Guiding sliding shoe fault prediction method and system based on intelligent algorithm

Abstract

The invention relates to the technical field of guide sliding shoe fault prediction, in particular to a guide sliding shoe fault prediction method and system based on an intelligent algorithm, and the method comprises the steps: collecting sliding shoe operation data in real time, carrying out the feature extraction of the sliding shoe operation data, and obtaining an operation state feature set; performing mode recognition on the running feature set, recognizing different modes in the running process of the sliding shoe, and classifying the recognized different modes to obtain a running mode classification result; based on the operation mode classification result, the sliding shoe operation data are grouped, and an operation data grouped data set is obtained; inputting the operation data packet data set into a pre-trained fault probability analysis model, calculating the fault probability of the guide sliding shoe in each operation mode, and sorting the fault probability of the guide sliding shoe in each operation mode to obtain a fault probability list; and comparing the fault probability of the guide sliding shoe in each operation mode in the fault probability list with a preset fault threshold value.

Description

Technical Field

[0001] This invention relates to the technical field of guide shoe failure prediction, and in particular to a guide shoe failure prediction method and system based on intelligent algorithms. Background Technology

[0002] In modern industrial and infrastructure construction, guide shoes, as key components, are widely used in various large-scale mechanical equipment, bridge sliding systems, building relocation projects, and other fields. Their operating status directly affects the safety, stability, and service life of the entire system. However, due to the long-term exposure to heavy loads, friction, vibration, and other complex working conditions, guide shoes are prone to wear, fatigue, deformation, and other failures, which can lead to serious consequences such as equipment downtime and engineering accidents. Therefore, real-time and accurate fault prediction of guide shoes is of great significance for ensuring system safety, improving operational efficiency, and reducing maintenance costs.

[0003] Traditional methods for detecting guide shoe failures mainly rely on periodic manual inspections. This approach is not only inefficient but also has significant limitations. Manual inspections struggle to provide real-time monitoring and cannot promptly capture subtle changes in the guide shoe during operation. Often, the problem is only discovered after it has progressed to a certain extent, which can lead to construction delays, increased project costs, and even safety accidents. Therefore, there is an urgent need for a guide shoe failure prediction method and system based on intelligent algorithms. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a guide shoe fault prediction method based on intelligent algorithms that can ensure system safety, improve operating efficiency, and reduce maintenance costs.

[0005] In a first aspect, the present invention provides a guide shoe fault prediction method based on intelligent algorithms, the method comprising:

[0006] Real-time acquisition of ski boot operation data, and feature extraction of ski boot operation data to obtain operation status feature set;

[0007] Pattern recognition is performed on the running feature set to identify different patterns in the running process of the skis, and the identified different patterns are classified to obtain the running pattern classification results;

[0008] Based on the classification results of the operating mode, the ski boot operating data is grouped to obtain the operating data group dataset;

[0009] The operational data group dataset is input into a pre-trained failure probability analysis model to calculate the probability of guide shoe failure under each operational mode. The probability of guide shoe failure under each operational mode is then compiled to obtain a failure probability list.

[0010] The probability of guide shoe failure in each operating mode in the failure probability list is compared with a preset failure threshold. The probability of failure in operating modes that exceed the preset failure threshold is sorted to obtain a failure factor priority list.

[0011] Based on the fault factor priority list, match the associated operating mode classification results and issue an early warning.

[0012] Furthermore, the skid running data includes temperature data, pressure data, vibration data, displacement data, speed data, and running time data.

[0013] Furthermore, methods for pattern recognition of running feature sets include cluster analysis and classification algorithms.

[0014] Furthermore, the operation mode classification results include normal operation mode, slight wear mode, severe wear mode, fatigue mode, and deformation mode.

[0015] Furthermore, the method for constructing the fault probability analysis model includes:

[0016] Collect operational data of the guide shoe and preprocess the collected data;

[0017] Choose a machine learning model as the basic architecture for the failure probability analysis model;

[0018] The preprocessed data is divided into a training set, a validation set, and a test set; the training set is used to train the model, the validation set is used to adjust the model's hyperparameters to avoid overfitting, and the test set is used to evaluate the model's final performance.

[0019] The trained failure probability analysis model is deployed to a real industrial system.

[0020] Furthermore, the factors influencing the setting of the preset fault threshold include historical fault data analysis, equipment performance specifications, operating environment conditions, system security requirements, and maintenance strategies.

[0021] Furthermore, the method for matching and issuing alerts based on the associated operational mode classification results includes:

[0022] Different warning levels are set based on the degree to which the probability of failure exceeds the threshold under different operating modes in the failure factor priority list;

[0023] For each warning level, generate corresponding warning information;

[0024] Choose the appropriate warning method based on the actual application scenario and needs;

[0025] The generated warning information will be sent to the relevant operators according to the selected warning method;

[0026] Record and archive relevant information for each warning.

