Sensor failure prediction method and program product

Abstract

The invention discloses a sensor fault prediction method and a program product. The method comprises the following steps: acquiring first operation data of a to-be-predicted sensor at a first moment, and determining a first fault prediction value of the to-be-predicted sensor according to the first operation data; acquiring second operation data of the to-be-predicted sensor at a plurality of second moments earlier than the first moment, determining a state deviation coefficient of the to-be-predicted sensor according to the second operation data, and determining a second fault prediction value according to the state deviation coefficient and the first fault prediction value; acquiring first signal strength data of the sensor to be predicted at the first moment and second signal strength data of the sensor to be predicted at multiple second moments, and determining a third fault prediction value according to the first signal strength data, the multiple second signal strength data and the second fault prediction value, and determining the fault prediction level of the to-be-predicted sensor according to the third fault prediction value. The accuracy of fault prediction of the sensor can be effectively improved, and the working reliability of the sensor is improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a sensor fault prediction method and program product. Background Art

[0002] Sensors can complete data monitoring and data collection tasks in various complex environments, and their applications in life play an important role.

[0003] In related technologies, sensor fault detection typically relies on a set output data range for the sensor. If the range is exceeded, an alarm is triggered. This involves directly comparing the acquired data with the pre-set data range to determine whether the sensor has failed and issuing a fault alarm if a fault occurs. However, current sensors cannot predict potential faults and can only provide fault warnings when a fault occurs, making it difficult to ensure sensor reliability during use. Summary of the Invention

[0004] The present invention provides a sensor fault prediction method and program product to solve the technical problem in the related art that it is difficult to accurately predict and diagnose potential faults in sensors.

[0005] According to one aspect of the present invention, a sensor fault prediction method is provided, the method comprising:

[0006] Acquiring first operating data of a sensor to be predicted at a first moment, and determining a first fault prediction value of the sensor to be predicted based on the first operating data;

[0007] obtaining second operating data of the sensor to be predicted at a plurality of second moments earlier than the first moment, determining a state deviation coefficient of the sensor to be predicted based on the second operating data, and determining a second fault prediction value based on the state deviation coefficient and the first fault prediction value; wherein the state deviation coefficient is used to indicate a degree of deviation between an abnormal operating event of the sensor to be predicted and a normal operating event;

[0008] Obtain the first signal strength data of the sensor to be predicted at the first moment and the second signal strength data at multiple second moments, determine a third fault prediction value based on the first signal strength data, multiple second signal strength data and the second fault prediction value, and determine the fault prediction level of the sensor to be predicted based on the third fault prediction value.

[0009] According to another aspect of the present invention, a sensor failure prediction device is provided, the device comprising:

[0010] a first fault prediction value determining module, configured to obtain first operating data of a sensor to be predicted at a first moment, and determine a first fault prediction value of the sensor to be predicted based on the first operating data;

[0011] a second fault prediction value determination module, configured to obtain second operating data of the sensor to be predicted at a plurality of second moments earlier than the first moment, determine a state deviation coefficient of the sensor to be predicted based on the second operating data, and determine a second fault prediction value based on the state deviation coefficient and the first fault prediction value; wherein the state deviation coefficient is used to indicate a degree of deviation between an abnormal operating event of the sensor to be predicted and a normal operating event;

[0012] A fault prediction level determination module is used to obtain the first signal strength data of the sensor to be predicted at the first moment and the second signal strength data at multiple second moments, determine a third fault prediction value based on the first signal strength data, multiple second signal strength data and the second fault prediction value, and determine the fault prediction level of the sensor to be predicted based on the third fault prediction value.

[0013] According to another aspect of the present invention, an electronic device is provided, comprising:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a determination machine program that can be executed by the at least one processor, and the determination machine program is executed by the at least one processor to enable the at least one processor to perform a sensor fault prediction method according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, a determination machine readable storage medium is provided, wherein the determination machine readable storage medium stores determination machine instructions, and the determination machine instructions are used to enable a processor to implement a sensor fault prediction method according to any embodiment of the present invention when executed.

[0018] According to another aspect of the present invention, an embodiment of the present disclosure further provides a determination machine program product, including a determination machine program, which, when executed by a processor, implements a sensor fault prediction method as described in any one of the embodiments of the present disclosure.

[0019] The technical solution of the embodiment of the present invention is to obtain the first operating data of the sensor to be predicted at a first moment, and determine the first fault prediction value of the sensor to be predicted based on the first operating data, so as to determine the benchmark fault prediction value data for determining the potential fault risk in the sensor; then, by obtaining the second operating data of the sensor to be predicted at multiple second moments earlier than the first moment, determine the state deviation coefficient of the sensor to be predicted based on the second operating data, and determine the second fault prediction value based on the state deviation coefficient and the first fault prediction value. Since the state deviation coefficient is used to indicate the degree of deviation between the abnormal operating event of the sensor to be predicted and the normal operating event, the state deviation coefficient can be used to determine the potential fault risk of the sensor. The first operating data is analyzed to determine the changes in sensor performance at different times, and the first fault prediction value is adjusted to obtain a more accurate sensor fault prediction value; by obtaining the first signal strength data of the sensor to be predicted at the first time and the second signal strength data at multiple second times, a third fault prediction value is determined based on the first signal strength data, multiple second signal strength data and the second fault prediction value, and the fault prediction level of the sensor to be predicted is determined based on the third fault prediction value. The fault prediction value of the sensor to be predicted can be further adjusted in combination with the signal strength of the sensor to be predicted at different times, effectively ensuring the accuracy and timeliness of the calculation of the sensor fault prediction value. The solution of the present invention solves the technical problem that it is difficult for current sensors to accurately predict and diagnose potential faults in sensors by calculating and analyzing the operating data and signal strength of the sensor at different times. It can significantly improve the reliability of the sensor during use, so that the sensor can be maintained in a timely manner, reducing production interruptions and economic losses caused by sensor failures.

[0020] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 This is a flowchart of a sensor fault prediction method provided according to the first embodiment of the present invention;

[0023] Figure 2 This is a flowchart of a sensor fault prediction method provided according to the second embodiment of the present invention;

[0024] Figure 3 This is a flowchart of a sensor fault prediction method provided in accordance with the third embodiment of the present invention;

[0025] Figure 4 This is a structural block diagram of a sensor fault prediction device provided according to a fourth embodiment of the present invention;

[0026] Figure 5 It is a structural diagram of an electronic device implementing the fifth embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first," "second," "target," and the like in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprises" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or apparatus.

[0029] It should be noted that the modifications of "one" and "multiple" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, they should be understood as "one or more".

[0030] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0031] It is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) must comply with the requirements of relevant laws, regulations and relevant provisions.

[0032] Example 1

[0033] Figure 1 A flowchart of a sensor fault prediction method is provided for the first embodiment of the present invention. This embodiment is applicable to scenarios of sensor fault prediction. The method can be executed by a sensor fault prediction device configured in an electronic device. The sensor fault prediction device can be implemented in the form of hardware and / or software. Optionally, it can be implemented by an electronic device, which can be a mobile terminal, PC or server. Figure 1 As shown, the method may specifically include:

[0034] S110 : Acquire first operating data of a sensor to be predicted at a first moment, and determine a first fault prediction value of the sensor to be predicted based on the first operating data.

