Multi-parameter acquisition fault detection method, system and device based on temperature vibration sensor

Abstract

The invention relates to the technical field of data processing, in particular to a multi-parameter acquisition fault detection method, system and device based on a temperature vibration sensor, and the method comprises the steps: carrying out the digital construction of a thermal deformation characteristic and a vibration characteristic of target equipment in an operation process, and obtaining a digital twin body; performing operation health diagnosis on the digital twinborn real-time simulation task to obtain a health state evaluation result; according to abnormal judgment of the health state evaluation result, performing diagnostic bidirectional calibration on the digital twinborn body to obtain a verification report; according to the verification report fault mode and the key parameters, attenuation process mapping is carried out in the digital twin body, and a performance attenuation time sequence graph is obtained; analyzing the health decline mode of the performance decline time sequence graph to obtain a predictive health conclusion; fusing the health state evaluation result, the verification report and the diagnosis decision information of the predictive health conclusion to obtain a deep fault diagnosis report; according to the invention, the multi-parameter acquisition fault detection efficiency of the temperature vibration sensor can be improved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a multi-parameter acquisition fault detection method, system and device based on temperature and vibration sensors. Background Technology

[0002] In the development of intelligent manufacturing and digital transformation in industry, the full lifecycle operation and maintenance management of equipment has become a core link in ensuring industrial production efficiency and improving equipment reliability. The thermal deformation and vibration characteristics of equipment during operation are core physical characteristics reflecting its mechanical structural state and operational performance, directly related to the equipment's working accuracy, operational stability, and service life. To achieve accurate perception and efficient control of equipment status, synchronous temperature and vibration monitoring technology has become the mainstream method for collecting equipment operating parameters. The deployment of temperature and vibration sensors and synchronous data acquisition provide a data foundation for obtaining equipment temperature and vibration-related operating parameters.

[0003] In the process of digitally constructing the thermal deformation and vibration characteristics of equipment, a deep correlation between thermal deformation features and vibration features in the time-frequency domain was not achieved. Only single-dimensional features were independently modeled and processed, and the modeling process lacked a deep integration with the physical mechanism of equipment operation. It was impossible to construct feature-driven logic that fits the actual operating law of the equipment, resulting in insufficient accuracy in replicating the operating characteristics of the physical equipment in the constructed virtual model. Furthermore, in the process of equipment health diagnosis and performance analysis, targeted calibration of the virtual model simulation state and dynamic digital mapping of the equipment performance degradation history were not achieved. Moreover, the integration and analysis of multi-source information for equipment status diagnosis was insufficient, and a full-dimensional linkage analysis of the current operating status of the equipment, the internal causes of the fault, and the future performance evolution trend was not achieved. This resulted in limited depth and foresight in equipment fault diagnosis, making it difficult to form fault diagnosis conclusions that are real-time, accurate, and instructive. Therefore, how to improve the fault detection efficiency of multi-parameter acquisition of temperature and vibration sensors has become an urgent problem to be solved. Summary of the Invention

[0004] This disclosure provides a multi-parameter acquisition fault detection method, system, and device based on temperature and vibration sensors.

[0005] Firstly, this disclosure provides a multi-parameter acquisition fault detection method based on a temperature and vibration sensor, including: S1. Digitally construct the thermal deformation and vibration characteristics of the target equipment during operation to obtain a digital twin of the target equipment; S2. Perform operational health diagnosis on the real-time simulation task of the digital twin to obtain the health status assessment results of the real-time simulation task; S3. Based on the anomaly determination in the health status assessment results, perform diagnostic bidirectional calibration on the digital twin to obtain a verification report for the target device; S4. Based on the failure modes and key parameters in the verification report, perform attenuation history mapping in the digital twin to obtain the performance attenuation time series map of the target device. S5. Perform health degradation mode analysis on the performance degradation time series map to obtain predictive health conclusions for the target device; S6. Integrate diagnostic decision information from health status assessment results, verification reports, and predictive health conclusions to obtain a deep fault diagnosis report for the target equipment.

[0006] In a preferred embodiment, the step of digitally constructing the thermal deformation and vibration characteristics of the target equipment during operation to obtain a digital twin of the target equipment includes: The temperature and vibration sensors deployed at key monitoring locations of the target equipment are synchronously collected to obtain the operating parameters of the target equipment; Feature extraction is performed on the original signals of the operating parameters to obtain the thermal deformation and vibration characteristics of the operating parameters; By correlating and fusing the time-frequency domain correlation between thermal deformation characteristics and vibration characteristics, a heterogeneous feature substrate of the target device is obtained; Based on real-time data streams of operating parameters and physical mechanism constraints, the heterogeneous feature base is embedded into the mechanism to obtain the driving rules of the target device. Based on driving rules, a virtual mapping is performed on the heterogeneous feature base to obtain a digital twin of the target device.

[0007] In a preferred embodiment, the step of performing a health diagnosis on the real-time simulation task of the digital twin to obtain a health status assessment result of the real-time simulation task includes: Based on the real-time operation data stream of the target device, the digital twin is synchronously driven to obtain the real-time simulation task of the digital twin; Health parameters are extracted from the real-time simulation task to obtain the health state flow of the simulation task. Based on the historical health operation data of the target device, the health status stream is calibrated to obtain the health operation baseline of the health status stream. Based on the health operation baseline, the health deviation of the health status flow is calculated to obtain the health quantification index of the health status flow; Based on health metrics, a comprehensive assessment of the real-time simulation task is conducted to obtain the health status evaluation result of the real-time simulation task.

[0008] In a preferred embodiment, the formula for calculating the health metric is as follows: ; In the formula, For the first A health status flow at any moment Health metrics For the first A health status flow at any moment The actual monitoring value, For the first A health status flow at any moment The baseline value for healthy operation, For the first Historical statistical standard deviation of a health status stream For the first A preset engineering allowable deviation threshold for each health status stream.

[0009] In a preferred embodiment, the step of performing diagnostic bidirectional calibration on the digital twin based on the anomaly determination in the health status assessment results to obtain a verification report for the target device includes: The results of the health status assessment are screened for abnormalities to obtain an abnormality judgment conclusion. Based on the anomaly determination results, anomaly correlation analysis is performed on the digital twin to obtain the anomaly correlation parameters of the digital twin; Based on real-time running data stream, retrospective fitting is performed on the anomaly correlation parameters to obtain the parameter correction vector of the anomaly correlation parameters; The parameter correction vector is injected into the digital twin to obtain the simulation results of the digital twin after calibration. The calibrated simulation results are cross-validated with the real-time running data stream, and the validation results are packaged into a validation report for the target device.

[0010] In a preferred embodiment, the step of mapping the degradation history in a digital twin based on the failure modes and key parameters in the verification report to obtain the performance degradation time series map of the target device includes: By tracing the root causes of anomalies in the verification report, the failure modes of the verification report can be obtained. Core variables were extracted from the validation report to obtain the key parameters of the validation report; Based on the failure modes and key parameters, the attenuation status of the digital twin is configured to obtain the attenuation evolution scenario of the digital twin. In the decay evolution scenario, dynamic degradation simulation is performed on the digital twin to obtain the state degradation trajectory of the digital twin; The performance degradation time series map of the target device is obtained by reconstructing the characteristic trajectory of the state degradation trajectory.