[0027] Furthermore, the warning information includes the operating mode of the guide shoe, the probability value of failure, the type of failure, and the recommended measures.

[0028] Furthermore, the warning methods include sound alarms, flashing lights, SMS notifications, email alerts, and system pop-up prompts.

[0029] On the other hand, this application also provides a guide shoe failure prediction system based on intelligent algorithms, the system comprising:

[0030] The data acquisition module collects real-time data on the operation of the skates and extracts features from the data to obtain a set of operational status features.

[0031] The pattern recognition and classification module performs pattern recognition on the running feature set, identifies different patterns in the running process of the skis, and classifies the identified different patterns to obtain the running pattern classification results.

[0032] The data grouping module groups the ski boot running data based on the running mode classification results to obtain the running data group dataset;

[0033] The fault probability list generation module and the fault probability analysis module input the grouped dataset of operating data into the pre-trained fault probability analysis model, calculate the probability of guide shoe failure under each operating mode, organize the probability of guide shoe failure under each operating mode, and obtain the fault probability list.

[0034] The fault threshold comparison module compares the probability of guide shoe failure in each operating mode in the fault probability list with a preset fault threshold, sorts the probability of failure in operating modes that exceed the preset fault threshold, and obtains a fault factor priority list.

[0035] The early warning module matches the associated operating mode classification results with the fault factor priority list and issues an early warning.

[0036] Compared with the prior art, the beneficial effects of the present invention are as follows: the method can grasp the working status of the skids in real time by collecting skid running data; the method can track the changes in its pressure, displacement and other data in real time, overcome the lag of manual inspection, discover potential fault hazards in time, avoid the faults from developing to a serious degree before they are detected, and effectively reduce the risk of construction delays and engineering accidents.

[0037] By extracting features from the collected data to obtain a set of operating status features, and through pattern recognition and classification, different operating modes of the skid can be accurately distinguished. In complex construction environments, this method can accurately distinguish different modes such as normal wear and wear caused by abnormal force, discover potential fault modes in advance, and provide a basis for taking measures to avoid faults in advance, thereby ensuring the safety and stability of the entire system.

[0038] Based on the classification results of the operating modes, the skid operation data is grouped to form an operating data group dataset, making subsequent analysis more targeted. Data under different operating modes are processed separately, which can deeply explore the patterns and characteristics of data under each mode, better analyze the correlation between different modes and faults, and improve the accuracy of fault prediction.

[0039] By using a pre-trained fault probability analysis model to calculate the probability of guide shoe failure under each operating mode, it is more scientific and reliable than manual judgment based on experience. By comparing and sorting the fault probabilities with preset thresholds, a fault factor priority list is obtained, allowing staff to clearly understand the probability of failure under different operating modes. In actual work, maintenance resources can be reasonably allocated based on this priority, and situations with high fault probabilities can be dealt with first, thereby improving fault handling efficiency, avoiding resource waste, and reducing maintenance costs.

[0040] By matching the associated operating mode classification results with the fault factor priority list and issuing early warnings, relevant personnel can be notified in a timely manner before a fault occurs. Through multiple early warning methods, staff can be ensured to take swift action to avoid serious consequences such as equipment downtime and engineering accidents caused by failure to detect and handle faults in a timely manner, thereby ensuring the safe operation of the system and reducing economic losses.

[0041] In summary, the above-mentioned intelligent algorithm-based guide shoe fault prediction method can ensure system safety, improve operating efficiency, and reduce maintenance costs. Attached Figure Description

[0042] Figure 1 This is a flowchart of the present invention;

[0043] Figure 2 This is a structural diagram of a guide shoe fault prediction method and system based on intelligent algorithms. Detailed Implementation

[0044] As will be apparent to those skilled in the art from the description of this application, this application can be implemented as a method, apparatus, electronic device, and computer-readable storage medium. Therefore, this application can be specifically implemented in the following forms: entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software. Furthermore, in some embodiments, this application can also be implemented as a computer program product contained in one or more computer-readable storage media, which includes computer program code.

[0045] The aforementioned computer-readable storage medium may be any combination of one or more computer-readable storage media. Computer-readable storage media include: electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media include: portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, flash memory, optical fiber, optical disc read-only memory, optical storage devices, magnetic storage devices, or any combination thereof. In this application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0046] The acquisition, storage, use, and processing of data in this application all comply with relevant national laws and regulations.

[0047] This application describes the provided methods, apparatus, and electronic devices using flowcharts and / or block diagrams.

[0048] It should be understood that each block of a flowchart and / or block diagram, as well as combinations of blocks in a flowchart and / or block diagram, can be implemented by computer-readable program instructions. These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine that, when executed by a computer or other programmable data processing apparatus, creates means for implementing the functions / operations specified in the blocks of the flowchart and / or block diagram.

[0049] These computer-readable program instructions may also be stored in a computer-readable storage medium that enables a computer or other programmable data processing device to function in a particular manner. In this way, the instructions stored in the computer-readable storage medium produce an instruction apparatus product that includes the functions / operations specified in the blocks of a flowchart and / or block diagram.

