System, method and product for predictive maintenance of equipment failure

Abstract

This disclosure relates to an equipment fault prediction and maintenance system, method, and product, belonging to the field of industrial control and automation technology. The equipment fault prediction and maintenance system includes a data acquisition module, an edge processor, a target processor, and an alarm execution module. The data acquisition module is used to collect equipment operating data and send the equipment operating data to the edge processor. The edge processor is used to process the equipment operating data to obtain preprocessed data and send the preprocessed data to the target processor. The target processor is used to input the preprocessed data into a pre-trained equipment fault prediction model, output equipment fault prediction data, generate equipment maintenance reference information based on the equipment fault prediction data and preset knowledge base data, and send the equipment maintenance reference information to the alarm execution module. The alarm execution module is used to generate and execute alarm prompt strategies based on the equipment maintenance reference information. This solves the problem of delayed maintenance of production line equipment.

Description

Technical Field

[0001] This disclosure relates to the field of industrial control and automation technology, and in particular to a system, method and product for predicting and maintaining equipment failures. Background Technology

[0002] In automated production lines, the stable operation of production line equipment (such as conveyor belt motors, automatic tightening machines, refrigerant charging machines, etc.) is crucial to production efficiency and product quality. Related technologies often rely on a reactive strategy of "repairing only when equipment breaks down," leading to frequent unplanned downtime, and equipment failure prediction depends on human experience; therefore, the solutions offered by these technologies suffer from a lag in production line equipment maintenance. Summary of the Invention

[0003] To overcome the problems existing in related technologies, this disclosure provides an equipment failure prediction and maintenance system, method and product.

[0004] According to a first aspect of the present disclosure, a device fault prediction and maintenance system is provided. The system includes a data acquisition module, an edge processor, a target processor, and an alarm execution module. The data acquisition module is used to acquire device operating data and send it to the edge processor. The edge processor receives the device operating data sent by the data acquisition module. It performs fusion calibration processing on the device operating data to obtain calibrated operating data. It preprocesses the calibrated operating data to obtain preprocessed data and sends the preprocessed data to the target processor. The target processor receives the preprocessed data sent by the edge processor. It inputs the preprocessed data into a pre-trained device fault prediction model and outputs device fault prediction data. Based on the device fault prediction data and preset knowledge base data, it generates device maintenance reference information. The device maintenance reference information includes device maintenance priority, device maintenance time window, spare parts data required for device maintenance, and device maintenance operation guidance information. The device maintenance reference information is sent to the alarm execution module. The alarm execution module generates and executes an alarm prompt strategy based on the device maintenance reference information.

[0005] In one possible embodiment, the alarm notification strategy includes a level 1 alarm notification strategy, a level 2 alarm notification strategy, or a level 3 alarm notification strategy; wherein, the level 1 alarm notification strategy includes a user interface module notification; the level 2 alarm notification strategy includes the operation of an audible and visual alarm and a user interface module notification; the level 3 alarm notification strategy includes the operation of an audible and visual alarm, a user interface module notification, execution of equipment shutdown protection, generation and sending of an equipment maintenance work order to the user interface module; the user interface module notification involves sending equipment maintenance reference information to a user terminal application and / or a factory monitoring system; generating and executing the alarm notification strategy based on the equipment maintenance reference information includes: determining the equipment maintenance alarm level based on the equipment maintenance reference information and a preset graded alarm threshold; and executing the level 1, level 2, or level 3 alarm notification strategy in response to the equipment maintenance alarm level.

[0006] In one possible embodiment, in response to the equipment maintenance alarm level, a Level 1 alarm notification strategy, a Level 2 alarm notification strategy, or a Level 3 alarm notification strategy is executed, including: in response to the equipment maintenance alarm level being a low-risk fault level, executing a Level 1 alarm notification strategy; in response to the equipment maintenance alarm level being a medium-risk fault level, executing a Level 2 alarm notification strategy; and in response to the equipment maintenance alarm level being a high-risk fault level, executing a Level 3 alarm notification strategy.

[0007] In one possible embodiment, in response to the equipment maintenance alarm level being a low-risk fault level, a level 1 alarm prompting strategy is executed, including: in response to the low-risk fault level, a user interface module prompt is executed to display equipment maintenance reference information on the display interface of the user terminal application and / or the factory monitoring system; wherein, the equipment maintenance reference information includes one or more of the following: equipment failure probability, equipment remaining lifespan, and equipment spare parts recommendations.

[0008] In response to the equipment maintenance alarm level being a medium-risk fault level, a two-level alarm prompting strategy is executed, including: in response to the medium-risk fault level, controlling the audible and visual alarm to emit an alarm sound at a first time interval and a first frequency, and to flash with a first color and a second frequency to prompt the production line user that the corresponding equipment has a medium-risk fault, and executing the user interface module reminder to display equipment maintenance reference information on the user terminal application and / or the display interface of the factory monitoring system.

[0009] In response to a high-risk fault level equipment maintenance alarm, a three-level alarm prompt strategy is implemented, including: in response to a high-risk fault level, controlling the audible and visual alarm to emit an alarm sound at a second time interval and a third frequency, and to flash with a second color and a fourth frequency to prompt the production line user that the corresponding equipment has a high-risk fault, and executing the user interface module reminder to display equipment maintenance reference information on the user terminal application and / or the display interface of the factory monitoring system, generating an equipment maintenance work order and sending the equipment maintenance work order to the user interface module, and sending the three-level alarm prompt strategy to the production operation system, so that the production operation system can perform cycle adjustment, workstation prompts or equipment shutdown protection according to the three-level alarm prompt strategy;

[0010] Among them, the first time interval is greater than or equal to the second time interval, the first frequency is less than or equal to the third frequency, and the second frequency is less than or equal to the fourth frequency.

[0011] In one possible embodiment, the data acquisition module includes a sensor unit and a data communication unit; the sensor unit is used to acquire device operation data based on a preset time interval; wherein the sensor unit includes one or more of a temperature sensor, a humidity sensor, a vibration sensor, and a current sensor; the data communication unit is used to send the device operation data to an edge processor.

[0012] In one possible embodiment, the data acquisition module further includes a data filtering unit; the data filtering unit is used to filter noise and remove outliers from the device operation data to obtain processed device operation data; and the data communication unit is used to send the processed device operation data to the edge processor.

[0013] In one possible embodiment, the data acquisition module further includes a data cache unit and a local storage unit; the data cache unit is used to store unsent device operation data; wherein the data cache unit adopts a circular buffer structure; the local storage unit is used to store the acquired device operation data.

[0014] In one possible embodiment, the equipment fault prediction and maintenance system further includes a data storage module; after generating and executing an alarm prompting strategy based on equipment maintenance reference information, it further includes: an alarm execution module, which is further configured to generate an equipment fault alarm log based on the equipment maintenance reference information and the alarm prompting strategy, and send the equipment fault alarm log to the data storage module; the data storage module is configured to store the equipment fault alarm log; the data storage module is further configured to store a model training set for training the equipment fault prediction model, and / or to encrypt and save the stored equipment fault alarm log and perform periodic backup and recovery; the data storage module is further configured to store the equipment operation data collected by the data acquisition module, and / or store the preprocessed data generated by the edge-side processor processing the equipment operation data.

[0015] In one possible embodiment, the data storage module is built on a high-performance relational database, which is used to implement data partitioning and indexing.

[0016] In one possible embodiment, the equipment fault prediction and maintenance system further includes a user interface module; the user interface module is used to receive equipment maintenance reference information and / or equipment repair work orders sent by the alarm execution module; the user interface module is also used to display the equipment maintenance reference information and / or equipment repair work orders; and / or, the user interface module is also used to receive equipment operation data sent by the data acquisition module, and display the equipment operation status according to the equipment operation data; and / or, the user interface module is also used to receive user operation instructions; and in response to the user operation instructions, execute one or more of the following policy configurations: confirming equipment maintenance reference information, confirming equipment repair work orders, and changing alarm prompting strategies; and / or, the user interface module is also used to receive user identity credentials and authentication credentials; and perform permission authentication according to the user identity credentials and authentication credentials to generate user authentication results; determine user role identifiers according to the user authentication results; and confirm user permissions according to the user role identifiers; wherein, user permissions include one or more of the following: viewing permissions for equipment maintenance reference information, confirmation permissions for equipment repair work orders, and configuration permissions for alarm prompting strategies.

[0017] In one possible embodiment, the equipment failure prediction and maintenance system further includes a model training module; the model training module is used to obtain a model training set from the data storage module, and train the equipment failure prediction model according to the model training set to obtain a pre-trained equipment failure prediction model, and deploy the pre-trained equipment failure prediction model on the target processor; and / or, the model training module is used to iteratively update the pre-trained equipment failure prediction model according to the model training set, so as to re-determine the updated equipment failure prediction model as the pre-trained equipment failure prediction model, and redeploy it in the target processor to replace the previous version of the pre-trained equipment failure prediction model deployed in the target processor.

[0018] In one possible embodiment, the equipment fault prediction and maintenance system further includes a communication interface module; the communication interface module is used to enable the alarm execution module to send the alarm prompting strategy to the equipment production and operation system when executing the alarm prompting strategy; the communication interface module is also used to enable data transmission between at least two of the data acquisition module, edge processor, target processor, alarm execution module, data storage module, user interface module, and model training module.

[0019] In one possible embodiment, the equipment fault prediction and maintenance system further includes a system monitoring module; the system monitoring module is used to monitor the operating status of one or more of the following modules: data acquisition module, edge processor, target processor, alarm execution module, data storage module, user interface module, model training module, and communication interface module; the system monitoring module is also used to generate corresponding module operation logs based on the module operating status; and in response to an abnormal module operating status, generate and output corresponding abnormal prompt information.

[0020] According to a second aspect of the present disclosure, a method for predicting and maintaining equipment failures is provided, which is applied to a target processor in the equipment failure prediction and maintenance system of any one of the first aspects.

[0021] In one possible embodiment, inputting preprocessed data into a pre-trained equipment failure prediction model and outputting equipment failure prediction data includes: inputting preprocessed data into a pre-trained equipment failure prediction model and outputting equipment remaining service life data; determining the equipment failure rate based on the equipment remaining service life data; and summarizing the equipment remaining service life data and the equipment failure rate to obtain equipment failure prediction data.

[0022] According to a third aspect of the present disclosure, a device fault prediction and maintenance method is provided, which is applied to an edge processor in a device fault prediction and maintenance system according to any one of the first aspects.

[0023] In one possible embodiment, the calibration operation data is preprocessed to obtain preprocessed data, including: performing sliding window analysis on the calibration operation data to obtain sliding window analysis results; performing dynamic threshold filtering on the calibration operation data based on the sliding window analysis results to obtain filtered operation data; and performing data cleaning on the filtered operation data to obtain preprocessed data.

[0024] In one possible embodiment, data cleaning is performed on the filtered running data to obtain preprocessed data, including: performing median filtering on the filtered running data to obtain filtered running data; and performing wavelet transform processing on the filtered running data to obtain preprocessed data.

[0025] In one possible embodiment, performing a sliding window analysis on the calibration operation data to obtain the sliding window analysis result includes: determining the window length of the sliding window based on the calibration operation data; and performing a sliding window analysis on the calibration operation data based on the window length of the sliding window to obtain the sliding window analysis result.

