A method, apparatus, device, and storage medium for equipment condition monitoring.

Abstract

This application discloses a method, apparatus, device, and storage medium for equipment status detection, applied in the field of equipment detection. The method includes: acquiring current monitoring data of the equipment under test; calculating a first evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model; calculating a second evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model; performing a weighted calculation on the first and second evaluation values ​​to obtain a comprehensive evaluation value; calculating the difference between the comprehensive evaluation value and the current monitoring data; and determining that the equipment under test is in an abnormal state when the difference exceeds a preset threshold. This application utilizes both a low false alarm rate algorithm model and a low false alarm rate algorithm model to obtain the first evaluation value / false alarm value and the second evaluation value / false alarm value, respectively, and performs a comprehensive evaluation by fusing / weighting the data from both models, taking advantage of their respective strengths. The comprehensive evaluation value is then used in conjunction with the current detection data to determine whether the equipment under test is abnormal; thus achieving simultaneous monitoring and ensuring both a low false alarm rate and a low false alarm rate.

Description

Technical Field

[0001] This application relates to the field of equipment testing, and in particular to a method, apparatus, equipment, and storage medium for equipment condition testing. Background Technology

[0002] A high false alarm rate leads to unnecessary maintenance work and increases operating costs; a high false alarm rate leads to equipment failures not being detected and handled in a timely manner, delaying troubleshooting and affecting production safety and efficiency. False alarms and false alarms are contradictory; a very low false alarm rate results in a high false alarm rate, and vice versa.

[0003] Currently, equipment condition monitoring relies on a single algorithm, meaning that only one of the false alarm rate or false negative rate can be guaranteed simultaneously; both cannot be guaranteed to be low. Therefore, the false alarm rate and false negative rate need to be adjusted and balanced based on on-site requirements. Enterprises need to configure and manage their equipment condition monitoring systems rationally and scientifically to ensure that the monitoring equipment can minimize false alarms and false negatives. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a device, apparatus, equipment and storage medium for device status detection, which solves the problem of the difficulty in balancing low false alarm rate and low false alarm rate in the prior art.

[0005] To address the aforementioned technical problems, this application provides a method for detecting equipment status, comprising:

[0006] Obtain the current monitoring data of the device under test;

[0007] The first evaluation value corresponding to the current monitoring data is calculated using a low false negative rate algorithm model;

[0008] The second evaluation value corresponding to the current monitoring data is calculated using a low false alarm rate algorithm model;

[0009] The first evaluation value and the second evaluation value are weighted and calculated to obtain a comprehensive evaluation value;

[0010] The difference between the comprehensive evaluation value and the current monitoring data is calculated. When the difference is greater than a preset threshold, the device under test is considered to be in an abnormal state.

[0011] Optionally, the step of calculating the first evaluation value corresponding to the current monitoring data using a low false negative rate algorithm model includes:

[0012] Acquire historical monitoring data of the device under test within a preset historical time period;

[0013] The distance vector is obtained by calculating the distance between the vectors using the current monitoring data and the historical monitoring data;

[0014] The distance vector is transformed into a false negative rate weighting base using a Gaussian kernel function.

[0015] Using the missed detection rate weight base and the historical monitoring data, the first evaluation value corresponding to the current detection data is calculated.

[0016] Optionally, the step of calculating the second evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model includes:

[0017] Acquire historical monitoring data of the device under test within a preset historical time period;

[0018] The false alarm rate weight base is obtained by using a nonlinear multivariate predictive diagnostic technique.

[0019] Using the false alarm rate weighting base and the historical monitoring data, the second evaluation value corresponding to the current detection data is calculated.

[0020] Optionally, the weighted calculation of the first evaluation value and the second evaluation value to obtain a comprehensive evaluation value includes:

[0021] Obtain a preset device importance weight library; the preset device importance weight library consists of weight values ​​corresponding to different device importance.

[0022] The first false alarm rate weight and the first false alarm rate weight corresponding to the device to be detected are determined according to the preset device importance weight library;

[0023] The comprehensive evaluation value is calculated based on the first false alarm rate weight and the first false negative rate weight.

[0024] Optionally, the weighted calculation of the first evaluation value and the second evaluation value to obtain a comprehensive evaluation value includes:

[0025] Obtain a preset weight library for equipment operation scale; the preset weight library for equipment operation rules consists of weight values ​​corresponding to different equipment operation scales.

