An intelligent inspection management platform and method

Abstract

This invention discloses an intelligent inspection management platform and method, relating to the field of management platform technology. The key technical points include the following steps: processing and analyzing the sound information of the production equipment to be inspected to obtain baseline feature information; evaluating the degree of sound correlation between the production equipment to be inspected and similar equipment to obtain a similarity correlation value; evaluating the degree of sound correlation between the production equipment to be inspected and supporting equipment to obtain a supporting correlation value; statistically analyzing the sound trends of the similarity correlation value and the supporting correlation value to obtain test trend feature one and test trend feature two; marking test trend feature two, which matches the baseline feature information, as preprocessed sound data one; comprehensively evaluating the sound data belonging to test trend feature one and the operation log data belonging to the production equipment to be inspected to obtain preprocessed sound data two and preprocessed sound data three; the effect is to improve the enterprise's operation and maintenance cost control capabilities and equipment reliability assurance level.

Description

Technical Field

[0001] This invention relates to the field of management platform technology, and more specifically, to an intelligent inspection management platform and method. Background Technology

[0002] In modern industrial production systems, stable equipment operation is the core foundation for ensuring capacity, quality, and efficiency. Traditional inspection and management methods lack effective means of integrating the sounds, vibrations, and operating parameters generated by equipment in the industrial environment. Therefore, they cannot determine whether equipment requires maintenance based on these factors. Traditional methods rely solely on experience to formulate equipment maintenance plans, potentially leading to over-maintenance or under-maintenance. Over-maintenance increases spare parts and labor costs, while under-maintenance accelerates equipment aging and shortens its lifespan. Consequently, traditional methods cannot dynamically adjust maintenance strategies based on the actual condition of the equipment, making it difficult to balance equipment reliability and maintenance costs. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the purpose of this invention is to provide an intelligent inspection management platform and method.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] An intelligent inspection management method, comprising the following steps:

[0006] The sound information of the production equipment to be tested is processed and analyzed to obtain the baseline feature information;

[0007] The degree of sound correlation between the production equipment to be tested and similar equipment is evaluated to obtain the similar correlation value, and the degree of sound correlation between the production equipment to be tested and supporting equipment is evaluated to obtain the supporting correlation value.

[0008] By statistically analyzing the sound trends of similar correlation values ​​and complementary correlation values, we obtain the first and second trend features to be tested.

[0009] The experimental trend feature two that matches the baseline feature information is marked as preprocessed sound data one;

[0010] The sound data belonging to the trend feature to be tested and the operation log data of the production equipment to be tested are comprehensively evaluated to obtain preprocessed sound data two and preprocessed sound data three;

[0011] Preprocessed sound feature data is obtained based on the historical status data of the production equipment to be tested, preprocessed sound data one, preprocessed sound data two, trend feature one to be tested, and preprocessed sound data three.

[0012] The maintenance priority of the equipment is obtained by assessing the degree of maintenance needs of the equipment under test based on the failure frequency value and pre-processed sound characteristic data.

[0013] Preferably, the sound information of the production equipment to be tested is processed and analyzed to obtain baseline feature information, specifically including the following steps:

[0014] Obtain the sound data of the production equipment under test to obtain the sound data to be analyzed;

[0015] The device audio data that shows abnormal status in the audio data to be analyzed is marked as abnormal audio data;

[0016] Mark other devices that are associated with the same type of device as the abnormal sound data as related devices of the same type;

[0017] A sound association dataset is obtained by evaluating the degree of sound association between similar devices and related devices of the same type;

[0018] Obtain auxiliary sound data by acquiring sound monitoring data from similar related devices;

[0019] The benchmark feature information is obtained by processing and analyzing the sound association dataset and the auxiliary test sound data.

[0020] Preferably, the benchmark feature information is obtained by processing and analyzing the sound association dataset and the auxiliary test sound data, specifically including the following steps:

[0021] Based on the sound association dataset and auxiliary test sound data, an auxiliary test sound database is obtained between devices of the same type and related devices of the same type.

[0022] The trend characteristics of the auxiliary test sound are obtained by performing comprehensive spectrum trend statistics on the auxiliary test sound database;

[0023] The first feature information is obtained by extracting sound trend features with stable frequencies less than or equal to a preset stable frequency threshold from the auxiliary sound trend feature conditions.

[0024] The device sound data that is in a normal state in the sound data to be analyzed is marked as the second feature information;

[0025] Among them, the combination of the first feature information and the second feature information constitutes the baseline feature information.

[0026] Preferably, the sound trends of the same-type correlation value and the matching correlation value are statistically analyzed to obtain the first trend feature to be tested and the second trend feature to be tested, which specifically includes the following steps:

[0027] By performing sound trend statistics on the correlation values ​​of the same type and the abnormal sound data, the first trend feature to be tested is obtained;

[0028] By performing sound trend statistics on the associated values ​​and abnormal sound data, we can obtain the second trend feature to be verified.

