Power metering equipment anomaly monitoring method and device, computer device and medium

Abstract

The application relates to the technical field of power equipment anomaly monitoring, and discloses a power metering equipment anomaly monitoring method and device, computer equipment and a medium. Prediction error data is determined based on equipment state information, equipment environment information and equipment use information, and then a prediction error curve is determined according to the prediction error data in a monitoring period. Actual error data is determined based on equipment metering data and target power station rated data, and then an actual error curve is determined according to the actual error data in the monitoring period. The power metering equipment anomaly monitoring is comprehensively and accurately analyzed, and the reliability of the prediction error data and the actual error data is improved. Then, a target abnormal value is determined based on the actual error data, the prediction error curve and the actual error curve. Finally, the monitoring abnormality degree of the power metering equipment is determined based on the target abnormal value. The abnormality degree of the power metering equipment is determined through the monitoring abnormality degree, and the accuracy of the power metering equipment anomaly monitoring is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power equipment anomaly monitoring, and in particular to a power metering equipment anomaly monitoring method and device, computer equipment and a medium. BACKGROUND

[0002] The power metering equipment is a set of devices for determining the amount of electrical energy used by users and monitoring parameters such as electrical energy power under the running state of the power system, so it is necessary to reasonably and effectively monitor the power metering equipment. However, in the related art, the power metering equipment anomaly monitoring technology has the problem of insufficient comprehensiveness when considering abnormal factors, resulting in the accuracy of the anomaly monitoring result to be improved. Therefore, a new power metering equipment anomaly monitoring method needs to be proposed. SUMMARY

[0003] The embodiments of the present specification aim to at least solve one of the technical problems in the related art. To this end, the embodiments of the present specification propose a power metering equipment anomaly monitoring method, device, computer equipment and medium.

[0004] The embodiments of the present specification provide a power metering equipment anomaly monitoring method, the method comprising:

[0005] According to the prediction error data in the monitoring period, a prediction error curve is determined; wherein the prediction error data is determined based on device state information, device environment information and device usage information;

[0006] According to the actual error data in the monitoring period, an actual error curve is determined; wherein the actual error data is determined based on device metering data and target power station rated data;

[0007] Based on the actual error data, the prediction error curve and the actual error curve, a target abnormal value is determined;

[0008] Based on the target abnormal value, a monitoring abnormality degree of the power metering equipment is determined.

[0009] In one of the embodiments, the device state information includes device temperature, device noise value and device vibration frequency, the device environment information includes environment temperature and environment humidity, and the device usage information includes device usage time length; based on the device state information, the device environment information and the device usage information, the prediction error data is determined, comprising:

[0010] According to the device temperature, the device noise value and the device vibration frequency, a first prediction influence coefficient is determined;

[0011] According to the environment temperature and the environment humidity, a second prediction influence coefficient is determined;

[0012] determining a third prediction influence coefficient according to the length of time the device has been used;

[0013] determining the prediction error data based on the first prediction influence coefficient, the second prediction influence coefficient, and the third prediction influence coefficient.

[0014] In one of the embodiments, the device metering data comprises power station metering power, and the target power station rated data comprises power station running power; determining the actual error data based on the device metering data and the target power station rated data comprises: determining the actual error data based on the power station metering power and the power station running power.

[0015] In one of the embodiments, the target abnormal value comprises a first abnormal value; determining the first abnormal value based on the prediction error curve comprises:

[0016] determining a prediction extreme value, a prediction curve slope, a prediction slope change amplitude, and a prediction curve change frequency based on the prediction error curve;

[0017] determining the first abnormal value based on the prediction extreme value, the prediction curve slope, the prediction slope change amplitude, and the prediction curve change frequency.

[0018] In one of the embodiments, the target abnormal value comprises a second abnormal value; determining the second abnormal value based on the actual error curve comprises:

[0019] determining an actual extreme value, an actual curve slope, an actual slope change amplitude, and an actual curve change frequency based on the actual error curve;

[0020] determining the second abnormal value based on the actual extreme value, the actual curve slope, the actual slope change amplitude, and the actual curve change frequency.

[0021] In one of the embodiments, the target abnormal value comprises a third abnormal value; determining the third abnormal value based on the prediction error curve and the actual error curve comprises:

[0022] determining an abnormal period proportion, an abnormal curve difference mean value, and an abnormal curve difference proportion based on the prediction error curve and the actual error curve;

[0023] determining the third abnormal value based on the abnormal period proportion, the abnormal curve difference mean value, and the abnormal curve difference proportion.

[0024] In one of the embodiments, the determining of the abnormal period proportion, the abnormal curve difference average value, and the abnormal curve difference proportion based on the predicted error curve and the actual error curve comprises:

[0025] The device further comprises a coincidence part curve determining module, configured to determine a coincidence part curve and a non-coincidence part curve based on the predicted error curve and the actual error curve.

[0026] The device further comprises an abnormal curve difference determining module, configured to determine an abnormal curve difference based on the coincidence part curve and the non-coincidence part curve.

[0027] The device further comprises an abnormal period proportion determining module, an abnormal curve difference average value determining module, and an abnormal curve difference proportion determining module, which are configured to determine the abnormal period proportion, the abnormal curve difference average value, and the abnormal curve difference proportion based on the abnormal curve difference.

[0028] In one of the embodiments, the determining of the abnormal curve difference based on the coincidence part curve and the non-coincidence part curve comprises:

[0029] The device further comprises a curve coincidence degree determining module, configured to determine a curve coincidence degree based on the coincidence part curve.

[0030] The device further comprises a curve difference value determining module, configured to determine a curve difference value of the non-coincidence part curve in a case where the curve coincidence degree does not satisfy a preset curve coincidence degree condition.

[0031] The abnormal curve difference determining module is further configured to determine the abnormal curve difference based on the curve difference value of the non-coincidence part curve.

[0032] In one of the embodiments, the target abnormal value comprises a fourth abnormal value, and the determining of the fourth abnormal value based on the actual error data comprises:

[0033] The device further comprises a fourth abnormal value determining module, configured to determine the fourth abnormal value based on the actual error data and a preset standard error threshold value in a case where the actual error data satisfies the preset standard error threshold value.

