Production line equipment health prediction and maintenance optimization method based on a honkong system

Abstract

The present application relates to the technical field of data processing, and especially relates to a production line equipment health prediction and maintenance optimization method based on a HongMeng system, which comprises the following steps: data acquisition; abnormality determination; degree determination; type determination I; type determination II; type correction; index calculation; threshold adjustment or generation of a maintenance management strategy. The present application dynamically adjusts the determination threshold or generates a targeted maintenance management strategy according to different abnormal types and their time distribution characteristics by comprehensively analyzing the change relationship between the key operation parameters of the production line equipment in real time. The health prediction index is formed by the step-by-step determination of the acceleration value, temperature value, current value, rotational speed value, diameter value, surface roughness, response delay value and voltage value, so that the prediction result is closer to the real operation state of the equipment, and the problem of low operation and maintenance efficiency caused by the lag of production line maintenance decision and the static maintenance suggestion due to the single monitoring index is effectively solved.

Description

Technical Field

[0001] This invention relates to the field of equipment maintenance technology, and in particular to a method for health prediction and maintenance optimization of production line equipment based on the HarmonyOS system. Background Technology

[0002] With the continuous development of industrial automation and intelligent manufacturing, the operational stability and reliability of production line equipment have become key factors affecting enterprise production efficiency, operating costs, and market competitiveness. In recent years, with the popularization of distributed operating system technology, industrial control and management architecture based on the HarmonyOS system has been gradually applied to the field of intelligent manufacturing. Its distributed task scheduling and multi-node data collaboration capabilities provide a technical foundation for realizing information exchange between equipment and dynamic collaboration between processes. Under the existing factory management model, equipment maintenance generally faces two major challenges: First, preventive maintenance based on fixed cycles or single parameter thresholds lacks accurate insight into the true health status of equipment, easily leading to unreasonable allocation of maintenance resources; second, traditional equipment monitoring systems suffer from severe data silos, making it impossible to conduct collaborative analysis and intelligent prediction of multi-dimensional operating parameters, making it difficult for production scheduling departments to make scientific maintenance decisions and production plan adjustments in advance, seriously affecting the smoothness and resource utilization of the entire production system.

[0003] Chinese Patent Application Publication No. CN113592179A discloses a predictive maintenance method, system, and storage medium. The method includes: receiving configuration information from a user for each of at least one target monitoring components via a workstation management human-machine interface. The configuration information includes: workstation, component name, data acquisition scheme and protocol, and collected data characteristic variables. Based on the configuration information, data from at least one target monitoring component is collected, and the collected data is managed uniformly and stored in a first database. When the configuration information further includes a fault alarm threshold for the data characteristic variables and a fault code representing fault information, the data of the target monitoring components collected in real time is monitored based on the fault alarm threshold. If, within a set time, the data of a data characteristic variable reaches the corresponding fault alarm threshold, a fault alarm is issued, and corresponding fault information is provided according to the fault code. Alternatively, when the configuration information further includes an adaptive threshold identifier and fault code for a data feature variable, the collected data of the data feature variable is analyzed to obtain the safe data range of the data feature variable. Based on the safe data range, a fault alarm threshold for the data feature variable is determined, or the fault alarm threshold for the data feature variable is determined based on the learned historical alarm threshold of the data feature variable. Based on the fault alarm threshold, the data of the target monitoring component collected in real time is monitored. When the data of the data feature variable reaches the fault alarm threshold within a set time, a fault alarm is issued and corresponding fault information is provided according to the fault code. The fault information includes at least one or any combination of fault name, fault description information, and fault resolution suggestions. The correspondence between fault codes and fault information is provided to users for learning and querying through a code information human-machine interaction interface.

[0004] Therefore, the predictive maintenance method has the following problems: the method only triggers an alarm when a single data feature variable continuously exceeds the standard, and its output "fault resolution suggestions" are usually static and predefined, resulting in a lack of priority judgment basis for maintenance decisions; its threshold "adaptation" is only limited to adjusting the statistical range of historical data, and it cannot dynamically adjust the sensitivity of anomaly monitoring according to the evolution trend of anomaly types and the overall health degradation law of equipment. Summary of the Invention

[0005] To address this, the present invention provides a method for health prediction and maintenance optimization of production line equipment based on the HarmonyOS system. This method overcomes the problems of lagging production line maintenance decisions and low operation and maintenance efficiency caused by single monitoring indicators and static maintenance recommendations in the prior art by realizing collaborative modeling and dynamic correlation of equipment operation data and production management data in a distributed operating system environment.

[0006] To achieve the above objectives, this invention provides a method for health prediction and maintenance optimization of production line equipment based on the HarmonyOS system, comprising:

[0007] The HarmonyOS distributed bus collects real-time data on the spindle acceleration and temperature values, drive unit current and speed values, workpiece diameter and surface roughness values, and control component response delay and voltage values ​​of each workstation in the production line equipment cluster.

[0008] An abnormal event is determined based on the acceleration value, the temperature value, and a preset anomaly determination threshold to obtain an anomaly determination result;

[0009] Based on the anomaly determination result, the degree of anomaly of the abnormal event is determined according to the current value and the rotation speed value, so as to obtain a first degree or a second degree;

[0010] Based on the first degree, the anomaly type of the abnormal event is determined according to the diameter value and the surface roughness value to obtain a first type;

[0011] Based on the second degree, the anomaly type of the abnormal event is determined according to the response delay value and the voltage value to obtain a second type or a third type;

[0012] Based on the time distribution characteristics of the first and second degrees within a preset correction period, the second type or the third type is corrected to the first type;

[0013] A health prediction index is determined based on the acceleration value and temperature value corresponding to the first type, the second type, and the third type within a preset prediction time.

[0014] Adjust the preset anomaly determination threshold based on the health prediction index and all the anomaly types, or generate a maintenance management strategy.

[0015] Further, the process of determining the degree of abnormality of the abnormal event based on the current value and the rotational speed value to obtain a first degree or a second degree includes:

[0016] Z-score calculation is performed on all the current values ​​within a predetermined time period to obtain several normalized currents;

[0017] Z-score calculation is performed on all the rotational speed values ​​within the predetermined time period to obtain several normalized rotational speeds;

[0018] Calculate the absolute value of the cosine similarity of all the normalized currents and all the normalized rotational speeds to obtain the degree determination value;

[0019] The degree of abnormality of the abnormal event is determined based on the degree determination value to obtain the first degree or the second degree.

[0020] Further, the process of determining the degree of abnormality of the abnormal event based on the degree determination value to obtain the first degree or the second degree includes:

[0021] When the degree determination value is less than the preset degree determination threshold, a correlation deviation is determined to occur, so as to determine the degree of abnormality of the abnormal event as the first degree;

[0022] When the degree determination value is greater than or equal to the preset degree determination threshold, it is determined that the association is maintained, so as to determine the degree of abnormality of the abnormal event as the second degree.