[0035] The sensor to be predicted may include at least one of a temperature sensor, a humidity sensor, a strain sensor, an electrochemical sensor, and a pressure sensor. The first moment may be the first moment of obtaining the first operating data. The first operating data may include at least one of information such as an operating frequency, a temperature range, a humidity range, and an operating time. Specifically, a corresponding fault warning threshold may be determined based on at least one of information such as the operating frequency, temperature range, humidity range, and operating time of different sensors. The first fault prediction value is a data result that can be used to characterize the likelihood of a potential fault in the sensor to be predicted. The first fault prediction value can be directly determined based on the first operating data.

[0036] Specifically, at least one of the following information, such as the operating frequency, temperature range, humidity range, and operating time of the sensor to be predicted at a first moment, can be obtained. Based on the first operating data, a corresponding fault prediction value for the sensor to be predicted can be determined. The present invention can rapidly determine the operating status of the sensor to be predicted based on the obtained first operating data, thereby promptly obtaining a fault prediction value for the sensor to be predicted.

[0037] S120. Obtain second operating data of the sensor to be predicted at multiple second moments earlier than the first moment, determine a state deviation coefficient of the sensor to be predicted based on the second operating data, and determine a second fault prediction value based on the state deviation coefficient and the first fault prediction value; wherein the state deviation coefficient is used to indicate the degree of deviation between the abnormal operating event of the sensor to be predicted and the normal operating event.

[0038] Among them, the second moment may refer to a moment earlier than the first moment. Specifically, the second moment may be a moment at the same time point of the sensor device in different periods; the second moment may also be a moment in the same time range as the first moment. The second operating data may include at least one of the information such as the operating frequency, temperature range, humidity range and operating time of the sensor, which is the same as the content included in the first operating data. The second fault prediction value can be used to characterize the data result of the size of the potential fault possibility of the sensor, which is the same as the content of the first fault prediction value. The second fault prediction value can be obtained by adjusting the first fault prediction value by the state deviation coefficient. The state deviation coefficient is used to indicate the degree of deviation between the abnormal operation event of the sensor to be predicted and the normal operation event. Generally speaking, the larger the state deviation coefficient, the farther the operating state of the sensor to be predicted deviates from the normal state, that is, the higher the possibility of the sensor to be predicted having a fault risk.

[0039] Specifically, at least one of the following information, such as the operating frequency, temperature range, and humidity range, of the sensor to be predicted before the first moment is obtained, and a state deviation coefficient of the sensor to be predicted is calculated based on the information in the second operating data. The first fault prediction value is then adjusted for deviation based on the state deviation coefficient to determine an adjusted fault prediction value, i.e., a second fault prediction value.

[0040] In one embodiment, determining the second fault prediction value based on the state deviation coefficient and the first fault prediction value may include: determining a first preset adjustment coefficient based on the state deviation coefficient and a preset coefficient threshold, multiplying the first fault prediction value and the first preset adjustment coefficient to obtain a second fault prediction value.

[0041] The preset coefficient threshold is a user-preset limit value for distinguishing state deviation coefficients, and may include multiple preset coefficient thresholds, such as a first preset coefficient threshold, a second preset coefficient threshold, and a third preset coefficient threshold. Multiple coefficient ranges may be determined based on the preset coefficient thresholds, with different coefficient ranges corresponding to different first preset adjustment coefficients. The first preset adjustment coefficient refers to multiple preset adjustment coefficients for adjusting the first fault prediction value, and may include multiple adjustment coefficients, such as a first adjustment coefficient, a second adjustment coefficient, and a third adjustment coefficient.

[0042] Specifically, the calculated state deviation coefficient is compared with multiple preset coefficient thresholds to determine the coefficient range of the state deviation coefficient, and then the first preset adjustment coefficient corresponding to the coefficient range is determined, and then the second fault prediction value is calculated based on the first preset adjustment coefficient.

[0043] For example, the number of first preset coefficient thresholds can be one, which can be used to determine whether the first fault prediction value needs to be adjusted. If the calculated state deviation coefficient is less than the first preset coefficient threshold, then the first fault prediction value does not need to be adjusted, and the first fault prediction value can be used as the second fault prediction value. If the calculated state deviation coefficient is greater than the first preset coefficient threshold, the first fault prediction value is multiplied by the first preset adjustment coefficient (the first preset adjustment coefficient is not 1) to obtain the second fault prediction value.

[0044] Alternatively, the number of first preset coefficient thresholds may be multiple, for example, including a first coefficient threshold and a second coefficient threshold, in which case three coefficient ranges can be determined. In this case, the first preset adjustment coefficient may include adjustment coefficient No. 1, adjustment coefficient No. 2, and adjustment coefficient No. 3. For example, if the state deviation coefficient is less than the second preset coefficient threshold, the first fault prediction value and the adjustment coefficient No. 1 are multiplied to obtain the second fault prediction value. If the calculated state deviation coefficient is greater than the second preset coefficient threshold and less than the third preset coefficient threshold, the first fault prediction value and the adjustment coefficient No. 2 are multiplied to obtain the second fault prediction value. If the calculated state deviation coefficient is greater than the third preset coefficient threshold, the first fault prediction value and the adjustment coefficient No. 3 are multiplied to obtain the second fault prediction value. It is understood that when the first preset adjustment coefficient is 1, the second fault prediction value is the same as the first fault prediction value.

[0045] In another embodiment, determining the second fault prediction value based on the state deviation coefficient and the first fault prediction value may include: determining the coefficient deviation ratio of the state deviation coefficient and the state deviation coefficient threshold, determining a first preset adjustment coefficient based on the coefficient deviation ratio and the preset ratio threshold, and multiplying the first fault prediction value and the first preset adjustment coefficient to obtain a second fault prediction value.

[0046] Specifically, the coefficient deviation ratio refers to the ratio of the state deviation coefficient to the state deviation coefficient threshold. Multiple ratio ranges can be determined based on the preset ratio threshold, with different ratio ranges corresponding to different first preset adjustment coefficients. For example, the preset ratio thresholds may include 1, 1.2, and 1.4. In this case, the multiple ratio ranges determined based on the preset ratio threshold may specifically include less than or equal to 1, greater than 1 and less than or equal to 1.2, greater than 1.2 and less than or equal to 1.4, and greater than 1.4. In this case, the first preset adjustment coefficient may include 1, as well as adjustment coefficients 1, 2, and 3 that are not 1. If the coefficient deviation ratio is less than or equal to 1, the first preset adjustment coefficient may be set to 1, meaning that no adjustment is required for the first fault prediction value, and the first fault prediction value becomes the second fault prediction value. If the coefficient deviation ratio is greater than 1 and less than or equal to 1.2, the first fault prediction value is multiplied by the adjustment coefficient 1 to obtain the second fault prediction value. If the coefficient deviation ratio is greater than 1.2 and less than or equal to 1.4, the first fault prediction value is multiplied by the second adjustment coefficient to obtain the second fault prediction value. If the coefficient deviation ratio is greater than 1.4, the first fault prediction value is multiplied by the third adjustment coefficient to obtain the second fault prediction value. This allows for flexible adjustment of the first fault prediction value based on the state deviation coefficient and the preset coefficient threshold, or the coefficient deviation ratio, making the adjustment more flexible and accurate, allowing for the rapid determination of the second fault prediction value and the fault level of the sensor to be predicted.