[0026] In one possible embodiment, the calibration operation data is dynamically threshold-filtered based on the sliding window analysis results to obtain filtered operation data, including: determining a dynamic threshold range based on the sliding window analysis results; and performing dynamic threshold filtering on the calibration operation data based on the dynamic threshold range to obtain filtered operation data.

[0027] In one possible embodiment, the device operation data includes multi-channel sensor data; the device operation data is fused and calibrated to obtain calibration operation data, including: performing time alignment processing on the multi-channel sensor data to obtain aligned multi-channel sensor data; constructing a state estimation sequence based on the aligned multi-channel sensor data; calculating the calibration value corresponding to each multi-channel sensor data according to the state estimation sequence; and updating each multi-channel sensor data according to the calibration value corresponding to each multi-channel sensor data to obtain calibration operation data.

[0028] According to a fourth aspect of the present disclosure, a first equipment fault prediction and maintenance apparatus is provided, which is applied to a target processor in an equipment fault prediction and maintenance system according to any one of the first aspects, comprising: a first receiving module, a first processing module, and a first sending module; wherein the first receiving module, the first processing module, and the first sending module implement the equipment fault prediction and maintenance method according to any one of the second aspects.

[0029] According to a fifth aspect of the present disclosure, a second equipment fault prediction and maintenance apparatus is provided, applied to an edge processor in an equipment fault prediction and maintenance system according to any one of the first aspects, comprising: a second receiving module, a second processing module, and a second sending module; wherein the second receiving module, the second processing module, and the second sending module implement the equipment fault prediction and maintenance method according to any one of the third aspects.

[0030] According to a sixth aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to perform a device fault prediction and maintenance method according to any one of the second aspects or a device fault prediction and maintenance method according to any one of the third aspects.

[0031] According to a seventh aspect of the present disclosure, a non-transitory computer-readable storage medium is provided, wherein when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the device failure prediction and maintenance method of any of the second aspects or the device failure prediction and maintenance method of any of the third aspects.

[0032] According to an eighth aspect of the present disclosure, a computer program product is provided, comprising: a computer program that, when executed by a processor, implements the equipment failure prediction and maintenance method of any one of the second aspects or the equipment failure prediction and maintenance method of any one of the third aspects.

[0033] According to a ninth aspect of the present disclosure, a chip system is provided, comprising: a processing unit and an interface circuit, wherein the processing unit obtains program instructions through the interface circuit, the program instructions are executed by the processing unit, and the processing unit is used to perform the equipment fault prediction and maintenance method of any one of the second aspects or the equipment fault prediction and maintenance method of any one of the third aspects.

[0034] The technical solutions provided by the embodiments of this disclosure can include the following beneficial effects: The equipment fault prediction and maintenance system includes a data acquisition module, an edge processor, a target processor, and an alarm execution module; the data acquisition module is used to collect equipment operation data and send the equipment operation data to the edge processor; the edge processor is used to receive the equipment operation data sent by the data acquisition module; the equipment operation data is fused and calibrated to obtain calibrated operation data; the calibrated operation data is preprocessed to obtain preprocessed data; the preprocessed data is sent to the target processor; the target processor is used to receive the preprocessed data sent by the edge processor; the preprocessed data is input into a pre-trained equipment fault prediction model and outputs equipment fault prediction data; equipment maintenance reference information is generated based on the equipment fault prediction data and preset knowledge base data; wherein, the equipment maintenance reference information includes equipment maintenance priority, equipment maintenance time window, spare parts data required for equipment maintenance, and equipment maintenance operation guidance information; the equipment maintenance reference information is sent to the alarm execution module; the alarm execution module is used to generate and execute alarm prompt strategies based on the equipment maintenance reference information; this solves the problem of lag in production line equipment maintenance in related technologies.

[0035] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0036] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0037] Figure 1 This is a system architecture of an equipment failure prediction and maintenance system illustrated according to some embodiments of the present disclosure. Figure 1 .

[0038] Figure 2 This is a system architecture of an equipment failure prediction and maintenance system illustrated according to some embodiments of the present disclosure. Figure 2 .

[0039] Figure 3 This is a flowchart illustrating a device failure prediction and maintenance method according to some embodiments of the present disclosure. Figure 1 .

[0040] Figure 4 This is a flowchart illustrating a device failure prediction and maintenance method according to some embodiments of the present disclosure. Figure 2 .

[0041] Figure 5 This is a block diagram illustrating an electronic device according to some embodiments of the present disclosure.

[0042] Figure 6 This is a block diagram of a chip system according to some embodiments of the present disclosure. Detailed Implementation

[0043] Some embodiments of this disclosure will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. Various changes, modifications, and equivalents of the methods, apparatus, and / or systems described herein will become apparent upon understanding this disclosure. For example, the order of operations described herein is merely illustrative and is not limited to those orders set forth herein, but can be changed as will become apparent upon understanding this disclosure, except for operations that must be performed in a particular order. Furthermore, for clarity and brevity, descriptions of features known in the art may be omitted.

[0044] The embodiments described in the following examples of this disclosure are not representative of all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0045] Figure 1 This is a system architecture diagram of an equipment failure prediction and maintenance system illustrated according to some embodiments of this disclosure, such as... Figure 1 As shown, the equipment fault prediction and maintenance system includes a data acquisition module, an edge processor, a target processor, and an alarm execution module.

[0046] The data acquisition module is used to collect device operation data and send the device operation data to the edge processor.

[0047] The edge processor receives equipment operation data sent by the data acquisition module; performs fusion calibration processing on the equipment operation data to obtain calibration operation data; preprocesses the calibration operation data to obtain preprocessed data; and sends the preprocessed data to the target processor.

[0048] The target processor receives preprocessed data sent by the edge processor; inputs the preprocessed data into a pre-trained equipment fault prediction model and outputs equipment fault prediction data; generates equipment maintenance reference information based on the equipment fault prediction data and preset knowledge base data; the equipment maintenance reference information includes equipment maintenance priority, equipment maintenance time window, spare parts data required for equipment maintenance, and equipment maintenance operation guidance information; and sends the equipment maintenance reference information to the alarm execution module.

[0049] The alarm execution module is used to generate and execute alarm prompt strategies based on equipment maintenance reference information. For example, the alarm execution module is an intelligent Andon terminal, which performs audible and visual alarms based on equipment maintenance reference information.

[0050] Among them, the edge processor is a processing unit deployed on the device site or close to the device data source side.

[0051] For example, equipment operation data such as temperature, current, and vibration are acquired through a data acquisition module, and then fused, calibrated, and preprocessed by an edge processor. This reduces the impact of sensor drift, noise contamination, sampling interruptions, and communication delays, thereby improving the consistency and reliability of the input data. The target processor outputs equipment fault prediction data based on a pre-trained equipment fault prediction model, and combines it with a preset knowledge base to generate equipment maintenance reference information including maintenance priorities, maintenance time windows, spare parts data, and operation guidance. This allows fault identification results to be directly linked to maintenance arrangements and resource preparation, which in turn enables the alarm execution module to form and execute more targeted alarm prompt strategies.

[0052] Therefore, the equipment failure prediction and maintenance system provided in this disclosure can reduce the probability of false alarms and missed alarms, improve the accuracy of equipment failure prediction, the timeliness of maintenance response, and the stability of continuous operation of the production line, and solve the problem of the lag in the maintenance of production line equipment in related technologies.

[0053] In some embodiments, the edge processor includes a data cleaning subunit, a noise filtering subunit, a data normalization subunit, and a feature extraction subunit. Therefore, the process of fusing and calibrating equipment operating data to obtain calibration operating data includes: using the data cleaning subunit to detect and correct outliers and missing points in the equipment operating data, thereby obtaining calibration operating data. Then, the calibration operating data is preprocessed to obtain preprocessed data, including: using the noise filtering subunit to filter the calibration operating data (e.g., using wavelet transform or low-pass filtering) to eliminate high-frequency interference signals; then, using the data normalization subunit to normalize the filtered calibration operating data to a uniform range (i.e., ensuring consistent dimensional ranges for all types of sensor data), thus obtaining normalized calibration operating data to ensure consistency in subsequent analysis; and using the feature extraction subunit to extract features from the normalized calibration operating data to obtain preprocessed data.

[0054] Furthermore, based on the above embodiments, the alarm notification strategy includes a level 1 alarm notification strategy, a level 2 alarm notification strategy, or a level 3 alarm notification strategy; wherein, the level 1 alarm notification strategy includes a user interface module reminder; the level 2 alarm notification strategy includes the operation of the audible and visual alarm and a user interface module reminder; the level 3 alarm notification strategy includes the operation of the audible and visual alarm, a user interface module reminder, execution of equipment shutdown protection, generation and sending of equipment maintenance work orders to the user interface module; the user interface module reminder involves sending equipment maintenance reference information to the user terminal application and / or the factory monitoring system; generating and executing the alarm notification strategy based on the equipment maintenance reference information includes: determining the equipment maintenance alarm level based on the equipment maintenance reference information and preset graded alarm thresholds; and executing the level 1, level 2, or level 3 alarm notification strategy in response to the equipment maintenance alarm level.

[0055] For example, the equipment maintenance priority, maintenance time window, spare parts data and operation instructions output by the target processor are uniformly mapped to a graded alarm level, so that alarm execution is no longer limited to a single alarm, but is matched with the risk level and maintenance needs. Thus, in the event of minor anomalies, targeted reminders are sent through user terminal applications or factory monitoring systems. When the risk increases, audible and visual alarms are triggered to improve on-site perceptibility. In the event of a serious fault trend, shutdown protection is implemented simultaneously and maintenance work orders are automatically generated, thereby forming a continuous closed loop from anomaly identification, maintenance prompts to maintenance dispatch.

[0056] Therefore, based on the above embodiments, the equipment failure prediction and maintenance system provided in this disclosure can reduce the interference of false alarms and missed alarms on maintenance decisions, make maintenance response more timely, schedule adjustment more proactive, and improve the operational safety and production continuity of critical equipment.

[0057] Furthermore, based on the aforementioned embodiments, in response to the equipment maintenance alarm level, a Level 1 alarm notification strategy, a Level 2 alarm notification strategy, or a Level 3 alarm notification strategy is executed, including: in response to the equipment maintenance alarm level being a low-risk fault level, a Level 1 alarm notification strategy is executed; in response to the equipment maintenance alarm level being a medium-risk fault level, a Level 2 alarm notification strategy is executed; and in response to the equipment maintenance alarm level being a high-risk fault level, a Level 3 alarm notification strategy is executed.

[0058] For example, the alarm execution module responds in tiers to the failure risks of conveyor belt motors, automatic tightening machines, or refrigerant charging machines based on maintenance reference information output by the target processor. At low risk, only a Level 1 alarm alert strategy is executed, reminding personnel to pay attention to early potential problems without interrupting production, thereby reducing excessive maintenance caused by short-term fluctuations. At medium risk, a Level 2 alarm alert strategy is executed, simultaneously triggering on-site audible and visual alerts and interface notifications, enabling maintenance personnel to intervene more quickly and arrange time windows and spare parts preparation. At high risk, a Level 3 alarm alert strategy is executed, which can link shutdown protection and the issuance of maintenance work orders, thereby suppressing the escalation of faults and the continued production of defective products.