[0034] In one of the embodiments, the determining of the monitoring abnormality degree of the power metering device based on the target abnormal value comprises:

[0035] The device further comprises a device monitoring abnormal value determining module, configured to determine a device monitoring abnormal value based on the target abnormal value.

[0036] The device further comprises a monitoring abnormality degree determining module, configured to determine the monitoring abnormality degree of the power metering device based on the device monitoring abnormal value.

[0037] The embodiments of the present specification provide a device for monitoring abnormality of a power metering device, which comprises:

[0038] The device further comprises a predicted error determining module, configured to determine a predicted error curve according to predicted error data in a monitoring period, wherein the predicted error data is determined based on device state information, device environment information, and device usage information.

[0039] The actual error determination module is configured to determine an actual error curve according to actual error data in a monitoring period, wherein the actual error data is determined based on device measurement data and target power station rated data.

[0040] The target abnormal value determination module is configured to determine a target abnormal value based on the actual error data, the predicted error curve and the actual error curve.

[0041] The monitoring abnormality determination module is configured to determine a monitoring abnormality of the power measurement device based on the target abnormal value.

[0042] In one of the embodiments, the device state information includes device temperature, device noise value and device vibration frequency, the device environment information includes environment temperature and environment humidity, and the device usage information includes device usage time length; the predicted error determination module is further configured to determine a first predicted influence coefficient according to the device temperature, the device noise value and the device vibration frequency, determine a second predicted influence coefficient according to the environment temperature and the environment humidity, determine a third predicted influence coefficient according to the device usage time length, and determine the predicted error data based on the first predicted influence coefficient, the second predicted influence coefficient and the third predicted influence coefficient.

[0043] In one of the embodiments, the device measurement data includes power station measurement power, and the target power station rated data includes power station operation power; the actual error determination module is further configured to determine the actual error data based on the power station measurement power and the power station operation power.

[0044] An embodiment of the present specification provides a computer device, which comprises a memory and one or more processors in communication connection with the memory; the memory stores instructions executable by the one or more processors, and the instructions are executed by the one or more processors to enable the one or more processors to implement the steps of the method according to any one of the above embodiments.

[0045] An embodiment of the present specification provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the method according to any one of the above embodiments.

[0046] An embodiment of the present specification provides a computer program product, which comprises instructions, and the instructions are executed by a processor of a computer device to enable the computer device to implement the steps of the method according to any one of the above embodiments.

[0047] In the above embodiment, the prediction error data is determined based on the device state information, the device environment information and the device usage information, and then the prediction error curve is determined according to the prediction error data in the monitoring period. The actual error data is determined based on the device metering data and the target power station rated data, and then the actual error curve is determined according to the actual error data in the monitoring period. Then, the target abnormal value is determined based on the actual error data, the prediction error curve and the actual error curve. Finally, the monitoring abnormality degree of the power metering device is determined based on the target abnormal value. The power metering device abnormality monitoring is comprehensively and accurately analyzed based on the device state information, the device environment information and the device usage information, and based on the device metering data and the target power station rated data, and the reliability of the prediction error data and the actual error data is improved. The abnormality degree of the power metering device is determined by the monitoring abnormality degree, and the accuracy of the power metering device abnormality monitoring is improved. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 A flowchart of a power metering device abnormality monitoring method provided by the embodiment of the present specification is shown in the figure;

[0049] Figure 2 A flowchart of determining prediction error data provided by the embodiment of the present specification is shown in the figure;

[0050] Figure 3 A flowchart of determining a first abnormal value provided by the embodiment of the present specification is shown in the figure;

[0051] Figure 4 A flowchart of determining a second abnormal value provided by the embodiment of the present specification is shown in the figure;

[0052] Figure 5 A flowchart of determining a third abnormal value provided by the embodiment of the present specification is shown in the figure;

[0053] Figure 6 A flowchart of determining abnormal data provided by the embodiment of the present specification is shown in the figure;

[0054] Figure 7 A flowchart of determining abnormal curve difference provided by the embodiment of the present specification is shown in the figure;

[0055] Figure 8 A flowchart of determining the monitoring abnormality degree of the power metering device provided by the embodiment of the present specification is shown in the figure;

[0056] Figure 9 A schematic diagram of a power metering device abnormality monitoring device provided by the embodiment of the present specification is shown in the figure;

[0057] Figure 10 An internal structure diagram of a computer device provided by the embodiment of the present specification is shown in the figure. DETAILED DESCRIPTION

[0058] Embodiments of the present application are described in detail below with reference to the attached drawing figures, wherein the same or like component have the same or similar designations or reference characters throughout the attached drawing figures. The embodiments described below are presented by way of example to explain the present application, and are not intended to limit the present application.

[0059] Power metering equipment is a set of equipment for determining the amount of electrical energy used by users and monitoring parameters such as electrical energy power under the operating state of the power system. Power metering equipment is an important part of the process of power generation, transmission, distribution and use, mainly including electric energy meter, transformer, voltage and current sensor for metering, acquisition terminal and other equipment. Therefore, reasonable and effective abnormal monitoring of power metering equipment can identify and diagnose potential problems in the operation of the equipment in a timely manner, thereby ensuring the stability and reliability of the equipment. This not only helps to ensure the normal operation of the power metering equipment itself, but also has important significance for the overall safety and stable operation of the power system. In related technologies, the power metering equipment abnormal monitoring technology has the problem of insufficient comprehensiveness when considering abnormal factors, resulting in the accuracy of the abnormal monitoring result to be improved.

[0060] Based on this, the embodiments of the present application provide a power metering equipment abnormal monitoring method. Based on the device state information, the device environment information and the device usage information, the prediction error data is determined, and then the prediction error curve is determined according to the prediction error data in the monitoring period. Based on the device metering data and the target power station rated data, the actual error data is determined, and then the actual error curve is determined according to the actual error data in the monitoring period. Then, based on the actual error data, the prediction error curve and the actual error curve, the target abnormal value is determined. Finally, based on the target abnormal value, the monitoring abnormality degree of the power metering equipment is determined. By comprehensively and accurately analyzing the power metering equipment abnormal monitoring based on the device state information, the device environment information and the device usage information, and based on the device metering data and the target power station rated data, the reliability of the prediction error data and the actual error data is improved. The abnormality degree of the power metering equipment is determined by the monitoring abnormality degree, and the accuracy of the power metering equipment abnormal monitoring is improved.