[0047] It's important to note that the adjustment coefficient is positively correlated with state deviation. For example, adjustment coefficient 1 < adjustment coefficient 2 < adjustment coefficient 3. When the state deviation coefficient deviates significantly from the preset coefficient threshold, the larger adjustment coefficient can be used to ensure timely and accurate fault prediction. This ensures prediction accuracy while reducing unnecessary maintenance operations and costs.

[0048] S130. Obtain the first signal strength data of the sensor to be predicted at the first moment and the second signal strength data at multiple second moments, determine a third fault prediction value based on the first signal strength data, multiple second signal strength data and the second fault prediction value, and determine the fault prediction level of the sensor to be predicted based on the third fault prediction value.

[0049] The first signal strength refers to at least one of the following information: amplitude, signal-to-noise ratio, dynamic range, sensitivity, and resolution, when the sensor being predicted is collecting data. The second signal strength includes the same information as the first signal strength. The first and second signal strength data specifically refer to the signal strength values ​​collected by the sensor during each fixed sampling time. The second signal strength data can also be used to construct a second signal strength database, providing a foundation for subsequent signal analysis and fault prediction.

[0050] The third fault prediction value can be used to indicate the likelihood of a potential fault in the sensor being predicted. It contains the same information as the first and second fault prediction values. The third fault prediction value can be calculated by adjusting the second fault prediction value based on signal strength data to determine the fault level. The fault prediction level refers to a pre-defined sensor fault level.

[0051] Specifically, the system obtains first signal strength data for the sensor to be predicted at a first moment, as well as second signal strength data at multiple second moments. The obtained first signal strength is compared with the multiple second signal strengths to determine an adjustment to the second fault prediction value based on the signal strength data, thereby obtaining an adjusted third fault prediction value. The third fault prediction value is then compared with a preset prediction value range to determine the fault prediction level of the sensor to be predicted. This ensures the accuracy and timeliness of the fault prediction level obtained based on the operating data of multiple sensors.

[0052] In one embodiment, determining the fault prediction level of the sensor to be predicted based on the third fault prediction value includes: determining the target prediction value range to which the third fault prediction value belongs from multiple preset prediction value ranges, and determining the preset fault level corresponding to the target prediction value range as the fault prediction level of the sensor to be predicted.

[0053] The preset prediction value range is user-defined and is used to classify the preset fault levels corresponding to different third fault prediction values. Fault prediction levels are categorized into three levels: first, second, and third. Different levels have varying degrees of fault severity or impact on sensor operation. Simply put, different fault levels have varying degrees of severity.

[0054] Specifically, the calculated third fault prediction value is compared with multiple preset prediction value ranges to determine the target prediction value range to which the third fault prediction value belongs. Based on the target prediction value, a corresponding preset fault level is determined, and the determined preset fault level is used as the fault prediction level for the sensor to be predicted. This hierarchical prediction approach enables early detection and timely resolution of sensors to be predicted, thereby reducing system failures caused by sensor failures.

[0055] In one embodiment, the target prediction value range to which the third fault prediction value belongs is determined from multiple preset prediction value ranges, and the preset fault level corresponding to the target prediction value range is determined as the fault prediction level of the sensor to be predicted, including: when the third fault prediction value is less than or equal to a first preset threshold, determining the fault prediction level of the sensor to be predicted to be the first level; when the third fault prediction value is greater than the first preset threshold and less than or equal to the second preset threshold, determining the fault prediction level of the sensor to be predicted to be the second level; when the third fault prediction value is greater than the second preset threshold, determining the fault prediction level of the sensor to be predicted to be the third level.

[0056] Among them, the first preset threshold and the second preset threshold can be pre-set by the user, or can be determined based on the historical operation data or historical fault data of one or more sensors of the same type. This application does not limit the determination method and specific data.

[0057] Specifically, by comparing the obtained third fault prediction value with the preset first preset threshold and the second preset threshold, the target prediction value range of the sensor to be predicted can be quickly and conveniently determined, and then the fault prediction level corresponding to the sensor to be predicted can be determined.

[0058] The technical solution of the embodiment of the present invention is to obtain the first operating data of the sensor to be predicted at a first moment, and determine the first fault prediction value of the sensor to be predicted based on the first operating data, so as to determine the benchmark fault prediction value data for determining the potential fault risk in the sensor; then, by obtaining the second operating data of the sensor to be predicted at multiple second moments earlier than the first moment, determine the state deviation coefficient of the sensor to be predicted based on the second operating data, and determine the second fault prediction value based on the state deviation coefficient and the first fault prediction value. Since the state deviation coefficient is used to indicate the degree of deviation between the abnormal operating event of the sensor to be predicted and the normal operating event, the state deviation coefficient can be used to determine the potential fault risk of the sensor. The first operating data is analyzed to determine the changes in sensor performance at different times, and the first fault prediction value is adjusted to obtain a more accurate sensor fault prediction value; by obtaining the first signal strength data of the sensor to be predicted at the first time and the second signal strength data at multiple second times, a third fault prediction value is determined based on the first signal strength data, multiple second signal strength data and the second fault prediction value, and the fault prediction level of the sensor to be predicted is determined based on the third fault prediction value. The fault prediction value of the sensor to be predicted can be further adjusted in combination with the signal strength of the sensor to be predicted at different times, effectively ensuring the accuracy and timeliness of the calculation of the sensor fault prediction value. The solution of the present invention solves the technical problem that it is difficult for current sensors to accurately predict and diagnose potential faults in sensors by calculating and analyzing the operating data and signal strength of the sensor at different times. It can significantly improve the reliability of the sensor during use, so that the sensor can be maintained in a timely manner, reducing production interruptions and economic losses caused by sensor failures.

[0059] Example 2

[0060] Figure 2A flowchart of a sensor fault prediction method provided in the second embodiment of the present invention, the solution in this embodiment is based on the above embodiment, and further refines how to determine the state deviation coefficient of the sensor to be predicted based on the second operating data. Optionally, the determination of the state deviation coefficient of the sensor to be predicted based on the second operating data includes: inputting the second operating data at multiple second moments into a pre-trained operating event recognition model to obtain the target operating event of the sensor to be predicted; wherein the target operating event is at least used to distinguish between normal operating events and abnormal operating events; respectively determining the state impact factor corresponding to each of the abnormal operating events, determining the first total duration of multiple abnormal operating events, and determining the second total duration of multiple normal operating events; determining the state deviation coefficient of the sensor to be predicted based on the state impact factor, the first total duration and the second total duration corresponding to each of the abnormal operating events. Wherein, the technical features that are the same or similar to those in the above embodiment are not repeated here. Figure 2 As shown, the method may specifically include:

[0061] S210: Acquire first operating data of a sensor to be predicted at a first moment, and determine a first fault prediction value of the sensor to be predicted based on the first operating data.

[0062] S220: Acquire second operating data of the sensor to be predicted at multiple second moments earlier than the first moment.

[0063] S230: Input the plurality of second operation data at the second moments into a pre-trained operation event recognition model to obtain a target operation event of the sensor to be predicted; wherein the target operation event is at least used to distinguish between normal operation events and abnormal operation events.