[0059] Therefore, based on the above embodiments, the equipment failure prediction and maintenance system provided in this disclosure can establish a correspondence between different risk levels and different prompting strategies, thereby making the alarm intensity more consistent with the actual state of the equipment and improving the timeliness of maintenance decisions and the safety of production operations.

[0060] In some embodiments, in response to the equipment maintenance alarm level being a low-risk fault level, a level 1 alarm prompting strategy is executed, including: in response to the low-risk fault level, a user interface module prompt is executed to display equipment maintenance reference information on the display interface of the user terminal application and / or the factory monitoring system; wherein, the equipment maintenance reference information includes one or more of the following: equipment failure probability, equipment remaining lifespan, and equipment spare parts recommendations.

[0061] In some embodiments, in response to the equipment maintenance alarm level being a medium-risk fault level, a two-level alarm prompting strategy is executed, including: in response to the medium-risk fault level, controlling the audible and visual alarm to emit an alarm sound at a first time interval and a first frequency, and to flash with a first color and a second frequency to prompt the production line user that the corresponding equipment has a medium-risk fault, and executing a user interface module reminder to display equipment maintenance reference information on the display interface of the user terminal application and / or the factory monitoring system.

[0062] In some embodiments, in response to a high-risk fault level in equipment maintenance alarms, a three-level alarm prompting strategy is executed, including: in response to a high-risk fault level, controlling an audible and visual alarm to emit an alarm sound at a second time interval and a third frequency, and to flash with a second color and a fourth frequency to prompt production line users that the corresponding equipment has a high-risk fault; executing a user interface module reminder to display equipment maintenance reference information on the display interface of the user terminal application and / or the factory monitoring system, generating an equipment maintenance work order and sending the equipment maintenance work order to the user interface module; and sending the three-level alarm prompting strategy to the production operation system, so that the production operation system performs cycle adjustment, workstation prompts, or equipment shutdown protection according to the three-level alarm prompting strategy.

[0063] In this system, the first time interval is greater than or equal to the second time interval, the first frequency is less than or equal to the third frequency, and the second frequency is less than or equal to the fourth frequency. For example, the first time interval is 1 second, the second time interval is 0.5 seconds, the first frequency is 800 Hz, the third frequency is 1200 Hz, the second frequency is 2 Hz, the fourth frequency is 4 Hz, the first color is yellow, and the second color is red.

[0064] Furthermore, based on the aforementioned embodiments, the data acquisition module includes a sensor unit and a data communication unit; the sensor unit is used to acquire device operation data based on a preset time interval; the data communication unit is used to send the device operation data to the edge processor.

[0065] The sensor unit includes one or more of temperature sensors, humidity sensors, vibration sensors, and current sensors to collect operating parameters and environmental data of the production line equipment, i.e., equipment operating data.

[0066] Specifically, sensor units can be deployed at target equipment such as conveyor belt motors, automatic tightening machines, and refrigerant charging machines to continuously collect equipment operation data such as temperature, current, and vibration at preset time intervals, thereby forming a raw data foundation with a unified time reference. The data communication unit promptly sends the collected equipment operation data to the edge processor, enabling the data to be quickly aggregated on the side closest to the field and enter the subsequent fusion calibration and preprocessing stages, thereby reducing the impact of data lag caused by long-link transmission.

[0067] Therefore, based on the above embodiments, the data acquisition module in the equipment fault prediction and maintenance system provided by this disclosure can not only meet the continuous monitoring needs of various equipment statuses in the production site, but also provide a more stable data input basis for subsequent fault prediction models, thereby improving the accuracy and timeliness of equipment status identification, abnormal trend judgment, and maintenance decision generation.

[0068] Furthermore, based on the aforementioned embodiments, the data acquisition module also includes a data filtering unit; the data filtering unit is used to filter noise and remove outliers from the device operation data to obtain processed device operation data; and the data communication unit is used to send the processed device operation data to the edge processor.

[0069] Specifically, for example, by setting up a data filtering unit within the data acquisition module, on-site electromagnetic interference, instantaneous spikes, and drift anomalies in raw signals such as temperature, current, and vibration can be filtered and eliminated before the equipment operation data enters the edge processor. This reduces the impact of invalid fluctuations and false anomalies on subsequent processing. Furthermore, the data communication unit sends processed equipment operation data, making the data received by the edge processor more stable, which in turn helps to maintain consistency between the fusion calibration processing and preprocessing results.

[0070] Therefore, based on the above embodiments, the data acquisition module in the equipment fault prediction and maintenance system provided in this disclosure can reduce fault misjudgment and omission, improve the reliability of input data for fault prediction models, and provide more accurate data support for subsequent maintenance reference information generation and alarm prompt strategy execution.

[0071] Furthermore, based on the aforementioned embodiments, the data acquisition module further includes a data caching unit and a local storage unit; the data caching unit is used to store unsent device operation data; wherein, the data caching unit adopts a circular buffer structure; the local storage unit is used to store the acquired device operation data.

[0072] Specifically, a data caching unit is set up in the data acquisition module. This allows for the prior storage of unsent equipment operation data in the event of communication delays, network jitter, or temporary unavailability of the edge processor, thus preventing the direct loss of critical status data from conveyor motors, automatic tightening machines, and refrigerant charging machines. The ring-shaped buffer structure can continuously cover and manage the cache space with limited storage resources, making the writing and reading of new and old data more stable and efficient, thus adapting to continuous sampling scenarios on the production line. The local storage unit continuously retains the collected equipment operation data, facilitating subsequent tracing of abnormal periods and verification of sensor drift. It also provides original evidence for fusion calibration, model verification, and maintenance analysis, thereby improving data link integrity and the reliability of analytical input, leading to more accurate fault prediction results and maintenance decisions.

[0073] Furthermore, based on the aforementioned embodiments, the equipment fault prediction and maintenance system also includes a data storage module; after generating and executing an alarm prompting strategy according to the equipment maintenance reference information, it further includes: an alarm execution module, which is also used to generate an equipment fault alarm log according to the equipment maintenance reference information and the alarm prompting strategy, and send the equipment fault alarm log to the data storage module; a data storage module, which is used to store the equipment fault alarm log; a data storage module, which is also used to store a model training set for training the equipment fault prediction model, and / or to encrypt and save the stored equipment fault alarm log and perform periodic backup and recovery; a data storage module, which is also used to store equipment operation data collected by the data acquisition module, and / or to store preprocessed data generated by the edge processor processing the equipment operation data.

[0074] Specifically, after an alarm is executed, an equipment fault alarm log is generated. This log records the corresponding equipment operating data, preprocessed data, fault prediction results, maintenance reference information, and actual alarm strategies. This creates a traceable closed loop for handling anomalies in conveyor motors, automatic tightening machines, and refrigerant charging machines in the air conditioning production line, facilitating subsequent verification of false alarms, missed alarms, and maintenance execution. The data storage module simultaneously saves alarm logs and model training sets, enabling continuous correlation between historical alarm events and training data, thus providing a basis for updating and optimizing the equipment fault prediction model.

[0075] Therefore, based on the above embodiments, the equipment fault prediction and maintenance system provided in this disclosure not only improves the data continuity and reusability of the prediction and maintenance process, but also improves the security, integrity and post-disaster recovery capability of fault information by encrypting and saving alarm logs and regularly backing them up and restoring them, making on-site maintenance decisions more reliable and stable.

[0076] Based on the above embodiments, the data storage module is built on a high-performance relational database, which is used to implement data partitioning and indexing.

[0077] For example, when constructing a data storage module using a high-performance relational database, equipment fault alarm logs, model training sets, and historical operating data corresponding to different equipment can be partitioned and stored according to production line, equipment type, and time interval. An index can be created using fields such as equipment number, alarm level, and timestamp. This ensures high data writing efficiency and query response speed even in scenarios with continuous writing of multi-source data such as conveyor motors, automatic tightening machines, and refrigerant charging machines. This facilitates the rapid archiving of logs generated by the alarm execution module and allows the target processor to retrieve training samples, maintenance records, and associated alarm information for analysis. This makes the retrieval, tracing, and correlation processing of massive amounts of operating data more stable, reducing analytical bias caused by data contamination and retrieval delays. Therefore, it is beneficial for supporting continuous optimization of fault prediction models and timely output of maintenance decisions.

[0078] Based on the above embodiments, the equipment fault prediction and maintenance system also includes a user interface module;

[0079] The user interface module is used to receive equipment maintenance reference information and / or equipment repair work orders sent by the alarm execution module.

[0080] The user interface module is also used to display equipment maintenance reference information and / or equipment repair work orders;

[0081] And / or, the user interface module is also used to receive device operation data sent by the data acquisition module, and display the device operation status based on the device operation data;

[0082] And / or, the user interface module is also used to receive user operation instructions; and in response to user operation instructions, to perform one or more of the following policy configurations: confirming equipment maintenance reference information, confirming equipment maintenance work orders, and changing alarm prompt policies;

[0083] And / or, the user interface module is also used to receive user identity credentials and authentication credentials; and to perform permission authentication based on user identity credentials and authentication credentials to generate user authentication results; to determine user role identifiers based on user authentication results; and to confirm user permissions based on user role identifiers; wherein, user permissions include one or more of the following: permission to view equipment maintenance reference information, permission to confirm equipment maintenance work orders, and permission to configure alarm prompt policy.

[0084] For example, the user interface module can be deployed as a production line central control terminal, a handheld terminal for maintenance personnel, or a large monitoring screen. It receives equipment maintenance reference information and equipment maintenance work orders from the alarm execution module and displays them visually, enabling maintenance personnel to simultaneously understand maintenance priorities, maintenance time windows, and spare parts preparation requirements. The user interface module further receives equipment operating data and displays the equipment operating status, allowing on-site personnel to correlate and verify real-time status with maintenance recommendations, thereby reducing the risk of mishandling based solely on a single alarm. By receiving user operation commands and executing them to confirm maintenance information, confirm work orders, or adjust alarm notification strategy configurations, a closed loop is formed for alarm handling, maintenance workflow, and strategy optimization. By combining user identity credentials and authentication credentials for permission authentication, and confirming corresponding permissions based on user role identifiers, viewing, confirming, and configuring operations can be limited to authorized personnel, thereby improving human-machine interaction security, maintenance execution standardization, and alarm management accuracy.

[0085] In some embodiments, the user interface module is specifically used to provide a human-machine interface that supports real-time data visualization, maintenance suggestion query, and alarm information management; provides a historical maintenance suggestion query function that supports filtering by device number, fault type, or timestamp; alarm information management includes a real-time alarm list, priority sorting, confirmation status update, and archiving operations; it also integrates a user access control system to ensure data security and operation auditing; and supports access from mobile and web terminals to facilitate remote monitoring and response by on-site personnel.

[0086] In some embodiments, the user interface module is specifically used to control user access permissions based on user roles, ensuring system operation security and data integrity; supporting fine-grained permission allocation, including operation level control for data viewing, editing, and deletion; implementing role-based access control policies, such as defining user roles like administrator, operator, and auditor; providing a user authentication mechanism, supporting password policies and multi-factor authentication; recording all user operation logs to ensure traceability and auditability of behavior; ensuring that data is not modified or leaked without authorization; and integrating a permission management interface to facilitate real-time configuration and monitoring of permission settings by administrators.