[0061] The embodiments of the present application provide a power metering equipment abnormal monitoring method, please refer to Figure 1 , the method can include the following steps:

[0062] S110, determining a prediction error curve according to prediction error data in a monitoring period.

[0063] Among them, the prediction error data is determined based on the device state information, the device environment information and the device usage information.

[0064] Specifically, since the device state information, the device environment information and the device usage information all belong to external factors of the power metering device, rather than real data directly measured by the power metering device itself. Therefore, the prediction error data corresponding to the power metering device is determined based on the device state information, the device environment information and the device usage information. First, according to the specific situation of the device operation and the monitoring requirement, a suitable monitoring period is determined. In the monitoring period, the device state information, the device environment information and the device usage information at multiple time points are collected. Then, at each time point, the prediction error data corresponding to each time point is determined based on the device state information, the device environment information and the device usage information. Finally, the multiple prediction error data obtained in the monitoring period are sorted in time sequence, and a prediction error curve is drawn.

[0065] It should be noted that the monitoring period of the power metering device corresponds to the power station operation time length of the target power station. For example, if the power station operation time length of the target power station is 1 hour, the monitoring period of the power metering device is also 1 hour.

[0066] S120, determining an actual error curve according to the actual error data in the monitoring period.

[0067] The actual error data is determined based on the device metering data and the target power station rated data.

[0068] Specifically, the device metering data is directly measured by the power metering device, and the target power station rated data is the design and operation standard of the power station. Therefore, the actual error data corresponding to the power metering device is determined based on the device metering data and the target power station rated data. In the monitoring period, the device metering data at multiple time points is collected. Then, at each time point, the actual error data corresponding to each time point is determined based on the device metering data and the target power station rated data. Finally, the multiple actual error data obtained in the monitoring period are sorted in time sequence, and an actual error curve is drawn.

[0069] S130, determining a target abnormal value based on the actual error data, the prediction error curve and the actual error curve.

[0070] S140, determining a monitoring abnormality degree of the power metering device based on the target abnormal value.

[0071] Specifically, when determining whether the power metering device is abnormal, the characteristics of the device itself and its external factors need to be considered comprehensively. Therefore, the target abnormal value measuring the abnormality of the power metering device can be determined in combination with the actual error data, the prediction error curve and the actual error curve. Then, the monitoring abnormality degree measuring the abnormality of the power metering device is determined through the target abnormal value.

[0072] In the power metering equipment anomaly monitoring method, the prediction error data is determined based on the equipment state information, the equipment environment information and the equipment use information, then the prediction error curve is determined according to the prediction error data in the monitoring period. The actual error data is determined based on the equipment metering data and the target power station rated data, then the actual error curve is determined according to the actual error data in the monitoring period. Then, the target anomaly value is determined based on the actual error data, the prediction error curve and the actual error curve. Finally, the monitoring anomaly degree of the power metering equipment is determined based on the target anomaly value. The power metering equipment anomaly monitoring is comprehensively and accurately analyzed based on the equipment state information, the equipment environment information and the equipment use information, and based on the equipment metering data and the target power station rated data, the reliability of the prediction error data and the actual error data is improved. The abnormality degree of the power metering equipment is determined through the monitoring anomaly degree, and the accuracy of the power metering equipment anomaly monitoring is improved.

[0073] In some embodiments, referring to Figure 2 , the equipment state information includes the equipment temperature, the equipment noise value and the equipment vibration frequency, the equipment environment information includes the environment temperature and the environment humidity, and the equipment use information includes the equipment used time length; the prediction error data is determined based on the equipment state information, the equipment environment information and the equipment use information, which can include the following steps:

[0074] S210, a first prediction influence coefficient is determined according to the equipment temperature, the equipment noise value and the equipment vibration frequency.

[0075] S220, a second prediction influence coefficient is determined according to the environment temperature and the environment humidity.

[0076] S230, a third prediction influence coefficient is determined according to the equipment used time length.

[0077] S240, the prediction error data is determined based on the first prediction influence coefficient, the second prediction influence coefficient and the third prediction influence coefficient.

[0078] Specifically, the device temperature, the device noise value and the device vibration frequency directly affect the operation stability of the power metering device. For example, high temperature can cause the device performance to decrease, and abnormal noise value and vibration frequency can indicate the wear or failure of mechanical components. During the monitoring period, the device temperature, the noise value and the vibration frequency at each time point are combined to calculate a comprehensive prediction influence coefficient, and a first prediction influence coefficient corresponding to each time point is obtained. Environmental factors have a significant impact on the performance of the device. For example, extreme environmental temperature and humidity can accelerate the aging of the device or affect its measurement accuracy. During the monitoring period, according to the environmental temperature and the environmental humidity at each time point, a comprehensive prediction influence coefficient is calculated, and a second prediction influence coefficient corresponding to each time point is obtained. The usage time of the device affects its performance. Long time use can cause the aging or performance decrease of components, thereby affecting the measurement accuracy. During the monitoring period, based on the usage time of the device at each time point, a prediction influence coefficient is calculated, and a third prediction influence coefficient corresponding to each time point is obtained. The first prediction influence coefficient, the second prediction influence coefficient and the third prediction influence coefficient corresponding to each time point are combined to calculate the prediction error data corresponding to each time point.

[0079] For example, during the monitoring period, the device temperature corresponding to each time point of the power metering device is denoted as , the device noise value is denoted as , and the device vibration frequency is denoted as . The first prediction influence coefficient corresponding to each time point is calculated by the following formula:

[0080]

[0081] wherein, is the first prediction influence coefficient, , , is a preset proportion factor.

