Fault identification method and system of intelligent electric meter

Abstract

The invention discloses a fault identification method and system of an intelligent electric meter, and belongs to the field of fault identification of the intelligent electric meter, and the method comprises the following steps: collecting the electric parameter data of a historical electric meter and the power consumption parameter of each electric device, and obtaining the standby power consumption parameter of the electric device; setting standby power consumption base lines in different time periods including working days and rest days, and triggering early warning when the deviation between the real-time power consumption and the base lines exceeds a preset threshold value; performing fault prediction according to the triggered early warning to obtain a specific fault; obtaining an abnormal reason of an abnormal point in the historical electricity meter electric parameter data, carrying out correlation analysis on the historical abnormal point and a predicted fault, obtaining the abnormal reason of the abnormal point with the highest correlation, and taking the abnormal reason as a fault reason after verification and outputting the abnormal reason; according to the invention, through obtaining the ammeter data, the power consumption of each device and the standby power consumption, the personalized baseline is established in combination with different time periods, time and seasons, the abnormality of the ammeter is accurately identified, and fault misjudgment caused by high standby power of the device is effectively avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of fault identification of smart meters, and in particular to a fault identification method and system for smart meters. Background Art

[0002] With the rapid development of smart grids, smart meters have been widely used in the field of electricity metering and monitoring. Existing smart meter fault identification technologies mostly use a unified standard baseline to judge meter anomalies. However, due to the significant differences in the types, quantities and usage habits of each household's electrical equipment, the power consumption and standby power consumption of the equipment are different. This "one-size-fits-all" fault identification setting method can easily lead to fault misjudgment. For example, when the standby power of the electrical equipment is high, it is often misjudged as equipment leakage, line short circuit or equipment aging. In addition, after the traditional method predicts potential faults, the lack of specific analysis is not conducive to extending the service life of the equipment and reducing operation and maintenance costs. In terms of fault troubleshooting, traditional technologies find it difficult to quickly locate the cause of the fault, resulting in low troubleshooting efficiency. Therefore, there is an urgent need for a smart meter fault identification method and system that can adapt to different power usage conditions, accurately identify faults, effectively predict potential faults and quickly locate the cause of the fault.

[0003] Currently, no effective solutions have been proposed for the problems in related technologies. Summary of the Invention

[0004] In response to the problems in the related art, the present invention proposes a fault identification method and system for a smart meter to overcome the problems of misjudgment of faults and time-consuming and labor-intensive fault problem analysis in the existing smart meter fault identification.

[0005] To this end, the specific technical solutions adopted in the present invention are as follows: A fault identification method for a smart meter, the method comprising the following steps: S1. Collect historical electric parameter data of electric meters and power consumption parameters of each electric device to obtain standby power consumption parameters of the electric device; S2. Set standby power consumption baselines for different time periods, including weekdays and weekends. When the deviation between the real-time power consumption and the baseline exceeds a preset threshold, an alert is triggered. S3. Perform fault prediction based on the triggered warning to obtain the specific fault; S4. Obtain the abnormal causes of abnormal points in the historical electric parameter data of the electric meter, and perform correlation analysis between the historical abnormal points and the predicted faults to obtain the abnormal cause of the abnormal point with the highest correlation, and output it as the fault cause after verification.

[0006] As a preferred embodiment, the collecting of historical electric meter electric parameter data and power consumption parameters of each electric device to obtain the standby power consumption parameters of the electric device includes the following steps: S11. Collect historical electrical parameter data in real time through the communication interface of the smart meter, including voltage, current, active power, reactive power, power factor and other parameters, and record the corresponding timestamp and meter identification information S12. Correlate the independent power consumption parameters of each electrical device through automated identification technology to obtain the power consumption characteristics of the device in the running, standby, and shutdown states; S13. For each electrical device, screen its power consumption records during a period without manual operation and activation by other automated programs, calculate the mean, standard deviation, and extreme value of the power consumption during this period, and form a standby power consumption benchmark parameter library for the device under different date types, including weekdays, weekends, and seasons.

[0007] By filtering power consumption data during periods without manual and automated intervention, interference from usage status can be eliminated, ensuring that the benchmark parameters truly reflect standby energy consumption.