[0087] Based on the above embodiments, the equipment fault prediction and maintenance system also includes a model training module;

[0088] The model training module is used to obtain the model training set from the data storage module, train the equipment fault prediction model based on the model training set to obtain the pre-trained equipment fault prediction model, and deploy the pre-trained equipment fault prediction model on the target processor.

[0089] And / or, the model training module is used to iteratively update the pre-trained device fault prediction model based on the model training set, so as to re-identify the updated device fault prediction model as the pre-trained device fault prediction model, and redeploy it in the target processor to replace the previous version of the pre-trained device fault prediction model deployed in the target processor.

[0090] For example, after retrieving historical operating data, alarm logs, and corresponding annotations from the data storage module to form a model training set, the model training module can perform initial training on the equipment fault prediction model and deploy the resulting pre-trained model to the target processor, enabling the target processor to perform fault trend identification based on the pre-processed data at the edge. Furthermore, when the operating conditions of the conveyor belt motor, automatic tightening machine, or refrigerant charging machine in the air conditioning production line change, or when new samples continuously accumulate, the model training module can use the model training set to iteratively update and train the existing pre-trained model, replacing the previous version with the updated model. This ensures that the model parameters continuously match the changes in the on-site equipment status, sensor drift characteristics, and noise distribution, reducing false alarms and missed alarms caused by model aging. This improves the timeliness and accuracy of fault prediction results and maintenance reference information generation, thus facilitating continuous optimization of maintenance decisions, spare parts preparation, and downtime arrangements.

[0091] In some embodiments, the model training module is used to iteratively optimize the accuracy and generalization ability of the equipment fault prediction model through machine learning algorithms, and supports the training and validation of supervised learning algorithms (e.g., random forest, support vector machine, long short-term memory network, etc.); it uses cross-validation technology to evaluate model performance, effectively prevents overfitting, and improves the model's adaptability to new data; it implements automated hyperparameter tuning, finding the optimal parameter combination through methods such as grid search or Bayesian optimization; and it integrates model training process monitoring and visualization tools to track changes in key indicators such as loss function and accuracy in real time; it provides a model performance evaluation report generation function, including analysis charts such as confusion matrix, precision-recall curve, and ROC curve, to facilitate the evaluation of model performance; and it supports model version management, recording the dataset, parameters, and results of each training session to ensure the traceability of the model iteration process.

[0092] Based on the above embodiments, the equipment fault prediction and maintenance system also includes a communication interface module;

[0093] The communication interface module is used to enable the alarm execution module to send the alarm prompting strategy to the production operation system when executing the alarm prompting strategy.

[0094] The communication interface module is also used to enable data transmission between at least two of the data acquisition module, edge processor, target processor, alarm execution module, data storage module, user interface module, and model training module.

[0095] For example, the communication interface module can use an industrial Ethernet interface, fieldbus interface, or wireless communication interface to establish a communication connection with the production line control system and the upper-level management system. After generating a hierarchical alarm prompt strategy, the alarm execution module synchronously sends the corresponding strategy to the production operation system via the communication interface module, enabling the production operation system to coordinate and execute cycle time adjustments, workstation prompts, or shutdown protection. Thus, alarm information is no longer limited to local output but enters the production scheduling chain. Simultaneously, the communication interface module provides a unified data channel between the data acquisition module, edge processor, target processor, data storage module, user interface module, and model training module, enabling stable transmission of acquired data, preprocessed data, prediction results, maintenance reference information, work order information, and model files, thereby improving the collaborative efficiency of each module. Therefore, on-site monitoring, fault prediction, maintenance decision-making, alarm execution, and model updates can be linked together, reducing the risk of response delays and mishandling caused by communication gaps.

[0096] In some embodiments, the communication interface module can also implement encrypted data transmission and access authentication to prevent unauthorized operations and ensure communication security; provide a real-time data synchronization mechanism to support millisecond-level latency to meet rapid response requirements; integrate error detection and automatic retry functions to ensure the reliability and integrity of data transmission; and support data format conversion and parsing to facilitate seamless integration with other system modules such as alarm management and data storage.

[0097] Based on the above embodiments, the equipment fault prediction and maintenance system also includes a system monitoring module;

[0098] Among them, the system monitoring module is used to monitor the operating status of one or more of the following modules: data acquisition module, edge processor, target processor, alarm execution module, data storage module, user interface module, model training module, and communication interface module.

[0099] The system monitoring module is also used to generate corresponding module operation logs based on the module's operating status; and to generate and output corresponding exception prompts when the module's operating status is abnormal.

[0100] For example, the system monitoring module continuously monitors the operational status of data acquisition, edge processing, fault prediction, alarm execution, storage, interaction, training, and communication links, and can promptly identify module offline, processing timeouts, communication interruptions, or resource anomalies. The system monitoring module further generates corresponding module operation logs, making the operation of each module in the air conditioning production line traceable, thereby facilitating the location of sensor drift, edge node failures, or model deployment anomalies. Furthermore, when an abnormal module operation status is detected, an anomaly prompt message is generated and output, enabling maintenance personnel to quickly intervene and prevent abnormal modules from continuously outputting erroneous data or interrupting maintenance processes. Therefore, it can improve the overall stability, reliability, and timeliness of maintenance response of the equipment fault prediction and maintenance system.

[0101] In some embodiments, the system monitoring module continuously monitors the processing units, memory, network traffic, and service processes of each module to obtain the module's operating status, and ensures that each module operates normally through heartbeat detection and resource utilization analysis; it also records detailed operation logs, including timestamps, operator identities, operation types, and execution results, to facilitate post-event auditing and problem tracing; it generates corresponding module operation logs, including system performance indicators, error rate statistics, potential fault warnings, and optimization suggestions, and displays them intuitively in chart form; it integrates a real-time alarm mechanism, automatically triggering email or SMS notifications when an abnormal state is detected in a module (e.g., resource overload or service interruption); and it provides a visual monitoring interface, allowing administrators to dynamically view system operating status and historical logs to ensure system stability and maintainability.

[0102] Based on the above implementation methods Figure 2 This is a system architecture of an equipment failure prediction and maintenance system illustrated according to some embodiments of the present disclosure. Figure 2 ,like Figure 2 As shown, the equipment fault prediction and maintenance system includes a data acquisition module, an edge processor, a target processor, an alarm execution module, a user interface module, a model training module, a data storage module, a communication interface module, and a system monitoring module. The equipment fault prediction and maintenance system communicates with the production operation system through the communication interface module. The data transmission process between the modules and the functions of each module are described in the aforementioned embodiments and will not be repeated here.

[0103] Figure 3 This is a flowchart illustrating a device failure prediction and maintenance method according to some embodiments of the present disclosure. Figure 1 ,like Figure 3 As shown, the equipment failure prediction and maintenance method is used for the target processor in the equipment failure prediction and maintenance system of the above embodiment, and includes the following steps.

[0104] Step S11: Receive preprocessed data sent by the edge processor.

[0105] The target processor is the main processing unit in the equipment fault prediction and maintenance system, responsible for executing subsequent fault prediction, knowledge base matching, and maintenance reference output processes. Edge processors are processing units deployed at the equipment site or near the equipment data source, used to perform front-end processing on the raw operational data collected by sensors before sending it to the target processor. Preprocessed data, the output of the edge processor, serves as the input basis for the subsequent equipment fault prediction model; its content corresponds to the state representation information during equipment operation. Preprocessed data can correspond to temperature, current, vibration, speed, start / stop status, runtime, or workstation cycle time data generated during the operation of conveyor motors, tightening equipment, refrigerant charging equipment, etc., in automated air conditioning production lines, and has already undergone at least one of the following processing steps in the edge processor: cleaning, filtering, noise reduction, time sequence alignment, missing data compensation, and data normalization.

[0106] Specifically, after establishing a data communication connection between the target processor and the edge processor, the target processor receives preprocessed data. This communication connection can be established via industrial Ethernet, fieldbus gateway, serial communication link, or wireless industrial communication link. Once the connection is established, the target processor listens to the corresponding data reporting port, message queue, or interface service of the edge processor. After the edge processor completes the preprocessing of the raw field data, it generates a data packet to be transmitted according to a preset data format and sends the packet to the target processor.

[0107] The data packet may contain device identifier, acquisition timestamp, workstation identifier, data type identifier, and corresponding preprocessed numerical sequence. After receiving the data packet, the target processor first performs integrity verification, device identity matching, and timestamp validity check. After confirming that the data source is consistent with the target device, the data length meets the prediction input requirements, and the time sequence is continuous, the data that passes the verification is written into the input buffer or time-series data storage area of ​​the target processor as the data source for subsequent model calls.

[0108] In one possible implementation, the data sent by the edge processor is batch-reported data at fixed intervals. The target processor archives the data according to the receiving time or data time window, for example, using 30 seconds, 1 minute, or a production cycle as a receiving unit for buffering. In another implementation, the data sent by the edge processor is event-driven reporting data. When the edge processor detects that the operating parameters fluctuate beyond a preset range, the local data loss rate exceeds a threshold, or the device state changes, it immediately sends the corresponding pre-processed data to the target processor. The target processor receives this type of data with high priority and marks it as an abnormal batch of concern.

[0109] Step S12: Input the preprocessed data into the pre-trained equipment fault prediction model and output the equipment fault prediction data.

[0110] The pre-trained equipment fault prediction model is a model that has been trained before the method execution and can be directly deployed and invoked. It is used to infer the equipment's operating status, fault trends, or failure risks based on the received preprocessed data. This model can be trained using historical equipment operating data, historical fault records, maintenance records, and downtime records. After training, it is deployed in the target processor or invoked by the target processor through the model service interface. The equipment fault prediction data is the output data of this model, used to characterize at least one of the following: the probability of fault occurrence, fault risk level, abnormal trend, or expected fault occurrence time in the future, thereby providing a basis for the generation of subsequent maintenance reference information.

[0111] Specifically, the target processor extracts preprocessed data corresponding to the target device from the input buffer and organizes it into model input vectors, input matrices, or time-series sample fragments according to the requirements of the pre-trained device fault prediction model. For example, if the model is a time-series prediction model, the target processor extracts multiple consecutive sampling points in chronological order to form an input sequence; if the model is a feature classification or regression model, the target processor extracts statistical features, frequency domain features, or trend features from the preprocessed data and then inputs them into the model.

[0112] Furthermore, after inputting the preprocessed data into the pre-trained equipment failure prediction model, equipment failure prediction data can be obtained. This data includes one or more of the following: failure probability value, risk level label, future failure trend curve, and expected failure occurrence interval. For multiple consecutive batches of preprocessed data from the same equipment, the target processor can iteratively call the model batch by batch, forming a rolling update sequence of equipment failure prediction data. The prediction results of the current batch are then associated and stored with the prediction results of the previous batch to characterize the trend of equipment status changes.

[0113] In one possible embodiment, the pre-trained equipment fault prediction model adopts a multi-input feature fusion structure. The input features include temperature features, current features, vibration features, and beat features. The target processor maps preprocessed data from different sources to a unified feature space and then inputs it into the model, so that the model can output equipment fault prediction data based on information from multiple operational dimensions during a single inference process.