[0082] During the monitoring period, the environmental temperature corresponding to each time point of the power metering device is denoted as , and the environmental humidity is denoted as . The second prediction influence coefficient corresponding to each time point is calculated by the following formula:

[0083]

[0084] wherein, is the second prediction influence coefficient, , is a preset proportion factor.

[0085] During the monitoring period, the usage time of the power metering device corresponding to each time point is denoted as The third predictive impact coefficient for each time point is calculated using the following formula:

[0086]

[0087] in, This is the third predictive impact coefficient. This is a preset scaling factor.

[0088] Then, the prediction error data for each time point is calculated using the following formula:

[0089]

[0090] in, For prediction error data, The first predictive impact coefficient, This is the second predictive impact coefficient. This is the third predictive impact coefficient. , , This is a preset scaling factor.

[0091] It should be noted that the preset scaling factor , , , , , , , , All of these can be determined based on historical experimental data and experience.

[0092] For example, if the temperature of the power metering equipment The temperature is 20 degrees Celsius, and the equipment noise level is... The vibration frequency of the equipment is 40 decibels. It is 20 Hz. If the preset scaling factor is... , , If the values ​​are 0.5, 0.5, and 1 respectively, then the first predictive influence coefficient... .

[0093] If the ambient temperature of the power metering equipment The temperature is 25 degrees Celsius, and the ambient humidity is... The RH is 20%. If the preset scaling factor is... , If the values ​​are 1 and 1 respectively, then the second predictive influence coefficient... .

[0094] If the power metering equipment has been in use for a long time For 1 year. If the preset scaling factor... is 20, the third prediction influence coefficient .

[0095] If the preset proportion factor , , is 0.5, 0.6, 0.5 respectively, the prediction error data .

[0096] In the above power metering equipment anomaly monitoring method, the first prediction influence coefficient is determined according to the equipment temperature, the equipment noise value and the equipment vibration frequency, the second prediction influence coefficient is determined according to the environment temperature and the environment humidity, multiple factors are comprehensively considered, the third prediction influence coefficient is determined according to the equipment usage time length, and the accuracy of the prediction error data is determined based on the first prediction influence coefficient, the second prediction influence coefficient and the third prediction influence coefficient.

[0097] In some embodiments, the equipment metering data includes power station metering power, and the target power station rated data includes power station operating power; the actual error data is determined based on the equipment metering data and the target power station rated data, which can include: determining the actual error data based on the power station metering power and the power station operating power.

[0098] Specifically, the power station metering power refers to the actual power value of the target power station counted by the power metering equipment, which represents the actual energy output of the power station at a specific time point. The metering power and the operating power of the power station are key data for evaluating the actual working state and performance of the power station equipment. By comparing these data, the running efficiency and accuracy of the equipment can be evaluated. Within the monitoring period, a comprehensive actual data is calculated based on the power station operating power and the power station metering power at each time point, to obtain the actual error data corresponding to each time point.

[0099] It should be noted that the monitoring period of the power metering equipment corresponds to the power station operating time length of the target power station, that is, the monitoring period of the power metering equipment is set according to the power station operating time length of the target power station, for example, if the power station operating time length of the target power station is 1 hour, the monitoring period of the power metering equipment is also 1 hour.

[0100] Exemplarily, within the monitoring period, the power station metering power corresponding to each time point of the power metering equipment is recorded as , and the power station operating power is recorded as . The actual error data corresponding to each time point is calculated by the following formula:

[0101]

[0102] wherein, is the actual error data, is a preset proportion factor.

[0103] For example, if the power station operation power of the target power station is 10 MW, the power station measurement power is 9.8 MW, and the preset ratio factor is 100, the actual error data is 0.2%.

[0104] In the power measurement equipment anomaly monitoring method, the reliability of the actual error data is improved based on the power station measurement power and the power station operation power.

[0105] In some embodiments, referring to Figure 3 , the target anomaly value includes a first anomaly value; and determining the first anomaly value based on the prediction error curve can include the following steps:

[0106] S310, determining a prediction extreme value, a prediction curve slope, a prediction slope change amplitude, and a prediction curve change frequency based on the prediction error curve.

[0107] S320, determining the first anomaly value based on the prediction extreme value, the prediction curve slope, the prediction slope change amplitude, and the prediction curve change frequency.

[0108] Specifically, in the monitoring period, the plurality of prediction error data obtained in the monitoring period are sorted in chronological order to draw the prediction error curve. By determining the key features of the prediction error curve, including the prediction extreme value, the prediction curve slope, the prediction slope change amplitude, and the prediction curve change frequency, the performance of the prediction model and its stability and reliability are evaluated.

[0109] The maximum and minimum values of the prediction error curve in the monitoring period are identified by analyzing the prediction error curve, and then the maximum and minimum values in the monitoring period are subtracted to obtain the prediction extreme value. The maximum curve slope of the prediction error curve in the monitoring period is calculated by differentiating the prediction error curve to obtain the prediction curve slope. The maximum and minimum curve slopes of the prediction error curve in the monitoring period are determined by differentiating the prediction error curve, and then the maximum and minimum curve slopes in the monitoring period are subtracted to obtain the prediction slope change amplitude. The inflection points in the prediction error curve, i.e. the points where the curve changes from rising to falling or from falling to rising, are analyzed to obtain the prediction curve change frequency. Finally, the prediction extreme value, the prediction curve slope, the prediction slope change amplitude, and the prediction curve change frequency are combined to calculate a comprehensive anomaly data to obtain the first anomaly value.

[0110] For example, the prediction extreme value of the prediction error curve is denoted as , and the prediction curve slope of the prediction error curve is denoted as ​​​​, the prediction curve change frequency of the prediction error curve is recorded as , the prediction slope change amplitude of the prediction error curve is recorded as The first abnormal value is calculated by the following formula:

[0111]

[0112] , wherein is the first abnormal value, , , , is a preset proportion factor.

[0113] For example, the maximum value of the prediction error curve is 4MW, and the minimum value is 2MW, so the prediction extreme value of the prediction error curve is 2MW. The maximum curve slope of the prediction error curve in the monitoring period is 5, and the minimum curve slope is -3, so the prediction slope change amplitude of the prediction error curve is 8.