[0011] In a preferred embodiment, the step of performing health degradation mode analysis on the performance degradation time series map to obtain predictive health conclusions for the target device includes: Evolution pattern identification was performed on the performance degradation time series graph to obtain the dominant degradation mode of the time series graph; Based on the dominant decay pattern, the trend elements of the performance degradation time series graph are deconstructed to obtain the trend primitives of the dominant decay pattern. Based on the historical health status of the target device, trend extrapolation is performed on the trend primitives to obtain the future health evolution trend of the target device; By assessing future health evolution trends and determining health risks, predictive health conclusions can be obtained for the target equipment.

[0012] In a preferred embodiment, diagnostic decision information fusion is performed on the health status assessment results, validation reports, and predictive health conclusions to obtain a deep fault diagnosis report for the target equipment, including: Diagnostic information is extracted from the health status assessment results to obtain the diagnostic features of the health status assessment results; By analyzing the development trends of predictive health conclusions, the risk trends of predictive health conclusions can be obtained. By fusing diagnostic features, failure modes, and risk trends from multiple sources, a comprehensive diagnostic decision for the target equipment can be obtained. The comprehensive diagnostic decision is structurally compiled to obtain a deep fault diagnosis report for the target equipment.

[0013] To address the aforementioned problems, the present invention also provides a multi-parameter acquisition fault detection system based on a temperature and vibration sensor, comprising: The twin construction module is used to digitally construct the thermal deformation and vibration characteristics of the target equipment during operation, thereby obtaining a digital twin of the target equipment; The simulation health diagnosis module is used to perform operational health diagnosis on the real-time simulation task of the digital twin and obtain the health status assessment results of the real-time simulation task. The twin bidirectional calibration module is used to perform diagnostic bidirectional calibration on the digital twin based on the anomaly determination in the health status assessment results, and obtain a verification report for the target device; The attenuation history mapping module is used to perform attenuation history mapping in the digital twin based on the failure modes and key parameters in the verification report, so as to obtain the performance attenuation time series map of the target device. The degradation mode analysis module is used to perform healthy degradation mode analysis on the performance degradation time series map to obtain predictive health conclusions for the target device. The diagnostic information fusion module is used to fuse health status assessment results, verification reports, and predictive health conclusions to obtain a deep fault diagnosis report for the target equipment.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention extracts features from the equipment operating parameters synchronously collected by temperature and vibration sensors, extracts thermal deformation and vibration features, and mines the time-frequency domain correlation between the two to form a heterogeneous feature base. Combining the real-time data stream of operating parameters with physical mechanism constraints, the heterogeneous feature base is mechanistically embedded to obtain driving rules. Based on the driving rules, a digital twin is constructed to achieve deep time-frequency domain correlation modeling of thermal deformation and vibration features. This makes the digital twin fit the actual operating law of the equipment, effectively improving the accuracy of the virtual model in replicating the characteristics of the physical equipment, and laying a solid foundation for accurate modeling of fault detection under multi-parameter temperature and vibration acquisition.

[0015] 2. This invention performs health diagnosis on a digital twin through real-time simulation tasks, performs diagnostic bidirectional calibration based on anomaly determination results and generates a verification report, combines the fault modes and key parameters in the verification report to complete the dynamic digital mapping of the performance degradation process, analyzes the performance degradation time series spectrum to obtain predictive health conclusions, integrates multi-source diagnostic information to form a deep fault diagnosis report, realizes virtual model calibration and performance degradation mapping, and completes a full-dimensional linkage analysis of the current status of the equipment, fault causes and future trends, breaking through the limitations of traditional fault diagnosis and improving the real-time performance, accuracy and guidance value of fault detection under the temperature and vibration multi-parameter acquisition mode. Attached Figure Description

[0016] The present disclosure will be described in more detail below based on embodiments and with reference to the accompanying drawings: Figure 1 The flowchart of the multi-parameter acquisition fault detection method based on temperature and vibration sensor according to Embodiment 1 of the present invention is shown. Figure 2 This diagram shows the functional block diagram of the multi-parameter acquisition fault detection system based on a temperature and vibration sensor according to Embodiment 2 of the present invention. Figure 3 The diagram shows the structural composition of the device for implementing the multi-parameter acquisition fault detection method based on a temperature and vibration sensor according to Embodiment 3 of the present invention. Detailed Implementation

[0017] To enable those skilled in the art to better understand the technical solutions of this disclosure, and to fully understand and implement the process of how this disclosure applies technical means to solve technical problems and achieve corresponding technical effects, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, not all embodiments. The embodiments of this disclosure and the various features within them can be combined with each other without conflict, and the resulting technical solutions are all within the protection scope of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort should fall within the protection scope of this disclosure.

[0018] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0019] Example 1 Figure 1 This is a flowchart illustrating a multi-parameter acquisition fault detection method based on a temperature and vibration sensor provided in an embodiment of this disclosure. Figure 1 As shown, the multi-parameter acquisition fault detection method based on temperature and vibration sensors includes: S1. Digitally construct the thermal deformation and vibration characteristics of the target equipment during operation to obtain a digital twin of the target equipment; In this embodiment of the invention, the step of digitally constructing the thermal deformation and vibration characteristics of the target device during operation to obtain a digital twin of the target device includes: The temperature and vibration sensors deployed at key monitoring locations of the target equipment are synchronously collected to obtain the operating parameters of the target equipment; Feature extraction is performed on the original signals of the operating parameters to obtain the thermal deformation and vibration characteristics of the operating parameters; By correlating and fusing the time-frequency domain correlation between thermal deformation characteristics and vibration characteristics, a heterogeneous feature substrate of the target device is obtained; Based on real-time data streams of operating parameters and physical mechanism constraints, the heterogeneous feature base is embedded into the mechanism to obtain the driving rules of the target device. Based on driving rules, a virtual mapping is performed on the heterogeneous feature base to obtain a digital twin of the target device.

[0020] Key monitoring locations are those where thermal deformation and vibration are most significant during the operation of the target equipment, and which have the greatest impact on equipment performance. Examples include the connection points of core transmission components and high-temperature operating areas. Temperature and vibration sensors are devices capable of simultaneously collecting temperature change data and vibration amplitude data. Before deployment, it is necessary to ensure that the sensors are closely fitted to the monitoring locations of the target equipment and that the sensor's acquisition frequency matches the operating frequency of the target equipment to avoid data omissions. Synchronous acquisition means that all temperature and vibration sensors deployed at key monitoring locations begin collecting data according to a unified time reference, ensuring that the temperature and vibration data at different locations correspond one-to-one in the time dimension. The collected content includes the real-time temperature values ​​and vibration amplitude changes at each monitoring location. This collection of raw data constitutes the operating parameters of the target equipment.

[0021] The raw signals of the operating parameters are the unprocessed raw temperature and vibration data collected by temperature and vibration sensors. These data contain various interference information during equipment operation. The feature extraction process first filters the raw signals to remove irrelevant information such as power supply interference and environmental noise. Then, for the temperature data, information reflecting the degree of thermal deformation of the equipment, such as the highest value, lowest value, rate of change, and stable time period, is extracted. For the vibration data, information reflecting the vibration state of the equipment, such as the vibration frequency, peak amplitude, vibration period, and vibration energy distribution, is extracted. This set of information related to thermal deformation extracted from the temperature data is the thermal deformation feature. The set of information related to vibration extracted from the vibration data is the vibration feature.