[0008] As a preferred embodiment, the standby power consumption baselines for different time periods, including working days and weekends, are set. When the deviation between the real-time power consumption and the baseline exceeds a preset threshold, further monitoring and determining whether to trigger an early warning includes the following steps: S21. Classify historical data by weekdays and weekends, distinguish electricity usage patterns on different date types, and further divide each day into time periods, with 00:00-06:00 as the low-activity period, 06:00-18:00 as the regular period, and 18:00-24:00 as the high-activity period. Dynamically adjust the time period based on seasonal factors. S22. For each date type and time period, calculate the mean and standard deviation of historical standby power consumption, remove extreme values, and generate a baseline. For power-consuming devices with large fluctuations, use the sliding window mean as the baseline, and introduce environmental parameters to correct the baseline. S23. Match the corresponding baseline value according to the current date type, time period and environmental parameters, collect the current standby power consumption in real time, calculate the deviation value from the baseline value, and further monitor and determine whether to trigger an early warning if the single value exceeds ±15%.

[0009] There are significant differences in personnel activities and equipment usage frequency between weekdays and weekends, and the standby power consumption characteristics of equipment are also completely different. Therefore, it is necessary to establish different power consumption baselines according to different times to make anomaly identification more accurate.

[0010] As a preferred embodiment, the real-time collection of current standby power consumption, calculation of the deviation value from the baseline value, and further monitoring and determining whether to trigger an early warning when a single value exceeds ±15% include the following steps: S231. First, calculate the current standby power consumption and the percentage deviation from the baseline value. The specific formula is: ; in, is the calculated deviation rate, is the current standby power consumption, is the baseline value; when If the deviation rate exceeds ±20%, a five-minute short time window is set for continuous monitoring. If the deviation rate exceeds the limit continuously within five minutes, it is judged as abnormal and an early warning is triggered.

[0011] As a preferred embodiment, the fault prediction based on the triggered early warning to obtain the specific fault includes the following steps: S31. Extract real-time electrical parameter data from the warning period, including voltage fluctuations, current harmonics, and power factor anomalies. Combined with the historical operating status of the equipment, including power on / off records and standby time, to screen key features related to the warning. S32, comparing the extracted features with a preset fault feature library, and using a rule engine to match the fault type with the highest probability; S33: Control the power-consuming equipment, briefly switch its state, and observe whether the power consumption returns to normal. Verify whether it is a transient interference or a permanent fault. Combined with the status data of other devices under the same meter, determine whether the fault is caused by a single device or a grid-side problem. S34. Update the fault library model based on the verification result and output the specific fault type.

[0012] As a preferred embodiment, the abnormal causes of abnormal points in historical electric meter electrical parameter data are obtained, and the historical abnormal points are correlated with predicted faults to obtain the abnormal cause of the abnormal point with the highest correlation, which is used as the fault cause after verification and outputted, including the following steps: S41. Obtain abnormal points and specific abnormal causes of the abnormal points in historical electric parameter data of electric meters; S42. Analyze the correlation between historical abnormal points and predicted faults, and verify them. After verification, output the fault cause.

[0013] As a preferred embodiment, the analysis of the correlation between historical abnormal points and predicted faults, and verification, and outputting the fault cause after verification, includes the following steps: S421. Build a historical anomaly feature database and store the historical anomaly points by feature classification; S422, using the DTW distance to calculate the similarity between the predicted fault and the data in the abnormal point feature library, presetting a similarity threshold, and outputting abnormal points that are higher than the threshold; S423: Verify the output abnormal points in combination with the environment and usage. If the verification passes, the abnormal points and the causes of the abnormalities are output. If the verification fails, irrelevant abnormal points are eliminated.