[0050] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus or other device to produce a computer-implemented process, such that the instructions that execute on the computer or other programmable data processing apparatus provide a process for implementing the functions / operations specified in the blocks of the flowchart and / or block diagram.

[0051] This application will now be described with reference to the accompanying drawings.

[0052] like Figure 1 As shown, the guide shoe fault prediction method based on intelligent algorithms of the present invention specifically includes the following steps:

[0053] Real-time acquisition of ski boot operation data, and feature extraction of ski boot operation data to obtain operation status feature set;

[0054] Pattern recognition is performed on the running feature set to identify different patterns in the running process of the skis, and the identified different patterns are classified to obtain the running pattern classification results;

[0055] Based on the classification results of the operating mode, the ski boot operating data is grouped to obtain the operating data group dataset;

[0056] The operational data group dataset is input into a pre-trained failure probability analysis model to calculate the probability of guide shoe failure under each operational mode. The probability of guide shoe failure under each operational mode is then compiled to obtain a failure probability list.

[0057] The probability of guide shoe failure in each operating mode in the failure probability list is compared with a preset failure threshold. The probability of failure in operating modes that exceed the preset failure threshold is sorted to obtain a failure factor priority list.

[0058] Based on the fault factor priority list, match the associated operating mode classification results and issue an early warning.

[0059] This method can grasp the working status of the skids in real time by collecting skid running data; it can track changes in pressure, displacement and other data in real time, overcome the lag of manual inspection, discover potential faults in time, and avoid the faults from developing to a serious level before they are discovered, effectively reducing construction delays and the risk of engineering accidents.

[0060] By extracting features from the collected data to obtain a set of operating status features, and through pattern recognition and classification, different operating modes of the skid can be accurately distinguished. In complex construction environments, this method can accurately distinguish different modes such as normal wear and wear caused by abnormal force, discover potential fault modes in advance, and provide a basis for taking measures to avoid faults in advance, thereby ensuring the safety and stability of the entire system.

[0061] Based on the classification results of the operating modes, the skid operation data is grouped to form an operating data group dataset, making subsequent analysis more targeted. Data under different operating modes are processed separately, which can deeply explore the patterns and characteristics of data under each mode, better analyze the correlation between different modes and faults, and improve the accuracy of fault prediction.

[0062] By using a pre-trained fault probability analysis model to calculate the probability of guide shoe failure under each operating mode, it is more scientific and reliable than manual judgment based on experience. By comparing and sorting the fault probabilities with preset thresholds, a fault factor priority list is obtained, allowing staff to clearly understand the probability of failure under different operating modes. In actual work, maintenance resources can be reasonably allocated based on this priority, and situations with high fault probabilities can be dealt with first, thereby improving fault handling efficiency, avoiding resource waste, and reducing maintenance costs.

[0063] By matching the associated operating mode classification results with the fault factor priority list and issuing early warnings, relevant personnel can be notified in a timely manner before a fault occurs. Through multiple early warning methods, staff can be ensured to take swift action to avoid serious consequences such as equipment downtime and engineering accidents caused by failure to detect and handle faults in a timely manner, thereby ensuring the safe operation of the system and reducing economic losses.

[0064] The skateboard operation data includes:

[0065] Temperature data: During operation, the skis generate heat due to friction and heavy loads, causing the temperature to rise. By monitoring the temperature of the skis in real time, it is possible to determine whether overheating is occurring, and thus predict possible wear or failure.

[0066] Pressure data: When a skate is subjected to heavy loads, it will generate pressure. Pressure data can reflect the stress on the skate and its interaction with the contact surface. Abnormal pressure changes may indicate problems with the skate's support structure or uneven load distribution.

[0067] Vibration data: The skates will vibrate during operation, and the vibration signals contain a wealth of operating status information. By collecting and analyzing vibration data, the vibration frequency, amplitude and other characteristics of the skates can be identified, and then it can be determined whether there is abnormal vibration.

[0068] Displacement data: The displacement of the skate during operation can reflect the skate's motion trajectory and stability; abnormal displacement means that there is a problem with the skate's guide mechanism, or that the skate itself is deformed or has other faults.

[0069] Speed ​​data: Changes in speed can reflect the working status of the skate and its matching with the drive mechanism; abnormal speed means that there is a problem with the drive mechanism of the skate, or that the skate itself has a malfunction such as jamming.

[0070] Running time data: By recording the running time of the skates, we can understand the usage and life cycle of the skates; combined with other running data, we can analyze the relationship between the failure rate of the skates and the running time, providing a basis for predicting the remaining life of the skates.