[0114] Step S13: Generate equipment maintenance reference information based on equipment failure prediction data and preset knowledge base data; wherein, the equipment maintenance reference information includes equipment maintenance priority, equipment maintenance time window, spare parts data required for equipment maintenance, and equipment maintenance operation guidance information.

[0115] The preset knowledge base data consists of a pre-stored and queryable set of rules, experiences, and related information. Its content includes at least one of the following: the correspondence between equipment type and fault mode, the mapping between fault risk level and maintenance level, equipment component life parameters, maintenance timing rules, spare parts list relationships, and maintenance operation process templates. Equipment maintenance reference information is maintenance result information generated by the target processor based on equipment fault prediction data and combined with the preset knowledge base data. It is used to transform the prediction conclusions into actionable maintenance guidelines on-site. Equipment maintenance priority characterizes the urgency of the current equipment in maintenance scheduling; equipment maintenance time window characterizes the time range within which maintenance should be performed; spare parts data required for equipment maintenance characterizes the name, model, quantity, or inventory requirements of spare parts corresponding to the predicted fault; and equipment maintenance operation guidance information characterizes the corresponding operation steps, inspection locations, disassembly / assembly sequence, or reset requirements for on-site maintenance.

[0116] For example, the target processor first reads the risk value, risk level, fault trend, or remaining service life information of the equipment fault prediction data, and then retrieves the corresponding rule set from the knowledge base based on the equipment identifier or equipment type. If a mapping relationship between risk level and maintenance priority is established in the knowledge base, the target processor directly determines the equipment maintenance priority according to the mapping result; if a mapping relationship between fault mode and time window is established in the knowledge base, the target processor calculates the equipment maintenance time window by combining the expected fault occurrence period, current production line scheduling status, and downtime rules in the prediction result. For spare parts data generation, the target processor outputs the spare parts data required for equipment maintenance corresponding to the currently predicted fault based on the fault mode and component association table, component model table, and spare parts quantity rules in the knowledge base. For maintenance operation guidance information generation, the target processor matches the current equipment type and fault type according to the maintenance operation process template in the knowledge base, extracts the corresponding inspection steps, execution actions, confirmation items, and recovery items, and writes them into the equipment maintenance reference information.

[0117] Specifically, for example, when equipment failure prediction data indicates a high probability of a conveyor belt motor experiencing bearing malfunction in the near future, the target processor, based on knowledge base rules, sets the equipment maintenance priority to a high level, sets the maintenance time window to before the end of the current shift or the next planned downtime, identifies the required spare parts for maintenance as the corresponding bearing components, lubricating materials, and fasteners, and extracts information from the knowledge base related to bearing replacement, such as power-off confirmation, disassembly / reassembly sequence, lubrication dosage, and trial operation checks, as equipment maintenance operation guidance information. If the equipment failure prediction data indicates that a tightening device only has a slight trend of misalignment and the expected failure time is far off, the target processor can output a lower equipment maintenance priority based on knowledge base rules, schedule the maintenance time window to a subsequent routine maintenance cycle, and output spare parts and operation guidance for sensor verification or torque calibration.

[0118] In one possible embodiment, when generating equipment maintenance reference information, the target processor can also perform rule-based filtering on multiple candidate maintenance schemes, retaining only the set of maintenance information with the highest matching degree to the current prediction result as the final output. This ensures that the prediction results for the same equipment in the same batch correspond to only one specific set of equipment maintenance reference information. In another possible embodiment, the preset knowledge base data contains pre-set maintenance parameters for different equipment types. For example, for refrigerant charging equipment, the focus is on rules for checking leak locations and replacing seals; for tightening equipment, the focus is on rules for maintaining torque mechanisms and drive units; and for conveyor belt motors, the focus is on rules for maintaining bearings, windings, and transmission structures. The target processor then calls different sub-knowledge bases to generate equipment maintenance reference information based on the equipment type.

[0119] This step further translates equipment failure prediction data into actionable maintenance schedules. The output includes maintenance sequence, implementation time, spare parts preparation, and operation methods, enabling the prediction results to be directly used for on-site maintenance scheduling and execution.

[0120] Step S14: Send the equipment maintenance reference information to the alarm execution module.

[0121] The alarm execution module is an execution unit that receives equipment maintenance reference information and triggers alarms or maintenance prompts. Its function is to convert the equipment maintenance reference information generated by the target processor into field-perceptible alarm results, message push results, or work order triggering results. The alarm execution module can be one or more combinations of Andon terminal control units, industrial monitoring platform alarm services, maintenance terminal message services, or work order system interface units. After generating the equipment maintenance reference information, the target processor encapsulates the information according to a data format recognizable by the alarm execution module. The encapsulated content includes at least the equipment identifier, equipment maintenance priority, equipment maintenance time window, spare parts data required for equipment maintenance, and equipment maintenance operation guidance information, and then sends it to the alarm execution module.

[0122] For example, the target processor generates a transmission message according to the interface protocol of the alarm execution module. If the alarm execution module receives information via a network interface, the target processor transmits the equipment maintenance reference information through interface calls, message publishing, or socket transmission. If the alarm execution module receives information via an industrial control link, the target processor converts the equipment maintenance reference information into a corresponding control message and writes it into the execution link.

[0123] Subsequently, after receiving the alarm, the alarm execution module determines the output strategy according to the equipment maintenance priority and the equipment maintenance time window. For example, when the equipment maintenance priority is high, the alarm execution module immediately displays the fault warning content on the Andon terminal along with an audible and visual prompt, and simultaneously pushes the spare parts data required for equipment maintenance and equipment maintenance operation guidance information to the maintenance personnel's terminal; when the equipment maintenance priority is medium, the alarm execution module generates a pending alarm and outputs a prompt when the time before the start of the equipment maintenance time window reaches a preset advance amount; when the equipment maintenance priority is low, the alarm execution module writes the equipment maintenance reference information into the maintenance task queue and displays it on the routine inspection interface.

[0124] Furthermore, if the spare parts required for equipment maintenance are in short supply, the alarm execution module can also simultaneously output a spare parts preparation reminder to the spare parts management terminal. If the equipment maintenance operation instructions include manual confirmation steps, the alarm execution module will display the corresponding instructions on the maintenance terminal and wait for maintenance personnel to confirm receipt or execution.

[0125] In one possible embodiment, the target processor adds generation time, information validity period, and response status fields to the equipment maintenance reference information before sending it. The alarm execution module uses this information to determine whether the current information needs to be executed immediately, whether an escalation reminder is needed, and whether a closed-loop feedback has been formed. In another possible embodiment, after completing the alarm output, the alarm execution module returns a receipt confirmation message to the target processor. The target processor associates the sending status with the equipment maintenance reference information and stores it for subsequent tracking of maintenance execution.

[0126] In this step, after the equipment maintenance reference information is output by the alarm execution module, the target processor forms a complete processing link from receiving preprocessed data, fault prediction, generating equipment maintenance reference information to connecting alarm execution. This enables the prediction results to be transmitted to the maintenance process in a timely manner, reducing the situation where the prediction results remain at the analysis level and are not used on site.

[0127] In this public disclosure Figure 3 In the equipment fault prediction and maintenance method provided in the illustrated embodiment, the edge processor first completes the front-end processing of the field-collected data, and the target processor then uses the pre-trained equipment fault prediction model to output equipment fault prediction data. It also combines the data from the preset knowledge base to generate directly executable equipment maintenance reference information. Finally, the alarm execution module implements corresponding alarms and prompts, ensuring that the data entering the prediction stage has high consistency and that the prediction results can be directly linked to maintenance scheduling, spare parts preparation, and work execution. This method forms a complete processing path to address the prediction deviation problems caused by sensor drift, noise interference, and discontinuous communication timing in the production environment. It solves the problem of lag in production line equipment maintenance in related technologies.

[0128] Furthermore, in some embodiments, the specific implementation steps of step S12 include:

[0129] Step S121: Input the preprocessed data into the pre-trained equipment failure prediction model and output the remaining service life data of the equipment.

[0130] Step S122: Determine the equipment failure rate based on the remaining service life data of the equipment.

[0131] Step S123: Summarize the remaining service life data of the equipment and the equipment failure rate to obtain equipment failure prediction data.

[0132] For example, the target processor receives preprocessed data, after fusion calibration and filtering by the edge processor, and inputs it as a feature vector into a pre-trained device failure prediction model. This allows the pre-trained model to output the corresponding remaining lifespan of the device based on the time span of the input data; the value can be expressed in hours, minutes, or operating cycles. Furthermore, the target processor generates a device failure rate based on the correspondence between the remaining lifespan value and a preset interval threshold, or based on the conversion relationship between the remaining lifespan value and a failure probability function. If historical failure statistics exist, historical statistical deviations can be used to correct the device failure rate. Finally, the target processor writes the remaining lifespan data and the device failure rate into the same result message or the same data structure to generate device failure prediction data.

[0133] In the steps disclosed herein, the equipment failure prediction data includes both lifespan information and failure probability information, which can more completely reflect the current degradation state of the equipment and provide a direct basis for subsequent equipment maintenance priority determination, equipment maintenance time window setting, and spare parts preparation, thereby improving the availability of failure prediction results and the accuracy of maintenance decisions.

[0134] Figure 4 This is a flowchart illustrating a device failure prediction and maintenance method according to some embodiments of the present disclosure. Figure 2 ,like Figure 4 As shown, the equipment failure prediction and maintenance method is used in the edge-side processor of the equipment failure prediction and maintenance system in the above embodiment, and includes the following steps.

[0135] Step S21: Receive device operation data sent by the data acquisition module.

[0136] The edge processor is deployed close to the industrial equipment to receive and process front-end data streams locally. The data acquisition module collects and transmits data on the operating status of the monitored equipment. In an automated air conditioning production line scenario, the data acquisition module includes temperature sensors, current sensors, vibration sensors, speed sensors, and equipment controllers corresponding to units such as conveyor motors, tightening equipment, and refrigerant charging equipment, thereby generating equipment operation data. This equipment operation data characterizes the actual state of the equipment during operation and can be either continuously sampled time-series data or data packets encapsulated according to a collection cycle.

[0137] For example, the edge processor establishes a data communication connection with the data acquisition module. After receiving the device operation data, it can perform basic access processing on the data for use in subsequent processes.

[0138] Step S22: Perform fusion calibration processing on the equipment operation data to obtain calibration operation data.

[0139] The fusion calibration process is used to perform consistency correction on the received equipment operation data. The edge processor combines relevant information from the equipment operation data to generate calibration operation data. The calibration operation data is the result of the fusion calibration process on the equipment operation data. Its numerical changes are closer to the actual operating conditions of the equipment and can be used as input for subsequent preprocessing.

[0140] For example, the edge processor corrects, merges, or performs consistency processing on the device operation data according to preset rules to output calibration operation data containing calibration values ​​and quality markers.

[0141] Step S23: Preprocess the calibration operation data to obtain preprocessed data.

[0142] Preprocessing is used to further remove residual interference, format the input data, and form a dataset that can be directly used by the target processor based on the calibration run data. Preprocessed data is data that has undergone anomaly processing, noise suppression, and data organization. Its data structure is consistent with the target processor's data interface definition and can be a fixed-length vector, a time-segment sequence, or a data frame with additional tags. The edge processor in this step processes the calibration run data that has already undergone consistency correction. The focus of preprocessing is not on recalibration, but on removing residual components that are ineffective or interfere with fault prediction and organizing the data according to the needs of subsequent prediction models.