[0022] The time-frequency domain correlation refers to the mutual influence and correspondence between thermal deformation features and vibration features in the time and frequency dimensions. For example, when the equipment temperature rises to a certain value, does the vibration frequency change in a specific way? Or, when the vibration is in a certain frequency range, does the rate of temperature change show a regular pattern? The feature correlation fusion process first sorts out the performance of each piece of information in the thermal deformation features at different time points and frequency ranges. Then, it finds the relevant information in the vibration features at the same time point and frequency range to determine the correspondence between the two. Finally, it integrates these correlated thermal deformation feature information and vibration feature information, removes duplicate and irrelevant information, and forms a unified feature set containing thermal deformation and vibration correlation information. This set is the heterogeneous feature base.

[0023] The real-time data stream of operating parameters consists of continuously collected and updated raw temperature and vibration data from temperature and vibration sensors. The physical mechanism constraints are the physical laws governing the operation of the target equipment, such as the laws of thermal expansion and contraction of materials, the mechanical balance principle of mechanical transmission, and the physical characteristics of vibration propagation. These laws have been verified through long-term practice and theory and can clearly define the intrinsic connection between thermal deformation and vibration of the equipment. The mechanism embedding process combines the latest temperature and vibration changes in the real-time data stream with the physical mechanism constraints to analyze whether the information in the heterogeneous feature substrate conforms to the physical laws. For feature information that conforms to the laws, it clarifies the role it plays in the equipment operation process and the causal relationship between different feature information based on the physical mechanism. Then, these causal relationships and mechanisms are organized into a set of guiding rules. These rules can clearly define the changes in the equipment operating state corresponding to different feature combinations, which are the driving rules of the target equipment.

[0024] Virtual mapping is the process of constructing a virtual model corresponding to a physical device in a computer virtual environment according to the actual structure, size, and operating logic of the target device. First, a basic device model framework is built in the virtual environment based on the actual hardware structure and component connection method of the target device. Then, the various feature information in the heterogeneous feature substrate is assigned to the various components of the virtual model according to the relationship defined by the driving rules. This enables the virtual model to simulate the operating state of the target device under different combinations of thermal deformation and vibration characteristics. For example, when a certain temperature change and vibration combination occurs in the heterogeneous feature substrate, the virtual model can present the corresponding degree of thermal deformation and vibration performance according to the driving rules. This virtual model that can accurately simulate the thermal deformation and vibration characteristics of the target device is the digital twin of the target device.

[0025] The beneficial effects include ensuring comprehensive, accurate, and consistent operating parameters, providing reliable raw data support for subsequent processes, improving the quality of basic data, eliminating irrelevant and interfering information, clarifying the core characteristics of thermal deformation and vibration, enhancing the effectiveness and relevance of characteristic information, integrating key related information on equipment operation, avoiding the limitations of single-feature analysis, providing a comprehensive feature basis for the formulation of driving rules, ensuring that driving rules conform to the objective laws of equipment operation, accurately reflecting the intrinsic relationship between characteristics and operating status, improving the accuracy of digital twin simulation, constructing a virtual model that accurately replicates equipment characteristics, providing a reliable simulation carrier for subsequent health diagnosis and fault detection, and ensuring the smooth implementation of the detection process and the accuracy of results.

[0026] S2. Perform operational health diagnosis on the real-time simulation task of the digital twin to obtain the health status assessment results of the real-time simulation task; In this embodiment of the invention, the step of performing a health diagnosis on the real-time simulation task of the digital twin to obtain a health status assessment result of the real-time simulation task includes: Based on the real-time operation data stream of the target device, the digital twin is synchronously driven to obtain the real-time simulation task of the digital twin; Health parameters are extracted from the real-time simulation task to obtain the health state flow of the simulation task. Based on the historical health operation data of the target device, the health status stream is calibrated to obtain the health operation baseline of the health status stream. Based on the health operation baseline, the health deviation of the health status flow is calculated to obtain the health quantification index of the health status flow; Based on health metrics, a comprehensive assessment of the real-time simulation task is conducted to obtain the health status evaluation result of the real-time simulation task.

[0027] The formulas for calculating the health metrics are as follows: ; In the formula, For the first A health status flow at any moment Health metrics For the first A health status flow at any moment The actual monitoring value, For the first A health status flow at any moment The baseline value for healthy operation, For the first Historical statistical standard deviation of a health status stream For the first A preset engineering allowable deviation threshold for each health status stream.

[0028] The real-time operating data stream of the target device refers to the data reflecting the operating status of the device continuously collected by temperature and vibration sensors deployed at key monitoring locations of the device. This includes information such as temperature changes and vibration amplitude that can present the actual operating conditions of the device in real time. These data streams are continuously input into the pre-built digital twin according to the time sequence of the device's operation. The digital twin simulates the current operating behavior of the target device based on this real-time data, restores the operating scenario of the device in the real environment, and finally forms a simulation operation that is completely synchronized with the real-time operating status of the target device. This simulation operation is the real-time simulation task of the digital twin.

[0029] The real-time simulation task accurately replicates the real-time operating status of the target equipment. Health parameters are key indicators that directly reflect the health status of the equipment, such as the stable temperature range during equipment operation, the normal fluctuation range of vibration frequency, and the reasonableness of thermal deformation. All operating status data contained in the real-time simulation task are screened and extracted one by one, and only the indicator data directly related to the health status of the equipment are retained. Then, these extracted health parameters are arranged in chronological order to form a continuous data sequence that can dynamically reflect the changes in the health status of the equipment. This data sequence is the health status stream of the simulation task.

[0030] The historical health operation data of the target equipment consists of all health-related data collected and stored by temperature and vibration sensors during the equipment's normal operation in the past. This data is a true record of the equipment under fault-free and stable performance conditions, and has clear reference value. First, statistical analysis is performed on this historical health operation data to determine the standard range of each health parameter during normal operation, including key data such as maximum, minimum, and average values. Then, these standard ranges are matched with the corresponding parameters in the current health status stream to set a fixed standard reference value for each health parameter in the health status stream that meets the requirements for normal equipment operation. The baseline data sequence formed by arranging all standard reference values ​​in chronological order is the health operation baseline of the health status stream.

[0031] The health operation baseline defines the standard reference values ​​for each health parameter at each time point. The health status stream records the actual data of each health parameter at the current time point. For each time point, the actual data of each health parameter in the health status stream is compared with the standard reference value of the corresponding node in the health operation baseline to determine the specific deviation of the actual data from the standard reference value. Then, combined with the importance of the health parameter in the overall health assessment of the equipment, the deviation is adjusted accordingly. Finally, the adjusted deviation is converted into a specific value that can intuitively reflect the health status of the equipment at that time point. This value is the health metric of the health status stream at that time point.

[0032] First, based on the equipment's design requirements, operating specifications, and historical fault data, clear health assessment criteria are established. Health metrics are divided into different health levels, such as fully healthy, slightly abnormal, moderately abnormal, and severely abnormal. Each level has a clearly defined numerical range. The health metrics at each time point are analyzed one by one to determine the health level of the equipment at that point. Then, the health levels at all time points are comprehensively reviewed to analyze the changing trends of health levels and determine whether the abnormal level continues to escalate or whether there are concentrated abnormalities in specific time periods. Finally, combined with the equipment's operating scenario and key operating links, a comprehensive judgment is made on the overall health status of the equipment, forming a conclusion that includes information such as the equipment's current health level, the time point when the abnormality occurred, and the trend of abnormality changes. This conclusion is the health status assessment result of the real-time simulation task.