[0029] Preferably, the sound data belonging to the trend feature to be tested and the operation log data of the production equipment to be tested are comprehensively evaluated to obtain preprocessed sound data two and preprocessed sound data three, specifically including the following steps:

[0030] The target operating trend is obtained by statistically analyzing the operating trend of the production equipment to be tested.

[0031] The operational trends of similar equipment are statistically analyzed to obtain the operational trends of the same type;

[0032] The preprocessed audio data 2 and preprocessed audio data 3 are obtained by comparing and analyzing the audio data of the first trend feature to be verified, the target running trend, and the same type running trend.

[0033] Preferably, the sound data to which the trend feature to be verified belongs, the target running trend, and the same type running trend are subjected to trend comparison and analysis to obtain preprocessed sound data two and preprocessed sound data three, specifically including the following steps:

[0034] If the target running trend is inconsistent with the running trend of the same type, and the trend feature to be verified does not conform to the benchmark feature information, then the abnormal sound data to which the trend feature to be verified belongs will be removed.

[0035] If the target operating trend is consistent with the operating trend of the same type and the first trend to be tested does not conform to the benchmark feature information, or if the target operating trend is inconsistent with the operating trend of the same type and the first trend to be tested conforms to the benchmark feature information, then obtain the operating log data of the production equipment to be tested, and evaluate the sound data of the first trend to be tested and the operating log data to obtain the second preprocessed sound data.

[0036] If the target running trend is consistent with the same type of running trend and the trend feature to be verified first meets the benchmark feature information, then the abnormal sound data to which the trend feature to be verified first belongs is marked as preprocessed sound data three.

[0037] Preferably, it further includes:

[0038] Obtain historical maintenance data of the production equipment to be tested;

[0039] The failure frequency value is obtained by evaluating the failure frequency value of the production equipment to be tested based on the historical maintenance data.

[0040] Preferably, the equipment maintenance priority is obtained by assessing the maintenance needs of the production equipment to be tested based on historical maintenance data and preprocessed sound feature data, specifically including the following steps:

[0041] Obtain the average maintenance cycle for similar equipment;

[0042] The degree of cycle matching is obtained by comparing the current operating cycle and the average maintenance cycle of the production equipment to be tested.

[0043] The maintenance priority of the equipment is obtained by assessing the maintenance needs of the production equipment to be tested based on the pre-processed sound characteristic data, fault frequency values, and cycle matching degree.

[0044] An intelligent inspection management platform includes:

[0045] Processing module: Processes and analyzes the sound information of the production equipment to be tested to obtain baseline feature information;

[0046] The first evaluation module evaluates the degree of sound correlation between the production equipment to be tested and similar equipment to obtain the similar correlation value, and evaluates the degree of sound correlation between the production equipment to be tested and supporting equipment to obtain the supporting correlation value.

[0047] Statistical module: Statistically analyze the sound trends of similar correlation values ​​and complementary correlation values ​​to obtain experimental trend feature one and experimental trend feature two;

[0048] Labeling module: Labels the second trend feature to be verified that matches the baseline feature information as the first preprocessed sound data;

[0049] The second evaluation module: comprehensively evaluates the sound data belonging to the trend feature to be tested and the operation log data belonging to the production equipment to be tested to obtain preprocessed sound data two and preprocessed sound data three;

[0050] Analysis module: Based on the historical status data of the production equipment to be tested, preprocessed sound data one, preprocessed sound data two, trend characteristics to be tested one, and preprocessed sound data three, preprocessed sound feature data is obtained;

[0051] The third assessment module evaluates the maintenance needs of the production equipment under test based on the failure frequency value and pre-processed sound characteristic data to obtain the equipment maintenance priority.

[0052] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the intelligent inspection management method.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] This invention performs multi-stage processing and analysis of equipment sound. It can identify early, subtle anomalies in equipment, such as subtle sound changes caused by bearing wear, preventing minor equipment malfunctions from escalating into production stoppages and reducing the impact of sudden equipment failures on production continuity. It integrates multi-source information such as equipment sound data, operation logs, and historical data to construct pre-processed sound feature data. The invention comprehensively reflects the equipment's health by analyzing its current acoustic state, historical fault patterns, and changes in operating conditions. For example, historical data can be used to correct for misjudgments of current sound anomalies, and operation logs can be used to explain reasonable operating conditions for sound trend fluctuations. Maintenance priorities are determined based on the fault frequency values ​​of the production equipment under inspection and the pre-processed sound feature data. High-risk and high-fault-frequency equipment is prioritized for maintenance, while low-risk equipment is appropriately delayed or included in routine inspections, thus achieving efficient allocation of maintenance resources. This avoids resource waste caused by over-maintenance and prevents critical equipment from failing due to delayed maintenance, improving the company's ability to control maintenance costs and ensure equipment reliability. Attached Figure Description

[0055] Figure 1 This is a schematic diagram illustrating the steps of an intelligent inspection management method proposed in this invention;

[0056] Figure 2 This is a schematic diagram of a module of an intelligent inspection management platform proposed in this invention;

[0057] Figure 3 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention.