[0114] If the prediction extreme value of the prediction error curve is 2MW, the prediction curve slope of the prediction error curve is 5, the prediction curve change frequency of the prediction error curve is 5 times / h, and the prediction slope change amplitude of the prediction error curve is 8. If the preset proportion factor , , , is 1, 1, 1, and 1 respectively, the first abnormal value .

[0115] In the above power metering equipment anomaly monitoring method, based on the prediction error curve, the prediction extreme value, the prediction curve slope, the prediction slope change amplitude, and the prediction curve change frequency are determined, and based on the prediction extreme value, the prediction curve slope, the prediction slope change amplitude, and the prediction curve change frequency, the first abnormal value is determined, which provides a basis for further determining the target abnormal value.

[0116] In some embodiments, referring to Figure 4 , the target abnormal value includes a second abnormal value; based on the actual error curve, the second abnormal value is determined, which can include the following steps:

[0117] S410, based on the actual error curve, determining an actual extreme value, an actual curve slope, an actual slope change amplitude, and an actual curve change frequency.

[0118] S420, based on the actual extreme value, the actual curve slope, the actual slope change amplitude, and the actual curve change frequency, determining the second abnormal value.

[0119] Specifically, in the monitoring period, the plurality of actual error data obtained in the monitoring period are sorted in chronological order to draw an actual error curve. By determining the key features of the actual error curve, including the actual extreme value, the actual curve slope, the actual slope change amplitude, and the actual curve change frequency, the performance of the actual model and its stability and reliability are evaluated.

[0120] The maximum and minimum values of the actual error curve in the monitoring period are identified by analyzing the actual error curve, and then the actual extreme value is obtained by subtracting the maximum value from the minimum value in the monitoring period. The maximum curve slope in the monitoring period is calculated by differentiating the actual error curve to obtain the actual curve slope. The maximum and minimum curve slopes in the monitoring period are determined by differentiating the actual error curve, and then the actual slope change amplitude is obtained by subtracting the maximum curve slope from the minimum curve slope in the monitoring period. The inflection points in the actual error curve, i.e. the points where the curve changes from rising to falling or from falling to rising, are analyzed to obtain the actual curve change frequency. Finally, the actual extreme value, the actual curve slope, the actual slope change amplitude, and the actual curve change frequency are combined to calculate a comprehensive abnormal data to obtain a second abnormal value.

[0121] Exemplarily, the actual extreme value of the actual error curve is denoted as , the actual curve slope of the actual error curve is denoted as , the actual curve change frequency of the actual error curve is denoted as , and the actual slope change amplitude of the actual error curve is denoted as The second abnormal value is calculated by the following formula:

[0122]

[0123] wherein, is the second abnormal value, , , , is a preset proportion factor.

[0124] For example, the maximum value of the actual error curve is 6 MW, and the actual error minimum value is 3 MW, so the actual extreme value of the actual error curve is 3 MW. The maximum curve slope of the actual error curve in the monitoring period is 7, and the minimum curve slope is -3, so the actual slope change amplitude of the actual error curve is 10.

[0125] If the actual extreme value of the actual error curve is is 3MW, actual curve slope of actual error curve is 6, actual curve frequency of actual error curve is 6 / h, actual slope change amplitude of actual error curve is 10. If preset proportion factors , , , are 1, 1, 1, 1 respectively, the second abnormal value .

[0126] In the above power metering equipment anomaly monitoring method, the actual extreme value, the actual curve slope, the actual slope change amplitude, and the actual curve frequency are determined based on the actual error curve, and the second abnormal value is determined by using the actual extreme value, the actual curve slope, the actual slope change amplitude, and the actual curve frequency, thereby providing a basis for further determining the target abnormal value.

[0127] In some embodiments, referring to Figure 5 , the target abnormal value includes a third abnormal value; the third abnormal value is determined based on the predicted error curve and the actual error curve, which can include the following steps:

[0128] S510, determining an abnormal period proportion, an abnormal curve difference mean value, and an abnormal curve difference proportion based on the predicted error curve and the actual error curve.

[0129] S520, determining the third abnormal value based on the abnormal period proportion, the abnormal curve difference mean value, and the abnormal curve difference proportion.

[0130] Specifically, by comparing the error between the predicted value and the actual value, the abnormal situation in the equipment operation can be identified. According to the predicted error curve and the actual error curve, the abnormal period proportion, the abnormal curve difference mean value, and the abnormal curve difference proportion are calculated. Then, the abnormal period proportion, the abnormal curve difference mean value, and the abnormal curve difference proportion are integrated to calculate a comprehensive abnormal data, and the third abnormal value is obtained.

[0131] Exemplarily, the abnormal curve difference mean value is denoted as , the abnormal curve difference proportion is denoted as , and the abnormal period proportion is denoted as . The third abnormal value is calculated by the following formula:

[0132]

[0133] wherein, is the second abnormal value, , , is a preset proportion factor.

[0134] For example, if the mean difference of the outlier curve For 2MW, the percentage of abnormal curve differences The percentage of abnormal cycles is 20%. It is 20%. If the preset scaling factor... , , If the values ​​are 10, 1, and 1 respectively, then the third outlier is... .

[0135] In the above-mentioned method for monitoring anomalies in power metering equipment, the percentage of abnormal cycles, the average difference of abnormal curves, and the percentage of abnormal curve differences are determined based on the predicted error curve and the actual error curve. Based on the percentage of abnormal cycles, the average difference of abnormal curves, and the percentage of abnormal curve differences, a third abnormal value is determined, providing a basis for further determining the target abnormal value.

[0136] In some implementations, please refer to Figure 6 Determining the percentage of abnormal periods, the mean difference between abnormal curves, and the percentage of abnormal curve differences based on the predicted error curve and the actual error curve may include the following steps:

[0137] S610. Based on the predicted error curve and the actual error curve, determine the overlapping and non-overlapping curves.

[0138] S620. Determine the difference between the abnormal curves based on the overlapping and non-overlapping curves.

[0139] S630. Determine the percentage of abnormal cycles, the mean of abnormal curve differences, and the percentage of abnormal curve differences based on the abnormal curve differences.