[0071] Real-time acquisition of sliding shoe operation data: Various types of sensors are installed on the sliding shoes and related equipment; pressure sensors are used to measure the pressure borne by the sliding shoes during operation. Due to the uneven pressure distribution and changes with working conditions in different parts of the sliding shoes during the sliding process of the steel structure, accurate pressure data helps to determine the load-bearing status of the sliding shoes; displacement sensors monitor the displacement changes of the sliding shoes in real time, and even small displacement anomalies may indicate potential faults; temperature sensors record the temperature of the sliding shoes during operation. Continuous heavy loads and friction will cause the temperature of the sliding shoes to rise, and temperature data can reflect its wear degree and working stability; at the same time, hydraulic system sensors collect pressure and flow data of the hydraulic system, because the hydraulic system provides power support for the guide sliding shoes, and its parameter changes can indirectly reflect the operating status of the sliding shoes; collecting equipment operating time and speed information can help analyze the operating characteristics of the sliding shoes under different working durations and speeds;

[0072] Feature extraction of skateboard running data includes: The raw data collected contains noise and is massive in volume, making direct analysis difficult; therefore, feature extraction is necessary. Filtering algorithms are used to remove noise, especially high-frequency noise interference, making the data smoother and more accurate, preventing noise from misleading subsequent analysis. Normalization maps data of different ranges and magnitudes to a specific interval, eliminating the influence of data dimensions and improving the convergence speed and accuracy of subsequent analysis models, making different types of data comparable. In the feature extraction stage, various features are obtained from the processed data. In the time domain, the mean of the pressure data is calculated, reflecting the average pressure the skateboard experiences over a period; variance is calculated to reflect the dispersion of pressure fluctuations, with larger variance indicating more unstable pressure changes. Frequency domain features are obtained through techniques such as Fourier transform; frequency distribution and power spectrum reveal the energy distribution of data at different frequency components, helping to analyze the impact of vibrations at different frequencies during skateboard operation.

[0073] In this embodiment, by acquiring various types of ski boot operation data in real time, such as temperature, pressure, vibration, displacement, speed, and running time, it is possible to comprehensively and meticulously monitor various state changes of the ski boot during operation. During data acquisition, multiple types of sensors are used to ensure the accuracy and reliability of the data. These sensors can accurately measure various operating parameters of the ski boot, providing high-quality data support for subsequent feature extraction and fault prediction. In the feature extraction stage, filtering algorithms are used to remove noise and interference from the data, making the data smoother and more accurate. This helps to avoid noise misleading subsequent analysis and improves the accuracy of fault prediction. Through normalization processing, the collected data of different ranges and magnitudes are uniformly mapped to a specific interval. This not only eliminates the influence of data dimensions but also improves the convergence speed and accuracy of subsequent analysis models, making different types of data comparable. Various features are extracted from the processed data, which can reflect key information in the ski boot operation process. Through real-time acquisition and feature extraction, a feature set reflecting the ski boot's operating state is obtained.

[0074] Pattern recognition is performed on the running feature set to identify different patterns in the running process of the skis, and the identified patterns are classified to obtain the running pattern classification results:

[0075] Pattern recognition methods include: clustering analysis, which divides data points into different clusters, such that data points within the same cluster have similar characteristics, while data points between different clusters have significant differences; in guide shoe fault prediction, clustering analysis can be used to perform preliminary pattern division of the operating feature set and identify different operating state modes; classification algorithms learn the characteristics of data points of known categories to predict the category of data points of unknown categories; in guide shoe fault prediction, classification algorithms can be used to classify the operating feature set and divide different operating state modes into normal mode, abnormal mode, or fault mode.

[0076] Before performing pattern recognition, the feature set is preprocessed, including data cleaning, normalization, and standardization, to ensure data quality and consistency.

[0077] The selected pattern recognition method is used to perform pattern recognition and classification on the preprocessed running feature set, and the classification results are evaluated.

[0078] After pattern recognition and classification, the operation mode classification results of the guide shoe are obtained;

[0079] The operation mode classification results include:

[0080] Normal operating mode: This is the mode in which the skates are in normal operating condition, indicating that all performance indicators of the skates are within the normal range and there are no signs of abnormal wear, fatigue or deformation. The normal operating mode is usually used as a benchmark for comparison and analysis with other abnormal modes.

[0081] Slight wear mode: When the skates show slight wear, their operating status will change slightly; the slight wear mode indicates that the skates have entered the initial wear stage and their development trend needs to be closely monitored;

[0082] Severe wear mode: As wear intensifies, the operating condition of the skates will change significantly, such as a substantial increase in friction, a significant increase in vibration frequency, and an increase in temperature. Severe wear mode indicates that the skates are about to reach the end of their service life and need to be replaced or repaired as soon as possible.

[0083] Fatigue mode: Long-term exposure to heavy loads and vibrations may cause fatigue in the slipper, which manifests as a decrease in material strength and the formation of cracks; fatigue mode can lead to sudden breakage or failure of the slipper, causing serious accidents.