[0023] Further, the process of determining the anomaly type of the abnormal event based on the diameter value and the surface roughness value to obtain the first type includes:

[0024] Calculate the average and variance of all diameter values ​​within the preset detection time period to obtain the average diameter and diameter variance;

[0025] Calculate the difference between the average diameter and the variance of the diameter to obtain the diameter deviation;

[0026] Calculate the average value and variance of all surface roughnesses within the preset detection time period to obtain the average roughness value and roughness variance;

[0027] Calculate the difference between the roughness average value and the roughness variance to obtain the roughness deviation;

[0028] The abnormality type of the abnormal event is determined based on the diameter deviation and the roughness deviation to obtain the first type.

[0029] Further, the process of determining the anomaly type of the abnormal event based on the diameter deviation and the roughness deviation to obtain the first type includes:

[0030] When the diameter deviation is greater than a preset diameter deviation threshold and the roughness deviation is greater than a preset roughness deviation threshold, the abnormality type is determined to be geometric wear, so as to obtain the first type.

[0031] Further, the process of determining the anomaly type of the abnormal event based on the response delay value and the voltage value to obtain a second or third type includes:

[0032] Calculate the average value of all response delay values ​​within the preset detection time to obtain the mean delay;

[0033] Calculate the average value of all voltage values ​​within the preset detection time to obtain the average voltage value;

[0034] When the average delay is greater than a preset delay threshold and the average voltage is less than a preset voltage threshold, the abnormality type is determined to be a control failure, thus obtaining the second type;

[0035] When the average delay is greater than the preset delay threshold and the average voltage is greater than or equal to the preset voltage threshold, the abnormality type is determined to be power supply fluctuation type, so as to obtain the third type.

[0036] Further, the process of correcting the second type or the third type to the first type based on the time distribution characteristics of the first degree and the second degree within a preset correction period includes:

[0037] Mark all timestamps corresponding to the first and second degrees within the preset correction period to obtain the first time series and the second time series;

[0038] Calculate the time difference between adjacent timestamps in the first time series to obtain the first time interval series;

[0039] Calculate the time difference between adjacent timestamps in the second time series to obtain the second time interval series;

[0040] Calculate the mean and variance of the first time interval sequence respectively to obtain the first interval mean and the first interval variance;

[0041] Calculate the mean and variance of the second time interval sequence to obtain the second interval mean and the second interval variance;

[0042] When the second interval variance is greater than a preset interval variance threshold, the second interval mean is less than the first interval mean, and the difference between the first interval mean and the second interval mean is greater than a preset mean difference threshold, the corresponding time period in the second time series is determined to be an irregularly distributed time period.

[0043] The second type or the third type in the irregular distribution period is identified as the first type, and then corrected to the first type.

[0044] Furthermore, the process of determining the health prediction index based on the acceleration value and temperature value corresponding to the first type, the second type, and the third type within the preset prediction time includes:

[0045] Calculate the average of the acceleration values ​​and the average of the temperature values ​​corresponding to all the first types within the preset prediction time period to obtain the first average acceleration value and the first average temperature value;

[0046] Calculate the average of the acceleration values ​​and the average of the temperature values ​​corresponding to all the second types within the preset prediction time period to obtain the second average acceleration value and the second average temperature value.

[0047] Calculate the average of the acceleration values ​​and the average of the temperature values ​​corresponding to all the third types within the preset prediction time period to obtain the third acceleration average and the third temperature average.

[0048] The health prediction index is determined based on the first average acceleration, the first average temperature, the second average acceleration, the second average temperature, the third average acceleration, and the third average temperature.

[0049] Furthermore, the process of adjusting the preset anomaly determination threshold or generating a maintenance management strategy based on the health prediction index and all the anomaly types includes:

[0050] When the health prediction index is greater than a preset prediction index threshold, the ratio of the number of the first type to the number of all the abnormal types is calculated to obtain a first proportion;

[0051] Calculate the ratio of the number of the second type to the total number of all said anomaly types to obtain a second proportion;

[0052] Calculate the ratio of the number of the third type to the total number of all the abnormal types to obtain the third proportion;

[0053] When the first proportion, the second proportion, and the third proportion are all greater than the preset proportion threshold, the preset abnormality judgment threshold is adjusted according to the first proportion, the second proportion, the third proportion, and the preset proportion threshold.

[0054] When any one or two of the first proportion, the second proportion, and the third proportion are greater than the preset proportion threshold, the maintenance management strategy is generated according to the corresponding anomaly type.

[0055] Furthermore, the process of determining the occurrence of an abnormal event based on the acceleration value, the temperature value, and the preset anomaly determination threshold includes:

[0056] Record the duration for which the acceleration value first exceeds a preset acceleration threshold and the temperature value first exceeds a preset temperature threshold to obtain an anomaly determination value;

[0057] When the anomaly determination value is greater than the preset anomaly determination threshold, the anomaly event is determined to have occurred.

[0058] Compared with existing technologies, the beneficial effects of this invention are that by collecting and comprehensively analyzing the dynamic changes between key operating parameters of production line equipment in real time, it can accurately identify and classify abnormal states, and dynamically adjust the judgment threshold or generate targeted maintenance management strategies according to different abnormality types and their time distribution characteristics, thereby effectively improving equipment operation stability and maintenance efficiency. Among them, acceleration and temperature values ​​reflect the mechanical vibration and thermal state of the equipment, current and speed values ​​jointly reveal the operating load characteristics of the drive unit, diameter and surface roughness values ​​reflect the machining accuracy, and response delay and voltage values ​​reflect the response performance of the control components. These parameters are interrelated and formed into a health prediction index through weighted summation, making the prediction results closer to the actual operating state of the equipment, enabling early warning and guidance for precise maintenance, extending equipment life and reducing production risks, effectively solving the problems of lagging production line maintenance decisions and low operation and maintenance efficiency caused by single monitoring indicators and static maintenance recommendations.

[0059] Furthermore, by normalizing the current and speed values ​​and calculating the cosine similarity, the dynamic correlation characteristics of the drive unit during operation can be accurately reflected, effectively revealing abnormal trends caused by load changes, mechanical resistance, or power supply fluctuations, thereby achieving a scientific distinction of the degree of abnormality. This judgment method based on multi-parameter correlation helps to capture subtle changes in the equipment's operating status, improves the accuracy of anomaly detection and early warning capabilities, ensures stable equipment operation, and extends its service life.