[0140] Specifically, the predicted error curve and the actual error curve are compared to identify the overlapping portion on the time axis, thus determining the overlapping curve. The non-overlapping portion is also identified. If the overlapping curve does not meet specific conditions (e.g., the error range exceeds a set threshold), the difference between the actual and predicted error curves is calculated using the non-overlapping curve to determine the abnormal curve difference. The abnormal monitoring period corresponding to the abnormal curve difference is determined. Then, the abnormal monitoring period is divided by the monitoring period to obtain the abnormal period percentage. The average of at least one determined abnormal monitoring period is calculated to obtain the average abnormal curve difference. Finally, the ratio of the number of abnormal curve differences to the total number of curve differences is calculated to obtain the abnormal curve difference percentage.

[0141] In some implementations, the predicted error curve and the actual error curve are compared. If the error between the predicted error curve and the actual error curve is within the allowable error range, the curves can be considered to coincide, i.e., the overlapping part of the curve.

[0142] In the power metering equipment anomaly monitoring method, the coincident part curve and the non-coincident part curve are determined based on the prediction error curve and the actual error curve, the anomaly curve difference is determined based on the coincident part curve and the non-coincident part curve, the anomaly period proportion, the anomaly curve difference mean value, and the anomaly curve difference proportion are determined based on the anomaly curve difference, which provides a basis for subsequent determination of the third anomaly value and further determination of the target anomaly value.

[0143] In some embodiments, referring to Figure 7 The anomaly curve difference can be determined based on the coincident part curve and the non-coincident part curve, which can include the following steps:

[0144] S710, determining the curve coincidence degree based on the coincident part curve.

[0145] S720, in the case where the curve coincidence degree does not satisfy the preset curve coincidence degree condition, determining the curve difference of the non-coincident part curve.

[0146] S730, determining the anomaly curve difference based on the curve difference of the non-coincident part curve.

[0147] Specifically, after determining the coincident part curve and the non-coincident part curve, the period corresponding to the coincident part curve is recorded as the coincident period, and the period corresponding to the non-coincident part curve is recorded as the non-coincident period. The coincident period is divided by the monitoring period to obtain the coincident period proportion, and the curve coincidence degree between the prediction error curve and the actual error curve is obtained according to the coincident period proportion. The curve coincidence degree is compared with the preset curve coincidence degree condition, and in the case where the curve coincidence degree satisfies the preset curve coincidence degree condition, subsequent operations are not required. In the case where the curve coincidence degree does not satisfy the preset curve coincidence degree condition, the difference between the prediction error curve and the actual error curve of the non-coincident part is calculated to obtain the curve difference of the non-coincident part curve. The curve difference of the non-coincident part curve is compared with the preset curve difference threshold, and the curve difference greater than the preset curve difference threshold is recorded as the anomaly curve difference.

[0148] In some embodiments, the preset curve coincidence degree condition can be a preset curve coincidence degree threshold. The curve coincidence degree is compared with the preset curve coincidence degree threshold, and in the case where the curve coincidence degree is greater than or equal to the preset curve coincidence degree threshold, the preset curve coincidence degree condition is not satisfied, and it is determined that the power metering equipment does not have the third anomaly value. In the case where the curve coincidence degree is less than the preset curve coincidence degree threshold, the preset curve coincidence degree condition is satisfied, and it is determined that the power metering equipment has the third anomaly value.

[0149] For example, the monitoring period is 50 minutes, and the coincidence period corresponding to the coincident part curve is 40 minutes. The coincidence period is divided by the monitoring period to obtain a coincidence period ratio of 80%. Then, the curve coincidence degree between the prediction error curve and the actual error curve according to the coincidence period ratio is 80.

[0150] In the monitoring period, the prediction error data of the non-coincident part curve at the first time is 4 MW, and the actual error data of the actual error curve of the non-coincident part is 2 MW. Therefore, the curve difference of the non-coincident part curve at the first time is 2 MW.

[0151] In the above power metering equipment anomaly monitoring method, the curve coincidence degree is determined based on the coincident part curve. In the case where the curve coincidence degree does not satisfy the preset curve coincidence degree condition, the curve difference of the non-coincident part curve is determined. Based on the curve difference of the non-coincident part curve, the abnormal curve difference is determined, which provides a basis for subsequent determination of the third abnormal value and further determination of the target abnormal value.

[0152] In some embodiments, the target abnormal value includes a fourth abnormal value. Determining the fourth abnormal value based on the actual error data can include: in the case where the actual error data satisfies the preset standard error threshold, determining the fourth abnormal value based on the actual error data and the preset standard error threshold.

[0153] Specifically, the preset standard error threshold is used to define the abnormal value according to the distribution characteristics of the data. In the monitoring period, the actual error data corresponding to each time point is compared with the preset standard error threshold. In the case where the actual error data does not satisfy the preset standard error threshold, i.e., the actual error data is less than the preset standard error threshold, it is considered that there is no fourth abnormal value. In the case where the actual error data satisfies the preset standard error threshold, i.e., the actual error data is greater than or equal to the preset standard error threshold, it is considered that there is a fourth abnormal value. The actual error data satisfying the preset standard error threshold and the preset standard error threshold are calculated to obtain a comprehensive abnormal data, and the fourth abnormal value is obtained.

[0154] For example, the actual error data satisfying the preset standard error threshold is denoted as , and the preset standard error threshold is denoted as . The fourth abnormal value is calculated by the following formula:

[0155]

[0156] wherein, is the first abnormal value, is a preset proportion factor, is the first actual error data satisfying the preset standard error threshold, and the fourth abnormal value is determined based on the actual error data, the preset standard error threshold, and the third abnormal value. the fourth abnormal value is determined based on the actual error data, the preset standard error threshold, and the third abnormal value.