[0058] 610. Processor; 620. Communication interface; 630. Memory; 640. Communication bus. Detailed Implementation

[0059] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0060] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0061] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0062] Reference Figures 1-3 As shown.

[0063] The embodiments further illustrate the intelligent inspection management platform and method proposed in this invention.

[0064] An intelligent inspection management method, comprising the following steps:

[0065] The sound information of the production equipment to be tested is processed and analyzed to obtain the baseline feature information;

[0066] The degree of sound correlation between the production equipment to be tested and similar equipment is evaluated to obtain the similar correlation value, and the degree of sound correlation between the production equipment to be tested and supporting equipment is evaluated to obtain the supporting correlation value.

[0067] By statistically analyzing the sound trends of similar correlation values ​​and complementary correlation values, we obtain the first and second trend features to be tested.

[0068] The experimental trend feature two that matches the baseline feature information is marked as preprocessed sound data one;

[0069] The sound data belonging to the trend feature to be tested and the operation log data of the production equipment to be tested are comprehensively evaluated to obtain preprocessed sound data two and preprocessed sound data three;

[0070] Preprocessed sound feature data is obtained based on the historical status data of the production equipment to be tested, preprocessed sound data one, preprocessed sound data two, trend feature one to be tested, and preprocessed sound data three.

[0071] The maintenance priority of the equipment is obtained by assessing the degree of maintenance needs of the equipment under test based on the failure frequency value and pre-processed sound characteristic data.

[0072] Production equipment will generate sound signals with specific patterns when operating stably. First, the sound information of the production equipment to be tested is collected. Through spectrum analysis and time-domain feature extraction, key features that can represent the normal operating state of the equipment are extracted from the complex sound data, such as the amplitude distribution of specific frequency bands and the periodicity of the sound signal. These features are then integrated into the baseline feature information.

[0073] Production equipment of the same type shares common sound characteristics during normal operation and exhibits similar acoustic behaviors when malfunctioning. By collecting sound data from the equipment under test and other similar equipment, the correlation between their sound fluctuations is determined using the spectral similarity of the sound signals, thus obtaining a similarity correlation value. For example, if a machine tool produces abnormal high-frequency sounds due to tool wear, the similarity correlation value can capture the correlation changes between the sounds of other machine tools and this abnormal sound, thereby helping to predict potential equipment failure trends.

[0074] In a production setting, the equipment under test and its supporting equipment (such as processing equipment and material conveying equipment) are physically connected, transfer energy, or coordinate processes. These relationships cause the sounds of the equipment to influence each other. By judging the coupling relationship between the sound signals of the equipment under test and its supporting equipment, such as phase difference changes and frequency modulation patterns, the degree of sound linkage between the two can be evaluated to obtain the correlation value.

[0075] Based on the continuous collection of device sound data using similar correlation values ​​and matching correlation values, trend judgment is performed. A time window is set to statistically analyze the changes in correlation values ​​within different time periods, including the upward slope, downward slope, fluctuation period, and frequency of extreme values, thereby generating two trend features to be verified: trend feature one and trend feature two.

[0076] The second trend feature to be tested is carefully compared with the baseline feature information. Sound data segments whose trend changes conform to the acoustic laws of normal equipment operation are selected and marked as preprocessed sound data one. Invalid noise data caused by brief abnormal fluctuations of supporting equipment and random environmental interference are removed, and valid acoustic information that matches the equipment's baseline state is retained.

[0077] For the sound data associated with the first trend feature under test, cross-validation is performed using the operating log data of the production equipment under test. The operating log data includes records of startup duration, load changes, and process parameter adjustments. If the sound trend is abnormal, but the operating log shows that the equipment is under special reasonable operating conditions (such as short-term full-load production), this type of sound data is marked as preprocessed sound data two, indicating that the abnormality is caused by the operating condition. If the sound trend is abnormal and the operating log has no reasonable explanation for the operating condition, it is marked as preprocessed sound data three, representing acoustic information of a suspected fault. By fusing operating logs, misjudging equipment faults solely based on sound anomalies is avoided, significantly improving data reliability.

[0078] Historical condition data includes historical fault records, changes in acoustic characteristics after each maintenance, and long-term acoustic trends. The historical condition data of the production equipment under test is integrated and deeply fused with pre-processed acoustic data one, pre-processed acoustic data two, pre-processed acoustic data three, and the trend feature one to be tested. This involves weighted combinations of features from different sources and feature splicing and reconstruction based on machine learning models to construct pre-processed acoustic feature data covering the entire lifecycle of the equipment.