[0014] A fault identification system for a smart meter, the system adopting any of the above-mentioned fault identification methods for a smart meter, comprising the following modules: a data acquisition module, a baseline comparison module, a fault prediction module, and a fault result output module, wherein the data acquisition module, the baseline comparison module, the fault prediction module, and the fault result output module are sequentially communicatively connected; The data acquisition module is used to collect the electrical parameter data of the electric meter and the power consumption parameters of each electrical device; The baseline comparison module establishes the standby power consumption baselines of different time periods as a benchmark to determine whether the power parameters are abnormal; The fault prediction module combines the triggered warning with the power consumption parameters to perform fault prediction and output the result; The fault result output module analyzes the correlation between historical abnormal points and predicted faults, and outputs the cause of the fault.

[0015] The beneficial effects of the present invention are: 1. The present invention can effectively identify specific meter anomalies by obtaining meter data, the power consumption and standby power consumption of each device, and establishing baselines according to different time periods. Since each household has different electrical equipment, the power consumption and power consumption parameters of each household's electrical equipment are also different. When the standby power of the electrical equipment is high, it may cause misjudgment of equipment leakage, line short circuit, equipment aging, etc. Therefore, by combining time and season to establish the standby power consumption parameters of the electrical equipment as a baseline, it can not only effectively adapt to the usage of different households, but also accurately reflect faults and prevent misjudgment of faults.

[0016] 2. The present invention can predict possible faults in advance by predicting fault points, allowing users to take protective measures in advance, thereby extending the service life of the equipment and reducing operation and maintenance costs. In addition, by analyzing the correlation between predicted faults and historical abnormal points, abnormal points associated with the predicted faults can be obtained. Outputting historical abnormal points can quickly obtain the abnormal cause of the predicted fault, simplifying the troubleshooting steps.

[0017] 3. By establishing baselines for different time periods, the present invention is not only applicable to different power usage situations, but also can prevent misjudgment of faults. In addition, by predicting faults based on the occurrence of anomalies, it is possible to understand in advance the possible faults that may occur in the later period. By analyzing the correlation between the predicted faults and historical abnormal points, the specific cause of the predicted fault can also be obtained, which is convenient for rapid troubleshooting. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 4 is a flow chart of a fault identification method for a smart meter according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention. They are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. By referring to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0021] According to an embodiment of the present invention, a fault identification method and system for a smart meter are provided.

[0022] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to the fault identification method and system of the smart meter according to the embodiment of the present invention, the method includes the following steps: S1. Collect historical electric parameter data of electric meters and power consumption parameters of each electric device to obtain standby power consumption parameters of the electric device; Furthermore, collecting historical electric parameter data of electric meters and power consumption parameters of each electric device to obtain standby power consumption parameters of the electric device includes the following steps: S11. Collect historical electrical parameter data in real time through the communication interface of the smart meter, including voltage, current, active power, reactive power, power factor and other parameters, and record the corresponding timestamp and meter identification information S12. Correlate the independent power consumption parameters of each electrical device through automated identification technology to obtain the power consumption characteristics of the device in the running, standby, and shutdown states; S13. For each electrical device, screen its power consumption records during a period without manual operation and activation by other automated programs, calculate the mean, standard deviation, and extreme value of the power consumption during this period, and form a standby power consumption benchmark parameter library for the device under different date types, including weekdays, weekends, and seasons.

[0023] It should be noted that for equipment with periodic changes, including air conditioners and water heaters, dynamic corrections need to be further combined with temperature and environmental factors. Through dynamic corrections, errors caused by temperature and environmental factors can be eliminated, making the benchmark parameters more scientific and practical.

[0024] S2. Set standby power consumption baselines for different time periods, including weekdays and weekends. When the deviation between the real-time power consumption and the baseline exceeds a preset threshold, an alert is triggered. Furthermore, different time periods, including standby power consumption baselines for weekdays and weekends, are set. When the deviation between the real-time power consumption and the baseline exceeds a preset threshold, further monitoring and determining whether to trigger an early warning includes the following steps: S21. Classify historical data by weekdays and weekends, distinguish electricity usage patterns on different date types, and further divide each day into time periods, with 00:00-06:00 as the low-activity period, 06:00-18:00 as the regular period, and 18:00-24:00 as the high-activity period. Dynamically adjust the time period based on seasonal factors. S22. For each date type and time period, calculate the mean and standard deviation of historical standby power consumption, remove extreme values, and generate a baseline. For power-consuming devices with large fluctuations, use the sliding window mean as the baseline, and introduce environmental parameters to correct the baseline. S23. Match the corresponding baseline value according to the current date type, time period and environmental parameters, collect the current standby power consumption in real time, calculate the deviation value from the baseline value, and further monitor and determine whether to trigger an early warning if the single value exceeds ±15%.