[0064] Among them, the operation event recognition model generally includes, but is not limited to, models with data analysis and processing, and can generally include one of the following models: a splicing fusion model, a graph neural network fusion model, and a time series feature fusion model. A normal operation event refers to an event in which the sensor operates within a preset data range, while an abnormal operation event refers to an event in which the sensor's operating data exceeds the preset data range, or an event in which the sensor's output signal exhibits irregular fluctuations, which is not in line with expectations. An abnormal operation event may refer to an event in which at least one of temperature data, humidity data, electromagnetic interference, pressure fluctuations, communication quality, and other information exceeds the preset data range, or an event in which the output signal exhibits irregular fluctuations, which is not in line with expectations. For example, for a temperature sensor, a normal operation event may be a fluctuation within a set temperature range, while an abnormal operation event may manifest as a sudden increase or decrease in temperature readings, which may be caused by a sensor failure or a change in the external environment; or, at a certain moment, the sensor's pressure fluctuation varies significantly, far exceeding the pressure fluctuation range during normal operation.

[0065] In an embodiment of the present invention, an operation event recognition model can be obtained by training a machine learning model based on multiple groups of sample operation data and their corresponding expected event labels. The expected event labels are used to distinguish normal operation events from abnormal operation events. Specifically, the operation event recognition model is trained in the following manner: by obtaining the second operation sample data at the second moment, inputting the second operation sample data into a pre-built machine learning model, obtaining a model recognition result output by the machine learning model, determining the model recognition loss based on the model recognition result and the expected event label corresponding to the second sample operation data, and adjusting the parameters of the machine learning model based on the model recognition loss to obtain an operation event recognition model. Furthermore, the operation event recognition model is used to obtain the normal operation events and abnormal operation events of the sensor to be predicted based on the second operation data.

[0066] Specifically, the acquired second operating data can be input into an operating event recognition model. Using a preset machine learning model, the second operating data is processed to obtain normal operating events and abnormal operating events for the sensor to be predicted. By identifying the operating status of the sensor to be predicted, potential problems with the sensor can be promptly identified, allowing appropriate maintenance measures to be taken promptly.

[0067] S240: Determine the status impact factor corresponding to each abnormal operation event, determine a first total duration of the plurality of abnormal operation events, and determine a second total duration of the plurality of normal operation events.

[0068] Among them, the state impact factor of the abnormal operation event can be determined based on the influencing factors that may interfere with the normal operation of the sensor, reduce its measurement accuracy or cause a malfunction. The influencing factors may specifically include at least one of temperature data, humidity data, electromagnetic interference, pressure fluctuations and communication quality information. Different attribute values ​​of different influencing factors have different state impact factors, which can be determined specifically based on the type and duration of the influencing factors. In one technical solution, the target influencing factor (main influencing factor) of the abnormal operation event can be determined, and the state impact factor of the abnormal operation event can be determined based on the state impact factor corresponding to the target influencing factor. Specifically, different state impact factor values ​​can be pre-set for different target influencing factors. For example, the state impact factor of temperature data can be 0.2, the state impact factor of humidity data can be 0.25, and the state impact factor of electromagnetic interference can be 0.3. In another technical solution, the attribute values ​​of the influencing factors corresponding to different abnormal operation events may be different, and the state impact factor corresponding to the abnormal operation event can be determined based on the state impact factors corresponding to the attribute values ​​of multiple influencing factors (such as weighted summation, etc.).

[0069] The first total duration refers to the total duration of the duration of each influencing factor when the sensor to be predicted is in an abnormal operating state. The second total duration refers to the total duration of the sensor to be predicted when it is in a normal operating state.

[0070] Specifically, by determining the state impact factor corresponding to the abnormal operation event and separately determining the first total duration of the abnormal operation event and the second total duration of the normal operation event, the time difference between the normal and abnormal operation of the predicted sensor can be calculated, thereby better quantifying the impact of the abnormal operation duration of the predicted sensor on sensor performance. For example, if the abnormal operation duration of a predicted sensor is significantly greater than its normal operation duration, it may be at risk of severe performance degradation or failure. Furthermore, by calculating the state impact factor corresponding to the abnormal operation event, the health status of the predicted sensor can be more accurately assessed, providing a basis for fault prediction and maintenance of the predicted sensor.

[0071] In one embodiment, the separately determining of the status impact factor corresponding to each of the abnormal operation events may specifically include: determining the ideal operation time based on the abnormal operation times corresponding to multiple abnormal operation events, and determining the minimum value of the normal operation times corresponding to multiple normal operation events as the minimum normal operation time; for each of the abnormal operation events, determining a first operation time difference between the abnormal operation time and the ideal operation time, and determining a second operation time difference between the abnormal operation time and the minimum normal operation time, and determining the status impact factor corresponding to the abnormal operation event based on the first operation time difference and the second operation time difference.

[0072] The ideal operating time is the expected operating time of the sensor under optimal operating conditions. The minimum operating time is the minimum operating time among the normal operating times of the sensor under prediction. The first operating time difference is the difference between the abnormal operating time of the sensor under prediction and the expected operating time under optimal operating conditions. The second operating time difference is the difference between the abnormal operating time of the sensor under prediction and the minimum operating time among the normal operating times.

[0073] Specifically, by calculating the ideal operating time and the minimum normal operating time, and then calculating the first line time difference between the abnormal operating time and the ideal operating time and the second operating time difference between the abnormal operating time and the minimum normal operating time during any abnormal operating event of the sensor to be predicted, by comparing these two time differences, the degree of influence of the abnormal event on the performance of the sensor to be predicted can be quantified, thereby determining the state influence factor corresponding to the abnormal operating event.

[0074] S250: Determine a state deviation coefficient of the sensor to be predicted according to the state impact factor, the first total duration, and the second total duration corresponding to each abnormal operation event.

[0075] Specifically, the state deviation coefficient can be quickly determined based on the state impact factor corresponding to each abnormal operation event in the predicted sensor, the total operation time in the abnormal operation state, and the total operation time in the normal operation state of the predicted sensor.

[0076] In one embodiment, the state deviation coefficient of the sensor to be predicted is determined based on the state influence factor, the first total duration and the second total duration corresponding to each abnormal operation event, including: for each abnormal operation event, multiplying the state influence factor corresponding to the abnormal operation event by the actual duration of the abnormal operation event to obtain the abnormal duration; summing the abnormal durations of multiple abnormal operation events to obtain a third total duration; and summing the first total duration and the second total duration to obtain a fourth total duration; and determining the ratio of the third total duration to the fourth total duration as the state deviation coefficient of the sensor to be predicted.

[0077] The abnormal duration refers to the comprehensive impact of the state impact factor of each abnormal operation event and the actual operation duration of the abnormality, which can be used to indicate the degree of impact of different abnormal operation events on the normal operation of the sensor.

[0078] Specifically, the state deviation coefficient can be obtained by the following formula:

[0079]

[0080] Among them, SDC represents the state deviation coefficient; T1 represents the first total duration; T2 represents the second total duration; Iai represents the state influence factor of the i-th abnormal event in the state influence factor; Tai represents the duration of the i-th abnormal event in the state influence factor sequence; n represents the number of abnormal events in the state influence factor sequence.