[0033] In the calculation formula of the aforementioned health measurement index, For the first A health status flow at any moment The health metric is used to intuitively quantify the health status of a given health state at a given moment; the value directly reflects the quality of health. For the first A health status flow at any moment The actual monitored values ​​are derived from the health parameter extraction results of the real-time simulation task of the digital twin, representing the health state flow at time [time value missing]. Real-world operational data is one of the core foundational data for calculating the degree of health deviation. For the first A health status flow at any moment The baseline value for healthy operation is derived from the historical healthy operation data of the target device and used as a benchmark for the health status stream. It represents the time when the health status stream is operating normally. The standard reference value provides a clear benchmark for determining whether the actual monitored value deviates from the standard. For the first The historical statistical standard deviation of a health status stream is derived from the statistical analysis of the corresponding health status stream data in the historical health operation data of the target device. It reflects the natural fluctuation range of the data during its historical normal operation and is an important basis for measuring whether the actual deviation is reasonable, thus avoiding the misjudgment of normal fluctuations as abnormalities. For the first The preset engineering allowable deviation threshold for each health status flow is set according to the target equipment design technical standards, actual operating requirements and industry engineering specifications. It clarifies the maximum acceptable deviation of the health parameter and is the key standard for distinguishing between normal and abnormal deviations.

[0034] The overall calculation logic is: first take and The absolute value of the difference is used to obtain the absolute degree of deviation without directional interference; then... and After summing, multiply by 0.5 to obtain a comprehensive reference threshold that integrates historical fluctuations and engineering allowable ranges; divide the absolute deviation by this threshold to obtain a proportion reflecting the relative severity of the actual deviation; square this proportion to amplify the impact weight of larger deviations; take the negative value of the squared result and combine it with an exponential function to achieve non-linear adjustment, making the relative deviation smaller. The closer to 1, the better the health condition; the greater the deviation, the better. The closer the value is to 0, the worse the health condition. Ultimately, the complex health status data will be transformed into intuitive and comparable quantitative indicators, providing accurate and reliable data support for comprehensive judgment in real-time simulation tasks, and ensuring the scientific nature and accuracy of health status assessment results.

[0035] The beneficial effects include: accurately replicating the operating status, strengthening the diagnostic foundation, improving diagnostic accuracy, extracting health parameters, eliminating redundancy, focusing on the core, improving the pertinence and efficiency of diagnosis, using historical data to calibrate the baseline, providing a fitting reference, avoiding deviation, improving the accuracy of assessment, quantifying health deviation, intuitively presenting the health status, objectively judging and reducing errors, improving the scientific nature of assessment, comprehensively judging health trends, grasping the health and abnormalities of equipment, forming reliable assessment results, and improving the efficiency of fault detection.

[0036] S3. Based on the anomaly determination in the health status assessment results, perform diagnostic bidirectional calibration on the digital twin to obtain a verification report for the target device; In this embodiment of the invention, the step of performing diagnostic bidirectional calibration on the digital twin based on the anomaly determination in the health status assessment results to obtain a verification report for the target device includes: The results of the health status assessment are screened for abnormalities to obtain an abnormality judgment conclusion. Based on the anomaly determination results, anomaly correlation analysis is performed on the digital twin to obtain the anomaly correlation parameters of the digital twin; Based on real-time running data stream, retrospective fitting is performed on the anomaly correlation parameters to obtain the parameter correction vector of the anomaly correlation parameters; The parameter correction vector is injected into the digital twin to obtain the simulation results of the digital twin after calibration. The calibrated simulation results are cross-validated with the real-time running data stream, and the validation results are packaged into a validation report for the target device.

[0037] When screening for abnormalities in the health status assessment results, the normal range of the health status is first defined. This range is derived from the historical health operation data of the target equipment and the preset health standards. The historical health operation data consists of various operation records accumulated by the target equipment during its previous normal operation. The preset health standards are determined by combining the equipment's design parameters and actual operating requirements. Then, each data point in the health status assessment results is checked one by one. The health status assessment results are the results containing health-related information obtained after performing operational health diagnosis on the real-time simulation task of the digital twin. It is determined whether each data point is within the normal range. If all data points are within the normal range, it is determined that there is no abnormality. If any data point exceeds the normal range, it is determined that there is an abnormality. The abnormality determination conclusion is a detailed record of the specific item with the abnormality and the specific circumstances under which the data of that item deviates from the normal range.

[0038] When performing anomaly correlation analysis on a digital twin based on anomaly determination conclusions, the anomaly determination conclusions have clearly identified specific items with anomalies. The digital twin is a digital mapping of the thermal deformation and vibration characteristics of the target equipment, including various parameters related to equipment operation and the relationships between these parameters. First, the basic parameters corresponding to the anomaly items in the anomaly determination conclusions are extracted. Then, all parameters in the digital twin that have a direct or indirect influence on these basic parameters are identified. A direct influence means that a change in one parameter directly leads to a change in another parameter. An indirect influence means that one parameter affects another parameter by influencing an intermediate parameter. For example, if the basic parameter is vibration frequency-related data, then the operating parameters of the components affecting the vibration frequency, power transmission parameters, etc., are identified. Next, the data change trajectories of these parameters during the digital twin simulation process are tracked. The data change trajectory is a record of the numerical changes of the parameters as the simulation progresses. It is then confirmed which parameter changes have synchronous fluctuations or causal relationships with the abnormal performance of the anomaly items. Synchronous fluctuations mean that the parameter change time is consistent with the time of the anomaly occurrence and the change trend is consistent. Causal relationships mean that the parameter change directly leads to the occurrence of the anomaly. These related parameters are the anomaly-related parameters.

[0039] When performing retrospective fitting of anomaly-related parameters based on real-time operational data streams, the real-time operational data streams are continuous operational data synchronously collected by temperature and vibration sensors during the operation of the target equipment. These data include complete records before and after the occurrence of an anomaly. First, all data of the anomaly-related parameters during the normal operation phase before the anomaly occurred, as well as relevant data after the anomaly, are extracted from the real-time operational data streams. The normal operation phase refers to the operational phase where the target equipment does not exhibit anomalies and all indicators meet the requirements. Then, using the normal data before the anomaly as a benchmark, the actual data of the anomaly-related parameters after the anomaly occurs are compared with the benchmark data time-by-time. Time-by-time comparison means comparing the actual data and benchmark data at each time point in chronological order to find the specific difference between each anomaly-related parameter and the benchmark data at different times. Then, based on the changing patterns of these differences (the increasing or decreasing trend and magnitude of the differences over time), an adjustment value sequence that can bring the anomaly-related parameters back to the normal range is determined. This adjustment value sequence is the parameter correction vector.

[0040] When injecting the parameter correction vector into the digital twin, the parameter correction vector is a sequence of adjustment values ​​for the abnormally associated parameters. The operation of the digital twin depends on the set values ​​of various internal parameters. According to the arrangement order of the abnormally associated parameters in the digital twin, which refers to the fixed order of parameter storage and retrieval in the digital twin, the corresponding adjustment values ​​in the parameter correction vector replace the current values ​​of the abnormally associated parameters in the digital twin one by one. After the replacement is completed, the simulation of the digital twin is started. The simulation process strictly follows the physical mechanism constraints and driving rules of the target device. The physical mechanism constraints refer to the physical laws followed by the target device during operation, and the driving rules are the simulation operation rules determined based on the device operating parameters and physical mechanisms. All the data output after the simulation is completed is the simulation result after calibration. These data cover the simulation situation of the digital twin in various operating indicators after calibration.