[0025] It should be noted that by establishing baselines at different times and using them as early warning standards, it is possible to finely distinguish electricity consumption behavior patterns of different date types and time periods, accurately capture the cyclical laws of equipment standby power consumption, and avoid misjudgments caused by ignoring differences in the time dimension based on fixed thresholds.

[0026] Furthermore, the current standby power consumption is collected in real time, and the deviation from the baseline value is calculated. If the single value exceeds ±15%, further monitoring and determination of whether to trigger an early warning include the following steps: S231. First, calculate the current standby power consumption and the percentage deviation from the baseline value. The specific formula is: ; in, is the calculated deviation rate, is the current standby power consumption, is the baseline value; when If the deviation rate exceeds ±20%, a five-minute short time window is set for continuous monitoring. If the deviation rate exceeds the limit continuously within five minutes, it is judged as abnormal and an early warning is triggered.

[0027] It should be noted that this short-term, high-frequency monitoring mode can quickly detect sudden failures, abnormal power consumption and other problems during equipment operation. Compared with long-period monitoring, it greatly shortens the problem exposure time, making it easier for operation and maintenance personnel to intervene and deal with it in the first time, reducing the risk of equipment damage and energy waste. At the same time, it can also effectively distinguish between real abnormal situations and short-term transient interference, reduce the occurrence of false alarms, and improve the accuracy of abnormality judgment.

[0028] S3. Perform fault prediction based on the triggered warning to obtain the specific fault; Furthermore, fault prediction is performed based on the triggered warning to obtain the specific fault, including the following steps: S31. Extract real-time electrical parameter data from the warning period, including voltage fluctuations, current harmonics, and power factor anomalies. Combined with the historical operating status of the equipment, including power on / off records and standby time, to screen key features related to the warning. S32, comparing the extracted features with a preset fault feature library, and using a rule engine to match the fault type with the highest probability; S33: Control the power-consuming equipment, briefly switch its state, and observe whether the power consumption returns to normal. Verify whether it is a transient interference or a permanent fault. Combined with the status data of other devices under the same meter, determine whether the fault is caused by a single device or a grid-side problem. S34. Update the fault library model based on the verification result and output the specific fault type.

[0029] S4. Obtain the abnormal causes of abnormal points in the historical electric parameter data of the electric meter, and perform correlation analysis between the historical abnormal points and the predicted faults to obtain the abnormal cause of the abnormal point with the highest correlation, and output it as the fault cause after verification.

[0030] Furthermore, the abnormal causes of abnormal points in the historical electric parameter data of the electric meter are obtained, and the historical abnormal points are correlated with the predicted faults to obtain the abnormal cause of the abnormal point with the highest correlation. After verification, it is used as the fault cause and outputted, including the following steps: S41. Obtain abnormal points and specific abnormal causes of the abnormal points in historical electric parameter data of electric meters; S42. Analyze the correlation between historical abnormal points and predicted faults, and verify them. After verification, output the fault cause. Furthermore, the correlation between historical abnormal points and predicted faults is analyzed and verified. After verification, the fault cause is output, which includes the following steps: S421. Build a historical anomaly feature database and store the historical anomaly points by feature classification; S422, using the DTW distance to calculate the similarity between the predicted fault and the data in the abnormal point feature library, presetting a similarity threshold, and outputting abnormal points that are higher than the threshold; S423: Verify the output abnormal points in combination with the environment and usage. If the verification passes, the abnormal points and the causes of the abnormalities are output. If the verification fails, irrelevant abnormal points are eliminated.

[0031] It should be noted that by calculating the similarity, we can accurately match predicted faults with historical anomalies and quickly identify potential related anomalies. At the same time, the preset similarity threshold to output highly correlated anomalies can narrow the scope of investigation. Finally, combined with the environment and usage verification, false matches caused by scenario differences can be eliminated to ensure that the output anomalies and causes are true and reliable, avoiding diagnostic errors.