[0079] A maintenance requirement assessment model is built by combining the failure frequency values ​​of the production equipment under test with pre-processed sound characteristic data. The degree of equipment maintenance requirement is quantified. If the equipment failure frequency is high and the pre-processed sound characteristic data shows that the acoustic anomaly is continuously deteriorating, it is determined to be high-priority maintenance; if the failure frequency is low and the acoustic characteristics are close to the baseline state, it is determined to be low-priority or routine inspection.

[0080] The sound information of the production equipment to be tested is processed and analyzed to obtain baseline feature information, specifically including the following steps:

[0081] Obtain the sound data of the production equipment under test to obtain the sound data to be analyzed;

[0082] The device audio data that shows abnormal status in the audio data to be analyzed is marked as abnormal audio data;

[0083] Mark other devices that are associated with the same type of device as the abnormal sound data as related devices of the same type;

[0084] A sound association dataset is obtained by evaluating the degree of sound association between similar devices and related devices of the same type;

[0085] Obtain auxiliary sound data by acquiring sound monitoring data from similar related devices;

[0086] The benchmark feature information is obtained by processing and analyzing the sound association dataset and the auxiliary test sound data.

[0087] By continuously capturing various sound signals generated during the operation of the equipment through sound acquisition devices around the equipment, the sound information, including the normal operation and abnormal vibration of the equipment, is summarized to form sound data to be analyzed.

[0088] This method uses spectral analysis and time-domain feature extraction techniques to identify segments in the sound data that differ significantly from the sound of normal equipment operation. For example, by comparing the frequency distribution, amplitude changes, and duration parameters of the sound, sound data containing potential faults such as wear and tear of equipment components or loose structure can be marked as abnormal sound data.

[0089] Other devices associated with the same type of device as the abnormal sound data are marked as related devices of the same type. For example, multiple machine tools of the same model on a production line are considered related devices of the same type because they share a material conveying system.

[0090] By calculating the cross-correlation coefficient and spectral matching degree of sound signals from similar and related devices, the degree of sound correlation between them is evaluated over time, thereby generating a sound correlation dataset. For example, the frequency and amplitude correlation values ​​of sound synchronization fluctuations of related devices of the same type can be statistically analyzed when the equipment load changes.

[0091] Collect real-time or historical sound monitoring data from related devices of the same type, and use this data as supplementary sound data. For example, changes in the sound characteristics of related devices due to changes in their own condition (such as bearing wear) will indirectly reflect changes in the operating environment of the device under test through device linkage.

[0092] By fusing sound association datasets with auxiliary sound data, and employing big data analytics and machine learning algorithms (such as cluster analysis and feature engineering), typical acoustic features representing the normal operation and abnormal correlation of similar devices and related devices are extracted. For example, the stable range and variation patterns of sound frequency, amplitude, and phase difference parameters of devices under different operating conditions and correlation states are selected to construct benchmark feature information.

[0093] The benchmark feature information is obtained by processing and analyzing the sound association dataset and the auxiliary test sound data, specifically including the following steps:

[0094] Based on the sound association dataset and auxiliary test sound data, an auxiliary test sound database is obtained between devices of the same type and related devices of the same type.

[0095] The trend characteristics of the auxiliary test sound are obtained by performing comprehensive spectrum trend statistics on the auxiliary test sound database;

[0096] The first feature information is obtained by extracting sound trend features with stable frequencies less than or equal to a preset stable frequency threshold from the auxiliary sound trend feature conditions.

[0097] The device sound data that is in a normal state in the sound data to be analyzed is marked as the second feature information;

[0098] Among them, the combination of the first feature information and the second feature information constitutes the baseline feature information.

[0099] First, based on the sound association dataset and auxiliary test sound data, the data is classified and integrated according to device type, association relationship and time dimension, so as to build an auxiliary test sound database between devices of the same type and related devices of the same type.

[0100] For the audio data in the auxiliary testing audio database, comprehensive spectral trend statistics are conducted using spectral analysis techniques (such as Fourier transform and short-time Fourier transform). From a frequency domain perspective, the frequency component distribution, amplitude variation trends, and frequency peak migration patterns of audio signals under different equipment and operating conditions are determined, thereby generating auxiliary testing audio trend characteristics. For example, the curves of the dominant frequency changes and the fluctuation trends of the energy proportion of each frequency band are statistically analyzed when the equipment is running from no-load to full-load, determining the inherent laws of sound changes with equipment status, thus providing trend basis for identifying abnormal acoustic characteristics of equipment.

[0101] A preset stable frequency threshold is set, which is determined based on the device's design parameters and historical acoustic data. Sound trend features with stable frequencies less than or equal to this threshold are selected and marked as the first feature information. These features indicate that the device's sound frequency remains relatively stable under normal operating conditions.