[0081] Specifically, the impact factors of different abnormal operating events are calculated and compared with the impact results corresponding to the duration of the abnormal operating events. The impact results corresponding to each abnormal event are then added together to obtain the combined impact result of all abnormal operating events. The total operating time of the sensor to be predicted is obtained by adding the abnormal operating time to the normal operating time. The combined impact results of all abnormal operating events for the sensor are calculated together with the operating time to determine the state deviation coefficient of the sensor to be predicted. This allows the impact of each abnormal operating event to be determined based on the operating time of multiple abnormal operating events. By calculating the state deviation degree, the degree of impact of the abnormal operating event on the sensor is determined. Based on the state deviation coefficient, the latest operating state of the sensor can be quickly and easily determined.

[0082] S260: Determine a second fault prediction value according to the state deviation coefficient and the first fault prediction value.

[0083] S270. Obtain the first signal strength data of the sensor to be predicted at the first moment and the second signal strength data at multiple second moments, determine a third fault prediction value based on the first signal strength data, multiple second signal strength data and the second fault prediction value, and determine the fault prediction level of the sensor to be predicted based on the third fault prediction value.

[0084] The technical solution of an embodiment of the present invention is to input the second operating data at multiple second moments into a pre-trained operating event recognition model to obtain the target operating event of the sensor to be predicted; wherein the target operating event is used to at least distinguish between normal operating events and abnormal operating events; thereby, normal operating events and abnormal operating events in the sensor to be predicted can be determined, thereby obtaining the status data of the abnormal operating event; the state impact factor corresponding to each abnormal operating event is determined, the first total duration of the multiple abnormal operating events is determined, and the second total duration of the multiple normal operating events is determined; the duration of the abnormal operating event and the duration of the normal operating event can be obtained respectively to better understand the working condition of the sensor. The state deviation coefficient of the sensor to be predicted is determined based on the state impact factor, the first total duration, and the second total duration corresponding to each abnormal operating event, so that the health status of the sensor can be timely assessed and the need for sensor maintenance or replacement can be determined based on the value of the state deviation coefficient, which helps to improve sensor maintenance efficiency.

[0085] Example 3

[0086] Figure 3A flowchart of a sensor fault prediction method provided in the second embodiment of the present invention, the solution in this embodiment is a refinement of the technical solution for determining a third fault prediction value based on the first signal strength data, multiple second signal strength data and the second fault prediction value on the basis of the above embodiment. Optionally, the third fault prediction value is determined based on the first signal strength data, multiple second signal strength data and the second fault prediction value, including: when the multiple second signal strength data include the second signal strength data consistent with the first signal strength data, adjusting the second fault prediction value according to the historical adjustment coefficient corresponding to the second signal strength data to obtain the third fault prediction value; when the multiple second signal strength data do not include the second signal strength data consistent with the first signal strength data, respectively determining the correlation index between the first signal strength data and each second signal strength data, adjusting the second fault prediction value according to the correlation index to obtain the third fault prediction value. For specific implementation methods, please refer to the description of this embodiment. Among them, the technical features that are the same or similar to the above embodiments are not repeated here. Figure 3 As shown, the method may specifically include:

[0087] S310: Acquire first operating data of a sensor to be predicted at a first moment, and determine a first fault prediction value of the sensor to be predicted based on the first operating data;

[0088] S320: Acquire second operating data of the sensor to be predicted at multiple second moments earlier than the first moment, determine a state deviation coefficient of the sensor to be predicted based on the second operating data, and determine a second fault prediction value based on the state deviation coefficient and the first fault prediction value; wherein the state deviation coefficient is used to indicate a degree of deviation between an abnormal operating event of the sensor to be predicted and a normal operating event;

[0089] S330: Acquire first signal strength data of the sensor to be predicted at the first moment and second signal strength data of the sensor at multiple second moments;

[0090] S340: If the plurality of second signal strength data includes second signal strength data that is consistent with the first signal strength data, adjust the second fault prediction value according to a historical adjustment coefficient corresponding to the second signal strength data to obtain a third fault prediction value;

[0091] The historical adjustment coefficient refers to a pre-set fault prediction adjustment value used to adjust the second fault prediction value when the second fault prediction value was adjusted prior to the current adjustment. This adjustment coefficient can adjust the second fault prediction value to a reasonable range based on the signal strength comparison result. The second signal strength data may correspond to one or more historical adjustment coefficients. Specifically, the historical adjustment coefficient corresponding to the second signal strength data may be the adjustment coefficient used in the most recent adjustment of the fault prediction value before the first moment; or, alternatively, the average or weighted sum of multiple historical adjustment coefficients within a preset time period corresponding to the first moment.

[0092] Specifically, by determining whether there is signal strength data consistent with the first signal strength among multiple second signal strengths, the second fault prediction value is adjusted according to the pre-set fault prediction adjustment to obtain an adjusted third fault prediction value. The third fault prediction is more consistent with the actual fault prediction value result of the sensor to be predicted, thereby making the calculation of the third fault prediction value of the sensor to be predicted more accurate.

[0093] S350. When the second signal strength data that is consistent with the first signal strength data is not included in the multiple second signal strength data, determine the correlation index between the first signal strength data and each second signal strength data respectively, adjust the second fault prediction value according to the correlation index, and obtain a third fault prediction value.

[0094] The correlation index may be determined based on the similarity between the signal strength at a first moment and the signal strengths at multiple second moments.

[0095] Specifically, if it is determined that none of the multiple second signal strength data is consistent with the first signal strength data, the second fault prediction value can be adjusted based on the similarity between the first signal strength data and the multiple second signal strength data to obtain an adjusted third fault prediction value. A quantitative indicator can be obtained to measure the degree of match between real-time data and historical data, and the second fault prediction value can be adjusted using the second signal strength data that is closest to the first signal strength data to ensure the accuracy of the adjusted fault prediction value.

[0096] In one embodiment, determining the correlation index between the first signal strength data and each of the second signal strength data includes: determining a first signal strength characteristic value of the first signal strength data, and respectively determining a second signal strength characteristic value of each of the second signal strength data; and determining the correlation index between the first signal strength data and each of the second signal strength data by the following formula, including:

[0097]

[0098] Among them, S q represents the correlation index between the first signal strength data and the qth second signal strength data in the second signal strength database, Dj represents the jth first signal strength characteristic value of the first signal strength data, Daj represents the first signal strength standard value corresponding to the jth first signal strength characteristic value of the first signal strength data, Hqj represents the jth second signal strength characteristic value of the qth second signal strength data in the second signal strength database, Hqaj represents the second signal strength standard value corresponding to the jth second signal strength characteristic value of the qth second signal strength data in the second signal strength database, m represents the number of second signal strength characteristic values ​​in each second signal strength data, j=1,2,…,m,q=1,2,…,k, k is the total number of second signal strength data in the second signal strength database.

[0099] The signal strength characteristic value is determined based on the characteristic values ​​of multiple signal strength influencing factors. Specifically, the signal strength influencing factors may include, but are not limited to, one or more of the amplitude, signal-to-noise ratio, dynamic range, sensitivity, and resolution in the first signal strength data. The characteristic value of the signal strength influencing factor can be understood as a quantitative indicator of the signal strength influencing factor.

[0100] Specifically, for a single signal strength data in the first signal strength data or the second signal strength data, the characteristic values ​​of multiple signal strength influencing factors in the signal strength data can be determined first; then, the characteristic sequence composed of the characteristic values ​​of multiple signal strength influencing factors of the signal strength data is determined as the signal strength characteristic value of the signal strength data, or the characteristic value obtained by summing or weighted summing the characteristic values ​​of multiple signal strength influencing factors is determined as the signal strength characteristic value of the signal strength data.