[0041] When cross-validating the calibrated simulation results with the real-time operational data stream, the calibrated simulation results are the simulation data after calibration of the digital twin, while the real-time operational data stream is the actual operational data of the target device. First, the two are mapped one-to-one according to the same time nodes and operational indicators. The time nodes refer to the specific moments in the device's operation process, and the operational indicators refer to various characteristic data reflecting the device's operating status. Then, the indicator data at each corresponding node are compared to determine whether the data in the calibrated simulation results are consistent with the corresponding data in the real-time operational data stream. The standard for consistency is that the difference between the two is within a preset reasonable error range. This reasonable error range is determined based on the operational accuracy requirements of the target device and the acquisition accuracy of the temperature and vibration sensors. Afterward, the comparison results of all indicators are recorded, including consistent items and items that still have differences, as well as the specific circumstances of the differences. These comparison results, the adjustment of abnormal correlation parameters, and the changes in simulation results before and after calibration are organized into a structured document, which is the verification report of the target device.

[0042] The beneficial effects include: accurately locating abnormal items, ensuring accurate anomaly judgments, providing clear targets for subsequent calibration, improving the pertinence of fault detection, comprehensively identifying anomaly-related parameters, avoiding omission of key influencing factors, providing complete parameter basis for correction, ensuring the comprehensiveness of calibration work, accurately obtaining parameter deviations, obtaining precise correction vectors, ensuring reasonable correction direction and magnitude, improving the accuracy of digital twin calibration, quickly obtaining post-calibration simulation results, realistically reflecting the post-calibration operating status of the digital twin, providing reliable data support, improving calibration effectiveness, comprehensively verifying simulation accuracy, clearly presenting calibration effects and anomaly handling, providing detailed diagnostic basis, and improving the reliability of fault detection.

[0043] S4. Based on the failure modes and key parameters in the verification report, perform attenuation history mapping in the digital twin to obtain the performance attenuation time series map of the target device. In this embodiment of the invention, the step of mapping the degradation history in a digital twin based on the fault modes and key parameters in the verification report to obtain the performance degradation time series map of the target device includes: By tracing the root causes of anomalies in the verification report, the failure modes of the verification report can be obtained. Core variables were extracted from the validation report to obtain the key parameters of the validation report; Based on the failure modes and key parameters, the attenuation status of the digital twin is configured to obtain the attenuation evolution scenario of the digital twin. In the decay evolution scenario, dynamic degradation simulation is performed on the digital twin to obtain the state degradation trajectory of the digital twin; The performance degradation time series map of the target device is obtained by reconstructing the characteristic trajectory of the state degradation trajectory.

[0044] The verification report is a structured document containing information such as the adjustment of abnormal project parameters and the changes in simulation results before and after calibration. To trace the root cause of the anomaly, all anomaly-related records must first be extracted from the verification report, including the time when the anomaly occurred, the changes in operating parameters before and after calibration, and the differences in data. Then, the relationships between these anomaly records are sorted out, the order and mutual influence of different anomaly records are clarified, and the root cause of the anomaly is analyzed and determined by combining the physical operating mechanism of the target equipment, such as the transmission relationship and heat transfer law between components. These root causes are classified according to the same attributes to form a clear fault type description. This clear fault type description is the fault mode.

[0045] The extraction of core variables requires first clarifying the core functional requirements for the normal operation of the target equipment. These core functional requirements are the main functions determined during the equipment design. Then, parameters directly related to these core functions are selected from the verification report. Changes in these parameters will directly affect the realization of the core functions. Next, the magnitude and scope of change of these parameters during the occurrence of anomalies are analyzed. The magnitude of change refers to the degree to which the parameter deviates from the normal range, and the scope of impact refers to the number of other related parameters that will change due to the change of this parameter. The parameters with the largest magnitude of change and the widest scope of impact are selected; these parameters are the key parameters.

[0046] A digital twin is a digital mapping of the thermal deformation and vibration characteristics of a target device. The configuration of the attenuation state requires first clarifying the direction of the fault's impact on the device's performance based on the fault mode. This impact direction refers to situations such as gradual performance decline or sudden failure. Then, combined with the characteristics of key parameters, the attenuation initiation conditions for each key parameter are determined. The attenuation initiation conditions refer to the specific operating state in which the key parameter begins to attenuate. These attenuation initiation conditions and the corresponding impact laws of the fault modes are then integrated into the operating rules of the digital twin. Simultaneously, the attenuation rate of the key parameters is set. The attenuation rate refers to the degree to which the parameter deviates from the normal range per unit time, and the setting of the attenuation rate must conform to the physical operating laws of the device. The operating scenario simulated by the digital twin after this configuration is the attenuation evolution scenario.

[0047] The degradation evolution scenario has clearly defined the degradation initiation conditions, degradation rate, and influence law of the digital twin. Dynamic degradation simulation needs to continuously drive the digital twin to run according to the actual running sequence of the target device. During the operation, the values ​​of key parameters are adjusted in real time according to the rules set by the degradation evolution scenario, so that the key parameters gradually deviate from the normal range according to the set degradation rate. At the same time, the complete data of all parameters related to the health status of the device in the digital twin are continuously recorded over time. These continuous data sequences arranged in chronological order are the state degradation trajectory.

[0048] The state degradation trajectory is a continuous sequence of all parameters related to the health status of the equipment changing over time. Feature trajectory reconstruction first requires extracting the key feature values ​​of each parameter at different time points from the state degradation trajectory. Key feature values ​​refer to values ​​that can reflect the core trend of parameter changes, including the values ​​corresponding to the inflection points of changes in maximum and minimum values. Then, based on these key feature values, the change trend of each parameter is sorted out. The change trend refers to the continuous rise, continuous fall, or fluctuation of the parameter. Then, the change trends of all parameters are integrated according to the time dimension, and duplicate and redundant information is eliminated. The core data that can comprehensively reflect the performance degradation of the equipment is retained. These core data are presented in a structured way with time as the horizontal axis and parameter feature values ​​as the vertical axis. This structured chart is the performance degradation time series map of the target equipment.

[0049] The beneficial effects include: identifying the root cause of anomalies, forming clear fault modes, providing accurate basis for configuring degradation trends, ensuring that evolution scenarios are realistic, improving the accuracy of degradation history mapping, focusing on key performance parameters, eliminating irrelevant interference, improving the relevance of degradation simulation, increasing the core information density of time series graphs, focusing on key changes in performance degradation, integrating fault modes and key parameter degradation characteristics, constructing realistic degradation evolution scenarios, providing a scientific environment for dynamic degradation simulation, ensuring the authenticity and reliability of degradation trajectories, driving digital twins and adjusting key parameters according to time sequence, fully recording changes in health parameters, forming continuous and comprehensive degradation trajectories, providing reliable data support for time series graph construction, extracting core feature values ​​and integrating parameter change trends, eliminating redundancy, forming structured time series graphs, clearly presenting the performance degradation process, providing accurate and easily analyzable data carriers for subsequent analysis, and improving analysis efficiency and accuracy.

[0050] S5. Perform health degradation mode analysis on the performance degradation time series map to obtain predictive health conclusions for the target device; In this embodiment of the invention, the step of performing health degradation mode analysis on the performance degradation time series map to obtain predictive health conclusions for the target device includes: Evolution pattern identification was performed on the performance degradation time series graph to obtain the dominant degradation mode of the time series graph; Based on the dominant decay pattern, the trend elements of the performance degradation time series graph are deconstructed to obtain the trend primitives of the dominant decay pattern. Based on the historical health status of the target device, trend extrapolation is performed on the trend primitives to obtain the future health evolution trend of the target device; By assessing future health evolution trends and determining health risks, predictive health conclusions can be obtained for the target equipment.