[0032] A fault identification system for a smart meter, the system adopting any of the above fault identification methods for a smart meter, comprising the following modules: a data acquisition module, a baseline comparison module, a fault prediction module, and a fault result output module, wherein the data acquisition module, the baseline comparison module, the fault prediction module, and the fault result output module are sequentially communicatively connected; Data acquisition module, used to collect electrical parameter data of the electric meter and power consumption parameters of each electrical device; The baseline comparison module establishes the standby power consumption baseline in different time periods as a benchmark to determine whether the power parameters are abnormal; The fault prediction module combines the triggered warning with the power consumption parameters to predict the fault and output the results; Fault result output module, which analyzes the correlation between historical abnormal points and predicted faults, and outputs the fault causes In summary, the present invention can effectively identify specific meter anomalies by obtaining meter data, power consumption and standby power consumption of each device, and establishing baselines according to different time periods. Since each household has different electrical equipment, the power consumption of each household's electrical equipment and the power consumption parameters of the electrical equipment are also different. When the standby power of the electrical equipment is high, it may cause misjudgment of equipment leakage, line short circuit, equipment aging, etc. Therefore, by combining time and season to establish the standby power consumption parameters of the electrical equipment as a baseline, it can not only effectively adapt to the usage of different households, but also accurately reflect faults and prevent misjudgment of faults; the present invention can predict possible faults in advance by predicting the fault point Faults can be predicted, so that users can take protective measures in advance, thereby extending the service life of the equipment and reducing operation and maintenance costs. In addition, by analyzing the correlation between the predicted faults and historical abnormal points, abnormal points associated with the predicted faults can be obtained. Outputting the historical abnormal points can quickly obtain the abnormal cause of the predicted fault, simplifying the troubleshooting steps. The present invention establishes baselines for different time periods, which is not only applicable to different power usage conditions, but also can prevent the occurrence of misjudgment of faults. In addition, fault prediction based on the occurring abnormalities can understand in advance the possible faults that may occur in the later period. By analyzing the correlation between the predicted faults and historical abnormal points, the specific cause of the predicted fault can also be obtained, which is convenient for rapid troubleshooting.

[0033] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A fault identification method and system for a smart meter, characterized in that: The method comprises the following steps: S1. Collect historical electric parameter data of electric meters and power consumption parameters of each electric device to obtain standby power consumption parameters of the electric device; S2. Set standby power consumption baselines for different time periods, including weekdays and weekends. When the deviation between the real-time power consumption and the baseline exceeds a preset threshold, an alert is triggered. S3. Perform fault prediction based on the triggered warning to obtain the specific fault; S4. Obtain the abnormal causes of abnormal points in the historical electric parameter data of the electric meter, and perform correlation analysis between the historical abnormal points and the predicted faults to obtain the abnormal cause of the abnormal point with the highest correlation, and output it as the fault cause after verification.

2. The fault identification method and system for a smart meter according to claim 1, characterized in that: The collecting of historical electric parameter data of electric meters and power consumption parameters of each electric device to obtain the standby power consumption parameters of the electric device includes the following steps: S11. Collect historical electrical parameter data in real time through the communication interface of the smart meter, including voltage, current, active power, reactive power, power factor and other parameters, and record the corresponding timestamp and meter identification information S12. Correlate the independent power consumption parameters of each electrical device through automated identification technology to obtain the power consumption characteristics of the device in the running, standby, and shutdown states; S13. For each electrical device, screen its power consumption records during a period without manual operation and activation by other automated programs, calculate the mean, standard deviation, and extreme value of the power consumption during this period, and form a standby power consumption benchmark parameter library for the device under different date types, including weekdays, weekends, and seasons.