[0101] For example, the correlation index can be calculated as follows: when m = 3 and q = 1, the first signal strength characteristic values ​​corresponding to the first signal strength data are: minimum first signal strength D1 = 5dBm, average first signal strength D2 = 7dBm, and maximum first signal strength D3 = 12dBm; the first signal strength standard values ​​corresponding to these first signal strength characteristic values ​​are Da1 = 6dBm, Da2 = 8dBm, and Da3 = 10dBm, respectively. The second signal characteristic values ​​corresponding to the first second signal strength data in the second signal strength database are: minimum second signal strength H11 = 4dBm, average second signal strength H12 = 6dBm, and maximum second signal strength H13 = 11dBm; the second signal standard values ​​corresponding to these second signal characteristic values ​​are H1a1 = 6dBm, H1a2 = 8dBm, and H1a3 = 10dBm, respectively. Substituting all these values ​​into the correlation index calculation formula, the calculated S value is approximately 0.816.

[0102] When m = 3 and q = 2, the first signal strength characteristic values ​​corresponding to the first signal strength data are: minimum first signal strength D1 = 5dBm, average first signal strength D2 = 7dBm, and maximum first signal strength D3 = 12dBm. The corresponding first signal standard values ​​for these first signal strength characteristic values ​​are Da1 = 6dBm, Da2 = 8dBm, and Da3 = 10dBm, respectively. The second signal strength characteristic values ​​corresponding to the second second signal strength data in the second signal strength database are: minimum second signal strength H21 = 3dBm, average second signal strength H22 = 5dBm, and maximum second signal strength H23 = 10dBm. The corresponding second signal standard values ​​for these second signal characteristic values ​​are H2a1 = 6dBm, H2a2 = 8dBm, and H2a3 = 10dBm, respectively. Based on this data, we can further calculate another set of correlation indices to assess the similarity between the first signal strength data and the second second signal strength data. This calculation yields another set of correlation indices, S, with a value of approximately 0.577. Thus, by calculating the correlation results between the first signal strength and the plurality of second signal strength data in this manner and comparing the correlation results, a closer correlation result can be obtained to adjust the fault prediction value.

[0103] In one embodiment, the second fault prediction value is adjusted according to the correlation index to obtain a third fault prediction value, including: determining the maximum value of the correlation index between the first signal strength data and multiple second signal strength data as the target correlation index; determining the target index range to which the target correlation index belongs from multiple preset index ranges, determining a second preset adjustment coefficient corresponding to the target index range, and multiplying the second fault prediction value and the second preset adjustment coefficient to obtain a third fault prediction value.

[0104] The target correlation index refers to the highest correlation index between the first signal strength data and the multiple second signal strength data. The multiple preset index ranges refer to different preset index ranges corresponding to different signal strength characteristic values, with each type of signal strength characteristic value having a corresponding preset index range. The second preset adjustment coefficients are multiple pre-set proportional coefficients used to calculate and adjust the second fault prediction value.

[0105] Specifically, multiple correlation index values ​​between the first signal strength and the second signal strength are calculated, and the maximum value among the multiple correlation indices is determined as a target correlation index. A corresponding signal strength characteristic value type is determined based on the target correlation index, and a target index range corresponding to the signal strength characteristic value type is selected from multiple preset index ranges. A corresponding preset adjustment coefficient is then determined based on the target index range, and a calculation is performed based on the preset adjustment coefficient and the second fault prediction value to obtain a calculated third fault prediction value.

[0106] The technical solution provided by the embodiment of the present invention is to adjust the second fault prediction value according to the historical adjustment coefficient corresponding to the second signal strength data when the second signal strength data is included in the plurality of second signal strength data, thereby obtaining a third fault prediction value; the historical adjustment coefficient is obtained by providing the performance data of the sensor to be predicted. By using the historical adjustment coefficient, the historical adjustment experience can be combined with the operating data of the sensor to be predicted at the first moment to improve the accuracy of the fault prediction; and when the plurality of second signal strength data does not include the second signal strength data consistent with the first signal strength data, the correlation index between the first signal strength data and each second signal strength data is determined respectively, and the second fault prediction value is adjusted according to the correlation index to obtain a third fault prediction value, thereby calculating the similarity between the signal strength data at the first moment and the signal strength at multiple second moments, and determining how to adjust the second fault prediction value according to the degree of similarity to obtain a relatively appropriate third fault prediction value. The present application can effectively improve the accuracy of the third fault prediction value through the above-mentioned technical method of adjusting the second fault prediction value to obtain a more accurate fault prediction level result.

[0107] Example 4

[0108] Figure 4 This is a structural block diagram of a sensor fault prediction device provided by the fourth embodiment of the present invention. The device can be implemented by software and / or hardware and can be configured in an electronic device. Figure 4 As shown, the device includes: a first fault prediction value determination module 401 , a second fault prediction value determination module 402 and a fault prediction level determination module 403 .

[0109] Among them, the first fault prediction value determination module 401 is used to obtain the first operating data of the sensor to be predicted at the first moment, and determine the first fault prediction value of the sensor to be predicted based on the first operating data; the second fault prediction value determination module 402 is used to obtain the second operating data of the sensor to be predicted at multiple second moments earlier than the first moment, determine the state deviation coefficient of the sensor to be predicted based on the second operating data, and determine the second fault prediction value based on the state deviation coefficient and the first fault prediction value; wherein, the state deviation coefficient is used to indicate the degree of deviation between the abnormal operating event of the sensor to be predicted and the normal operating event; the fault prediction level determination module 403 is used to obtain the first signal strength data of the sensor to be predicted at the first moment and the second signal strength data at multiple second moments, determine the third fault prediction value based on the first signal strength data, multiple second signal strength data and the second fault prediction value, and determine the fault prediction level of the sensor to be predicted based on the third fault prediction value.

[0110] The technical solution of the embodiment of the present invention is to obtain the first operating data of the sensor to be predicted at a first moment through the first fault prediction value determination module 401, and determine the first fault prediction value of the sensor to be predicted based on the first operating data, so as to determine the latest fault prediction value data for predicting the potential fault risk in the sensor; obtain the second operating data of the sensor to be predicted at multiple second moments earlier than the first moment through the second fault prediction value determination module 402, determine the state deviation coefficient of the sensor to be predicted based on the second operating data, and determine the second fault prediction value based on the state deviation coefficient and the first fault prediction value; wherein, the state deviation coefficient is used to indicate the degree of deviation between the abnormal operating event of the sensor to be predicted and the normal operating event; by using comparative analysis with historical data, By determining the state deviation coefficient and adjusting the first fault prediction value, it is possible to determine subtle changes in sensor performance at different times, thereby obtaining a more accurate fault prediction value for the sensor, ensuring the accuracy and timeliness of the fault prediction value. The fault prediction level determination module 403 obtains the first signal strength data of the sensor to be predicted at the first time and the second signal strength data at multiple second times, determines a third fault prediction value based on the first signal strength data, multiple second signal strength data, and the second fault prediction value, and determines the fault prediction level of the sensor to be predicted based on the third fault prediction value. By adjusting the fault prediction value in combination with the sensor's signal strength, the sensor's health status can be more accurately assessed, improving the accuracy and timeliness of the calculation of the sensor's fault prediction value. The present invention solves the technical problem of the difficulty in accurately predicting and diagnosing potential faults in sensors by performing computational analysis on the sensor's operating data at different times. This can significantly improve the reliability and maintenance efficiency of the sensor, reducing production interruptions and economic losses caused by sensor failures.