[0102] Audio signal recognition technology is used to filter out sound data from devices operating normally, and this data is then marked as secondary feature information. This information directly reflects the acoustic characteristics of the device under test during normal operation.

[0103] The first and second feature information are fused together. A baseline feature information system is constructed through feature concatenation and weight allocation, encompassing individual devices and related device groups, stable frequencies, and their own normal acoustic modes. This information serves as the standard for determining whether a device's sound is abnormal during intelligent inspection. Subsequent equipment inspections compare real-time sound data with this baseline to identify potential equipment malfunctions and operational anomalies.

[0104] The sound trends of homogeneous correlation values ​​and complementary correlation values ​​are statistically analyzed to obtain the first and second trends to be tested, which specifically includes the following steps:

[0105] By performing sound trend statistics on the correlation values ​​of the same type and the abnormal sound data, the first trend feature to be tested is obtained;

[0106] By performing sound trend statistics on the associated values ​​and abnormal sound data, we can obtain the second trend feature to be verified.

[0107] The homomorphic correlation value reflects the degree of acoustic correlation and quantitative relationship between the tested production equipment and similar equipment. Abnormal sound data is segment from the sound data to be analyzed that deviates from the acoustic performance of the equipment during normal operation. When performing sound trend statistics, the homomorphic correlation value is continuously tracked over time, while the frequency, amplitude, and frequency characteristics of abnormal sound data in the corresponding time periods are observed. For example, within a set continuous time window, the fluctuation range and slope of the homomorphic correlation value, as well as the proportion of abnormal sound data and the evolution of acoustic characteristics within each window, are statistically analyzed. The dynamic correlation trend between the homomorphic correlation value and abnormal sound data is determined; for example, whether the abnormal sound data increases synchronously when the homomorphic correlation value drops sharply, and whether its frequency components exhibit a specific pattern.

[0108] The correlation value reflects the quantitative relationship between the sound of the production equipment under test and its supporting equipment. Similarly, it combines abnormal sound data to conduct sound trend statistics, determining the changing trend of the correlation value under different production conditions and time stages, and whether the generation and changes of abnormal sound data are related to the fluctuations of the correlation value. For example, when the load of the supporting equipment is adjusted, the correlation value will change. Simultaneously, the acoustic characteristics (such as amplitude and frequency) of the abnormal sound data of the equipment under test are monitored to determine how they change, and the correlation trend between the two over time is statistically analyzed, which could be positive, negative, or lagged. The correlation patterns between the correlation value and the abnormal sound data at the trend level are extracted and summarized to form the second trend feature to be tested.

[0109] The two trend features to be verified, namely, the first and second trend features, determine the dynamic trend relationship between abnormal sound data and the sound correlation values ​​between devices from two dimensions: correlation with similar devices and correlation with supporting devices, respectively. These features present the development trend of abnormal sounds in the device correlation environment, and at the same time provide a basis for subsequent judgment on whether abnormal sounds belong to real equipment failure hazards and assess equipment maintenance needs, thereby improving the comprehensiveness and accuracy of intelligent inspection in judging equipment status.

[0110] The sound data belonging to the trend feature to be tested and the operation log data of the production equipment to be tested are comprehensively evaluated to obtain preprocessed sound data two and preprocessed sound data three, which specifically includes the following steps:

[0111] The target operating trend is obtained by statistically analyzing the operating trend of the production equipment to be tested.

[0112] The operational trends of similar equipment are statistically analyzed to obtain the operational trends of the same type;

[0113] The preprocessed audio data 2 and preprocessed audio data 3 are obtained by comparing and analyzing the audio data of the first trend feature to be verified, the target running trend, and the same type running trend.

[0114] For the production equipment under test, operational parameters are collected, such as the equipment's start-up and shutdown status, operating load, and process parameter data. By analyzing the patterns of these parameters' changes over time, such as using time series analysis and sliding window statistical methods, the start-up-stable operation-shutdown cycle, load fluctuation patterns, and process parameter change trends are determined, thereby constructing the target operational trend of the equipment under test. The operational trends of similar equipment are then statistically analyzed to obtain the operational trends for that type of equipment.

[0115] The sound data associated with the first trend feature to be verified, the target operating trend, and the similar operating trends are compared and processed. The sound data associated with the first trend feature to be verified includes trend information on equipment sound correlation and anomalies. Combining the first trend feature to be verified with the equipment operating trend is to determine the matching degree between the sound trend and the actual operating status trend of the equipment.

[0116] The preprocessed audio data 2 and preprocessed audio data 3 are obtained by performing trend comparison analysis on the audio data of the first trend feature to be verified, the target running trend, and the same type running trend. The specific steps include:

[0117] If the target running trend is inconsistent with the running trend of the same type, and the trend feature to be verified does not conform to the benchmark feature information, then the abnormal sound data to which the trend feature to be verified belongs will be removed.