[0084] Deformation mode: In some cases, the skate may deform due to uneven stress or material problems, resulting in changes in its shape and size. The deformation mode will affect the fit accuracy and operational stability of the skate and its matching components, and timely adjustment or replacement is required.

[0085] In this embodiment, pattern recognition transforms complex data from the skate's operation into different operating modes. Classifying these modes allows for a clear distinction between normal, abnormal, or fault modes, leading to more accurate predictions of skate malfunctions. Pattern recognition and classification facilitate a deeper understanding of the skate's characteristics under different operating conditions, providing strong support for subsequent fault analysis and maintenance. Analyzing the characteristics of different modes reveals potential problems and risks during skate operation, providing a basis for preventative maintenance. Based on the results of pattern recognition and classification, real-time early warnings of skate malfunctions can be achieved, enabling timely detection and handling of potential problems. When the slipper operates abnormally or malfunctions, the system can quickly issue an alarm to notify relevant personnel to take measures and avoid serious consequences such as equipment downtime and engineering accidents. Through pattern recognition and classification, maintenance plans can be formulated more accurately, avoiding unnecessary maintenance operations and reducing maintenance costs. Different maintenance strategies can be adopted for different operating modes, improving maintenance efficiency and quality and extending the service life of the slipper. Accurate fault prediction and real-time early warning help improve the safety and stability of the entire system. By promptly detecting and handling slipper malfunctions, chain reactions caused by slipper failures can be avoided, ensuring the normal operation of the entire system.

[0086] Based on the classification results of the operating modes, the ski boot operating data is grouped to obtain the operating data group dataset:

[0087] For each collected data point, a corresponding label is assigned based on its pattern recognition result; if the skis are in normal working condition during a certain period of time, the data for that period of time will be marked as "normal"; if signs of initial wear are detected, the corresponding data segment will be marked as "initial wear".

[0088] Based on the assigned label information, the entire dataset is divided into multiple subsets, each representing a specific operating mode;

[0089] After completing the initial grouping, check whether the data in each group is complete and consistent, and ensure that the data within each group truly reflects the characteristics of the labeled operating mode;

[0090] Organize the collected grouped data to construct a runtime data grouped dataset. Use a database table format, with each group as a data table, where each row represents a data record and each column represents a different parameter. During the dataset construction process, ensure the integrity and consistency of the data, while also considering the efficiency of data storage and retrieval, so that this data can be efficiently input into the fault probability analysis model for calculation.

[0091] In this embodiment, data points are labeled according to the classification results of operating modes, ensuring that each data point clearly belongs to a specific operating mode. This facilitates independent analysis and processing for each operating mode, improving the targeting and efficiency of data processing. Dividing the data into subsets representing different operating modes ensures that the data within each subset reflects the characteristics of that operating mode. This helps the fault probability analysis model learn the fault characteristics under each operating mode more accurately, thereby improving the accuracy of fault prediction. The use of database tables to construct the operating data grouped dataset makes data management and maintenance more convenient. Each group serves as a data table, with each row representing a data record and each column representing different parameters, facilitating data query, update, and deletion operations. During dataset construction, checking the completeness and consistency of data in each group, and ensuring that the data within each group truly reflects the characteristics of the labeled operating mode, ensures data integrity and consistency. Considering data storage and retrieval efficiency during dataset construction, employing appropriate database technologies and storage structures significantly improves data retrieval and storage efficiency.

[0092] The operational data sets are grouped and input into a pre-trained failure probability analysis model. The probability of guide shoe failure under each operational mode is calculated, and the failure probabilities of guide shoe failure under each operational mode are compiled to obtain a failure probability list.

[0093] The method for constructing the failure probability analysis model includes:

[0094] Collect a large amount of operational data on guide shoes. This data should cover different working conditions, including different load levels, operating speeds, and working environments. Data sources can include equipment monitoring data and experimental test data from actual industrial production. At the same time, record the corresponding guide shoe failure information as label data for supervised learning.

[0095] The collected data is preprocessed, including data cleaning to remove noise, outliers and missing values; data with different features are normalized to the same scale to avoid some features having too large an impact on model training due to their large numerical range.

[0096] Based on the characteristics and data properties of guide shoe failure prediction, a machine learning model is selected as the basic architecture of the failure probability analysis model; the machine learning model includes artificial neural networks, support vector machines, decision trees, and random forests.

[0097] The preprocessed data is divided into training, validation, and test sets. The training set is used for model training, the validation set is used to adjust the model's hyperparameters to avoid overfitting, and the test set is used to evaluate the model's final performance. The training data is input into the model, and the prediction results are calculated through forward propagation. Then, the error between the prediction results and the true labels is calculated based on the loss function, and the error is backpropagated to each layer of the model to update the model's parameters. During training, the loss and evaluation metrics on the validation set are monitored to adjust the model's hyperparameters. The trained model is evaluated using the test set data, and the values ​​of various evaluation metrics are calculated. If the model's performance meets the requirements, it can be applied to actual guide shoe failure prediction. If the model's performance is substandard, the reasons need to be analyzed, and the model needs to be adjusted and improved.