[0026] The second false alarm rate weight and the second false alarm rate weight corresponding to the device to be detected are determined according to the preset device operation scale weight library;

[0027] The comprehensive evaluation value is calculated based on the second false alarm rate weight and the second false negative rate weight.

[0028] Optionally, after acquiring the current monitoring data of the device under test and acquiring the historical monitoring data of the device under test within a preset historical time period, the method further includes:

[0029] The current monitoring data and the historical monitoring data are normalized and cleaned.

[0030] Optionally, after acquiring the historical monitoring data of the device under test within a preset historical time period, the method further includes:

[0031] Delete the data from the historical monitoring data that shows abnormal states and abnormal operating conditions.

[0032] This application also provides a device for detecting equipment status, including:

[0033] The current monitoring data acquisition module is used to acquire the current monitoring data of the device under test;

[0034] The first evaluation value calculation module is used to calculate the first evaluation value corresponding to the current monitoring data using a low false negative rate algorithm model.

[0035] The second evaluation value calculation module is used to calculate the second evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model.

[0036] The comprehensive evaluation value calculation module is used to perform a weighted calculation on the first evaluation value and the second evaluation value to obtain a comprehensive evaluation value;

[0037] The difference calculation module is used to calculate the difference between the comprehensive evaluation value and the current monitoring data. When the difference is greater than a preset threshold, the device under test is in an abnormal state.

[0038] This application also provides a device for detecting device status, including:

[0039] Memory, used to store computer programs;

[0040] A processor is used to execute the steps of the computer program to implement the device status detection method described above.

[0041] This application also provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the device status detection method described above.

[0042] As can be seen, this application obtains the current monitoring data of the device under test; calculates the first evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model; calculates the second evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model; performs a weighted calculation on the first and second evaluation values ​​to obtain a comprehensive evaluation value; calculates the difference between the comprehensive evaluation value and the current monitoring data, and when the difference is greater than a preset threshold, the device under test is considered to be in an abnormal state. This application uses both the low false alarm rate algorithm model and the low false alarm rate algorithm model to obtain the first evaluation value / false alarm value and the second evaluation value / false alarm value, respectively, and performs a comprehensive evaluation by fusing / weighting the data of the two models, taking advantage of each to compensate for their shortcomings. The comprehensive evaluation value is then used to calculate whether the device under test is abnormal based on the current monitoring data; thus achieving simultaneous monitoring and ensuring low false alarm rate and low false alarm rate.

[0043] In addition, this application also provides a device for monitoring device status, equipment, and storage medium, which also have the above-mentioned beneficial effects. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0045] Figure 1 A flowchart of a device status detection method provided in an embodiment of this application;

[0046] Figure 2 This is a schematic diagram of the structure of a device status detection device provided in an embodiment of this application;

[0047] Figure 3 This is a schematic diagram of the structure of a device status detection device provided in an embodiment of this application. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0049] In the field of industrial equipment monitoring, if the monitoring results of equipment sensors show abnormalities, the detection system will issue alarm information. Sometimes the alarm information reflects a real problem and is considered a valid alarm; other times, the alarm information is a false alarm, indicating that the equipment is operating normally without any fault or abnormality, and is therefore invalid. A genuine alarm will promptly notify engineers to address any existing faults or abnormalities, while a false alarm will cause engineers to be unnecessarily busy, wasting time and resources, and may also lead to unnecessary factory shutdowns for maintenance, increasing production costs.

[0050] In reality, false alarm rate and false negative rate are contradictory. When the desired false negative rate is very low, the false alarm rate will be high; conversely, when the desired false alarm rate is very low, the false negative rate will be high. Adjustments and balances need to be made based on actual on-site requirements. Furthermore, to minimize false alarms and false negatives, industrial enterprises need to rationally and scientifically configure and manage their detection systems to ensure that equipment can issue alarms promptly and accurately, and take appropriate action. This application proposes a method for equipment status detection; please refer to [reference needed]. Figure 1 , Figure 1 A flowchart illustrating a device status detection method provided in this application embodiment. The method may include:

[0051] S101: Obtain the current monitoring data of the device under test.

[0052] The execution subject of this embodiment is any device capable of performing the device testing method.