[0118] If the target operating trend is consistent with the operating trend of the same type and the first trend to be tested does not conform to the benchmark feature information, or if the target operating trend is inconsistent with the operating trend of the same type and the first trend to be tested conforms to the benchmark feature information, then obtain the operating log data of the production equipment to be tested, and evaluate the sound data of the first trend to be tested and the operating log data to obtain the second preprocessed sound data.

[0119] If the target running trend is consistent with the same type of running trend and the trend feature to be verified first meets the benchmark feature information, then the abnormal sound data to which the trend feature to be verified first belongs is marked as preprocessed sound data three.

[0120] If the target operating trend of the production equipment under test is inconsistent with the operating trend of similar equipment of the same type, it indicates that the operating status of the equipment under test differs from the typical status of similar equipment. Furthermore, if the first characteristic of the trend under test does not conform to the baseline characteristic information, the abnormal sound data is caused by non-real fault factors such as special interference and erroneous acquisition, and therefore has no value for equipment fault diagnosis. Thus, the abnormal sound data belonging to the first characteristic of the trend under test is discarded.

[0121] If the target operating trend is consistent with the operating trend of similar equipment, but the first characteristic of the trend to be verified does not conform to the baseline characteristic information, it indicates that the operating status of the equipment under test conforms to the normal mode of similar equipment. However, the first characteristic of the trend to be verified does not conform to the baseline characteristic information, meaning that there is an anomaly in the sound data. In this case, the abnormal sound may be a potential minor fault in the equipment, or it may be caused by special operating conditions (such as short-term process adjustments) in the operation log.

[0122] If the target operating trend is inconsistent with the operating trend of the same type, but the first trend feature to be tested conforms to the benchmark feature information, the inconsistency between the target operating trend and the operating trend of the same type indicates that the operating state of the device to be tested is special, but the first trend feature to be tested conforms to the benchmark feature information, that is, the sound data seems normal.

[0123] In both sub-cases, it is necessary to obtain the operation log data of the production equipment under test (covering equipment start-up and shutdown times, load changes, process parameter adjustments, manual operation records, etc.). The sound data corresponding to the first trend characteristic to be tested and the operation log data are cross-evaluated. For example, it is determined whether abnormal sound periods correspond to process parameter adjustments, and whether there are hidden abnormal operations during normal sound periods. Through evaluation and filtering, sound fluctuations caused by reasonable operating conditions, as well as sound data that, while conforming to some trend characteristics, still require attention based on the log data, are integrated to obtain the second preprocessed sound data.

[0124] When the target's operating trend aligns with that of similar equipment, it indicates that the equipment's operating status conforms to the norm for such equipment. Simultaneously, the first characteristic of the tested trend matches the baseline characteristic information, meaning the sound data also fits the equipment's normal acoustic model. The abnormal sound data associated with the first characteristic of the tested trend represents early, weak signals of a real equipment malfunction, or key acoustic information reflecting potential equipment problems. This is marked as preprocessed sound data three, serving as a basis for determining whether the equipment requires early maintenance and troubleshooting.

[0125] Also includes:

[0126] Obtain historical maintenance data of the production equipment to be tested;

[0127] The failure frequency value is obtained by evaluating the failure frequency value of the production equipment to be tested based on historical maintenance data.

[0128] Historical data related to the equipment is extracted from the enterprise's equipment management system, maintenance work order records, maintenance files, and other media, including the time of each equipment failure, the type of failure, the duration of failure repair, maintenance measures, and maintenance cycle information.

[0129] Set a time period and count the total number of failures of the production equipment under test within that time period. Weight the number of failures based on factors such as the severity of the failure type and the weight of different failures' impact on production; for example, assign higher weights to major failures and lower weights to minor failures.

[0130] For example, if the equipment experienced 5 general failures and 2 major failures that caused production line shutdowns in the past year, the major failures were given twice the weight of the general failures when calculating the failure frequency. The failure frequency value per unit time was obtained by ÷ 12 (months) using the method of (5×1+2×2).

[0131] Based on historical maintenance data and pre-processed sound characteristic data, the maintenance needs of the production equipment to be tested are assessed to determine the equipment maintenance priority. This includes the following steps:

[0132] Obtain the average maintenance cycle for similar equipment;

[0133] The degree of cycle matching is obtained by comparing the current operating cycle and the average maintenance cycle of the production equipment to be tested.

[0134] The maintenance priority of the equipment is obtained by assessing the maintenance needs of the production equipment to be tested based on the pre-processed sound characteristic data, fault frequency values, and cycle matching degree.

[0135] This application selects a group of equipment of the same type and specifications as the production equipment to be tested from the enterprise's equipment management system or industry equipment operation and maintenance database, and collects the historical maintenance records of these devices, including the time of each maintenance and the maintenance interval. The average maintenance cycle of similar equipment under normal use is calculated by statistics.