[0157] For example, if the actual error data satisfying the preset standard error threshold is 5MW, the preset standard error threshold is 3MW, and the preset ratio factor is 10, the fourth abnormal value is 50MW. For example, if the actual error data satisfying the preset standard error threshold is 5MW, the preset standard error threshold is 3MW, and the preset ratio factor is 10, the fourth abnormal value is 50MW. For example, if the actual error data satisfying the preset standard error threshold is 5MW, the preset standard error threshold is 3MW, and the preset ratio factor is 10, the fourth abnormal value is 50MW. For example, if the actual error data satisfying the preset standard error threshold is 5MW, the preset standard error threshold is 3MW, and the preset ratio factor is 10, the fourth abnormal value is 50MW. .

[0158] In the power metering equipment anomaly monitoring method, when the actual error data satisfies the preset standard error threshold, the fourth abnormal value is determined based on the actual error data, the preset standard error threshold, and the third abnormal value, which provides a basis for further determining the target abnormal value.

[0159] In some embodiments, referring to Figure 8 based on the target abnormal value, the monitoring abnormality degree of the power metering equipment can include the following steps:

[0160] S810, based on the target abnormal value, determining the equipment monitoring abnormal value.

[0161] S820, based on the equipment monitoring abnormal value, determining the monitoring abnormality degree of the power metering equipment.

[0162] Specifically, the target abnormal value includes the first abnormal value, the second abnormal value, the third abnormal value, and the fourth abnormal value. By using the target abnormal value, the equipment monitoring abnormal value is determined. Then, according to the equipment monitoring abnormal value, the monitoring abnormality degree reflecting the operation stability and reliability of the power metering equipment is determined.

[0163] It should be noted that the greater the monitoring abnormality degree of the power metering equipment, the more serious the monitoring abnormality degree of the power metering equipment, indicating that the equipment has a greater operation risk.

[0164] For example, the equipment monitoring abnormal value is calculated by the following formula:

[0165]

[0166] wherein, the equipment monitoring abnormal value, , , , is a preset ratio factor, is the first abnormal value, is the second abnormal value, is the third abnormal value, is the fourth abnormal value.

[0167] The monitoring anomaly degree is calculated using the following formula:

[0168]

[0169] in, To monitor anomalies, This is a preset scaling factor.

[0170] For example, if the first abnormal value of the power metering equipment The value is 20, the second outlier. The value is 25, the third outlier. The value is 50, the fourth outlier. The default scaling factor is 10. , , , If the values ​​are 0.5, 1, 0.7, and 1 respectively, then the abnormal values ​​monitored by the equipment are... .

[0171] If the power metering equipment monitoring shows abnormal values The preset scaling factor is 80. If the value is 0.1, then the anomaly detection rate is 0.1. .

[0172] In the above-mentioned method for monitoring anomalies in power metering equipment, the abnormal value of the equipment is determined based on the target abnormal value, the monitoring anomaly degree of the power metering equipment is determined based on the monitoring anomaly value, and the degree of anomaly of the power metering equipment is determined by the monitoring anomaly degree, thereby improving the accuracy of anomaly monitoring of power metering equipment.

[0173] This specification provides an embodiment of a power metering equipment anomaly monitoring device 900. Please refer to [link / reference]. Figure 9 The power metering equipment anomaly monitoring device 900 includes: a prediction error determination module 910, an actual error determination module 920, a target anomaly value determination module 930, and a monitoring anomaly degree determination module 940.

[0174] The prediction error determination module 910 is used to determine the prediction error curve based on the prediction error data within the monitoring period; wherein the prediction error data is determined based on equipment status information, equipment environment information, and equipment usage information.

[0175] The actual error determination module 920 is used to determine the actual error curve based on the actual error data within the monitoring period; wherein the actual error data is determined based on equipment metering data and the rated data of the target power station.

[0176] The target outlier determination module 930 is used to determine the target outlier based on the actual error data, the prediction error curve, and the actual error curve.

[0177] The monitoring abnormality determination module 940 is configured to determine a monitoring abnormality of the power metering device based on the target abnormal value.

[0178] In some embodiments, the device state information includes a device temperature, a device noise value, and a device vibration frequency, the device environment information includes an environment temperature and an environment humidity, and the device usage information includes a device usage duration; the prediction error determination module 910 is further configured to determine a first prediction influence coefficient according to the device temperature, the device noise value, and the device vibration frequency, determine a second prediction influence coefficient according to the environment temperature and the environment humidity, and determine a third prediction influence coefficient according to the device usage duration; and determine the prediction error data based on the first prediction influence coefficient, the second prediction influence coefficient, and the third prediction influence coefficient.

[0179] In some embodiments, the device metering data includes a power station metering power, and the target power station rated data includes a power station running power; and the actual error determination module 920 is further configured to determine the actual error data based on the power station metering power and the power station running power.

[0180] For a specific description of the power metering device abnormality monitoring apparatus, refer to the description of the power metering device abnormality monitoring method above, which will not be repeated here.

[0181] An embodiment of the present specification provides a computer device, including a memory and a processor, the memory stores a computer program, and the processor implements the method steps in the above embodiments when executing the computer program.

[0182] An embodiment of the present specification provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the method in any one of the above embodiments.

[0183] An embodiment of the present specification provides a computer program product, which includes instructions, and the instructions are executed by a processor of a computer device to enable the computer device to perform the steps of the method in any one of the above embodiments.

[0184] In some embodiments, a computer device is provided, which can be a terminal, and an internal structure diagram of the terminal can be as shown in FIG. 8. Figure 10As shown in the figure. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium, an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. Wireless mode can be achieved through WIFI, operator network, NFC (near field communication) or other technologies. The computer program is executed by the processor to implement a power metering device anomaly monitoring method. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0185] Those skilled in the art can understand that, Figure 10 The skilled in the art can understand that,

[0186] It is to be appreciated that the logical and / or steps represented in the flow diagrams, or otherwise described herein, can be considered as a sequence of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions, or a combination thereof. For purposes of this specification, a "computer-readable medium" can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The computer-readable medium can be a product of "transitory fabrication" that transits from one place to another place. More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electronic connection having one or more wires (electronic devices), a portable computer diskette (magnetic device), a random-access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable medium upon which the program is printed, as the program can be electronically captured, for example via an optical scanner, then compiled, interpreted, or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.