[0053] Assume the current monitoring data (data to be tested) obtained by the device under test is as follows:

[0054] x obs =[x obs1 ,x obs2 ,…,x obsn-1 ,x obsn ];

[0055] Where, x obs The matrix representation of the current monitoring data; x obs1 This represents the monitoring data collected by the first sensor; x obs2 This represents the monitoring data collected by the second sensor; x obsn-1 This represents the monitoring data collected by the (n-1)th sensor at time i; x obsn This represents the monitoring data collected by the nth sensor.

[0056] S102: Calculate the first evaluation value corresponding to the current monitoring data using a low false negative rate algorithm model.

[0057] A low false negative rate algorithm refers to an algorithm that achieves a low false negative rate in equipment condition monitoring and fault diagnosis. This algorithm needs to ensure high sensitivity to potential faults while maintaining a low false negative rate to a certain extent. This embodiment does not limit the low false negative rate algorithm model. For example, a low false negative rate algorithm can be a kernel regression algorithm; or a support vector machine; or a neural network algorithm; or a decision tree algorithm; or a random forest, etc. The key is to ensure a low false positive rate and a low false negative rate during model training. Specifically, when the low false negative rate algorithm model is a kernel regression algorithm, this embodiment does not limit the specific kernel function. For example, the kernel function can be a Gaussian kernel function / radial basis function; or a linear kernel function; or a Sigmoid function.

[0058] Furthermore, to ensure the reliability and convenience of low false negative rate calculation, the above-mentioned calculation of the first evaluation value corresponding to the current monitoring data using the low false negative rate algorithm model may include the following steps, specifically:

[0059] Step 21: Obtain historical monitoring data of the device under test within a preset historical time period;

[0060] Step 22: Calculate the distance between vectors using current monitoring data and historical monitoring data to obtain the distance vector;

[0061] Step 23: Use the Gaussian kernel function to convert the distance vector into a false negative rate weighting base;

[0062] Step 24: Calculate the first evaluation value corresponding to the current detection data using the missed detection rate weight base and historical monitoring data.

[0063] This embodiment uses kernel regression algorithm as the low false negative rate algorithm model, and uses Gaussian kernel function to transform the distance vector between current monitoring data and historical monitoring data into the weight of false negative rate, i.e. false negative rate weight base. Using this false negative rate weight base and historical monitoring data, the first evaluation value corresponding to the current detection data is calculated, i.e. false negative rate value.

[0064] Furthermore, to ensure the accuracy of detection and the reliability and accuracy of model training, after obtaining the current monitoring data of the device under test and the historical monitoring data of the device under test within a preset historical time period, the following steps may be included, specifically:

[0065] The current and historical monitoring data are normalized and cleaned.

[0066] Delete data from historical monitoring data that shows abnormal states and abnormal operating conditions.

[0067] This embodiment normalizes the collected data (historical and current monitoring data). To ensure data quality, the collected data also needs to be cleaned, including removing missing values ​​and outliers. Secondly, the cleaned historical monitoring data undergoes anomaly removal, meaning data collected under abnormal conditions / operating circumstances is deleted. This ensures that the historical monitoring data used for subsequent modeling is collected entirely under normal operating conditions, allowing the established model to learn from historical information under normal operating conditions without being affected by abnormal conditions.

[0068] The specific process of calculating the first evaluation value using the kernel regression algorithm in this embodiment includes:

[0069] In this embodiment, the device under test may include multiple sensors to monitor the device, that is, to obtain monitoring data collected by multiple sensors at the current time. For example, if n sensors monitor the device under test, the monitoring data matrix obtained from the sensors at time i is represented as:

[0070] x i =[x i1 ,x i2 ,…,x in-1 ,x in ];

[0071] Where, x i The matrix representation of the monitoring data at time i; x i1 This represents the monitoring data collected by the first sensor at time i; x i2 This represents the monitoring data collected by the second sensor at time i; x in-1 This represents the monitoring data collected by the (n-1)th sensor at time i; x in This represents the monitoring data collected by the nth sensor at time i.

[0072] The historical data from time 1 to time m are now integrated and represented in matrix form, as follows:

[0073]

[0074] Where D is the matrix form of historical monitoring data; x1, x2, ..., x i ,…,x m This refers to historical monitoring data from time 1 to time m.

[0075] This embodiment does not limit the method of calculating the distance between vectors using current monitoring data and historical monitoring data. For example, Euclidean distance can be used to calculate the distance between the vectors of historical monitoring data and current monitoring data; or Manhattan distance can be used to calculate the distance between the vectors of historical monitoring data and current monitoring data. To make the calculation more accurate and efficient, this embodiment uses Euclidean distance to calculate the distance between the two, as shown in the following formula:

[0076]

[0077] Where Dis represents the distance vector, dis1 dis2…dis m This represents the distance vector from the i-th to the m-th distance vector.