[0060] Furthermore, by performing similarity analysis on the normalized values ​​of current and rotational speed, the operating status of the equipment is quantified into a severity judgment value, thereby accurately distinguishing different degrees of anomaly. When the severity judgment value is below a preset threshold, it reflects a weakened correlation between current and rotational speed, indicating that the equipment may have a sudden or unstable anomaly, i.e., the first degree. When the severity judgment value is above the threshold, it reflects a stable correlation between current and rotational speed, suggesting that the anomaly is a continuous or systematic change, i.e., the second degree. Based on the real-time change characteristics of current and rotational speed, the system can effectively capture the dynamic changes in equipment operation, thereby improving the accuracy and response efficiency of anomaly detection.

[0061] Furthermore, by performing differential calculations on the average and variance of diameter and surface roughness values, it is helpful to accurately capture minute fluctuations in dimensional changes and surface conditions during processing. This statistical characteristic-based judgment can reflect the intrinsic relationship between equipment processing accuracy and stability, thereby enabling accurate identification of abnormal types such as geometric wear, improving the reliability of equipment health prediction and the scientific nature of maintenance decisions.

[0062] Furthermore, by simultaneously analyzing diameter deviation and roughness deviation and comparing them with preset thresholds, the intrinsic relationship between changes in the geometric features and surface quality of the processed parts can be comprehensively reflected. This allows for accurate identification of geometric wear anomalies, ensuring that anomaly determination is targeted and reliable. It also helps to take targeted maintenance measures in advance, extend equipment lifespan, and improve processing accuracy.

[0063] Furthermore, through statistical analysis of response delay and voltage values, the control state and power supply state of the equipment can be accurately distinguished, enabling effective determination of anomaly types and ensuring the pertinence and accuracy of anomaly diagnosis. The average response delay reflects the real-time response capability of the control system, while the average voltage reflects the stability of the power supply. The coordinated changes of the two reveal the essential characteristics of control performance degradation or power supply fluctuations during system operation, thereby providing a scientific basis for subsequent maintenance strategies and improving the reliability and safety of equipment operation.

[0064] Furthermore, by comparing and analyzing the mean and variance of the time intervals between the first and second degrees within the preset correction period, the temporal distribution characteristics of abnormal events can be accurately identified, and the reliability of the abnormality type can be judged based on the differences in the time series. This effectively corrects misjudgments that may be caused by short-term fluctuations or occasional interference. By utilizing the inherent relationship between the statistical characteristics of time intervals and the correlation between abnormality types, the abnormality judgment becomes more robust, improving the accuracy and reliability of health prediction and maintenance decisions.

[0065] Furthermore, by comprehensively analyzing the mean changes in spindle acceleration and temperature under different anomaly types, a weighted health prediction index is constructed. The deviation of acceleration and temperature is quantified into a numerical indicator of equipment health status, reflecting the operating characteristics and trends of equipment under different anomaly types. This can scientifically capture the coupling relationship between equipment vibration and temperature rise, and reflect the differentiated impact of different anomaly types on equipment health, thereby achieving more accurate health prediction and maintenance decisions.

[0066] Furthermore, through dynamic analysis of the health prediction index and the proportion of each anomaly type, the overall operating status of the equipment is combined with specific anomaly patterns to achieve intelligent adjustment of anomaly thresholds and precise maintenance management strategies. Changes in different anomaly types and their proportions reveal the coupling relationship between equipment vibration, temperature, current, speed, and machining accuracy, reflecting the health status of equipment operation from the overall to the local level. This allows maintenance decisions to be targeted at key components and systems, thereby effectively preventing fault propagation, extending equipment life, and improving production stability.

[0067] Furthermore, by simultaneously monitoring acceleration and temperature values ​​and combining this with the duration of the fluctuations, it is possible to effectively distinguish between transient fluctuations and genuine anomalies, thereby improving the accuracy of anomaly detection. Acceleration reflects the vibration state of mechanical components, while temperature reflects the thermal load of the equipment. The combined changes of these two values ​​can accurately reflect the occurrence of equipment malfunctions. Using duration as a criterion for judgment can avoid misjudgments caused by occasional fluctuations, thus enabling timely and reliable identification of abnormal events. This helps ensure stable equipment operation and extend its service life. Attached Figure Description

[0068] Figure 1 This is a flowchart of the production line equipment health prediction and maintenance optimization method based on the HarmonyOS system in this embodiment;

[0069] Figure 2 This is a logic diagram for determining the degree of abnormality in this embodiment;

[0070] Figure 3 This is a logic diagram for determining the anomaly type in this embodiment;

[0071] Figure 4 This is a logic diagram for determining the occurrence of abnormal events in this embodiment. Detailed Implementation

[0072] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0073] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0074] Please see Figure 1The diagram shows a flowchart of a production line equipment health prediction and maintenance optimization method based on the HarmonyOS system in this embodiment. This embodiment provides a production line equipment health prediction and maintenance optimization method based on the HarmonyOS system, including: real-time acquisition of the acceleration and temperature values ​​of the spindles, the current and rotational speed values ​​of the drive units, the diameter and surface roughness values ​​of the workpieces, and the response delay and voltage values ​​of the control components of each workstation in the production line equipment cluster using the HarmonyOS distributed bus; determining an abnormal event based on the acceleration value, the temperature value, and a preset abnormality judgment threshold to obtain an abnormality judgment result; and determining the degree of abnormality of the abnormal event based on the abnormality judgment result, according to the current value and the rotational speed value, to obtain a first degree or a second degree; based on... The first degree is used to determine the anomaly type of the abnormal event based on the diameter value and the surface roughness value to obtain a first type; based on the second degree, the anomaly type of the abnormal event is determined based on the response delay value and the voltage value to obtain a second type or a third type; the second type or the third type is corrected to the first type based on the time distribution characteristics of the first degree and the second degree within a preset correction time; a health prediction index is determined based on the acceleration value and the temperature value corresponding to the first type, the second type, and the third type within a preset prediction time; the preset anomaly judgment threshold is adjusted based on the health prediction index and all the anomaly types, or a maintenance management strategy is generated.

[0075] In this embodiment, the production line equipment cluster refers to a multi-station collaborative production system built according to the production task process under the support of the HarmonyOS distributed system. Each station equipment realizes unified collection and cross-node sharing of operating status data through the HarmonyOS distributed bus, which is used to realize the processing of precision mechanical parts. Based on the concept of HarmonyOS distributed bus, each acquisition node on the production line, including the vibration / temperature acquisition module on the spindle side, the current / speed acquisition module on the drive unit side, the workpiece measurement module, and the control component interface module, is first registered and declared under the HarmonyOS distributed registration and discovery mechanism. Each acquisition node is connected to the local node of the HarmonyOS distributed bus through common fieldbus or field interface, such as analog quantity sampled by ADC after anti-aliasing filtering, MEMS accelerometer acquired by SPI / I²C, RTD / thermometer read by ADC or digital temperature sensor, speed obtained by encoder pulse counter or frequency measurement module, current directly sampled by shunt + amplifier + ADC or Hall current sensor, and laser / contact diameter gauge and roughness gauge output through industrial Ethernet or serial port.