[0051] The performance degradation time series graph is a structured chart that presents the core data of equipment performance degradation. The horizontal axis represents time, and the vertical axis represents parameter characteristic values. To identify the evolution pattern, it is necessary to first identify the common types of performance degradation, including continuous linear degradation, staged jump degradation, and slow increase followed by rapid degradation. Next, the complete trajectory of all parameter characteristic values ​​changing over time is extracted from the graph. The magnitude and direction of change of parameter characteristic values ​​in each time period are analyzed segment by segment in chronological order. The magnitude of change is obtained by calculating the difference between characteristic values ​​at adjacent time nodes, and the direction of change is divided into rising, falling, or stable. Then, the trajectory of each parameter is compared with the preset common types one by one. The number of parameters corresponding to each type and the degree of fit between each parameter and the type are counted. The degree of fit is determined by the trend of the trajectory, the range of change, and the matching of the type. Finally, the type with the highest degree of fit and the largest number of corresponding parameters is selected as the dominant degradation mode. The dominant degradation mode is the mode type that can reflect the core trend of the overall performance degradation of the equipment.

[0052] The dominant degradation pattern clearly defines the core trend of overall equipment degradation. The performance degradation time series map contains data on the changes of various parameters over time. The deconstruction of trend elements requires first identifying the time points when the magnitude of parameter characteristic value changes suddenly changes or the direction of change reverses, based on the characteristics of the dominant degradation pattern, to determine the key change nodes in this pattern, such as the starting degradation point in linear degradation and the boundary points of each stage in staged degradation. Then, using these key nodes as boundaries, the change trajectory of each parameter in the performance degradation time series map is divided into multiple continuous segments. The change trend of parameter characteristic values ​​within each segment is completely consistent with the characteristics of the dominant degradation pattern at that stage. Finally, features are extracted from each segment, including the time span of the segment, the initial value, the final value, and the average rate of change of the parameter characteristic values ​​within that time period. The average rate of change is obtained by dividing the total change of characteristic values ​​within the segment by the time span. These independent segments containing the time span, initial value, final value, and average rate of change are the trend primitives, and each trend primitive corresponds to a continuous degradation stage in the dominant degradation pattern.

[0053] The historical health status of the target equipment consists of health-related data recorded during its past normal operation and degradation processes. This data includes trend primitives at different degradation stages and their corresponding subsequent degradation. Trend extrapolation requires first filtering historical trend primitives from the historical health status data that differ from the current trend primitive in terms of time span, initial value, average rate of change, and other characteristics within a preset reasonable range. Then, it involves extracting the subsequent degradation trajectory data of these similar historical trend primitives, including the magnitude and direction of change in each subsequent time period, as well as the time interval between key nodes. Finally, based on these subsequent historical degradation trajectory data, combined with the termination value and time node of the current trend primitive, the parameter characteristic values ​​corresponding to each subsequent time node are calculated sequentially. These calculated characteristic values ​​are then arranged in chronological order to form a continuous change trajectory. This trajectory represents the future health evolution trend of the target equipment, reflecting the subsequent performance changes of the equipment under the current degradation state.

[0054] The future health evolution trend is a projected trajectory of subsequent equipment performance changes. The pre-set health risk level standard for the target equipment is formulated based on the equipment's design and operation requirements, safe operation thresholds, and performance failure thresholds. It clearly defines low-risk, medium-risk, high-risk, and failure risk levels corresponding to different parameter characteristic value ranges. Each level has a clearly defined characteristic value range. Health risk assessment requires first analyzing the range of each parameter characteristic value in the future health evolution trend at each time point, and determining the risk level corresponding to each time point according to the health risk level standard. Then, it is necessary to statistically analyze the duration percentage of each risk level in the entire future evolution cycle, as well as the specific time points when high-risk and failure risks occur. Finally, combined with the key performance indicators that the equipment must meet, i.e., core requirements, it is determined whether there are situations in the future health evolution trend where parameter characteristic values ​​exceed safe operation thresholds or reach performance failure thresholds, and whether these situations occur within the equipment's planned maintenance cycle. Based on the above analysis results, the potential risk levels, durations of each risk, occurrence times of high risks, existence of failure risks, and related warning times of the equipment are compiled to form a clear predictive health conclusion.

[0055] The beneficial effects include: locking onto the core trend pattern of performance degradation, avoiding interference from single parameters, ensuring a reflection of the overall degradation state, providing direction for trend deconstruction, improving the pertinence and accuracy of health degradation analysis, breaking down complex degradation trajectories into structured basic units, clearly presenting the degradation characteristics of each stage, eliminating analytical complexity, providing operable objects for trend extrapolation, improving prediction accuracy, establishing trend continuity based on historical health data, ensuring that future health evolution trend predictions conform to equipment degradation laws, avoiding subjective bias, providing accurate data support for risk assessment, clarifying the specific situation of future health risks, providing clear and implementable references for maintenance decisions, providing early warning of failure risks, avoiding operational interruptions, and improving equipment operation safety and stability.

[0056] S6. Integrate diagnostic decision information from health status assessment results, verification reports, and predictive health conclusions to obtain a deep fault diagnosis report for the target equipment.

[0057] In this embodiment of the invention, diagnostic decision information is fused from the health status assessment results, verification report, and predictive health conclusions to obtain a deep fault diagnosis report for the target device, including: Diagnostic information is extracted from the health status assessment results to obtain the diagnostic features of the health status assessment results; By analyzing the development trends of predictive health conclusions, the risk trends of predictive health conclusions can be obtained. By fusing diagnostic features, failure modes, and risk trends from multiple sources, a comprehensive diagnostic decision for the target equipment can be obtained. The comprehensive diagnostic decision is structurally compiled to obtain a deep fault diagnosis report for the target equipment.

[0058] When extracting diagnostic information from the health status assessment results, the health status assessment results are conclusions formed after conducting operational health diagnoses on the real-time simulation task of the digital twin. These results include specific content such as the current health level of the equipment, the time point of the anomaly occurrence, and the trend of anomaly changes. First, the core information dimensions directly related to equipment failure in the results are identified. These dimensions need to be developed around whether the equipment is operating normally and the specific manifestations of the anomaly. Then, information that meets these dimensions is selected from the health status assessment results, and redundant content that is irrelevant to the fault diagnosis is eliminated. Next, the selected information is classified and organized. For example, anomaly-related information is arranged in chronological order of occurrence, and health level-related information is classified according to the judgment criteria. The final diagnostic features are a set of information that can accurately reflect the core attributes of the current health anomaly of the equipment, covering the anomaly type, anomaly degree, and key operational scenarios in which the anomaly occurred.

[0059] When analyzing the development trend of predictive health conclusions, these conclusions are obtained through health degradation mode analysis of the performance degradation time series map. They include the future health evolution trend of the target equipment and the health risk assessment results. First, all descriptions of future health status changes in the conclusions are thoroughly reviewed to clarify whether the future health status will continue to deteriorate, slowly recover, or remain at the current level. Then, the specific manifestations corresponding to each change are analyzed. For example, if it is continuous deterioration, it is necessary to identify which health-related manifestations will gradually worsen. Next, the temporal patterns of these changes are determined, including the starting point of the change, the rate of change, and the degree of change corresponding to different time periods. Integrating these changes and temporal patterns forms a risk trend that is a coherent trajectory description that clearly shows the future health risk development direction, change rhythm, and specific manifestations of the equipment.