[0078] Then, using the Gaussian kernel function transformation, the distance vector is converted into weights, i.e., the false negative rate weight base W. The transformation formula is as follows:

[0079]

[0080] Where h represents the bandwidth of the kernel function; k h Let represent the Gaussian kernel transformation function; W represents the weights of the m vectors in the historical monitoring data. The first evaluation value x is obtained by multiplying these weights by the historical monitoring data according to the following formula. est_1 :

[0081]

[0082] S103: Calculate the second evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model.

[0083] A low false alarm rate algorithm refers to an algorithm with a low false alarm rate in equipment condition monitoring and fault diagnosis, aiming to avoid frequent false alarms and thus reduce operation and maintenance costs. This embodiment does not limit the low false alarm rate algorithm model. For example, it can be a support vector machine; or it can be a neural network model; or it can be the MSET (Multivariate State Estimation Technique, a nonlinear multivariate predictive diagnostic technique) algorithm.

[0084] Furthermore, to ensure the accuracy and reliability of the low false alarm rate calculation, the above-mentioned calculation of the second evaluation value corresponding to the current monitoring data using the low false alarm rate algorithm model may include the following steps, specifically:

[0085] Step 31: Obtain historical monitoring data of the device under test within a preset historical time period;

[0086] Step 32: Use nonlinear multivariate predictive diagnostic techniques to obtain the false alarm rate weight base;

[0087] Step 33: Calculate the second evaluation value corresponding to the current detection data using the false alarm rate weight base and the historical monitoring data.

[0088] The specific process of calculating the second evaluation value using nonlinear multivariate predictive diagnostic technology in this embodiment includes:

[0089] Based on current monitoring data x obs The second evaluation value x is obtained. est_2 for:

[0090] x est_2 =Y T D;

[0091] Where Y is the false alarm rate weighting base, and D is the matrix form of historical monitoring data.

[0092] The intermediate solution process includes:

[0093] Assume ε = [ε1, ε2, ..., ε n ] T It is the residual vector, which is the difference between the current monitoring data and the second evaluation value, i.e.

[0094] ε=x obs -x est_2 ;

[0095] To ensure that the residual values ​​of the model are within a small range, a commonly used residual function S(Y) needs to be assumed, and its formula is defined as follows:

[0096]

[0097] Using the least squares method to find the minimum value of S(Y), and differentiating the above formula for S(Y), we get:

[0098]

[0099] According to the principle of least squares, When the minimum value of S(Y) can be obtained, the above formula becomes... It can be deduced that:

[0100]

[0101] The above Rewritten in matrix form, we get:

[0102] Dx obs =DD T Y;

[0103] The solution yields:

[0104] x est_2=Y T D=(DD T ) -1 (Dx obs )) T D.

[0105] S104: The first and second evaluation values ​​are weighted and calculated to obtain the comprehensive evaluation value.

[0106] The comprehensive evaluation value x in this embodiment est The calculation formula is:

[0107] x est =ax est_2 +(1-a)x est_1 ,a∈[0,1];

[0108] This embodiment allows for a balance between low false positive and low false positive rates in the final evaluation result by adjusting the value of 'a'. This embodiment does not impose a limitation on the weight 'a'. For example, it can be set based on the importance of the device under test; or it can be set based on the size of the device under test.

[0109] Furthermore, to consider whether the equipment under test is critical equipment, i.e., the importance level of the equipment, the above-mentioned weighted calculation of the first evaluation value and the second evaluation value to obtain a comprehensive evaluation value may include the following steps, specifically including:

[0110] Step 41: Obtain the preset equipment importance weight library; the preset equipment importance weight library contains the weight values ​​corresponding to different equipment importance.

[0111] Step 42: Determine the first false alarm rate weight and the first false alarm rate weight corresponding to the device to be detected based on the preset device importance weight library;

[0112] Step 43: Calculate the comprehensive evaluation value based on the weight of the first false alarm rate and the weight of the first false alarm rate.