[0098] The trained fault probability analysis model is deployed into a real industrial system and integrated with the skid running data acquisition system and fault early warning system.

[0099] In this embodiment, by collecting a large amount of guide shoe operation data covering different working conditions and recording the corresponding guide shoe failures as label data, a rich and accurate data foundation is provided for supervised learning. This enables the model to learn the relationship between guide shoe operation status and failures under various conditions, thereby improving the model's accuracy and generalization ability in predicting failures under different working conditions. Preprocessing operations such as data cleaning and normalization are performed on the collected data to effectively remove noise, outliers, and missing values, avoiding the adverse effects of large differences in the range of data feature values ​​on model training, improving data quality, and thus enhancing the efficiency and accuracy of model training. Based on the characteristics of guide shoe failure prediction and data features, a suitable infrastructure is selected from various machine learning models, providing multiple possible solutions for model construction. The most suitable model is selected based on the actual situation to achieve better fault prediction results. The data is divided into training, validation, and test sets. By monitoring the loss and evaluation metrics on the validation set during training, the model hyperparameters are adjusted to effectively avoid overfitting and ensure the model's performance stability on different datasets. At the same time, the test set is used to perform final evaluation of the model to ensure that the model's performance meets the requirements in practical applications, thereby improving the model's reliability and practicality. The trained fault probability analysis model is deployed to the actual industrial system and integrated with the guide shoe operation data acquisition system and fault early warning system, realizing a complete process from data acquisition and model analysis to fault early warning. It can predict guide shoe faults in real time, providing effective technical support for ensuring system safety, improving operating efficiency, and reducing maintenance costs.

[0100] The probability of guide shoe failure in each operating mode of the failure probability list is compared with a preset failure threshold. The probability of failure in operating modes that exceeds the preset failure threshold is sorted to obtain a failure factor priority list:

[0101] The factors influencing the setting of the preset fault threshold include:

[0102] Historical Failure Data Analysis: By analyzing historical failure records, we can understand the frequency, type, and severity of failures of guide shoes under different operating conditions. Based on this data, we can statistically determine the probability distribution of failures, providing a benchmark for setting failure thresholds. By analyzing abnormal signs before failures in historical data, we can determine the time window between these signs and the occurrence of failures, thereby helping to set thresholds that can provide early warnings.

[0103] Equipment performance specifications: The design parameters, material properties, and expected service life of guide shoes will affect their failure rate; high-performance or specially designed guide shoes may have higher wear resistance and fatigue resistance, so their failure threshold can be set accordingly higher;

[0104] Operating environment conditions: The ambient temperature, humidity, dust concentration, vibration level, etc. of the skis will affect their performance and service life; harsh environmental conditions will accelerate the wear and fatigue of the skis, so it is necessary to set a more sensitive fault threshold; consider seasonal changes or environmental differences under specific working conditions, and adjust the fault threshold in a timely manner to adapt to different conditions.

[0105] System security requirements: Based on the system's security requirements, the fault threshold should be set to ensure sufficient warning time before a fault occurs, so that preventive measures can be taken;

[0106] Maintenance strategy: The maintenance strategy will affect the setting of the failure threshold; preventive maintenance may allow for a higher failure threshold because regular inspections can compensate for the inaccuracy of prediction; considering maintenance costs and downtime, the setting of the failure threshold should balance the accuracy of failure prediction with the economy of maintenance intervention.

[0107] In this embodiment, by setting a reasonable fault threshold and monitoring the fault probability in real time, potential fault risks can be detected in a timely manner, providing a sufficient time window for taking preventive measures and effectively reducing the likelihood and impact of faults. Ranking the operating modes that exceed the fault threshold by probability and generating a fault factor priority list allows for precise identification of the operating modes most likely to cause faults, providing clear guidance for maintenance and repair work, and improving maintenance efficiency and accuracy. The setting of the preset fault threshold comprehensively considers factors such as historical fault data analysis, equipment performance specifications, operating environment conditions, system safety requirements, and maintenance strategies, ensuring the scientific and reasonable nature of the fault threshold and improving the accuracy and reliability of fault prediction. By considering seasonal changes or environmental differences under specific operating conditions, adjusting the fault threshold in a timely manner, and balancing the accuracy of fault prediction according to maintenance strategies and economic considerations, this step has strong adaptability and flexibility, meeting the fault prediction needs of different scenarios. This step, through scientific and reasonable fault threshold setting and fault probability ranking, achieves early warning and precise location of guide shoe faults, improving system safety and reliability, and reducing maintenance costs and downtime.

[0108] Based on the fault factor priority list, match the associated operating mode classification results and issue an alert:

[0109] The fault factor priority list is sorted from high to low fault probability. Each item in the list represents the probability of guide shoe failure under a specific operating mode. By interpreting the fault factor priority list, it is possible to quickly identify which operating modes are currently in a high-risk state and require immediate attention.