[0076] The HarmonyOS edge service running on the local node is responsible for: synchronously sampling the raw sensor signals according to the preset sampling strategy, including sampling rate, sampling window, and timestamp accuracy requirements; performing minimal local preprocessing, including DC component removal, anti-aliasing filtering, CRC / checking, abnormal sample removal, and segmented compression, and attaching a unified timestamp generated by the HarmonyOS distributed clock synchronization module to each sample or sample segment, using network time synchronization / distributed clock protocols to ensure cross-node timing consistency; and securely distributing the packaged raw timing data to the designated edge aggregator or cloud data bus through the HarmonyOS distributed communication channel in a lightweight message format, such as binary encapsulation or JSON / Protobuf.

[0077] The technical functions of HarmonyOS distributed bus include: unified device registration and capability description, distributed clock and timestamp synchronization, point-to-point or multicast data transmission based on service discovery, and execution of permissions and transmission encryption between nodes, such as message authentication based on secure channels / TLS. The edge aggregator performs time-series alignment, short-term buffering, fault code annotation, and writes the raw sample streams from each node to a time-series database or distributed file storage for direct reading by the upper-layer decision engine.

[0078] The response delay between control commands and execution feedback is obtained by adding a distributed timestamp to the control end when issuing commands and by sending feedback back from the execution end and recording the feedback timestamp. Voltage values ​​are collected in real time through sampling channels and reported by the distributed bus. All communication and data storage are subject to permission verification and auditing records under the HarmonyOS distributed security policy to ensure the integrity and traceability of the collected data.

[0079] The preset anomaly judgment threshold is a reference value for judging the duration of acceleration and temperature anomalies. It depends on the equipment operating characteristics and process stability, and is usually set between 10 and 120 seconds. In this embodiment, it is set to 30 seconds, which can effectively filter short-term fluctuations and ensure the accuracy of anomaly judgment. The preset correction duration is the length of the time window used to judge the anomaly type correction. It depends on the equipment failure evolution pattern and data acquisition frequency, and is usually set between 1 and 30 minutes. In this embodiment, it is set to 10 minutes, which can reasonably identify the time distribution characteristics of anomaly types and make corrections. The preset prediction duration is the time range used to calculate the health prediction index. It depends on the equipment operating cycle and maintenance plan requirements, and is usually set between 30 minutes and 24 hours. In this embodiment, it is set to 2 hours, which can balance prediction accuracy and response timeliness and support effective maintenance decisions.

[0080] By collecting and comprehensively analyzing the dynamic relationship between key operating parameters of production line equipment in real time, the system can accurately identify and classify abnormal states. Based on different abnormality types and their time distribution characteristics, it can dynamically adjust judgment thresholds or generate targeted maintenance management strategies, thereby effectively improving equipment operational stability and maintenance efficiency. Specifically, acceleration and temperature values ​​reflect the mechanical vibration and thermal state of the equipment; current and rotational speed values ​​reveal the operating load characteristics of the drive unit; diameter and surface roughness values ​​reflect machining accuracy; and response delay and voltage values ​​reflect the response performance of control components. These parameters are interrelated and, through weighted summation, form a health prediction index. This makes the prediction results closer to the actual operating state of the equipment, enabling early warning and guidance for precise maintenance, extending equipment lifespan and reducing production risks. This effectively solves the problems of delayed production line maintenance decisions and low operational efficiency caused by single monitoring indicators and static maintenance recommendations.

[0081] Specifically, the process of determining the degree of abnormality of the abnormal event based on the current value and the rotational speed value to obtain a first degree or a second degree includes: performing Z-score calculation on all the current values ​​within a preset degree determination time period to obtain a number of normalized currents; performing Z-score calculation on all the rotational speed values ​​within the preset degree determination time period to obtain a number of normalized rotational speeds; calculating the absolute value of the cosine similarity between all the normalized currents and all the normalized rotational speeds to obtain a degree determination value; and determining the degree of abnormality of the abnormal event based on the degree determination value to obtain the first degree or the second degree.

[0082] The preset time limit is a time window used to calculate the normalization of current and speed values. It depends on the operating characteristics of the equipment and the response speed to abnormal changes. It is usually set between 10 and 120 seconds. In this embodiment, it is set to 60 seconds to balance data stability and the timeliness of abnormal response.

[0083] By normalizing current and speed values ​​and calculating cosine similarity, the dynamic correlation characteristics of the drive unit during operation can be accurately reflected, effectively revealing abnormal trends caused by load changes, mechanical resistance, or power supply fluctuations, thereby enabling a scientific distinction of the degree of anomaly. This judgment method based on multi-parameter correlation helps to capture subtle changes in equipment operating status, improves the accuracy of anomaly detection and early warning capabilities, ensures stable equipment operation, and extends service life.

[0084] Please see Figure 2As shown, this is a logic diagram for determining the degree of abnormality in this embodiment. In this embodiment, the process of determining the degree of abnormality of the abnormal event based on the degree determination value to obtain the first degree or the second degree includes: when the degree determination value is less than a preset degree determination threshold, determining that a correlation deviation has occurred, so as to determine that the degree of abnormality of the abnormal event is the first degree; when the degree determination value is greater than or equal to the preset degree determination threshold, determining that a correlation has been maintained, so as to determine that the degree of abnormality of the abnormal event is the second degree.

[0085] The preset severity threshold is a reference value used to distinguish different degrees of anomaly. It depends on the operating characteristics of the equipment and the stability of historical operating data. It is usually set between 0.7 and 0.9. In this embodiment, it is set to 0.85, which can effectively distinguish between short-term fluctuation anomalies and persistent anomalies and improve the accuracy of anomaly judgment.

[0086] By performing similarity analysis on the normalized values ​​of current and rotational speed, the operating status of the equipment is quantified into a severity judgment value, thereby accurately distinguishing different degrees of anomaly. When the severity judgment value is below a preset threshold, it reflects a weakened correlation between current and rotational speed, indicating that the equipment may have a sudden or unstable anomaly, i.e., the first degree. When the severity judgment value is above the threshold, it reflects a stable correlation between current and rotational speed, suggesting that the anomaly is a continuous or systematic change, i.e., the second degree. Based on the real-time change characteristics of current and rotational speed, it can effectively capture the dynamic changes in equipment operation, thereby improving the accuracy and response efficiency of anomaly detection.