[0113] This embodiment does not specifically limit the determination of the first false alarm rate weight and the first false alarm rate weight corresponding to the device to be detected based on the pre-device importance weight library. The following example can be used as a reference:

[0114] (1) For critical equipment that needs to be alarmed once there is a problem, i.e. the equipment to be tested is very important and is a critical equipment: set a to 0 to ensure that its false alarm rate is the lowest and allow it to generate more false alarms. If there are too many false alarms, you can consider setting a to a range of <0.1.

[0115] (2) For non-critical equipment with multiple standby units, i.e., the equipment to be tested is less important and is not critical: set a to 1 (or > 0.3). This can reduce many false alarms caused by changes in normal operating conditions and reduce maintenance costs to a great extent.

[0116] (3) For key equipment that is expected to reduce the false alarm rate to a certain extent (maintain a low false alarm rate) and at the same time reduce the false alarm rate to a certain extent (maintain a low false alarm rate): set a in the range of <0.3 to maintain a low false alarm rate.

[0117] Furthermore, to take into account the operational scale of the equipment under test, the first evaluation value and the second evaluation value are weighted and calculated to obtain a comprehensive evaluation value. This may include the following steps, specifically:

[0118] Step 51: Obtain the preset equipment operation scale weight library; the preset equipment operation rule weight library contains the weight values ​​corresponding to different equipment operation scales;

[0119] Step 52: Determine the second false alarm rate weight and the second false alarm rate weight corresponding to the device to be detected based on the preset equipment operation scale weight library;

[0120] Step 52: Calculate the comprehensive evaluation value based on the second false alarm rate weight and the second false alarm rate weight.

[0121] This embodiment does not specifically limit the determination of the second false alarm rate weight and the second false alarm rate weight corresponding to the device to be detected based on the preset device operation scale weight library. The following example can be used as a reference:

[0122] (1) For small-scale maintenance operations: At this time, it is necessary to adjust the operating conditions of several devices, which can easily generate a large number of false alarms. Therefore, the weight parameter a of the monitoring model of these devices can be set to 1 to reduce the number of false alarms of these devices as much as possible. After the maintenance is completed, the a value of the monitoring model of these devices will be adjusted back to the original value.

[0123] (2) For large-scale periodic maintenance operations: At this time, it is necessary to adjust the operating conditions of a large number of devices. Each device is prone to generating a large number of alarms. Therefore, it is recommended to turn off the device status monitoring model and restart the monitoring model after the large-scale periodic maintenance operation is completed.

[0124] (3) Operation for switching between two similar devices. At this time, the sensor data related to these two devices will fluctuate greatly. Therefore, it is recommended to set 'a' to 1 to reduce false alarms. After the device switching is completed, 'a' should be adjusted back to its original value.

[0125] S105: Calculate the difference between the comprehensive evaluation value and the current monitoring data. If the difference is greater than the preset threshold, the device under test is in an abnormal state.

[0126] The difference can be obtained by subtracting the comprehensive assessment value from the current monitoring data. The calculation formula is as follows:

[0127] res = x est -x obs .

[0128] The preset threshold in this embodiment can be set based on experience, statistical methods, or by experienced engineers. When the difference exceeds this threshold, the status of the device under test is abnormal, and the system issues an alarm. This completes the operation of the device status monitoring model with a false alarm / false alarm balancing mechanism.

[0129] The device status detection method provided in this application involves acquiring the current monitoring data of the device under test; calculating a first evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model; calculating a second evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model; weighting the first and second evaluation values ​​to obtain a comprehensive evaluation value; and calculating the difference between the comprehensive evaluation value and the current monitoring data. If the difference is greater than a preset threshold, the device under test is considered to be in an abnormal state. This application uses both a low false alarm rate algorithm model and a low false alarm rate algorithm model to obtain the first evaluation value / false alarm value and the second evaluation value / false alarm value, respectively. A comprehensive evaluation is then performed by fusing and weighting the data from both models, taking advantage of their respective strengths. The comprehensive evaluation value is then used in conjunction with the current detection data to determine whether the device under test is abnormal. This method enables simultaneous monitoring and ensures both a low false alarm rate and a low false alarm rate. Furthermore, this application proposes the concepts of a low false alarm rate algorithm and a low false alarm rate algorithm, and uses them in combination to determine the equipment status. It also proposes an adjustment strategy for the weight coefficient 'a' based on different types of equipment (equipment importance) or different operating operations (equipment scale), achieving a balance between the false alarm rate and the false alarm rate. Additionally, it uses a kernel regression algorithm as the low false alarm rate algorithm model, ensuring the reliability and convenience of low false alarm rate calculation. Furthermore, it performs normalization, data cleaning, and deletion of data under abnormal states and operating conditions on the collected data, improving the accuracy of detection and the reliability and accuracy of model training. Finally, it uses a nonlinear multivariate prediction and diagnostic technique to calculate the false alarm rate value (second evaluation value), ensuring the accuracy and reliability of low false alarm rate calculation.