Claims

1. A method for monitoring anomalies in power metering equipment, characterized in that, The method includes: A prediction error curve is determined based on the prediction error data within the monitoring period. The prediction error data is determined based on equipment status information, equipment environment information, and equipment usage information. The equipment status information includes equipment temperature, equipment noise level, and equipment vibration frequency. The equipment environment information includes ambient temperature and ambient humidity. The equipment usage information includes the equipment's usage time. Determining the prediction error data based on the equipment status information, the equipment environment information, and the equipment usage information includes: determining a first prediction influence coefficient based on the equipment temperature, the equipment noise level, and the equipment vibration frequency; determining a second prediction influence coefficient based on the ambient temperature and the ambient humidity; determining a third prediction influence coefficient based on the equipment's usage time; and determining the prediction error data based on the first prediction influence coefficient, the second prediction influence coefficient, and the third prediction influence coefficient. Based on the actual error data within the monitoring period, an actual error curve is determined; wherein, the actual error data is determined based on equipment metering data and target power plant rated data, the equipment metering data includes the power plant's metered power, and the target power plant's rated data includes the power plant's operating power; determining the actual error data based on the equipment metering data and target power plant rated data includes: determining the actual error data based on the power plant's metered power and the power plant's operating power; Based on the actual error data, the predicted error curve, and the actual error curve, a target outlier is determined. The target outlier includes a first outlier. Determining the first outlier based on the predicted error curve includes: determining the predicted extreme value, the predicted curve slope, the magnitude of the predicted slope change, and the predicted curve change frequency based on the predicted error curve; and determining the first outlier based on the predicted extreme value, the predicted curve slope, the magnitude of the predicted slope change, and the predicted curve change frequency. Based on the target outlier, the monitoring anomaly degree of the power metering equipment is determined.

2. The method for abnormal monitoring of power metering equipment according to claim 1, characterized in that, The target outlier includes a second outlier; Based on the actual error curve, the second outlier is determined, including: Based on the actual error curve, determine the actual extreme value, the actual curve slope, the actual slope change range, and the actual curve change frequency. The second outlier is determined based on the actual extreme value, the actual curve slope, the magnitude of the actual slope change, and the actual curve change frequency.

3. The method for abnormal monitoring of power metering equipment according to claim 1, characterized in that, The target outlier includes a third outlier; Based on the predicted error curve and the actual error curve, the third outlier is determined, including: Based on the predicted error curve and the actual error curve, the percentage of abnormal periods, the mean difference of abnormal curves, and the percentage of abnormal curve differences are determined. The third outlier is determined based on the percentage of outlier cycles, the average difference of outlier curves, and the percentage of outlier curve differences.

4. The method for abnormal monitoring of power metering equipment according to claim 3, characterized in that, The step of determining the percentage of abnormal periods, the mean difference of abnormal curves, and the percentage of abnormal curve differences based on the predicted error curve and the actual error curve includes: Based on the predicted error curve and the actual error curve, the overlapping and non-overlapping curves are determined. Based on the overlapping curves and the non-overlapping curves, the difference between the abnormal curves is determined; The percentage of abnormal cycles, the mean of the abnormal curve differences, and the percentage of abnormal curve differences are determined based on the abnormal curve differences.

5. The method for abnormal monitoring of power metering equipment according to claim 4, characterized in that, The step of determining the difference between the overlapping and non-overlapping curves includes: Based on the overlapping portion of the curves, the degree of curve overlap is determined; If the curve overlap does not meet the preset curve overlap condition, determine the curve difference of the non-overlapping part of the curve. The abnormal curve difference is determined based on the curve difference of the non-overlapping part of the curve.

6. The method for abnormal monitoring of power metering equipment according to claim 1, characterized in that, The target outlier includes a fourth outlier; Determining the fourth outlier based on the actual error data includes: If the actual error data meets the preset standard error threshold, the fourth outlier is determined based on the actual error data and the preset standard error threshold.

7. The method for abnormal monitoring of power metering equipment according to claim 1, characterized in that, The determination of the monitoring anomaly degree of the power metering equipment based on the target anomaly value includes: Based on the target anomaly, determine the equipment monitoring anomaly; Based on the abnormal values ​​monitored by the device, the monitoring abnormality degree of the power metering device is determined.

8. A device for monitoring abnormalities in power metering equipment, characterized in that, The power metering equipment anomaly monitoring device includes: A prediction error determination module is used to determine a prediction error curve based on prediction error data within a monitoring period. The prediction error data is determined based on equipment status information, equipment environment information, and equipment usage information. The equipment status information includes equipment temperature, equipment noise level, and equipment vibration frequency. The equipment environment information includes ambient temperature and ambient humidity. The equipment usage information includes the equipment's usage time. Determining the prediction error data based on the equipment status information, the equipment environment information, and the equipment usage information includes: determining a first prediction influence coefficient based on the equipment temperature, the equipment noise level, and the equipment vibration frequency; determining a second prediction influence coefficient based on the ambient temperature and the ambient humidity; determining a third prediction influence coefficient based on the equipment's usage time; and determining the prediction error data based on the first prediction influence coefficient, the second prediction influence coefficient, and the third prediction influence coefficient. The actual error determination module is used to determine the actual error curve based on the actual error data within the monitoring period. The actual error data is determined based on equipment metering data and the target power station's rated data. The equipment metering data includes the power station's metered power, and the target power station's rated data includes the power station's operating power. Determining the actual error data based on the equipment metering data and the target power station's rated data includes determining the actual error data based on the power station's metered power and the power station's operating power. The target outlier determination module is used to determine target outliers based on the actual error data, the prediction error curve, and the actual error curve. The target outliers include a first outlier. Determining the first outlier based on the prediction error curve includes: determining a prediction extreme value, a prediction curve slope, a prediction slope change magnitude, and a prediction curve change frequency based on the prediction error curve; and determining the first outlier based on the prediction extreme value, the prediction curve slope, the prediction slope change magnitude, and the prediction curve change frequency. The monitoring anomaly determination module is used to determine the monitoring anomaly degree of the power metering equipment based on the target anomaly value.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images