[0087] Specifically, the process of determining the anomaly type of the abnormal event based on the diameter value and the surface roughness value to obtain a first type includes: calculating the average value and variance of all diameter values ​​within a preset detection period to obtain the average diameter value and diameter variance; calculating the difference between the average diameter value and the diameter variance to obtain the diameter deviation; calculating the average value and variance of all surface roughness values ​​within the preset detection period to obtain the roughness average value and roughness variance; calculating the difference between the average roughness value and the roughness variance to obtain the roughness deviation; and determining the anomaly type of the abnormal event based on the diameter deviation and the roughness deviation to obtain the first type.

[0088] The preset detection duration is a time window used to collect diameter and surface roughness values. It depends on the equipment processing cycle and the rate of change of abnormal features. It is usually set between 10 minutes and 1 hour. In this embodiment, it is set to 30 minutes to ensure that the collected data is representative and supports accurate anomaly type determination.

[0089] By differentially calculating the average and variance of diameter and surface roughness values, it is helpful to accurately capture the minute fluctuations in dimensional changes and surface conditions during processing. This statistical characteristic-based judgment can reflect the intrinsic relationship between equipment processing accuracy and stability, thereby enabling accurate identification of abnormal types such as geometric wear, improving the reliability of equipment health prediction and the scientific nature of maintenance decisions.

[0090] Please see Figure 3 As shown, it is a logic diagram for determining the abnormality type in this embodiment. In this embodiment, the process of determining the abnormality type of the abnormal event based on the diameter deviation and the roughness deviation to obtain the first type includes: when the diameter deviation is greater than a preset diameter deviation threshold and the roughness deviation is greater than a preset roughness deviation threshold, determining the abnormality type as geometric wear to obtain the first type.

[0091] The preset diameter deviation threshold is a reference value used to determine whether the workpiece diameter deviation exceeds the normal range. It depends on the precision of the processing equipment and the process requirements, and is usually set between ±0.01 mm and ±0.05 mm. In this embodiment, it is set to 0.02 mm, which can effectively distinguish between normal processing fluctuations and abnormal geometric wear. The preset roughness deviation threshold is a reference value used to determine whether the workpiece surface roughness deviation is abnormal. It depends on the processing technology and quality standards, and is usually set between ±0.1 μm and ±0.5 μm. In this embodiment, it is set to 0.2 μm, which can accurately identify surface quality abnormalities during the processing.

[0092] In this embodiment, the anomaly type determination only applies to geometric wear. That is, when both the diameter deviation and the roughness deviation are greater than the corresponding preset thresholds, the anomaly type is determined to be geometric wear, thus obtaining the first type. For other cases that do not meet the above conditions, they are determined by default to no anomaly. By limiting the range of anomaly types, the focus can be placed on geometric wear problems that have the most significant impact on machining accuracy, simplifying the determination logic, improving computational efficiency, and ensuring accurate identification of key risk points and generation of maintenance management strategies.

[0093] By simultaneously analyzing diameter deviation and roughness deviation and comparing them with preset thresholds, the intrinsic relationship between changes in the geometric features and surface quality of the processed parts can be comprehensively reflected. This allows for accurate identification of geometric wear anomalies, ensuring that anomaly determination is targeted and reliable. It also helps to take targeted maintenance measures in advance, extend equipment lifespan, and improve processing accuracy.

[0094] Specifically, the process of determining the anomaly type of the abnormal event based on the response delay value and the voltage value to obtain a second type or a third type includes: calculating the average value of all the response delay values ​​within the preset detection time to obtain a delay average; calculating the average value of all the voltage values ​​within the preset detection time to obtain a voltage average; when the delay average is greater than a preset delay threshold and the voltage average is less than a preset voltage threshold, determining the anomaly type as a control failure to obtain the second type; and when the delay average is greater than the preset delay threshold and the voltage average is greater than or equal to the preset voltage threshold, determining the anomaly type as a power supply fluctuation type to obtain the third type.

[0095] The preset delay threshold is a time standard for judging whether the control system response is abnormal. It depends on the equipment control response requirements and process precision, and is usually set between 5 milliseconds and 50 milliseconds. In this embodiment, it is set to 20 milliseconds, which can effectively distinguish between normal response and abnormal delay state. The preset voltage threshold is an important reference value for judging power supply stability. It depends on the equipment rated voltage and power supply fluctuation tolerance, and is usually set between ±3% and ±10% of the rated voltage. In this embodiment, it is set to ±5% of the rated voltage, which can accurately identify the abnormal type caused by power supply fluctuation.

[0096] By statistically analyzing response delay and voltage values, the control state and power supply state of the equipment can be accurately distinguished, enabling effective determination of anomaly types and ensuring the pertinence and accuracy of anomaly diagnosis. The average response delay reflects the real-time response capability of the control system, while the average voltage reflects the stability of the power supply. The coordinated changes of the two reveal the essential characteristics of control performance degradation or power supply fluctuations during system operation, thus providing a scientific basis for subsequent maintenance strategies and improving the reliability and safety of equipment operation.

[0097] Specifically, the process of correcting the second type or the third type to the first type based on the time distribution characteristics of the first and second degrees within a preset correction period includes: marking all timestamps corresponding to the first and second degrees within the preset correction period to obtain a first time series and a second time series; calculating the time difference between adjacent timestamps in the first time series to obtain a first time interval series; calculating the time difference between adjacent timestamps in the second time series to obtain a second time interval series; calculating the mean and variance of the first time interval series to obtain a first interval mean and a first interval variance; calculating the mean and variance of the second time interval series to obtain a second interval mean and a second interval variance; when the second interval variance is greater than a preset interval variance threshold, the second interval mean is less than the first interval mean, and the difference between the first interval mean and the second interval mean is greater than a preset mean difference threshold, determining that the corresponding time period in the second time series is an irregularly distributed time period; and determining the second type or the third type in the irregularly distributed time period as the first type to correct it to the first type.

[0098] The preset interval variance threshold is a reference value used to judge the degree of fluctuation of the time interval. It depends on the stability of the equipment operation and the process requirements. It is usually set between 0.5 seconds and 5 seconds. In this embodiment, it is set to 2 seconds, which can effectively distinguish between normal operation and abnormal time distribution, and improve the accuracy of abnormality type correction.

[0099] By comparing and analyzing the mean and variance of the time intervals between the first and second levels within the preset correction period, the distribution characteristics of abnormal events over time can be accurately identified. Based on the differences in time series, the reliability of the abnormality type can be judged, thereby effectively correcting misjudgments that may be caused by short-term fluctuations or occasional interferences. By utilizing the inherent relationship between the statistical characteristics of time intervals and the correlation between abnormality types, the abnormality judgment is more robust, improving the accuracy and reliability of health prediction and maintenance decisions.