[0130] The following describes the equipment status detection device provided in the embodiments of this application. The equipment status detection device described below can be referred to in correspondence with the equipment status detection method described above.

[0131] Please refer to the details. Figure 2 , Figure 2 A schematic diagram of a device status detection apparatus provided in this application embodiment may include:

[0132] The current monitoring data acquisition module 100 is used to acquire the current monitoring data of the device under test;

[0133] The first evaluation value calculation module 200 is used to calculate the first evaluation value corresponding to the current monitoring data using a low false negative rate algorithm model.

[0134] The second evaluation value calculation module 300 is used to calculate the second evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model.

[0135] The comprehensive evaluation value calculation module 400 is used to perform a weighted calculation on the first evaluation value and the second evaluation value to obtain a comprehensive evaluation value;

[0136] The difference calculation module 500 is used to calculate the difference between the comprehensive evaluation value and the current monitoring data. When the difference is greater than a preset threshold, the device under test is in an abnormal state.

[0137] Based on the above embodiments, the first evaluation value calculation module 200 may include:

[0138] The first historical monitoring data acquisition unit is used to acquire the historical monitoring data of the device under test within a preset historical time period.

[0139] The distance vector calculation unit is used to calculate the distance between vectors using the current monitoring data and the historical monitoring data to obtain a distance vector.

[0140] The transformation unit is used to convert the distance vector into a false negative rate weighting base using a Gaussian kernel function.

[0141] The first evaluation value calculation unit is used to calculate the first evaluation value corresponding to the current detection data using the missed detection rate weight base and the historical monitoring data.

[0142] Based on the above embodiments, the second evaluation value calculation module 300 may include:

[0143] The second historical monitoring data acquisition unit is used to acquire the historical monitoring data of the device under test within a preset historical time period.

[0144] The solution unit is used to solve for the false alarm rate weight base using nonlinear multivariate prediction and diagnostic techniques;

[0145] The second evaluation value calculation unit is used to calculate the second evaluation value corresponding to the current detection data using the false alarm rate weight base and the historical monitoring data.

[0146] Based on the above embodiments, the comprehensive evaluation value calculation module 400 may include:

[0147] The first acquisition unit is used to acquire a preset device importance weight library; the preset device importance weight library consists of weight values ​​corresponding to different device importance.

[0148] The first determining unit is configured to determine the first false alarm rate weight and the first false alarm rate weight corresponding to the device to be detected based on the preset device importance weight library;

[0149] The first comprehensive evaluation value calculation unit is used to calculate the comprehensive evaluation value based on the first false alarm rate weight and the first false negative rate weight.

[0150] Based on the above embodiments, the comprehensive evaluation value calculation module 400 may include:

[0151] The second acquisition unit is used to acquire a preset equipment operation scale weight library; the preset equipment operation rule weight library consists of weight values ​​corresponding to different equipment operation scales.

[0152] The second determining unit is used to determine the second false alarm rate weight and the second false alarm rate weight corresponding to the device to be detected based on the preset device operation scale weight library;

[0153] The second comprehensive evaluation value calculation unit is used to calculate the comprehensive evaluation value based on the second false alarm rate weight and the second false negative rate weight.

[0154] Based on the above embodiments, the equipment status detection device may further include:

[0155] The processing module is used to perform normalization and cleaning processing on the current monitoring data and the historical monitoring data.

[0156] Based on the above embodiments, the equipment status detection device may further include:

[0157] The deletion module is used to delete data under abnormal states and abnormal operating conditions from the historical monitoring data.

[0158] It should be noted that the order of the modules and units in the above-mentioned equipment status detection device can be changed without affecting the logic.