[0136] The current operating cycle data of the production equipment under test is compared with the average maintenance cycle of similar equipment. The degree of cycle matching is calculated. For example, if the average maintenance cycle is 180 days and the current operating cycle of the equipment under test is 150 days, the degree of cycle matching can be reflected as the current cycle being 83% of the average cycle.

[0137] If the data contains a large number of high-frequency abnormal sound features and the trend deviates significantly from the baseline, it indicates that the current acoustic condition of the equipment is poor and the potential risk of failure is high; conversely, a stable acoustic condition indicates a low risk. Equipment with a high historical failure frequency indicates a high susceptibility to failure, thus requiring more attention to maintenance; a low frequency indicates relative stability. Equipment nearing or exceeding its service life should be prioritized for maintenance according to standard operating procedures.

[0138] By assigning appropriate weights to different dimensions of data (e.g., 40% for acoustic features, 30% for failure frequency, and 30% for cycle matching; these weights can be adjusted based on industry characteristics and equipment importance), a weighted summation method is used to quantitatively assess the maintenance needs of the equipment under test. Equipment with high maintenance needs is marked as high-priority maintenance; equipment with low maintenance needs is marked as low-priority maintenance.

[0139] An intelligent inspection management platform includes:

[0140] Processing module: Processes and analyzes the sound information of the production equipment to be tested to obtain baseline feature information;

[0141] The first evaluation module evaluates the degree of sound correlation between the production equipment to be tested and similar equipment to obtain the similar correlation value, and evaluates the degree of sound correlation between the production equipment to be tested and supporting equipment to obtain the supporting correlation value.

[0142] Statistical module: Statistically analyze the sound trends of similar correlation values ​​and complementary correlation values ​​to obtain experimental trend feature one and experimental trend feature two;

[0143] Labeling module: Labels the second trend feature to be verified that matches the baseline feature information as the first preprocessed sound data;

[0144] The second evaluation module: comprehensively evaluates the sound data belonging to the trend feature to be tested and the operation log data belonging to the production equipment to be tested to obtain preprocessed sound data two and preprocessed sound data three;

[0145] Analysis module: Based on the historical status data of the production equipment to be tested, preprocessed sound data one, preprocessed sound data two, trend characteristics to be tested one, and preprocessed sound data three, preprocessed sound feature data is obtained;

[0146] The third assessment module evaluates the maintenance needs of the production equipment under test based on the failure frequency value and pre-processed sound characteristic data to obtain the equipment maintenance priority.

[0147] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement an intelligent inspection management method.

[0148] like Figure 3 As shown, the electronic device may include a processor 610, a communication interface 620, a memory 630, and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other through the communication bus 640. The processor 610 can call logical instructions in the memory 630 to execute an intelligent inspection management method.

[0149] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks.

[0150] On the other hand, the present invention also provides a computer program product, the computer program product including a computer program, the computer program being able to be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer is able to execute an intelligent inspection management method.

[0151] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform an intelligent inspection management method.

[0152] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0153] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0154] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent inspection management method, characterized in that, The method includes the following steps: The sound information of the production equipment to be tested is processed and analyzed to obtain the baseline feature information; The degree of sound correlation between the production equipment to be tested and similar equipment is evaluated to obtain the similar correlation value, and the degree of sound correlation between the production equipment to be tested and supporting equipment is evaluated to obtain the supporting correlation value. By statistically analyzing the sound trends of similar correlation values ​​and complementary correlation values, we obtain the first and second trend features to be tested. The experimental trend feature two that matches the baseline feature information is marked as preprocessed sound data one; The sound data belonging to the trend feature to be tested and the operation log data of the production equipment to be tested are comprehensively evaluated to obtain preprocessed sound data two and preprocessed sound data three; Preprocessed sound feature data is obtained based on the historical status data of the production equipment to be tested, preprocessed sound data one, preprocessed sound data two, trend feature one to be tested, and preprocessed sound data three. The maintenance priority of the equipment is obtained by assessing the degree of maintenance needs of the equipment under test based on the failure frequency value and pre-processed sound characteristic data.

2. The intelligent inspection management method according to claim 1, characterized in that, The sound information of the production equipment to be tested is processed and analyzed to obtain baseline feature information, specifically including the following steps: Obtain the sound data of the production equipment under test to obtain the sound data to be analyzed; The device audio data that shows abnormal status in the audio data to be analyzed is marked as abnormal audio data; Mark other devices that are associated with the same type of device as the abnormal sound data as related devices of the same type; A sound association dataset is obtained by evaluating the degree of sound association between similar devices and related devices of the same type; Obtain auxiliary sound data by acquiring sound monitoring data from similar related devices; The benchmark feature information is obtained by processing and analyzing the sound association dataset and the auxiliary test sound data.