[0100] Specifically, the process of determining the health prediction index based on the acceleration values ​​and temperature values ​​corresponding to the first type, the second type, and the third type within the preset prediction time includes: calculating the mean of the acceleration values ​​and the mean of the temperature values ​​corresponding to all first types within the preset prediction time to obtain a first acceleration mean and a first temperature mean; calculating the mean of the acceleration values ​​and the mean of the temperature values ​​corresponding to all second types within the preset prediction time to obtain a second acceleration mean and a second temperature mean; calculating the mean of the acceleration values ​​and the mean of the temperature values ​​corresponding to all third types within the preset prediction time to obtain a third acceleration mean and a third temperature mean; and determining the health prediction index based on the first acceleration mean, the first temperature mean, the second acceleration mean, the second temperature mean, the third acceleration mean, and the third temperature mean, where P = w1 × {a1 × max[0, (A1 - A') / A'] + b1×︱B1-B'︱ / B'}+w2×{a2×max[0,(A2-A') / A']+b2×︱B2-B'︱ / B'}+w3×{a3×max[0,(A3-A') / A']+b3×︱B3-B'︱ / B'}, where P is the health prediction index, w1 is the preset first type weight, a1 is the preset first acceleration mean weight, A1 is the first acceleration mean, A' is the preset acceleration mean threshold, b1 is the preset first temperature mean weight, and B1 is the first temperature mean weight. The average temperature is B', the average temperature threshold is B', the average temperature threshold is B2, the average temperature threshold is B2, the average temperature threshold is B2, the average temperature threshold is B3 ...

[0101] The preset first-type weight is a coefficient that controls the contribution of the first type of anomaly to the health prediction index. It depends on the importance of the first-type anomaly and is usually set between 0 and 1. In this embodiment, it is set to 0.4 to highlight the impact of the first-type anomaly on the overall health status. The preset first acceleration mean weight controls the proportion of the acceleration mean in the first-type health prediction. It depends on the sensitivity of acceleration anomalies to equipment health and is usually set between 0 and 1. In this embodiment, it is set to 0.7 to enhance the recognition effect of acceleration changes. The preset acceleration mean threshold is a reference standard for judging acceleration anomalies. It depends on the equipment operating characteristics and is usually set to ±10% to ±20% of the normal operating acceleration mean. In this embodiment, the weight is set to ±15%, which can effectively distinguish between normal and abnormal states. The preset first temperature mean weight controls the proportion of the temperature mean in the first type of health prediction, depending on the degree of influence of temperature changes on the equipment, and is usually set between 0 and 1. In this embodiment, it is set to 0.3, which can assist in the judgment of acceleration anomalies. The preset temperature mean threshold is the reference standard for judging temperature anomalies, depending on the temperature control requirements of the equipment, and is usually set to ±5% to ±15% of the normal operating temperature mean. In this embodiment, it is set to ±10%, which can effectively filter out the influence of small fluctuations. The preset second type weight controls the coefficient of the contribution of the second type to the health prediction index, depending on the second type of anomaly. Importance is typically set between 0 and 1; in this embodiment, it is set to 0.35, which reasonably reflects the impact of type II anomalies on overall health. The preset second acceleration mean weight controls the proportion of the acceleration mean in type II health prediction, depending on the sensitivity to acceleration changes; it is typically set between 0 and 1; in this embodiment, it is set to 0.6, which appropriately reflects the impact of acceleration changes. The preset second temperature mean weight controls the proportion of the temperature mean in type II health prediction, depending on the impact of temperature changes on this type; it is typically set between 0 and 1; in this embodiment, it is set to 0.4, which comprehensively considers the impact of temperature on anomaly judgment. The preset third type weight controls the third type... The coefficient contributing to the health prediction index depends on the importance of the third type of anomaly and is usually set between 0 and 1. In this embodiment, it is set to 0.25 to ensure the overall balance of the prediction. The preset weight of the third acceleration mean controls the proportion of the acceleration mean in the third type of health prediction. It depends on the sensitivity of this type to acceleration changes and is usually set between 0 and 1. In this embodiment, it is set to 0.5 to reasonably reflect acceleration changes. The preset weight of the third temperature mean controls the proportion of the temperature mean in the third type of health prediction. It depends on the impact of temperature changes on this type and is usually set between 0 and 1. In this embodiment, it is set to 0.5 to balance the effects of acceleration and temperature.

[0102] By comprehensively analyzing the mean changes in spindle acceleration and temperature under different anomaly types, a weighted health prediction index is constructed. The deviation of acceleration and temperature is quantified into a numerical indicator of equipment health status, reflecting the operating characteristics and trends of equipment under different anomaly types. This index can scientifically capture the coupling relationship between equipment vibration and temperature rise, and reflect the differentiated impact of different anomaly types on equipment health, thereby achieving more accurate health prediction and maintenance decisions.

[0103] Specifically, the process of adjusting the preset anomaly determination threshold or generating a maintenance management strategy based on the health prediction index and all the anomaly types includes: when the health prediction index is greater than the preset prediction index threshold, calculating the ratio of the number of the first type to the number of all the anomaly types to obtain a first proportion; calculating the ratio of the number of the second type to the number of all the anomaly types to obtain a second proportion; calculating the ratio of the number of the third type to the number of all the anomaly types to obtain a third proportion; when the first proportion, the second proportion, and the third proportion are all greater than the preset proportion threshold, adjusting the preset anomaly determination threshold based on the first proportion, the second proportion, the third proportion, and the preset anomaly determination threshold to obtain a maintenance management strategy. The percentage threshold is reduced by the preset anomaly determination threshold, where Y' = Y × [1 - i1 × (U1 - U) / U - i2 × (U2 - U) / U - i3 × (U3 - U) / U], Y' is the reduced preset anomaly determination threshold, Y is the original preset anomaly determination threshold, i1 is the preset first adjustment coefficient, U1 is the first percentage, U is the preset percentage threshold, i2 is the preset second adjustment coefficient, U2 is the second percentage, i3 is the preset third adjustment coefficient, U3 is the third percentage; when any one or two of the first percentage, the second percentage, and the third percentage are greater than the preset percentage threshold, the maintenance management strategy is generated according to the corresponding anomaly type.

[0104] In this embodiment, when only the first proportion is greater than the preset proportion threshold, a maintenance management strategy is generated: "A high proportion of the first type of anomaly is detected. It is recommended to calibrate the accelerometer and temperature sensor, check the vibration state and temperature changes of key mechanical components, and perform bearing lubrication, structural tightening, and temperature control system maintenance to prevent equipment damage caused by increased vibration and abnormal temperature."

