of coal mining equipment also includes adaptive updating of anomaly data based on Kalman filtering. Kalman filtering optimizes the current equipment state estimate by fusing new data with previous state estimates. Whenever new data enters the system, Kalman filtering recalculates the optimal estimate of the current state, enabling the model to continuously adapt to the dynamically changing equipment environment.

6. The coal mining equipment operation and maintenance method based on a data integration platform as described in claim 5, characterized in that: The process of generating personalized operation and maintenance strategies and automatically updating the strategy library includes analyzing the health status of the equipment, the estimated maintenance time, and the spare parts requirements. By comparing various operation and maintenance options, the strategy with the lowest cost and the best efficiency is selected. When generating operation and maintenance strategies, the execution cost of various operation and maintenance activities, the price of equipment spare parts, and the necessity of spare parts replacement are evaluated. Based on the current status of the equipment and historical data, it is determined whether spare parts need to be replaced or whether replacement should be postponed.

7. A system employing the coal mining equipment operation and maintenance method based on a data integration platform as described in any one of claims 1 to 6, characterized in that: It includes a data acquisition and fusion module, a fault prediction module, and an operation and maintenance strategy module; The acquisition and fusion module is used to acquire and fuse sensor data and external data from coal mining equipment in real time through a data integration platform; The fault prediction module is used to perform anomaly detection and fault prediction on coal mining equipment based on the fault prediction model. The operation and maintenance strategy module is used to generate personalized operation and maintenance strategies based on the fault prediction results and automatically update the strategy library.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the coal mining equipment operation and maintenance method based on the data integration platform as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the coal mining equipment operation and maintenance method based on the data integration platform as described in any one of claims 1 to 6.

Images