[0060] When integrating diagnostic features, failure modes, and risk trends from multiple sources, diagnostic features are a collection of information reflecting the core attributes of the equipment's current health anomalies. Failure modes are clear descriptions of failure types obtained from the verification report through root cause tracing. Risk trends are descriptions of the trajectory of future health risks of the equipment. First, the logical relationship between the three is established, clarifying the failure mode corresponding to each abnormal manifestation in the diagnostic features and how the failure mode affects the development of the risk trend. Then, the information from the three is integrated complementaryly. Diagnostic features provide the current actual abnormal situation, failure modes reveal the root cause of the abnormality, and risk trends show the future development trend. This ensures that the integrated information is non-duplicative and comprehensive, covering all key aspects of fault diagnosis. Based on the integrated information, the severity and scope of the equipment failure are determined, and targeted diagnostic conclusions and treatment directions are formulated. The final comprehensive diagnostic decision is a complete decision plan that includes the core situation of the failure, the root cause, the future impact, and the preliminary treatment ideas.

[0061] When structuring a comprehensive diagnostic decision, which includes the core fault situation, root cause, future impact, and preliminary handling approach, a fixed structural framework for the in-depth fault diagnosis report is first established. This framework must meet the needs of equipment maintenance personnel and is divided into five parts: fault overview, current health status analysis, fault cause analysis, future risk prediction, and handling suggestions. Then, the corresponding content from the comprehensive diagnostic decision is filled into each part. The fault overview briefly explains the core conclusions of the fault; the current health status analysis presents the relevant content of the diagnostic characteristics in detail; the fault cause analysis comprehensively explains the fault mode and its formation logic; the future risk prediction fully displays the risk trend; and the handling suggestions clearly define the specific maintenance measures, implementation steps, and precautions. Subsequently, the language of each part is standardized to ensure accurate expression, clear organization, and avoid logical gaps or information contradictions. The final in-depth fault diagnosis report is a standardized document with a complete structure, detailed information, and can be directly used to guide equipment maintenance.

[0062] The beneficial effects include: accurately extracting core diagnostic information, providing high-quality and highly targeted basic data, improving the accuracy of comprehensive diagnostic decisions, sorting out future change patterns, comprehensively and coherently presenting the future development of health risks, providing forward-looking support, helping to predict fault situations in advance, establishing related logic and complementary integration, realizing comprehensive linkage between current anomalies, root causes and future trends, covering key information throughout the fault diagnosis process, improving the comprehensiveness and depth of diagnosis, compiling comprehensive diagnostic decisions with a standardized structure, making reports clear and well-structured, providing intuitive and operable decision-making basis, and improving the targeting and efficiency of fault handling.

[0063] Example 2 like Figure 2As shown in the figure, this embodiment also provides a functional block diagram of a multi-parameter acquisition fault detection system based on a temperature and vibration sensor.

[0064] The multi-parameter acquisition fault detection system 100 based on a temperature and vibration sensor described in this embodiment can be installed in an electronic device. Depending on the functions implemented, the multi-parameter acquisition fault detection system 100 based on a temperature and vibration sensor may include a twin construction module 101, a simulation health diagnosis module 102, a twin bidirectional calibration module 103, a decay history mapping module 104, a decay mode analysis module 105, and a diagnostic information fusion module 106. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0065] In this embodiment, the functions of each module / unit are as follows: The twin construction module 101 is used to digitally construct the thermal deformation characteristics and vibration characteristics of the target equipment during operation, so as to obtain a digital twin of the target equipment; The simulation health diagnosis module 102 is used to perform operational health diagnosis on the real-time simulation task of the digital twin and obtain the health status assessment results of the real-time simulation task. The twin bidirectional calibration module 103 is used to perform diagnostic bidirectional calibration on the digital twin based on the anomaly determination in the health status assessment results, and obtain a verification report of the target device. The attenuation history mapping module 104 is used to perform attenuation history mapping in the digital twin based on the failure modes and key parameters in the verification report, so as to obtain the performance attenuation time series map of the target device. The degradation mode analysis module 105 is used to perform healthy degradation mode analysis on the performance degradation time series map to obtain predictive health conclusions of the target device. The diagnostic information fusion module 106 is used to fuse the health status assessment results, verification reports and predictive health conclusions to obtain a deep fault diagnosis report for the target equipment.

[0066] In detail, each module in the multi-parameter acquisition fault detection system 100 based on temperature and vibration sensors described in the embodiments of the present invention adopts the same technical means as the multi-parameter acquisition fault detection method based on temperature and vibration sensors described in Embodiment 1 and Embodiment 2, and can produce the same technical effect, which will not be repeated here.

[0067] Example 3 like Figure 3As shown, this embodiment also provides a computer device, which may include a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a multi-parameter acquisition fault detection program based on a temperature and vibration sensor.

[0068] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the device, connecting various components of the device via various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., executing a multi-parameter acquisition fault detection program based on a temperature and vibration sensor) and calls data stored in the memory 11 to perform various functions of the device and process data.

[0069] The memory 11 includes at least one type of medium, including flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the device, such as the portable hard drive of the device. In other embodiments, the memory 11 can also be an external storage device of the device, such as a plug-in portable hard drive, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc. Furthermore, the memory 11 can include both internal storage units and external storage devices. The memory 11 can be used not only to store application software and various types of data installed on the device, such as the code of a multi-parameter acquisition fault detection program based on a temperature and vibration sensor, but also to temporarily store data that has been output or will be output.

[0070] The communication bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable communication between the memory 11 and at least one processor 10, etc.

[0071] The communication interface 13 is used for communication between the aforementioned device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, Bluetooth interface, etc.), typically used to establish communication connections between the device and other devices. The user interface may be a display, an input unit (such as a keyboard), or optionally, a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed within the device and to display a visual user interface.

[0072] The figure only shows the device with components. Those skilled in the art will understand that the structure shown in the figure does not constitute a limitation on the device and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0073] For example, although not shown, the device may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 via a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power sources, a recharging device, a power fault detection circuit, a power converter or inverter, a power status indicator, or any other components. The device may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0074] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.

[0075] The memory 11 in the device stores a multi-parameter acquisition fault detection program based on a temperature and vibration sensor, which is a combination of multiple instructions. When run in the processor 10, it can achieve the following: S1. Digitally construct the thermal deformation and vibration characteristics of the target equipment during operation to obtain a digital twin of the target equipment; S2. Perform operational health diagnosis on the real-time simulation task of the digital twin to obtain the health status assessment results of the real-time simulation task; S3. Based on the anomaly determination in the health status assessment results, perform diagnostic bidirectional calibration on the digital twin to obtain a verification report for the target device; S4. Based on the failure modes and key parameters in the verification report, perform attenuation history mapping in the digital twin to obtain the performance attenuation time series map of the target device. S5. Perform health degradation mode analysis on the performance degradation time series map to obtain predictive health conclusions for the target device; S6. Integrate diagnostic decision information from health status assessment results, verification reports, and predictive health conclusions to obtain a deep fault diagnosis report for the target equipment.

[0076] Specifically, the specific implementation method of the processor 10 for the above instructions can be referred to the description of the relevant steps in the corresponding embodiment of the accompanying drawings, and will not be repeated here.