[0105] When only the proportion of the second type of abnormality exceeds the preset proportion threshold, a maintenance management strategy is generated: "A high proportion of the second type of abnormality has been detected. It is recommended to check the motor, current acquisition system, and speed detection device, investigate abnormal motor load or unstable power supply, perform current sampling calibration, speed sensor detection, and maintain the relevant drive system and control unit."

[0106] When only the proportion of the third type of anomaly exceeds the preset threshold, a maintenance management strategy is generated: "A high proportion of the third type of anomaly has been detected. It is recommended to check the machining accuracy of key components of the equipment, especially the component diameter and surface roughness, perform mechanical component measurement and calibration, replace worn components, and perform precision correction on the machining equipment to prevent the machining error from expanding and causing equipment failure."

[0107] When both the first and second proportions are greater than the preset proportion threshold, a maintenance management strategy is generated: "If the proportions of both the first and second types of anomalies are found to be high, it is recommended to simultaneously check the vibration and temperature status, current and speed data, and implement comprehensive maintenance, including sensor calibration, motor repair, bearing lubrication, and temperature control adjustment."

[0108] When only the proportions of the first and third types of anomalies are both greater than the preset proportion threshold, a maintenance management strategy is generated: "High proportions of the first and third types of anomalies have been detected. It is recommended to perform vibration and temperature detection on the equipment, as well as mechanical processing accuracy checks, and to perform sensor calibration, component inspection and replacement, mechanical structure tightening, and temperature control system maintenance."

[0109] When both the second and third proportions exceed the preset proportion thresholds, a maintenance management strategy is generated: "High proportions of the second and third types of anomalies have been detected. It is recommended to simultaneously check the motor operating status and the precision of the machined parts, implement current sampling calibration, speed detection, and part diameter and roughness detection, and adjust or replace the relevant parts."

[0110] The preset prediction index threshold is a benchmark value used to determine whether the health status of the equipment requires maintenance adjustments. It depends on the equipment's historical operating data and fault sensitivity, and is usually set between 0.5 and 0.9. In this embodiment, it is set to 0.75, which can effectively balance the timeliness of fault warnings and the false alarm rate. The preset proportion threshold is a benchmark value used to determine whether the proportion of each anomaly type reaches the threshold for generating a maintenance management strategy. It depends on the distribution of the equipment's anomaly patterns and maintenance strategies, and is usually set between 0.2 and 0.6. In this embodiment, it is set to 0.4, which can accurately identify the dominant influence of anomaly types and guide targeted maintenance. The preset first adjustment coefficient is a coefficient used to adjust the influence of the proportion of the first type of anomaly on the anomaly judgment threshold. It depends on the severity of the first type of anomaly and historical experience, and is usually... The first adjustment coefficient is set between 0.01 and 0.2, and in this embodiment it is set to 0.05, which can balance the sensitivity of the first type of anomaly to the threshold adjustment. The second adjustment coefficient is a coefficient used to adjust the influence of the proportion of the second type of anomaly on the anomaly judgment threshold. It depends on the influence range and frequency of the second type of anomaly, and is usually set between 0.01 and 0.2. In this embodiment it is set to 0.04, which can appropriately reflect the adjustment needs of the judgment due to the changes in the second type of anomaly. The third adjustment coefficient is a coefficient used to adjust the influence of the proportion of the third type of anomaly on the anomaly judgment threshold. It depends on the occurrence characteristics and risk level of the third type of anomaly, and is usually set between 0.01 and 0.2. In this embodiment it is set to 0.03, which can reasonably reflect the adjustment effect of the third type of anomaly on the overall judgment threshold.

[0111] By dynamically analyzing the health prediction index and the proportion of each anomaly type, the overall operating status of the equipment is combined with specific anomaly patterns to achieve intelligent adjustment of anomaly thresholds and precise generation of maintenance management strategies. Changes in different anomaly types and their proportions reveal the coupling relationship between equipment vibration, temperature, current, speed, and machining accuracy, reflecting the health status of equipment operation from the overall to the local level. This allows maintenance decisions to be targeted at key components and systems, thereby effectively preventing fault propagation, extending equipment life, and improving production stability.

[0112] Please see Figure 4 As shown, it is a logic diagram for determining the occurrence of an abnormal event in this embodiment. In this embodiment, the process of determining the occurrence of an abnormal event based on the acceleration value, the temperature value, and the preset abnormality determination threshold includes: recording the duration for which the acceleration value first exceeds the preset acceleration threshold and the temperature value first exceeds the preset temperature threshold to obtain an abnormality determination value; when the abnormality determination value exceeds the preset abnormality determination threshold, the abnormal event is determined to have occurred.

[0113] The preset acceleration threshold is a critical value used to determine abnormal equipment vibration. It depends on the equipment type, operating conditions, and historical vibration data, and is usually set between 0.5 m / s² and 5 m / s². In this embodiment, it is set to 2.0 m / s², which can effectively distinguish between normal and abnormal vibration and reduce misjudgment. The preset temperature threshold is a critical value used to determine abnormal equipment thermal conditions. It depends on the equipment material, working environment, and historical temperature characteristics, and is usually set between 50°C and 120°C. In this embodiment, it is set to 85°C, which can promptly identify potential failure risks caused by overheating and ensure safe equipment operation.

[0114] By simultaneously monitoring acceleration and temperature values ​​and combining this with the duration of the fluctuations, it is possible to effectively distinguish between transient fluctuations and genuine anomalies, thereby improving the accuracy of anomaly detection. Acceleration reflects the vibration state of mechanical components, while temperature reflects the thermal load of the equipment. The combined changes of these two values ​​can accurately reflect the occurrence of equipment malfunctions. Using the duration of the fluctuations as a criterion can avoid misjudgments caused by occasional fluctuations, thus enabling timely and reliable identification of abnormal events. This helps ensure stable equipment operation and extend its service life.