[0143] For example, the edge processor can process the calibration operation data according to a preset data organization method, identify and process abnormal data, suppress noise or interference components, and then construct preprocessed data according to the field order specified by the target processor. The fields include at least the device identifier, time information, processing results of each parameter, and corresponding quality labels. If the target processor uses a fault prediction model, the preprocessed data can also be mapped according to the model input requirements.

[0144] Step S24: Send the preprocessed data to the target processor.

[0145] The target processor is a subsequent processing unit following the edge processor. It receives preprocessed data and performs calculations related to equipment fault prediction and maintenance. The target processor can be deployed on a local server on the production line, a workshop-level industrial computing platform, or a central analysis node. The data it receives is no longer the unprocessed raw sample values, but preprocessed data processed by the edge processor. After completing preprocessing, the edge processor pushes the data to the target processor according to a preset communication protocol and transmission strategy. This enables the target processor to perform fault prediction, maintenance priority generation, maintenance time window calculation, spare parts information matching, and operation guidance output based on more stable data.

[0146] For example, the edge processor first encapsulates the preprocessed data into a standard message body, which includes a device identifier, data batch number, time window information, processing parameter version, preprocessing result field, quality flag, and integrity check code. The edge processor can send data to the target processor through message queues, industrial gateway forwarding interfaces, long-lived connections based on TCP / IP (Transmission Control Protocol / Internet Protocol), or service-oriented interfaces.

[0147] Furthermore, to ensure link reliability, the edge processor waits for confirmation information from the target processor after sending data. This confirmation information includes at least the batch number and reception status. If no confirmation is received within a preset time limit, the edge processor retransmits the corresponding batch of data according to the retransmission strategy and records the number of retransmissions. When the retransmission count reaches its limit, the edge processor writes the batch of data into its local queue for retransmission and simultaneously generates a link anomaly flag for the system monitoring module to read. For devices with high real-time requirements, the edge processor immediately sends the corresponding preprocessed data after each sliding window ends. For scenarios sensitive to bandwidth usage, the edge processor can send data in batches when data volume thresholds, time thresholds, or anomaly triggering conditions are met. Anomaly triggering conditions include a quality label indicating a continuous abnormal trend, a window feature value exceeding the predicted input alarm threshold, or the device switching from standby to operation.

[0148] In one possible embodiment, the edge processor appends a digest checksum and a processing serial number to the preprocessed data before transmission. The target processor then performs deduplication reception based on this information to prevent network retransmissions from causing the same window of data to be analyzed repeatedly. In another possible embodiment, the edge processor also archives local transmission records in correspondence with the reception acknowledgments returned by the target processor, forming a data traceability chain from data acquisition and edge processing to central analysis.

[0149] In this public disclosure Figure 4 In the equipment fault prediction and maintenance method provided in the illustrated embodiment, the edge processor is located near the equipment data source, receives field-collected data, and generates structurally unified preprocessed data through fusion calibration and preprocessing. This preprocessed data is then sent to the target processor. As a result, the data entering the subsequent fault prediction and maintenance chain is no longer directly affected by original noise, drift, and timing discontinuities. The data obtained by the target processor has higher consistency with the actual operating status of the equipment, enabling the generation of subsequent maintenance priorities, maintenance time windows, spare parts information, and operation instructions to be based on stable input.

[0150] Furthermore, in some embodiments, the specific implementation steps of step S23 include:

[0151] Step S231: Perform sliding window analysis on the calibration operation data to obtain the sliding window analysis results.

[0152] Step S232: Dynamic threshold filtering is performed on the calibration operation data based on the sliding window analysis results to obtain filtered operation data.

[0153] Step S233: Perform data cleaning on the filtered running data to obtain preprocessed data.

[0154] For example, the edge processor first receives the calibration operation data after fusion calibration, performs local statistical analysis within a limited time window, dynamically adjusts the filtering range based on the statistical results, and then cleans and shapes the retained data. Through this process, short-term disturbances and non-steady-state interferences in the calibration operation data are gradually reduced, enhancing the continuity and consistency of the output data. This allows subsequent fault prediction models to make judgments based on data that more closely approximates real-world operating conditions, thus making the generation of maintenance reference information more stable.

[0155] Furthermore, in some embodiments, the specific implementation steps of step S233 include: performing median filtering on the filtered running data to obtain filtered running data; and performing wavelet transform processing on the filtered running data to obtain preprocessed data.

[0156] For example, the filtered operating data first undergoes median filtering to suppress spike fluctuations, then wavelet transform processing to denoise the frequency domain and preserve effective trends, ultimately outputting preprocessed data. This processing method enables the data entering the subsequent prediction model to more closely approximate the actual operating state of the equipment, reduces the impact of short-term interference on the state identification results, and makes the data used for subsequent fault prediction and maintenance decisions more stable and consistent.

[0157] Furthermore, in some embodiments, the specific implementation steps of step S231 include: determining the window length of the sliding window based on the calibration operation data; performing sliding window analysis on the calibration operation data based on the window length of the sliding window to obtain the sliding window analysis result.

[0158] For example, after receiving calibration operation data, the edge processor first calculates the window length according to preset statistical rules or adaptive rules, and then slides the data sequence segment by segment according to the window length. It performs mean, extreme value, variance, or fluctuation amplitude statistics on the data within each window to form a sliding window analysis result. This sliding window analysis result can characterize the local variation features of the calibration operation data in different time intervals and can be used as an input parameter for subsequent dynamic threshold filtering. Because the window length matches the variation rhythm of the calibration operation data, the sliding window analysis results corresponding to different window lengths will exhibit different local sensitivities, thus adapting to calibration operation data under different operating conditions. The specific value of the window length can be a fixed time length, a fixed number of sampling points, or a length adaptively adjusted according to statistical characteristics.

[0159] In the steps of this disclosure, sliding window analysis can extract effective features from calibration operation data while taking into account both local anomaly detection capability and statistical stability. This makes the obtained sliding window analysis results closer to the actual operating state of the equipment and provides a reliable basis for subsequent dynamic threshold filtering and data cleaning, thereby improving the accuracy and consistency of preprocessed data.

[0160] Furthermore, in some embodiments, the specific implementation steps of step S232 include: determining a dynamic threshold range based on the sliding window analysis results; and performing dynamic threshold filtering on the calibration operation data based on the dynamic threshold range to obtain filtered operation data.

[0161] For example, after obtaining the dynamic threshold range, the system compares the calibration operation data point by point or segment by segment with the range. Data falling outside the range is identified as abnormal data and filtered, while data remaining within the range is output as filtered operation data. Dynamic threshold filtering can use a bilateral threshold determination method, or it can combine the current load, speed, or operating condition of the equipment to synchronously correct the upper and lower limits of the range, so that the filtering results match the actual operating characteristics.

[0162] In the steps of this disclosure, the threshold boundary can be updated in real time based on the sliding window analysis results, and abnormal fluctuations in the calibration operation data can be selectively removed, so that the output filtered operation data more accurately reflects the actual operating status of the equipment, providing a stable input for subsequent data cleaning and fault prediction.

[0163] Furthermore, in some embodiments, the device operation data includes multi-channel sensor data; the specific implementation steps of step S22 include: performing time alignment processing on the multi-channel sensor data to obtain aligned multi-channel sensor data; constructing a state estimation sequence based on the aligned multi-channel sensor data; calculating the calibration values ​​corresponding to each of the multi-channel sensor data according to the state estimation sequence; and updating each of the multi-channel sensor data according to the calibration values ​​corresponding to each of the multi-channel sensor data to obtain calibration operation data.

[0164] For example, after receiving the raw sampled values ​​from multiple sensors, the edge processor first reads the timestamps carried in each sample frame and recalibrates the timestamps using a unified clock source. Then, it performs interpolation or resampling processing on adjacent sample points according to a preset time granularity to obtain aligned multi-sensor data. Subsequently, the aligned data of each channel is input into the state estimation model, and a state estimation sequence is constructed by combining the device's historical operating baseline, short-term change trends, and inter-channel correlations. The state estimation sequence can be a set of state vectors arranged in chronological order. Based on the state estimation sequence, the edge processor calculates the deviation between each acquisition channel and the estimated state, and converts the deviation into the calibration value corresponding to each sensor. The calibration value can be an additive correction or a proportional correction coefficient. Then, based on their respective calibration values, each sensor data is corrected channel by channel, and updated calibration operation data is output for subsequent preprocessing and fault prediction.

[0165] In this disclosed procedure, data from different acquisition channels are first made time-series consistent, and then calibrated separately based on the unified state estimation results, thereby forming calibration operation data that can be directly used by subsequent models. Since the correction values ​​for each acquisition channel are calculated separately and updated one channel at a time, the calibration operation data can maintain the consistency and continuity of multi-source information, reducing the impact of sampling asynchrony and sensor bias on subsequent prediction results.

[0166] Based on the implementation process provided in the above embodiments, this disclosure takes equipment monitoring of an automated air conditioning production line as an application scenario for illustrative purposes.

[0167] Specifically, the data acquisition module is deployed around the refrigerant charging machine or conveyor motor to collect real-time data on equipment operating under high loads for extended periods. Taking the refrigerant charging machine as an example, temperature sensors, current sensors, and vibration sensors are installed on the surface and inside the equipment. Data from these multiple sensors can include motor housing temperature, motor power supply current, and bearing vibration spectrum data, with a sampling frequency of once per second. The data acquisition module transmits the equipment's operating data to the edge processor via a wireless communication network and can first cache it in a circular buffer or local database. For example, if a sudden increase in current from the normal 12A to 25A is detected during equipment startup, this instantaneous data is cached before entering the subsequent processing flow.

[0168] Subsequently, after receiving multi-channel sensor data, the edge processor processes the multi-channel sensor data based on the aforementioned implementation method to generate pre-processed data, and then sends the pre-processed data to the target processor.

[0169] Subsequently, after receiving the preprocessed data, the target processor processes the preprocessed data based on the aforementioned implementation method to generate equipment fault prediction data.

[0170] Subsequently, the target processor combines equipment fault prediction data and preset knowledge base data to generate equipment maintenance reference information. The preset knowledge base can be pre-configured with rules such as "sudden current increase and abnormal vibration spectrum → check bearing wear." Based on the current prediction results of the refrigerant charging machine, it generates equipment maintenance priorities, maintenance time windows, required spare parts data, and maintenance operation guidance information. The maintenance time window corresponds to a 10-hour downtime maintenance period, spare parts data can include bearings, lubricants, and related replacement parts, and maintenance operation guidance information can include checking the refrigerant charging machine bearings and replacing the lubricant. The generated equipment maintenance reference information is then sent to the alarm execution module.