[0111] On the basis of the above-mentioned optional technical solutions, optionally, the second fault prediction value determination module 402 may further include: an operation event determination unit, an operation event total duration determination unit, and a state deviation coefficient determination unit. The operation event determination unit is used to input the second operation data at the plurality of second moments into a pre-trained operation event recognition model to obtain the target operation event of the sensor to be predicted; wherein the target operation event is at least used to distinguish between normal operation events and abnormal operation events; the operation event total duration determination unit is used to respectively determine the state impact factor corresponding to each of the abnormal operation events, determine the first total duration of the plurality of abnormal operation events, and determine the second total duration of the plurality of normal operation events; the state deviation coefficient determination unit is used to determine the state deviation coefficient of the sensor to be predicted based on the state impact factor, the first total duration, and the second total duration corresponding to each of the abnormal operation events.

[0112] Based on the above-mentioned optional technical solutions, the operation event total duration determination unit may further include: an operation time determination subunit and a state impact factor determination subunit. The operation time determination subunit is configured to determine an ideal operation time based on the abnormal operation times corresponding to the plurality of abnormal operation events, and to determine the minimum normal operation time among the normal operation times corresponding to the plurality of normal operation events; and the state impact factor determination subunit is configured to determine, for each abnormal operation event, a first operation time difference between the abnormal operation time and the ideal operation time, and a second operation time difference between the abnormal operation time and the minimum normal operation time, and to determine the state impact factor corresponding to the abnormal operation event based on the first and second operation time differences.

[0113] On the basis of the above-mentioned optional technical solutions, optionally, the state deviation coefficient determination unit may further include: an abnormal duration integration unit, a duration summing unit and a state deviation coefficient determination subunit. The abnormal duration integration unit is used to multiply the state impact factor corresponding to each abnormal operation event by the actual duration of the abnormal operation event for each abnormal operation event to obtain the abnormal duration; the duration summing unit is used to sum the abnormal durations of multiple abnormal operation events to obtain a third total duration; and to sum the first total duration and the second total duration to obtain a fourth total duration; the state deviation coefficient determination subunit is used to determine the ratio of the third total duration to the fourth total duration as the state deviation coefficient of the sensor to be predicted.

[0114] Based on the above optional technical solutions, the second fault prediction value determination module 402 may optionally further include: a second fault prediction value integration unit. The second fault prediction value integration unit is configured to determine a first preset adjustment coefficient based on the state deviation coefficient and a preset coefficient threshold, and multiply the first fault prediction value by the first preset adjustment coefficient to obtain a second fault prediction value; or to determine a coefficient deviation ratio between the state deviation coefficient and the state deviation coefficient threshold, determine a first preset adjustment coefficient based on the coefficient deviation ratio and a preset ratio threshold, and multiply the first fault prediction value by the first preset adjustment coefficient to obtain a second fault prediction value.

[0115] Based on the above-mentioned optional technical solutions, optionally, the fault prediction level determination module 403 may further include: a historical adjustment coefficient adjustment unit and a correlation index adjustment unit. The historical adjustment coefficient adjustment unit is configured to adjust the second fault prediction value according to the historical adjustment coefficient corresponding to the second signal strength data to obtain a third fault prediction value when the plurality of second signal strength data includes the second signal strength data that is consistent with the first signal strength data; and the correlation index adjustment unit is configured to determine the correlation index between the first signal strength data and each of the second signal strength data when the plurality of second signal strength data does not include the second signal strength data that is consistent with the first signal strength data, and adjust the second fault prediction value according to the correlation index to obtain a third fault prediction value.

[0116] Based on the above optional technical solutions, the correlation index adjustment unit may optionally further include: a signal strength characteristic value calculation unit and a correlation index calculation unit. The signal strength characteristic value calculation unit is configured to determine a first signal strength characteristic value of the first signal strength data and to determine a second signal strength characteristic value of each second signal strength data; and the correlation index determination unit is configured to obtain a correlation index using the following formula:

[0117]

[0118] Among them, S qrepresents the correlation index between the first signal strength data and the qth second signal strength data in the second signal strength database, Dj represents the jth first signal strength characteristic value of the first signal strength data, Daj represents the first signal strength standard value corresponding to the jth first signal strength characteristic value of the first signal strength data, Hqj represents the jth second signal strength characteristic value of the qth second signal strength data in the second signal strength database, Hqaj represents the second signal strength standard value corresponding to the jth second signal strength characteristic value of the qth second signal strength data in the second signal strength database, m represents the number of second signal strength characteristic values ​​in each second signal strength data, j=1,2,…,m,q=1,2,…,k, k is the total number of second signal strength data in the second signal strength database.

[0119] Based on the above-mentioned optional technical solutions, the fault prediction level determination module 403 may optionally further include: a maximum correlation index determination unit and a third fault prediction value integration unit. The maximum correlation index determination unit is configured to determine the maximum value of the correlation indexes between the first signal strength data and the plurality of second signal strength data as a target correlation index; and the third fault prediction value integration unit is configured to determine a target index range to which the target correlation index belongs from a plurality of preset index ranges, determine a second preset adjustment coefficient corresponding to the target index range, and multiply the second fault prediction value by the second preset adjustment coefficient to obtain a third fault prediction value.

[0120] Based on the above optional technical solutions, the fault prediction level determination module 403 may optionally further include a fault prediction level determination unit. The fault prediction level determination unit is configured to determine a target prediction value range to which the third fault prediction value belongs from a plurality of preset prediction value ranges, and determine the preset fault level corresponding to the target prediction value range as the fault prediction level of the sensor to be predicted.

[0121] The sensor fault prediction device provided in the embodiments of the present invention can execute a sensor fault prediction method provided in any of the embodiments of the present invention, and includes the corresponding functional modules and beneficial effects of executing a sensor fault prediction method. For technical details not fully described in this embodiment, please refer to the sensor fault prediction method described in any of the embodiments of the present invention.

[0122] Example 5

[0123] Figure 5A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital determination machines, such as laptop determination machines, desktop determination machines, workstations, personal digital assistants, servers, blade servers, large determination machines, and other suitable determination machines. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar determination devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0124] like Figure 5 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by the at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0125] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a network such as the Internet and / or various telecommunication networks.

[0126] The processor 11 can be any general-purpose and / or specialized processing component with processing and determination capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) determination chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any other suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as executing a sensor fault prediction method.

[0127] In some embodiments, a sensor fault prediction method may be implemented as a determination machine program, which is tangibly contained in a determination machine-readable storage medium, such as storage unit 18. In some embodiments, part or all of the determination machine program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the determination machine program is loaded into RAM 13 and executed by processor 11, one or more steps of a sensor fault prediction method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform a sensor fault prediction method in any other appropriate manner (e.g., by means of firmware).