[0110] Match the high-risk operating modes in the failure factor priority list with the operating mode classification results obtained in step S2; each mode in the operating mode classification results represents the operating state of the guide shoe under specific working conditions; through matching, it is possible to determine which actual operating modes are associated with high-risk failure factors.

[0111] Once a high-risk operating mode is identified, an early warning mechanism needs to be triggered to notify relevant personnel.

[0112] Methods for matching and associating operational mode classification results and issuing alerts include:

[0113] Based on the degree to which the probability of failure exceeds the threshold under different operating modes in the failure factor priority list, and the degree of harm that may be caused, different warning levels are set, such as general warning, important warning and emergency warning.

[0114] For each warning level, detailed warning information is generated; the warning information includes the operating mode of the guide shoe, the failure probability value, the type of failure, and the recommended measures.

[0115] Based on the actual application scenario and needs, select the warning method to issue a warning; the warning methods include sound alarm, flashing light, mobile phone SMS notification, email reminder and system pop-up prompt;

[0116] The generated warning information will be sent to the relevant operators according to the selected warning method; ensuring that the information can be delivered to all personnel who need to know in a timely and accurate manner so that they can take appropriate measures quickly.

[0117] Recording and archiving relevant information for each warning, including warning time, warning level, warning information content, and recipients, helps in subsequent performance evaluation of the fault prediction system and analysis and summary of guide shoe failures, providing a basis for optimizing fault prediction models and maintenance strategies.

[0118] In this embodiment, the fault factor priority list can quickly identify which operating modes are currently in a high-risk state and require immediate attention; this helps to promptly detect potential faults, shorten fault response time, and reduce losses caused by faults; accurately matching high-risk operating modes with operating mode classification results can determine which actual operating modes are associated with high-risk fault factors; this ensures the accuracy and relevance of warnings, avoids false alarms and missed alarms, and improves the reliability of the warning system; different warning levels are set, and detailed warning information is generated according to the warning level; this helps operators quickly judge the fault situation and take corresponding measures based on the severity and specific content of the warning; multiple warning methods such as sound alarms, flashing lights, SMS notifications, email reminders, and system pop-up prompts can be selected according to actual application scenarios and needs; this ensures that warning information can be delivered to all personnel who need to know in a timely and accurate manner, regardless of their location, so that they can quickly receive the warning and react; recording and archiving relevant information for each warning helps to evaluate the performance of the fault prediction system and analyze and summarize the fault situation of the guide shoe, providing a basis for optimizing the fault prediction model and maintenance strategy.

[0119] like Figure 2 As shown, the guide shoe fault prediction method and system based on intelligent algorithms of the present invention specifically includes the following modules;

[0120] The data acquisition module collects real-time data on the operation of the skates and extracts features from the data to obtain a set of operational status features.

[0121] The pattern recognition and classification module performs pattern recognition on the running feature set, identifies different patterns in the running process of the skis, and classifies the identified different patterns to obtain the running pattern classification results.

[0122] The data grouping module groups the ski boot running data based on the running mode classification results to obtain the running data group dataset;

[0123] The fault probability list generation module and the fault probability analysis module input the grouped dataset of operating data into the pre-trained fault probability analysis model, calculate the probability of guide shoe failure under each operating mode, organize the probability of guide shoe failure under each operating mode, and obtain the fault probability list.

[0124] The fault threshold comparison module compares the probability of guide shoe failure in each operating mode in the fault probability list with a preset fault threshold, sorts the probability of failure in operating modes that exceed the preset fault threshold, and obtains a fault factor priority list.

[0125] The early warning module matches the associated operating mode classification results with the fault factor priority list and issues an early warning.

[0126] The system collects real-time data on the operation of the skis through a data acquisition module, ensuring continuous monitoring of the skis' operating status and avoiding the lag of manual inspection. The efficient processing of modules such as feature extraction, pattern recognition and classification, and data grouping enables the system to quickly analyze the skis' operating status and improve the efficiency of fault prediction.

[0127] The system employs intelligent algorithms for pattern recognition and fault probability analysis, which can accurately identify different modes during the operation of the skid and calculate the probability of a fault occurring in each mode, thus improving the accuracy of fault prediction. By comparing with preset fault thresholds, the system can reliably determine which operating modes pose a fault risk, providing a strong basis for early warning.

[0128] The system automates the entire process from data acquisition, feature extraction, pattern recognition to fault warning, reducing manual intervention and the possibility of human error. The application of intelligent algorithms enables the system to automatically learn and adapt to changes in the operating state of the skis, improving the system's intelligence level.

[0129] The early warning module issues early warning signals in a timely manner based on the fault factor priority list, ensuring the timely transmission and processing of fault information; the system can match the associated operation mode classification results, providing specific fault modes and possible causes for the early warning, enabling maintenance personnel to quickly locate the problem and take corresponding measures;

[0130] Through real-time and accurate fault prediction, the system can detect potential faults in advance, avoiding serious consequences such as equipment downtime and engineering accidents, thereby reducing maintenance costs; the application of the system improves the safety of guide shoe operation and ensures the stability and service life of the entire system.