[0171] Subsequently, after receiving equipment maintenance reference information, the alarm execution module can perform multi-level linkage on the production line. For situations where the failure rate exceeds 90%, the on-site Andon device can illuminate a red light and emit a buzzer. For refrigerant charging machines with a failure rate of 30% and a remaining service life of less than 50 hours, a work order can be pushed to the maintenance engineer's mobile application. The work order includes the failure rate, remaining service life, maintenance priority, maintenance time window, and spare parts preparation information. Simultaneously, for refrigerant charging machines with a failure rate exceeding 90%, the alarm execution module can also link with the production operation system. The production operation system can adjust the refrigerant charging process schedule, suspend relevant workstations, and notify the warehouse to prepare stock. Relevant alarm logs can be written to the database, and the equipment can be marked as high-risk equipment.

[0172] Furthermore, for the equipment fault prediction and maintenance system, the operating status of one or more of the following modules can be continuously monitored during system operation: data acquisition module, edge processor, target processor, alarm execution module, data storage module, user interface module, model training module, and communication interface module. For example, the status of the server where the edge processor is located, such as a central processing unit utilization rate of 75% and an average transmission latency of less than 100ms, can be monitored to support long-term monitoring and maintenance scheduling of the production line.

[0173] Some embodiments of this disclosure also provide a first equipment fault prediction and maintenance device, applied to... Figures 1 to 2 The target processor in the equipment fault prediction and maintenance system shown in the embodiment includes: a first receiving module, a first processing module, and a first transmitting module; wherein, the first receiving module, the first processing module, and the first transmitting module implement as follows: Figure 3 The equipment failure prediction and maintenance method shown in the embodiment.

[0174] In one possible embodiment, when the first processing module inputs preprocessed data into a pre-trained equipment failure prediction model and outputs equipment failure prediction data, it is specifically used to: input preprocessed data into a pre-trained equipment failure prediction model and output equipment remaining service life data; determine the equipment failure rate based on the equipment remaining service life data; and summarize the equipment remaining service life data and the equipment failure rate to obtain equipment failure prediction data.

[0175] Regarding the first equipment fault prediction and maintenance device in the above embodiments, the specific way in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated here.

[0176] Some embodiments of this disclosure also provide a second equipment failure prediction and maintenance device, applied to... Figures 1 to 2The edge processor in the equipment fault prediction and maintenance system shown in the embodiment includes: a second receiving module, a second processing module, and a second transmitting module; wherein, the second receiving module, the second processing module, and the second transmitting module implement as follows: Figure 4 The equipment failure prediction and maintenance method shown in the embodiment.

[0177] In one possible embodiment, when the second processing module preprocesses the calibration operation data to obtain preprocessed data, it specifically performs the following: performs sliding window analysis on the calibration operation data to obtain sliding window analysis results; performs dynamic threshold filtering on the calibration operation data based on the sliding window analysis results to obtain filtered operation data; and performs data cleaning on the filtered operation data to obtain preprocessed data.

[0178] In one possible embodiment, when the second processing module cleans the filtered running data to obtain preprocessed data, it specifically performs the following: performs median filtering on the filtered running data to obtain filtered running data; and performs wavelet transform processing on the filtered running data to obtain preprocessed data.

[0179] In one possible embodiment, when the second processing module performs sliding window analysis on the calibration operation data to obtain the sliding window analysis result, it is specifically used to: determine the window length of the sliding window based on the calibration operation data; and perform sliding window analysis on the calibration operation data based on the window length of the sliding window to obtain the sliding window analysis result.

[0180] In one possible embodiment, when the second processing module performs dynamic threshold filtering on the calibration operation data based on the sliding window analysis results to obtain filtered operation data, it specifically performs the following: determining the dynamic threshold range based on the sliding window analysis results; and performing dynamic threshold filtering on the calibration operation data based on the dynamic threshold range to obtain filtered operation data.

[0181] In one possible embodiment, the device operation data includes multi-channel sensor data; when the second processing module performs fusion calibration processing on the device operation data to obtain calibrated operation data, it specifically performs the following: performs time alignment processing on the multi-channel sensor data to obtain aligned multi-channel sensor data; constructs a state estimation sequence based on the aligned multi-channel sensor data; calculates the calibration value corresponding to each of the multi-channel sensor data according to the state estimation sequence; and updates each of the multi-channel sensor data according to the calibration value corresponding to each of the multi-channel sensor data to obtain calibrated operation data.

[0182] Regarding the second equipment fault prediction and maintenance device in the above embodiments, the specific way in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated here.

[0183] Figure 5This is a block diagram illustrating an electronic device according to some embodiments of the present disclosure. (Refer to...) Figure 5 The electronic device 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input / output (I / O) interface 812, a sensor component 814, and a communication component 816.

[0184] Processing component 802 typically controls the overall operation of electronic device 800, such as operations associated with display, telephone calls, data communication, camera operation, and recording operations. Processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the methods described above. Furthermore, processing component 802 may include one or more modules to facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802.

[0185] Memory 804 is configured to store various types of data to support the operation of electronic device 800. Examples of such data include instructions for any application or method operating on electronic device 800, contact data, phonebook data, messages, pictures, videos, etc. Memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0186] Power component 806 provides power to various components of electronic device 800. Power component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 800.

[0187] Multimedia component 808 includes a screen that provides an output interface between electronic device 800 and user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of touch or swipe actions but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 808 includes a front-facing camera and / or a rear-facing camera. When electronic device 800 is in an operating mode, such as a shooting mode or video mode, the front-facing camera and / or rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0188] Audio component 810 is configured to output and / or input audio signals. For example, audio component 810 includes a microphone (MIC) configured to receive external audio signals when electronic device 800 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 804 or transmitted via communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

[0189] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0190] Sensor assembly 814 includes one or more sensors for providing state assessments of various aspects of electronic device 800. For example, sensor assembly 814 can detect the on / off state of electronic device 800, the relative positioning of components such as the display and keypad of electronic device 800, changes in position of electronic device 800 or a component of electronic device 800, the presence or absence of user contact with electronic device 800, orientation or acceleration / deceleration of electronic device 800, and temperature changes of electronic device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 814 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0191] Communication component 816 is configured to facilitate wired or wireless communication between electronic device 800 and other devices. Electronic device 800 can access wireless networks based on communication standards, such as WiFi, 3G, 4G, 5G, other communication standards, or combinations thereof. In some embodiments of this disclosure, communication component 816 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In some embodiments of this disclosure, communication component 816 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0192] In some embodiments of this disclosure, the electronic device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0193] In some embodiments of this disclosure, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions that can be executed by a processor 820 of an electronic device 800 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0194] A non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by the processor of an electronic device, enables the electronic device to perform the device failure prediction and maintenance method provided in any of the above embodiments.

[0195] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the device fault prediction and maintenance method provided in any of the above embodiments.

[0196] Figure 6 This is a block diagram illustrating a chip system according to some embodiments of the present disclosure. (Refer to...) Figure 6The chip system 500 includes at least one processor 501 and at least one interface circuit 502. The processor 501 and the interface circuit 502 are interconnected via lines. For example, the interface circuit 502 can be used to receive signals from other devices (e.g., the memory of an electronic device). As another example, the interface circuit 502 can be used to send signals to other devices (e.g., the processor 501). Exemplarily, the interface circuit 502 can read instructions stored in memory and send those instructions to the processor 501. When the instructions are executed by the processor 501, the processor 501 can perform the steps in the above embodiments. Of course, the chip system 500 may also include other discrete devices, and some embodiments of this disclosure do not specifically limit this.

[0197] In some embodiments of this disclosure, the interface circuit 502 can obtain data, program instructions, and / or information from the internal storage area of ​​the chip system 500; it can also obtain data, program instructions, and / or information from outside the chip system 500.

[0198] Optionally, the chip system 500 also includes a memory for storing necessary computer programs and data.

[0199] Those skilled in the art will also understand that the various illustrative logical blocks and steps listed in the embodiments of this application can be implemented by electronic hardware, computer software, or a combination of both. Whether such functionality is implemented through hardware or software depends on the specific application and the overall system design requirements. Those skilled in the art can implement the functionality using various methods for each specific application, but such implementation should not be construed as exceeding the scope of protection of the embodiments of this application.

[0200] In the above detailed description, reference has been made to the accompanying drawings, which illustrate specific aspects of this disclosure by way of illustration. In this regard, terms indicating direction or positional relationship, such as “center,” “longitudinal,” “lateral,” “length,” “width,” “thickness,” “upper,” “lower,” “front,” “rear,” “left,” “right,” “vertical,” “horizontal,” “top,” “bottom,” “inner,” “outer,” “clockwise,” “counterclockwise,” “axial,” “radial,” and “circumferential,” are used with reference to the orientation of the described figures. Since components of the described device can be positioned in multiple different orientations, directional terms are used for illustrative purposes and not for limitation. It should be understood that other aspects can be utilized and structural or logical changes can be made without departing from the concept of this disclosure. Therefore, the following detailed description should not be considered limiting.

[0201] It should be understood that, unless otherwise specifically indicated, features of various embodiments of this disclosure described herein can be combined with each other. As used herein, the term “and / or” includes any one of the relevant listed items and any combination of any two or more; similarly, “at least one of…” includes any one of the relevant listed items and any combination of any two or more.

[0202] It should be understood that, unless otherwise expressly specified and limited, the terms "joining," "attaching," "installing," "connecting," "linking," "fixing," etc., used in the embodiments of this disclosure should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a connection that allows communication between them; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms herein based on the specific circumstances.

[0203] Furthermore, the term "above" as used herein with respect to components, elements, or material layers formed or located "above" a surface may be used to indicate that the component, element, or material layer is "indirectly" positioned (e.g., placed, formed, deposited, etc.) on the surface such that one or more additional components, elements, or layers are arranged between the surface and the component, element, or material layer. However, the term "above" as used with respect to components, elements, or material layers formed or located "above" a surface may also optionally have a specific meaning: that the component, element, or material layer is "directly" positioned (e.g., placed, formed, deposited, etc.) on the surface, for example, in direct contact with the surface.

[0204] Although terms such as “first,” “second,” and “third” may be used herein to describe various components, parts, regions, layers, or sections, these components, parts, regions, layers, or sections are not limited to these terms. Rather, these terms are used only to distinguish one component, part, region, layer, or section from another. Therefore, without departing from the teachings of the examples described herein, the first component, part, region, layer, or section mentioned in the examples may also be referred to as the second component, part, region, layer, or section. Furthermore, the terms “first” and “second” are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as “first” or “second” may explicitly or implicitly include at least one of that feature. In the description herein, “a plurality” means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0205] It should be understood that spatial relative terms, such as “above,” “upper,” “below,” and “lower,” are used herein to describe the relationship between one element and another shown in the figures. In addition to the orientation depicted in the figures, these spatial relative terms are also intended to encompass different orientations of the device in use or operation. For example, if the device in the figures is flipped, an element described as “above” or “upper” relative to another element would be “below” or “lower” relative to that other element. Thus, depending on the spatial orientation of the device, the term “above” encompasses both above and below orientations. Devices may have other orientations (e.g., rotated 90 degrees or in other orientations), and the spatial relative terms used herein should be interpreted accordingly.

[0206] Furthermore, the term “exemplary” is used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as “exemplary” is not necessarily to be construed as advantageous compared to other aspects or designs. Rather, the use of the term “exemplary” is intended to present the concept in a concrete manner. As used herein, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or.” That is, unless otherwise specified or clear from the context, “X applies A or B” is intended to mean any of the natural inclusive arrangements. That is, “X applies A or B” satisfies any of the foregoing instances if X applies A; X applies B; or both X applies A and B. Additionally, unless otherwise specified or clear from the context to refer to the singular form, the articles “a” and “an” as used in this application and the appended claims are generally understood to mean “one or more.”