[0115] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for health prediction and maintenance optimization of production line equipment based on the HarmonyOS system, characterized in that, include: The HarmonyOS distributed bus collects real-time data on the spindle acceleration and temperature values, drive unit current and speed values, workpiece diameter and surface roughness values, and control component response delay and voltage values ​​of each workstation in the production line equipment cluster. An abnormal event is determined based on the acceleration value, the temperature value, and a preset anomaly determination threshold to obtain an anomaly determination result; Based on the anomaly determination result, the degree of anomaly of the abnormal event is determined according to the current value and the rotation speed value, so as to obtain a first degree or a second degree; By performing similarity analysis on the normalized values ​​of current and rotation speed, the operating status of the equipment is quantified into a degree judgment value, thereby accurately distinguishing different degrees of anomaly. When the degree judgment value is lower than the preset threshold, it reflects a weakening correlation between current and rotation speed, indicating that the equipment has a sudden or unstable anomaly, i.e., the first degree; when the degree judgment value is higher than the threshold, it reflects a stable correlation between current and rotation speed, indicating that the anomaly is a continuous or systematic change, i.e., the second degree. Based on the first degree, the anomaly type of the abnormal event is determined according to the diameter value and the surface roughness value to obtain a first type; Based on the second degree, the anomaly type of the abnormal event is determined according to the response delay value and the voltage value to obtain a second type or a third type; Based on the time distribution characteristics of the first and second degrees within a preset correction period, the second type or the third type is corrected to the first type; A health prediction index is determined based on the acceleration value and temperature value corresponding to the first type, the second type, and the third type within a preset prediction time. Adjust the preset anomaly determination threshold based on the health prediction index and all the anomaly types, or generate a maintenance management strategy; The process of determining the anomaly type of the abnormal event based on the diameter value and the surface roughness value to obtain the first type includes: Calculate the average and variance of all diameter values ​​within the preset detection time period to obtain the average diameter and diameter variance; Calculate the difference between the average diameter and the variance of the diameter to obtain the diameter deviation; Calculate the average value and variance of all surface roughnesses within the preset detection time period to obtain the average roughness value and roughness variance; Calculate the difference between the roughness average value and the roughness variance to obtain the roughness deviation; The anomaly type of the abnormal event is determined based on the diameter deviation and the roughness deviation to obtain the first type; The process of determining the anomaly type of the abnormal event based on the diameter deviation and the roughness deviation to obtain the first type includes: When the diameter deviation is greater than a preset diameter deviation threshold and the roughness deviation is greater than a preset roughness deviation threshold, the abnormality type is determined to be geometric wear, so as to obtain the first type; The process of determining the anomaly type of the abnormal event based on the response delay value and the voltage value, to obtain a second or third type, includes: Calculate the average value of all response delay values ​​within the preset detection time to obtain the mean delay; Calculate the average value of all voltage values ​​within the preset detection time to obtain the average voltage value; When the average delay is greater than a preset delay threshold and the average voltage is less than a preset voltage threshold, the abnormality type is determined to be a control failure, thus obtaining the second type; When the average delay is greater than the preset delay threshold and the average voltage is greater than or equal to the preset voltage threshold, the anomaly type is determined to be power supply fluctuation type, so as to obtain the third type; The process of correcting the second type or the third type to the first type based on the time distribution characteristics of the first and second degrees within a preset correction period includes: Mark all timestamps corresponding to the first and second degrees within the preset correction period to obtain the first time series and the second time series; Calculate the time difference between adjacent timestamps in the first time series to obtain the first time interval series; Calculate the time difference between adjacent timestamps in the second time series to obtain the second time interval series; Calculate the mean and variance of the first time interval sequence respectively to obtain the first interval mean and the first interval variance; Calculate the mean and variance of the second time interval sequence to obtain the second interval mean and the second interval variance; When the second interval variance is greater than a preset interval variance threshold, the second interval mean is less than the first interval mean, and the difference between the first interval mean and the second interval mean is greater than a preset mean difference threshold, the corresponding time period in the second time series is determined to be an irregularly distributed time period. The second type or the third type in the irregular distribution period is identified as the first type, and then corrected to the first type.

2. The method for health prediction and maintenance optimization of production line equipment based on HarmonyOS according to claim 1, characterized in that, The process of determining the degree of abnormality of the abnormal event based on the current value and the rotational speed value, to obtain a first degree or a second degree, includes: Z-score calculation is performed on all the current values ​​within a predetermined time period to obtain several normalized currents; Z-score calculation is performed on all the rotational speed values ​​within the predetermined time period to obtain several normalized rotational speeds; Calculate the absolute value of the cosine similarity of all the normalized currents and all the normalized rotational speeds to obtain the degree determination value; The degree of abnormality of the abnormal event is determined based on the degree determination value to obtain the first degree or the second degree.

3. The method for health prediction and maintenance optimization of production line equipment based on HarmonyOS according to claim 2, characterized in that, The process of determining the degree of abnormality of the abnormal event based on the degree determination value to obtain the first degree or the second degree includes: When the degree determination value is less than the preset degree determination threshold, a correlation deviation is determined to occur, so as to determine the degree of abnormality of the abnormal event as the first degree; When the degree determination value is greater than or equal to the preset degree determination threshold, it is determined that the association is maintained, so as to determine the degree of abnormality of the abnormal event as the second degree.

4. The method for health prediction and maintenance optimization of production line equipment based on HarmonyOS according to claim 3, characterized in that, The process of determining the health prediction index based on the acceleration value and temperature value corresponding to the first type, the second type, and the third type within the preset prediction time includes: Calculate the average of the acceleration values ​​and the average of the temperature values ​​corresponding to all the first types within the preset prediction time period to obtain the first average acceleration value and the first average temperature value; Calculate the average of the acceleration values ​​and the average of the temperature values ​​corresponding to all the second types within the preset prediction time period to obtain the second average acceleration value and the second average temperature value. Calculate the average of the acceleration values ​​and the average of the temperature values ​​corresponding to all the third types within the preset prediction time period to obtain the third acceleration average and the third temperature average. The health prediction index is determined based on the first average acceleration, the first average temperature, the second average acceleration, the second average temperature, the third average acceleration, and the third average temperature.

5. The method for health prediction and maintenance optimization of production line equipment based on HarmonyOS according to claim 4, characterized in that, The process of adjusting the preset anomaly determination threshold or generating a maintenance management strategy based on the health prediction index and all the anomaly types includes: When the health prediction index is greater than a preset prediction index threshold, the ratio of the number of the first type to the number of all the abnormal types is calculated to obtain a first proportion; Calculate the ratio of the number of the second type to the total number of all said anomaly types to obtain a second proportion; Calculate the ratio of the number of the third type to the total number of all the abnormal types to obtain the third proportion; When the first proportion, the second proportion, and the third proportion are all greater than the preset proportion threshold, the preset abnormality judgment threshold is adjusted according to the first proportion, the second proportion, the third proportion, and the preset proportion threshold. When any one or two of the first proportion, the second proportion, and the third proportion are greater than the preset proportion threshold, the maintenance management strategy is generated according to the corresponding anomaly type.

6. The method for health prediction and maintenance optimization of production line equipment based on HarmonyOS according to claim 5, characterized in that, The process of determining the occurrence of an abnormal event based on the acceleration value, the temperature value, and the preset anomaly determination threshold includes: Record the duration for which the acceleration value first exceeds a preset acceleration threshold and the temperature value first exceeds a preset temperature threshold to obtain an anomaly determination value; When the anomaly determination value is greater than the preset anomaly determination threshold, the anomaly event is determined to have occurred.

Images