[0131] In summary, the above-mentioned guide shoe fault prediction system based on intelligent algorithms has advantages such as real-time performance, high efficiency, accuracy, reliability, automation, intelligence, timely and targeted early warning, as well as reduced maintenance costs and improved safety. It can effectively solve the limitations of traditional guide shoe fault detection methods.

[0132] In addition, this application also provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are respectively connected via the bus. When the computer program is executed by the processor, it implements the various processes of the above-described method embodiment for controlling output data and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0133] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for predicting guide shoe failures based on intelligent algorithms, characterized in that, The method includes: Real-time acquisition of ski boot operation data, and feature extraction of ski boot operation data to obtain operation status feature set; Pattern recognition is performed on the running feature set to identify different patterns in the running process of the skis, and the identified different patterns are classified to obtain the running pattern classification results; Based on the classification results of the operating mode, the ski boot operating data is grouped to obtain the operating data group dataset; The operational data group dataset is input into a pre-trained failure probability analysis model to calculate the probability of guide shoe failure under each operational mode. The probability of guide shoe failure under each operational mode is then compiled to obtain a failure probability list. The probability of guide shoe failure in each operating mode in the failure probability list is compared with a preset failure threshold. The probability of failure in operating modes that exceed the preset failure threshold is sorted to obtain a failure factor priority list. Based on the fault factor priority list, match the associated operating mode classification results and issue an early warning.

2. The guide shoe fault prediction method based on intelligent algorithms as described in claim 1, characterized in that, The ski boot operation data includes temperature data, pressure data, vibration data, displacement data, speed data, and operation time data.

3. The guide shoe fault prediction method based on intelligent algorithms as described in claim 1, characterized in that, Methods for pattern recognition of runtime feature sets include cluster analysis and classification algorithms.

4. The guide shoe fault prediction method based on intelligent algorithms as described in claim 1, characterized in that, The operational mode classification results include normal working mode, slight wear mode, severe wear mode, fatigue mode, and deformation mode.

5. The guide shoe fault prediction method based on intelligent algorithms as described in claim 1, characterized in that, The method for constructing the failure probability analysis model includes: Collect operational data of the guide shoe and preprocess the collected data; Choose a machine learning model as the basic architecture for the failure probability analysis model; The preprocessed data is divided into a training set, a validation set, and a test set; the training set is used to train the model, the validation set is used to adjust the model's hyperparameters to avoid overfitting, and the test set is used to evaluate the model's final performance. The trained failure probability analysis model is deployed to a real industrial system.

6. The guide shoe fault prediction method based on intelligent algorithms as described in claim 1, characterized in that, The factors influencing the setting of the preset fault threshold include historical fault data analysis, equipment performance specifications, operating environment conditions, system security requirements, and maintenance strategies.

7. The guide shoe fault prediction method based on intelligent algorithms as described in claim 1, characterized in that, Methods for matching and associating operational mode classification results and issuing alerts include: Different warning levels are set based on the degree to which the probability of failure exceeds the threshold under different operating modes in the failure factor priority list; For each warning level, generate corresponding warning information; Choose the appropriate warning method based on the actual application scenario and needs; The generated warning information will be sent to the relevant operators according to the selected warning method; Record and archive relevant information for each warning.

8. The guide shoe fault prediction method based on intelligent algorithms as described in claim 7, characterized in that, The warning information includes the operating mode of the guide shoe, the probability value of failure, the type of failure, and the recommended measures.

9. The guide shoe fault prediction method based on intelligent algorithms as described in claim 7, characterized in that, The warning methods include sound alarms, flashing lights, SMS notifications, email alerts, and system pop-up prompts.

10. A guide shoe fault prediction system based on intelligent algorithms, characterized in that, The system includes: The data acquisition module collects real-time data on the operation of the skates and extracts features from the data to obtain a set of operational status features. The pattern recognition and classification module performs pattern recognition on the running feature set, identifies different patterns in the running process of the skis, and classifies the identified different patterns to obtain the running pattern classification results. The data grouping module groups the ski boot running data based on the running mode classification results to obtain the running data group dataset; The fault probability list generation module and the fault probability analysis module input the grouped dataset of operating data into the pre-trained fault probability analysis model, calculate the probability of guide shoe failure under each operating mode, organize the probability of guide shoe failure under each operating mode, and obtain the fault probability list. The fault threshold comparison module compares the probability of guide shoe failure in each operating mode in the fault probability list with a preset fault threshold, sorts the probability of failure in operating modes that exceed the preset fault threshold, and obtains a fault factor priority list. The early warning module matches the associated operating mode classification results with the fault factor priority list and issues an early warning.

Images