[0077] Furthermore, if the modules / units integrated into the device are implemented as software functional units and sold or used as independent products, they can be stored in a medium. The medium can be volatile or non-volatile. For example, the medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0078] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0079] The modules described as separate components may or may not be physically separate. The components shown as modules 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.

[0080] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0081] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0082] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A multi-parameter acquisition fault detection method based on a temperature and vibration sensor, characterized in that, The method includes: S1. Digitally construct the thermal deformation and vibration characteristics of the target equipment during operation to obtain a digital twin of the target equipment; S2. Perform operational health diagnosis on the real-time simulation task of the digital twin to obtain the health status assessment results of the real-time simulation task; S3. Based on the anomaly determination in the health status assessment results, perform diagnostic bidirectional calibration on the digital twin to obtain a verification report for the target device; S4. Based on the failure modes and key parameters in the verification report, perform attenuation history mapping in the digital twin to obtain the performance attenuation time series map of the target device. S5. Perform health degradation mode analysis on the performance degradation time series map to obtain predictive health conclusions for the target device; S6. Integrate diagnostic decision information from health status assessment results, verification reports, and predictive health conclusions to obtain a deep fault diagnosis report for the target equipment.

2. The multi-parameter acquisition fault detection method based on a temperature and vibration sensor as described in claim 1, characterized in that, The process of digitally constructing the thermal deformation and vibration characteristics of the target equipment during operation to obtain a digital twin of the target equipment includes: The temperature and vibration sensors deployed at key monitoring locations of the target equipment are synchronously collected to obtain the operating parameters of the target equipment; Feature extraction is performed on the original signals of the operating parameters to obtain the thermal deformation and vibration characteristics of the operating parameters; By correlating and fusing the time-frequency domain correlation between thermal deformation characteristics and vibration characteristics, a heterogeneous feature substrate of the target device is obtained; Based on real-time data streams of operating parameters and physical mechanism constraints, the heterogeneous feature base is embedded into the mechanism to obtain the driving rules of the target device. Based on driving rules, a virtual mapping is performed on the heterogeneous feature base to obtain a digital twin of the target device.

3. The multi-parameter acquisition fault detection method based on a temperature and vibration sensor as described in claim 1, characterized in that, The operational health diagnosis of the real-time simulation task of the digital twin, to obtain the health status assessment results of the real-time simulation task, includes: Based on the real-time operation data stream of the target device, the digital twin is synchronously driven to obtain the real-time simulation task of the digital twin; Health parameters are extracted from the real-time simulation task to obtain the health state flow of the simulation task. Based on the historical health operation data of the target device, the health status stream is calibrated to obtain the health operation baseline of the health status stream. Based on the health operation baseline, the health deviation of the health status flow is calculated to obtain the health quantification index of the health status flow; Based on health metrics, a comprehensive assessment of the real-time simulation task is conducted to obtain the health status evaluation result of the real-time simulation task.

4. The multi-parameter acquisition fault detection method based on a temperature and vibration sensor as described in claim 3, characterized in that, The formulas for calculating the health metrics are as follows: ； In the formula, For the first A health status flow at any moment Health metrics For the first A health status flow at any moment The actual monitoring value, For the first A health status flow at any moment The baseline value for healthy operation, For the first Historical statistical standard deviation of a health status stream For the first A preset engineering allowable deviation threshold for each health status stream.

5. The multi-parameter acquisition fault detection method based on a temperature and vibration sensor as described in claim 3, characterized in that, The process of determining anomalies based on health status assessment results, performing diagnostic bidirectional calibration on the digital twin, and obtaining a verification report for the target device includes: The results of the health status assessment are screened for abnormalities to obtain an abnormality judgment conclusion. Based on the anomaly determination results, anomaly correlation analysis is performed on the digital twin to obtain the anomaly correlation parameters of the digital twin; Based on real-time running data stream, retrospective fitting is performed on the anomaly correlation parameters to obtain the parameter correction vector of the anomaly correlation parameters; The parameter correction vector is injected into the digital twin to obtain the simulation results of the digital twin after calibration. The calibrated simulation results are cross-validated with the real-time running data stream, and the validation results are packaged into a validation report for the target device.

6. The multi-parameter acquisition fault detection method based on a temperature and vibration sensor as described in claim 1, characterized in that, Based on the failure modes and key parameters in the verification report, a degradation history mapping is performed in the digital twin to obtain the performance degradation time series map of the target device, including: By tracing the root causes of anomalies in the verification report, the failure modes of the verification report can be obtained. Core variables were extracted from the validation report to obtain the key parameters of the validation report; Based on the failure modes and key parameters, the attenuation status of the digital twin is configured to obtain the attenuation evolution scenario of the digital twin. In the decay evolution scenario, dynamic degradation simulation is performed on the digital twin to obtain the state degradation trajectory of the digital twin; The performance degradation time series map of the target device is obtained by reconstructing the characteristic trajectory of the state degradation trajectory.

7. The multi-parameter acquisition fault detection method based on a temperature and vibration sensor as described in claim 1, characterized in that, The step of performing health degradation mode analysis on the performance degradation time series spectrum to obtain predictive health conclusions for the target device includes: Evolution pattern identification was performed on the performance degradation time series graph to obtain the dominant degradation mode of the time series graph; Based on the dominant decay pattern, the trend elements of the performance degradation time series graph are deconstructed to obtain the trend primitives of the dominant decay pattern. Based on the historical health status of the target device, trend extrapolation is performed on the trend primitives to obtain the future health evolution trend of the target device; By assessing future health evolution trends and determining health risks, predictive health conclusions can be obtained for the target equipment.

8. The multi-parameter acquisition fault detection method based on a temperature and vibration sensor as described in claim 6, characterized in that, By fusing diagnostic decision information from health status assessment results, validation reports, and predictive health conclusions, a deep fault diagnosis report for the target equipment is obtained, including: Diagnostic information is extracted from the health status assessment results to obtain the diagnostic features of the health status assessment results; By analyzing the development trends of predictive health conclusions, the risk trends of predictive health conclusions can be obtained. By fusing diagnostic features, failure modes, and risk trends from multiple sources, a comprehensive diagnostic decision for the target equipment can be obtained. The comprehensive diagnostic decision is structurally compiled to obtain a deep fault diagnosis report for the target equipment.

9. A multi-parameter acquisition fault detection system based on a temperature and vibration sensor, characterized in that, The system for implementing the multi-parameter acquisition fault detection method based on a temperature and vibration sensor as described in claim 1 includes: The twin construction module is used to digitally construct the thermal deformation and vibration characteristics of the target equipment during operation, thereby obtaining a digital twin of the target equipment; The simulation health diagnosis module is used to perform operational health diagnosis on the real-time simulation task of the digital twin and obtain the health status assessment results of the real-time simulation task. The twin bidirectional calibration module is used to perform diagnostic bidirectional calibration on the digital twin based on the anomaly determination in the health status assessment results, and obtain a verification report for the target device; The attenuation history mapping module is used to perform attenuation history mapping in the digital twin based on the failure modes and key parameters in the verification report, so as to obtain the performance attenuation time series map of the target device. The degradation mode analysis module is used to perform healthy degradation mode analysis on the performance degradation time series map to obtain predictive health conclusions for the target device. The diagnostic information fusion module is used to fuse health status assessment results, verification reports, and predictive health conclusions to obtain a deep fault diagnosis report for the target equipment.

10. An apparatus comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.

Images