3. The intelligent inspection management method according to claim 2, characterized in that, The benchmark feature information is obtained by processing and analyzing the sound association dataset and the auxiliary test sound data, specifically including the following steps: Based on the sound association dataset and auxiliary test sound data, an auxiliary test sound database is obtained between devices of the same type and related devices of the same type. The trend characteristics of the auxiliary test sound are obtained by performing comprehensive spectrum trend statistics on the auxiliary test sound database; The first feature information is obtained by extracting sound trend features with stable frequencies less than or equal to a preset stable frequency threshold from the auxiliary sound trend feature conditions. The device sound data that is in a normal state in the sound data to be analyzed is marked as the second feature information; Among them, the combination of the first feature information and the second feature information constitutes the baseline feature information.

4. The intelligent inspection management method according to claim 3, characterized in that, The sound trends of homogeneous correlation values ​​and complementary correlation values ​​are statistically analyzed to obtain the first and second trends to be tested, which specifically includes the following steps: By performing sound trend statistics on the correlation values ​​of the same type and the abnormal sound data, the first trend feature to be tested is obtained; By performing sound trend statistics on the associated values ​​and abnormal sound data, we can obtain the second trend feature to be verified.

5. The intelligent inspection management method according to claim 4, characterized in that, The sound data belonging to the trend feature to be tested and the operation log data of the production equipment to be tested are comprehensively evaluated to obtain preprocessed sound data two and preprocessed sound data three, which specifically includes the following steps: The target operating trend is obtained by statistically analyzing the operating trend of the production equipment to be tested. The operational trends of similar equipment are statistically analyzed to obtain the operational trends of the same type; The preprocessed audio data 2 and preprocessed audio data 3 are obtained by comparing and analyzing the audio data of the first trend feature to be verified, the target running trend, and the same type running trend.

6. The intelligent inspection management method according to claim 5, characterized in that, The preprocessed audio data 2 and preprocessed audio data 3 are obtained by performing trend comparison analysis on the audio data of the first trend feature to be verified, the target running trend, and the same type running trend. The specific steps include: If the target running trend is inconsistent with the running trend of the same type, and the trend feature to be verified does not conform to the benchmark feature information, then the abnormal sound data to which the trend feature to be verified belongs will be removed. If the target operating trend is consistent with the operating trend of the same type and the first trend to be tested does not conform to the benchmark feature information, or if the target operating trend is inconsistent with the operating trend of the same type and the first trend to be tested conforms to the benchmark feature information, then obtain the operating log data of the production equipment to be tested, and evaluate the sound data of the first trend to be tested and the operating log data to obtain the second preprocessed sound data. If the target running trend is consistent with the same type of running trend and the trend feature to be verified first meets the benchmark feature information, then the abnormal sound data to which the trend feature to be verified first belongs is marked as preprocessed sound data three.

7. The intelligent inspection management method according to claim 6, characterized in that, Also includes: Obtain historical maintenance data of the production equipment to be tested; The failure frequency value is obtained by evaluating the failure frequency value of the production equipment to be tested based on the historical maintenance data.

8. The intelligent inspection management method according to claim 7, characterized in that, Based on historical maintenance data and pre-processed sound characteristic data, the maintenance needs of the production equipment to be tested are assessed to determine the equipment maintenance priority. This includes the following steps: Obtain the average maintenance cycle for similar equipment; The degree of cycle matching is obtained by comparing the current operating cycle and the average maintenance cycle of the production equipment to be tested. The maintenance priority of the equipment is obtained by assessing the maintenance needs of the production equipment to be tested based on the pre-processed sound characteristic data, fault frequency values, and cycle matching degree.

9. An intelligent inspection management platform, applied to the intelligent inspection management method according to any one of claims 1 to 8, characterized in that, include: Processing module: Processes and analyzes the sound information of the production equipment to be tested to obtain baseline feature information; The first evaluation module evaluates the degree of sound correlation between the production equipment to be tested and similar equipment to obtain the similar correlation value, and evaluates the degree of sound correlation between the production equipment to be tested and supporting equipment to obtain the supporting correlation value. Statistical module: Statistically analyze the sound trends of similar correlation values ​​and complementary correlation values ​​to obtain experimental trend feature one and experimental trend feature two; Labeling module: Labels the second trend feature to be verified that matches the baseline feature information as the first preprocessed sound data; The second evaluation module: comprehensively evaluates the sound data belonging to the trend feature to be tested and the operation log data belonging to the production equipment to be tested to obtain preprocessed sound data two and preprocessed sound data three; Analysis module: Based on the historical status data of the production equipment to be tested, preprocessed sound data one, preprocessed sound data two, trend characteristics to be tested one, and preprocessed sound data three, preprocessed sound feature data is obtained; The third assessment module evaluates the maintenance needs of the production equipment under test based on the failure frequency value and pre-processed sound characteristic data to obtain the equipment maintenance priority.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements an intelligent inspection management method as described in any one of claims 1 to 8.

Images