[0207] Similarly, although this disclosure has been shown and described with respect to one or more implementations, equivalent variations and modifications will occur to those skilled in the art upon reading and understanding the specification and drawings. This disclosure includes all such modifications and variations and is limited only by the scope of the claims. In particular, with respect to the various functions performed by the components described above (e.g., elements, resources, etc.), unless otherwise indicated, the terminology used to describe such components is intended to correspond to any component (functionally equivalent) that performs the specific function of the described component, even if structurally not equivalent to the disclosed structure. Furthermore, although specific features of this disclosure may have been disclosed with respect to only one of several implementations, such features may be combined with one or more other features of other implementations, as may be desired and advantageous for any given or particular application. Moreover, with regard to the terms “comprising,” “owning,” “having,” “having,” or variations thereof as used in the detailed description or claims, such terms are intended to be inclusive in a manner similar to the term “including.”

[0208] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0209] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A device fault prediction and maintenance system, characterized in that, The equipment fault prediction and maintenance system includes a data acquisition module, an edge processor, a target processor, and an alarm execution module; The data acquisition module is used to collect device operation data and send the device operation data to the edge processor; The edge-side processor is used to receive the device operation data sent by the data acquisition module; The equipment operation data is fused and calibrated to obtain calibration operation data; The calibration operation data is preprocessed to obtain preprocessed data; The preprocessed data is sent to the target processor; The target processor is used to receive preprocessed data sent by the edge-side processor; The preprocessed data is input into a pre-trained equipment fault prediction model, which outputs equipment fault prediction data. Based on the equipment fault prediction data and preset knowledge base data, equipment maintenance reference information is generated. The equipment maintenance reference information includes equipment maintenance priority, equipment maintenance time window, spare parts data required for equipment maintenance, and equipment maintenance operation guidance information. The equipment maintenance reference information is then sent to the alarm execution module. The alarm execution module is used to generate and execute alarm prompt strategies based on the equipment maintenance reference information.

2. The equipment fault prediction and maintenance system according to claim 1, characterized in that, The alarm notification strategy includes a level 1 alarm notification strategy, a level 2 alarm notification strategy, or a level 3 alarm notification strategy; wherein, the level 1 alarm notification strategy includes a user interface module notification; the level 2 alarm notification strategy includes the operation of the audible and visual alarm and a user interface module notification; the level 3 alarm notification strategy includes the operation of the audible and visual alarm, a user interface module notification, execution of equipment shutdown protection, generation and sending of equipment maintenance work orders to the user interface module; the user interface module notification involves sending the equipment maintenance reference information to the user terminal application and / or the factory monitoring system; The step of generating and executing an alarm notification strategy based on the equipment maintenance reference information includes: The equipment maintenance alarm level is determined based on the equipment maintenance reference information and the preset graded alarm threshold. In response to the equipment maintenance alarm level, execute the first-level alarm notification strategy, the second-level alarm notification strategy, or the third-level alarm notification strategy.

3. The equipment fault prediction and maintenance system according to claim 2, characterized in that, The step of responding to the equipment maintenance alarm level by executing the first-level alarm notification strategy, the second-level alarm notification strategy, or the third-level alarm notification strategy includes: In response to the equipment maintenance alarm level being a low-risk fault level, the first-level alarm prompting strategy is executed; In response to the equipment maintenance alarm level being a medium-risk fault level, the level 2 alarm prompting strategy is executed; In response to the equipment maintenance alarm level being classified as a high-risk fault level, the three-level alarm notification strategy is executed.

4. The equipment fault prediction and maintenance system according to claim 3, characterized in that, In response to the equipment maintenance alarm level being a low-risk fault level, the first-level alarm notification strategy is executed, including: In response to the low-risk fault level, the user interface module prompts the user to display the equipment maintenance reference information on the user terminal application and / or the display interface of the factory monitoring system; wherein the equipment maintenance reference information includes one or more of the following: equipment failure probability, equipment remaining lifespan, and equipment spare parts recommendations. In response to the equipment maintenance alarm level being classified as a medium-risk fault level, the secondary alarm notification strategy is executed, including: In response to the medium-risk fault level, the audible and visual alarm is controlled to emit an alarm sound at a first time interval and a first frequency, and to flash with a first color and a second frequency to alert the production line user that the corresponding equipment has a medium-risk fault, and to execute the user interface module reminder to display the equipment maintenance reference information on the user terminal application and / or the display interface of the factory monitoring system. In response to the equipment maintenance alarm level being classified as a high-risk fault level, the three-level alarm notification strategy is executed, including: In response to the high-risk fault level, the audible and visual alarm is controlled to emit an alarm sound at a second time interval and a third frequency, and to flash with a second color and a fourth frequency to alert production line users that the corresponding equipment has a high-risk fault. The user interface module is also executed to remind users to display the equipment maintenance reference information on the user terminal application and / or the display interface of the factory monitoring system, generate an equipment maintenance work order, send the equipment maintenance work order to the user interface module, and send the three-level alarm prompt strategy to the production operation system, so that the production operation system can perform cycle adjustment, workstation prompt, or equipment shutdown protection according to the three-level alarm prompt strategy. Wherein, the first time interval is greater than or equal to the second time interval, the first frequency is less than or equal to the third frequency, and the second frequency is less than or equal to the fourth frequency.

5. The equipment fault prediction and maintenance system according to claim 1, characterized in that, The data acquisition module includes a sensor unit and a data communication unit; The sensor unit is used to collect the device operation data based on a preset time interval; wherein, the sensor unit includes one or more of a temperature sensor, a humidity sensor, a vibration sensor, and a current sensor; The data communication unit is used to send the device operation data to the edge processor.

6. The equipment fault prediction and maintenance system according to claim 5, characterized in that, The data acquisition module also includes a data filtering unit; The data filtering unit is used to filter noise and remove outliers from the equipment operation data to obtain processed equipment operation data. The data communication unit is used to send the processed device operation data to the edge processor.

7. The equipment fault prediction and maintenance system according to claim 1, characterized in that, The data acquisition module also includes a data caching unit and a local storage unit; The data cache unit is used to store the device operation data that has not been sent; wherein, the data cache unit adopts a circular buffer structure; The local storage unit is used to store the collected device operation data.

8. The equipment fault prediction and maintenance system according to claim 1, characterized in that, The equipment fault prediction and maintenance system also includes a data storage module; After generating and executing the alarm notification strategy based on the equipment maintenance reference information, the method further includes: The alarm execution module is also used to generate an equipment fault alarm log based on the equipment maintenance reference information and the alarm prompt strategy, and send the equipment fault alarm log to the data storage module; The data storage module is used to store the device fault alarm logs; The data storage module is also used to store the model training set used to train the equipment fault prediction model, and / or to encrypt and save the stored equipment fault alarm logs and perform periodic backup and recovery. The data storage module is also used to store the device operation data collected by the data acquisition module, and / or store the preprocessed data generated by the edge processor processing the device operation data.

9. The equipment fault prediction and maintenance system according to claim 8, characterized in that, The data storage module is built on a high-performance relational database, which is used to implement data partitioning and indexing.

10. The equipment fault prediction and maintenance system according to claim 1, characterized in that, The equipment fault prediction and maintenance system also includes a user interface module; The user interface module is used to receive the equipment maintenance reference information and / or equipment repair work order sent by the alarm execution module; The user interface module is also used to display the equipment maintenance reference information and / or the equipment repair work order; And / or, the user interface module is further configured to receive the device operation data sent by the data acquisition module, and display the device operation status according to the device operation data; And / or, the user interface module is further configured to receive user operation instructions; In response to the user operation command, perform one or more of the following: confirm the equipment maintenance reference information, confirm the equipment repair work order, and change the alarm prompt strategy configuration. And / or, the user interface module is further configured to receive user identity credentials and authentication credentials; and perform permission authentication based on the user identity credentials and authentication credentials to generate a user authentication result; and determine a user role identifier based on the user authentication result; Based on the user role identifier, confirm the user's permissions; wherein, the user permissions include one or more of the following: the permission to view the equipment maintenance reference information, the permission to confirm the equipment maintenance work order, and the permission to configure the alarm prompt policy.

11. The equipment fault prediction and maintenance system according to claim 1, characterized in that, The equipment fault prediction and maintenance system also includes a model training module; The model training module is used to obtain a model training set from the data storage module, and train the equipment fault prediction model according to the model training set to obtain the pre-trained equipment fault prediction model, and deploy the pre-trained equipment fault prediction model to the target processor. And / or, the model training module is configured to iteratively update the pre-trained device fault prediction model based on the model training set, so as to re-determine the updated device fault prediction model as the pre-trained device fault prediction model, and redeploy it in the target processor to replace the previous version of the pre-trained device fault prediction model deployed in the target processor.

12. The equipment fault prediction and maintenance system according to claim 1, characterized in that, The equipment fault prediction and maintenance system also includes a communication interface module; The communication interface module is used to enable the alarm execution module to send the alarm prompting strategy to the production operation system when executing the alarm prompting strategy; The communication interface module is also used to enable data transmission between at least two of the data acquisition module, the edge processor, the target processor, the alarm execution module, the data storage module, the user interface module, and the model training module.

13. The equipment fault prediction and maintenance system according to claim 1, characterized in that, The equipment fault prediction and maintenance system also includes a system monitoring module; The system monitoring module is used to monitor the operating status of one or more of the following modules: the data acquisition module, the edge processor, the target processor, the alarm execution module, the data storage module, the user interface module, the model training module, and the communication interface module. The system monitoring module is also used to generate corresponding module operation logs based on the module's operating status; and to generate and output corresponding abnormal prompt information in response to the module's operating status being abnormal.

14. A method for predicting and maintaining equipment failures, characterized in that, The target processor is applied to the equipment failure prediction and maintenance system as described in any one of claims 1 to 13.

15. A method for predicting and maintaining equipment failures, characterized in that, An edge-side processor used in the equipment failure prediction and maintenance system as described in any one of claims 1 to 13.

16. A first equipment fault prediction and maintenance device, characterized in that, The target processor applied in the equipment fault prediction and maintenance system as described in any one of claims 1 to 13 includes: a first receiving module, a first processing module, and a first transmitting module; The first receiving module, the first processing module, and the first sending module implement the equipment fault prediction and maintenance method as described in claim 14.

17. A second equipment fault prediction and maintenance device, characterized in that, An edge processor applied in the equipment fault prediction and maintenance system as described in any one of claims 1 to 13 includes: a second receiving module, a second processing module, and a second transmitting module; The second receiving module, the second processing module, and the second sending module implement the equipment fault prediction and maintenance method as described in claim 15.

18. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to execute the equipment fault prediction and maintenance method of claim 14 or the equipment fault prediction and maintenance method of claim 15.

19. A non-transitory computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device is able to perform the device fault prediction and maintenance method of claim 14 or the device fault prediction and maintenance method of claim 15.

20. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the equipment failure prediction and maintenance method as described in claim 14, or the equipment failure prediction and maintenance method as described in claim 15.

Images