[0128] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), determination machine hardware, firmware, software, and / or combinations thereof. These various embodiments may include: being implemented in one or more determination machine programs, which can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a dedicated or general programmable processor, which can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device. The determination machine program for implementing the method of the present invention can be written in any combination of one or more programming languages. These determination machine programs can be provided to a processor of a general determination machine, a dedicated determination machine, or other programmable data processing device so that when the determination machine program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The machine program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine, or entirely on the remote machine or server.

[0129] In the context of the present invention, it is determined that the machine-readable storage medium can be a tangible medium that can contain or store a machine program for use by an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. It is determined that the machine-readable storage medium can include but is not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the above. Alternatively, it is determined that the machine-readable storage medium can be a machine-readable signal medium. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable determination disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0130] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0131] The systems and techniques described herein can be implemented in a system that includes backend components (e.g., as a data server), or includes middleware components (e.g., an application server), or includes frontend components (e.g., a user interface with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0132] A system can include a client and a server. The client and server are typically remote from each other and typically interact via a communication network. This client-server relationship is established by server programs running on the respective servers, creating a client-server relationship. The server can be a cloud server, also known as a cloud server or cloud host. This server is a hosting product within a cloud hosting service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0133] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a determination machine software program. For example, an embodiment of the present invention includes a determination machine program product, which includes a determination machine program carried on a non-transitory determination machine readable medium, and the determination machine program includes program code for executing the method shown in the flowchart. In such an embodiment, the determination machine program can be downloaded and installed from the network via the communication unit 19, or installed from the storage unit 18, or installed from the ROM 12. When the determination machine program is executed by the processor 11, the above-mentioned functions defined in the method of the embodiment of the present invention are performed.

[0134] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0135] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A sensor failure prediction method, characterized in that: include: Acquiring first operating data of a sensor to be predicted at a first moment, and determining a first fault prediction value of the sensor to be predicted based on the first operating data; obtaining second operating data of the sensor to be predicted at a plurality of second moments earlier than the first moment, determining a state deviation coefficient of the sensor to be predicted based on the second operating data, and determining a second fault prediction value based on the state deviation coefficient and the first fault prediction value; wherein the state deviation coefficient is used to indicate a degree of deviation between an abnormal operating event of the sensor to be predicted and a normal operating event; Obtain the first signal strength data of the sensor to be predicted at the first moment and the second signal strength data at multiple second moments, determine a third fault prediction value based on the first signal strength data, multiple second signal strength data and the second fault prediction value, and determine the fault prediction level of the sensor to be predicted based on the third fault prediction value.

2. The method according to claim 1, characterized in that The determining the state deviation coefficient of the sensor to be predicted according to the second operating data includes: Inputting the plurality of second operating data at the second moments into a pre-trained operating event recognition model to obtain a target operating event of the sensor to be predicted; wherein the target operating event is used at least to distinguish between a normal operating event and an abnormal operating event; respectively determining a state impact factor corresponding to each of the abnormal operation events, determining a first total duration of the plurality of abnormal operation events, and determining a second total duration of the plurality of normal operation events; The state deviation coefficient of the sensor to be predicted is determined according to the state impact factor, the first total duration, and the second total duration corresponding to each abnormal operation event.

3. The method according to claim 2, characterized in that The respectively determining the status impact factor corresponding to each of the abnormal operation events includes: determining an ideal operating time according to the abnormal operating times corresponding to the plurality of abnormal operating events, and determining a minimum value of the normal operating times corresponding to the plurality of normal operating events as a minimum normal operating time; For each of the abnormal operation events, a first operation time difference between the abnormal operation time and the ideal operation time is determined, and a second operation time difference between the abnormal operation time and the minimum normal operation time is determined; and a state impact factor corresponding to the abnormal operation event is determined based on the first operation time difference and the second operation time difference.

4. The method according to claim 3, characterized in that The determining the state deviation coefficient of the sensor to be predicted according to the state impact factor corresponding to each abnormal operation event, the first total duration, and the second total duration includes: For each abnormal operation event, multiplying the state impact factor corresponding to the abnormal operation event by the actual duration of the abnormal operation event to obtain the abnormal duration; Summing the abnormal durations of the plurality of abnormal operation events to obtain a third total duration; and summing the first total duration and the second total duration to obtain a fourth total duration; The ratio of the third total duration to the fourth total duration is determined as a state deviation coefficient of the sensor to be predicted.

5. The method according to claim 1, characterized in that The determining of the second fault prediction value according to the state deviation coefficient and the first fault prediction value includes: Determine a first preset adjustment coefficient according to the state deviation coefficient and a preset coefficient threshold, and multiply the first fault prediction value by the first preset adjustment coefficient to obtain a second fault prediction value; or Determine the coefficient deviation ratio of the state deviation coefficient and the state deviation coefficient threshold, determine a first preset adjustment coefficient based on the coefficient deviation ratio and the preset ratio threshold, multiply the first fault prediction value and the first preset adjustment coefficient to obtain a second fault prediction value.

6. The method according to claim 1, characterized in that The determining a third fault prediction value according to the first signal strength data, a plurality of the second signal strength data, and the second fault prediction value includes: When the plurality of second signal strength data include second signal strength data that is consistent with the first signal strength data, adjusting the second fault prediction value according to a historical adjustment coefficient corresponding to the second signal strength data to obtain a third fault prediction value; When the second signal strength data that is consistent with the first signal strength data is not included in the multiple second signal strength data, the correlation index between the first signal strength data and each second signal strength data is determined respectively, and the second fault prediction value is adjusted according to the correlation index to obtain a third fault prediction value.

7. The method according to claim 6, characterized in that The determining of the correlation index between the first signal strength data and each of the second signal strength data includes: Determine a first signal strength characteristic value of the first signal strength data, and respectively determine a second signal strength characteristic value of each second signal strength data; The correlation index between the first signal strength data and each of the second signal strength data is determined respectively by the following formula, including: Among them, S q represents the correlation index between the first signal strength data and the qth second signal strength data in the second signal strength database, Dj represents the jth first signal strength characteristic value of the first signal strength data, Daj represents the first signal strength standard value corresponding to the jth first signal strength characteristic value of the first signal strength data, Hqj represents the jth second signal strength characteristic value of the qth second signal strength data in the second signal strength database, Hqaj represents the second signal strength standard value corresponding to the jth second signal strength characteristic value of the qth second signal strength data in the second signal strength database, m represents the number of second signal strength characteristic values ​​in each second signal strength data, j=1,2,…,m,q=1,2,…,k, k is the total number of second signal strength data in the second signal strength database.

8. The method according to claim 6, characterized in that The adjusting the second fault prediction value according to the correlation index to obtain a third fault prediction value includes: determining a maximum value among correlation indices between the first signal strength data and a plurality of second signal strength data as a target correlation index; Determine a target index range to which the target correlation index belongs from multiple preset index ranges, determine a second preset adjustment coefficient corresponding to the target index range, and multiply the second fault prediction value and the second preset adjustment coefficient to obtain a third fault prediction value.

9. The method according to claim 1, characterized in that Determining the fault prediction level of the sensor to be predicted according to the third fault prediction value includes: A target prediction value range to which the third fault prediction value belongs is determined from a plurality of preset prediction value ranges, and a preset fault level corresponding to the target prediction value range is determined as the fault prediction level of the sensor to be predicted.

10. A determination machine program product, comprising a determination machine program, characterized in that: When the determination machine program is executed by a processor, it implements a sensor fault prediction method according to any one of claims 1 to 9.

Images