[0159] The device status detection device provided in this application includes: a current monitoring data acquisition module 100 for acquiring current monitoring data of the device under test; a first evaluation value calculation module 200 for calculating a first evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model; a second evaluation value calculation module 300 for calculating a second evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model; a comprehensive evaluation value calculation module 400 for weighted calculation of the first evaluation value and the second evaluation value to obtain a comprehensive evaluation value; and a difference calculation module 500 for calculating the difference between the comprehensive evaluation value and the current monitoring data. When the difference is greater than a preset threshold, the device under test is considered to be in an abnormal state. This application uses a low false alarm rate algorithm model and a low false alarm rate algorithm model to obtain the first evaluation value / false alarm value and the second evaluation value / false alarm value, respectively, and performs a comprehensive evaluation by fusing / weighting the data of the two models to take advantage of each. The comprehensive evaluation value is then used to calculate whether the device under test is abnormal based on the current detection data. This achieves simultaneous monitoring and ensures both a low false alarm rate and a low false alarm rate. Furthermore, this application proposes the concepts of a low false alarm rate algorithm and a low false alarm rate algorithm, and uses them in combination to determine the equipment status. It also proposes an adjustment strategy for the weight coefficient 'a' based on different types of equipment (equipment importance) or different operating operations (equipment scale), achieving a balance between the false alarm rate and the false alarm rate. Additionally, it uses a kernel regression algorithm as the low false alarm rate algorithm model, ensuring the reliability and convenience of low false alarm rate calculation. Furthermore, it performs normalization, data cleaning, and deletion of data under abnormal states and operating conditions on the collected data, improving the accuracy of detection and the reliability and accuracy of model training. Finally, it uses a nonlinear multivariate prediction and diagnostic technique to calculate the false alarm rate value (second evaluation value), ensuring the accuracy and reliability of low false alarm rate calculation.

[0160] The device status detection device provided in the embodiments of this application is described below. The device status detection device described below and the device status detection method described above can be referred to in correspondence.

[0161] Please refer to Figure 3 , Figure 3 A schematic diagram of a device status detection device provided in this application embodiment may include:

[0162] Memory 10 is used to store computer programs;

[0163] The processor 20 is used to execute a computer program to implement the device status detection method described above.

[0164] The memory 10, processor 20, and communication interface 31 all communicate with each other through the communication bus 32.

[0165] In this embodiment, the memory 10 is used to store one or more programs. The programs may include program code, which includes computer operation instructions. In this embodiment, the memory 10 may store programs for implementing the following functions:

[0166] Obtain the current monitoring data of the device under test;

[0167] The first evaluation value corresponding to the current monitoring data is calculated using a low false alarm rate algorithm model; the second evaluation value corresponding to the current monitoring data is calculated using a low false alarm rate algorithm model.

[0168] The first and second assessment values ​​are weighted to obtain the comprehensive assessment value.

[0169] Calculate the difference between the comprehensive evaluation value and the current monitoring data. If the difference is greater than the preset threshold, the device under test is considered to be in an abnormal state.

[0170] In one possible implementation, the memory 10 may include a program storage area and a data storage area, wherein the program storage area may store the operating system and applications required for at least one function; and the data storage area may store data created during use.

[0171] Furthermore, memory 10 may include read-only memory and random access memory, providing instructions and data to the processor. A portion of the memory may also include NVRAM. The memory stores operating systems and operating instructions, executable modules, or data structures, or subsets thereof, or extended sets thereof, wherein the operating instructions may include various operating instructions for implementing various operations. The operating system may include various system programs for implementing various basic tasks and handling hardware-based tasks.

[0172] Processor 20 can be a central processing unit (CPU), an application-specific integrated circuit, a digital signal processor, a field-programmable gate array, or other programmable logic device. Processor 20 can be a microprocessor or any conventional processor. Processor 20 can call programs stored in memory 10.

[0173] The communication interface 31 can be the interface of the communication module, used to connect with other devices or systems.

[0174] Of course, it should be noted that, Figure 3 The structure shown does not constitute a limitation on the device status detection device in the embodiments of this application. In practical applications, the device status detection device may include more than Figure 3 More or fewer components as shown, or combinations of certain components.

[0175] The storage medium provided in the embodiments of this application is described below. The storage medium described below can be referred to in correspondence with the device status detection method described above.

[0176] This application also provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the device status detection method described above.