3. The fault identification method and system for a smart meter according to claim 1, characterized in that: The standby power consumption baselines are set for different time periods, including weekdays and weekends. When the deviation between the real-time power consumption and the baseline exceeds a preset threshold, further monitoring and determining whether to trigger an early warning includes the following steps: S21. Classify historical data by weekdays and weekends, distinguish electricity usage patterns on different date types, and further divide each day into time periods, with 00:00-06:00 as the low-activity period, 06:00-18:00 as the regular period, and 18:00-24:00 as the high-activity period. Dynamically adjust the time period based on seasonal factors. S22. For each date type and time period, calculate the mean and standard deviation of historical standby power consumption, remove extreme values, and generate a baseline. For power-consuming devices with large fluctuations, use the sliding window mean as the baseline, and introduce environmental parameters to correct the baseline. S23. Match the corresponding baseline value according to the current date type, time period and environmental parameters, collect the current standby power consumption in real time, calculate the deviation value from the baseline value, and further monitor and determine whether to trigger an early warning if the single value exceeds ±15%.

4. The fault identification method and system for a smart meter according to claim 3, characterized in that: The real-time collection of current standby power consumption, calculation of the deviation from the baseline value, and further monitoring and determining whether to trigger an early warning if the single value exceeds ±15% include the following steps: S231. First, calculate the current standby power consumption and the percentage deviation from the baseline value. The specific formula is: ； in, is the calculated deviation rate, is the current standby power consumption, is the baseline value; when If the deviation rate exceeds ±20%, a five-minute short time window is set for continuous monitoring. If the deviation rate exceeds the limit continuously within five minutes, it is judged as abnormal and an early warning is triggered.

5. The fault identification method and system for a smart meter according to claim 1, characterized in that: The fault prediction based on the triggered early warning to obtain the specific fault includes the following steps: S31. Extract real-time electrical parameter data from the warning period, including voltage fluctuations, current harmonics, and power factor anomalies. Combined with the historical operating status of the equipment, including power on / off records and standby time, to screen key features related to the warning. S32, comparing the extracted features with a preset fault feature library, and using a rule engine to match the fault type with the highest probability; S33: Control the power-consuming equipment, briefly switch its state, and observe whether the power consumption returns to normal. Verify whether it is a transient interference or a permanent fault. Combined with the status data of other devices under the same meter, determine whether the fault is caused by a single device or a grid-side problem. S34. Update the fault library model based on the verification result and output the specific fault type.

6. The fault identification method and system for a smart meter according to claim 1, characterized in that: The method of obtaining the abnormal causes of abnormal points in the historical electric parameter data of the electric meter, performing correlation analysis between the historical abnormal points and the predicted faults, obtaining the abnormal cause of the abnormal point with the highest correlation, and outputting it as the fault cause after verification includes the following steps: S41. Obtain abnormal points and specific abnormal causes of the abnormal points in historical electric parameter data of electric meters; S42. Analyze the correlation between historical abnormal points and predicted faults, and verify them. After verification, output the fault cause.

7. The fault identification method and system for a smart meter according to claim 6, characterized in that: The analysis of the correlation between historical abnormal points and predicted faults, and verification thereof, and outputting the fault cause after verification, includes the following steps: S421. Build a historical anomaly feature database and store the historical anomaly points by feature classification. S422, using the DTW distance to calculate the similarity between the predicted fault and the data in the abnormal point feature library, presetting a similarity threshold, and outputting abnormal points that are higher than the threshold; S423: Verify the output abnormal points in combination with the environment and usage. If the verification passes, output the abnormal points and the abnormal reasons. If the verification fails, remove the irrelevant abnormal points.

8. The fault identification system of the smart meter is characterized by: The system adopts the fault identification method of the smart meter according to any one of claims 1 to 7, comprising the following modules: a data acquisition module, a baseline comparison module, a fault prediction module and a fault result output module, wherein the data acquisition module, the baseline comparison module, the fault prediction module and the fault result output module are sequentially communicatively connected; The data acquisition module is used to collect the electrical parameter data of the electric meter and the power consumption parameters of each electrical device; The baseline comparison module establishes the standby power consumption baselines of different time periods as a benchmark to determine whether the power parameters are abnormal; The fault prediction module combines the triggered warning with the power consumption parameters to perform fault prediction and output the result; The fault result output module analyzes the correlation between historical abnormal points and predicted faults, and outputs the cause of the fault.

Images