[0177] The storage medium can include various media that can store program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0178] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0179] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0180] Finally, it should be noted that in this document, relationships such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0181] The above provides a detailed description of the device status detection method, apparatus, equipment, and storage medium provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for detecting equipment status, characterized in that, include: Obtain the current monitoring data of the device under test; The first evaluation value corresponding to the current monitoring data is calculated using a low false negative rate algorithm model; The second evaluation value corresponding to the current monitoring data is calculated using a low false alarm rate algorithm model; The first evaluation value and the second evaluation value are weighted and calculated to obtain a comprehensive evaluation value; Calculate the difference between the comprehensive evaluation value and the current monitoring data. When the difference is greater than a preset threshold, the device under test is in an abnormal state. The calculation of the first evaluation value corresponding to the current monitoring data using a low false negative rate algorithm model includes: Acquire historical monitoring data of the device under test within a preset historical time period; The distance vector is obtained by calculating the distance between the vectors using the current monitoring data and the historical monitoring data; The distance vector is transformed into a false negative rate weighting base using a Gaussian kernel function. Using the missed detection rate weight base and the historical monitoring data, the first evaluation value corresponding to the current monitoring data is calculated; The calculation of the second evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model includes: Acquire historical monitoring data of the device under test within a preset historical time period; The false alarm rate weight base is obtained by using a nonlinear multivariate predictive diagnostic technique. Using the false alarm rate weighting base and the historical monitoring data, the second evaluation value corresponding to the current monitoring data is calculated.

2. The equipment status detection method according to claim 1, characterized in that, The weighted calculation of the first evaluation value and the second evaluation value to obtain the comprehensive evaluation value includes: Obtain a preset device importance weight library; the preset device importance weight library consists of weight values ​​corresponding to different device importance. The first false alarm rate weight and the first false alarm rate weight corresponding to the device to be detected are determined according to the preset device importance weight library; The comprehensive evaluation value is calculated based on the first false alarm rate weight and the first false negative rate weight.

3. The equipment status detection method according to claim 1, characterized in that, The weighted calculation of the first evaluation value and the second evaluation value to obtain the comprehensive evaluation value includes: Obtain a preset equipment operation scale weight library; the preset equipment operation rule weight library consists of weight values ​​corresponding to different equipment operation scales; The second false alarm rate weight and the second false alarm rate weight corresponding to the device to be detected are determined according to the preset device operation scale weight library; The comprehensive evaluation value is calculated based on the second false alarm rate weight and the second false negative rate weight.

4. The equipment status detection method according to any one of claims 1 to 3, characterized in that, After acquiring the current monitoring data of the device under test and acquiring the historical monitoring data of the device under test within a preset historical time period, the method further includes: The current monitoring data and the historical monitoring data are normalized and cleaned.

5. The equipment status detection method according to claim 4, characterized in that, After acquiring the historical monitoring data of the device under test within a preset historical time period, the method further includes: Delete the data from the historical monitoring data that shows abnormal states and abnormal operating conditions.

6. A device for detecting equipment status, characterized in that, include: The current monitoring data acquisition module is used to acquire the current monitoring data of the device under test; The first evaluation value calculation module is used to calculate the first evaluation value corresponding to the current monitoring data using a low false negative rate algorithm model. The second evaluation value calculation module is used to calculate the second evaluation value corresponding to the current monitoring data using a low false alarm rate algorithm model. The comprehensive evaluation value calculation module is used to perform a weighted calculation on the first evaluation value and the second evaluation value to obtain a comprehensive evaluation value; The difference calculation module is used to calculate the difference between the comprehensive evaluation value and the current monitoring data. When the difference is greater than a preset threshold, the device under test is in an abnormal state. The first evaluation value calculation module includes: The first historical monitoring data acquisition unit is used to acquire the historical monitoring data of the device under test within a preset historical time period. The distance vector calculation unit is used to calculate the distance between vectors using the current monitoring data and the historical monitoring data to obtain a distance vector. The transformation unit is used to convert the distance vector into a false negative rate weighting base using a Gaussian kernel function. The first evaluation value calculation unit is used to calculate the first evaluation value corresponding to the current monitoring data using the missed detection rate weight base and the historical monitoring data. The second evaluation value calculation module includes: The second historical monitoring data acquisition unit is used to acquire the historical monitoring data of the device under test within a preset historical time period. The solution unit is used to solve for the false alarm rate weight base using nonlinear multivariate prediction and diagnostic techniques; The second evaluation value calculation unit is used to calculate the second evaluation value corresponding to the current monitoring data using the false alarm rate weight base and the historical monitoring data.

7. A device for detecting equipment status, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the device status detection method as described in any one of claims 1 to 5.

8. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the device status detection method as described in any one of claims 1 to 5.

Images