Power transformation operation and maintenance monitoring and early warning system and early warning method

Abstract

The invention discloses a power transformation operation and maintenance monitoring and early warning system and method, and relates to the technical field of power transformation equipment early warning, and the system comprises a partial discharge sensor, a temperature sensor, a vibration sensor, and a data processing unit. The partial discharge sensor is used for monitoring a partial discharge signal of the equipment; the vibration sensor is used for monitoring a vibration acceleration signal of the equipment; the temperature sensor is used for detecting a temperature sequence signal of the equipment; and the data processing unit is used for carrying out data conversion, feature extraction, threshold comparison and data prediction on the various signals and generating alarm information based on the relevance experimental data of the power transformation equipment. The method has the effect of reducing the misjudgment probability of the substation operation and maintenance monitoring and early warning system.

Description

Technical Field

[0001] The present application relates to the technical field of early warning technology for substation equipment, and in particular to a substation operation and maintenance monitoring and early warning system and an early warning method. Background Art

[0002] With the rapid development of the power industry and the rapid iteration of related technologies, substations and substation equipment are becoming increasingly important in the power system. As key nodes in power transmission and distribution, the operating status of substations and substation equipment is directly related to the stability and reliability of power system operations.

[0003] In order to improve the stability of substation operation, substations are usually equipped with relevant operation and maintenance monitoring and early warning systems to monitor the operation process of the substation in real time. This system can not only immediately activate relevant emergency plans for power failures that have occurred, but also issue early warnings for possible power accidents based on relevant power monitoring data.

[0004] However, existing substation operation and maintenance early warning systems can usually only issue early warning information based on a single power data combined with relevant thresholds for comparison, especially between interrelated power data. For example, during the normal operation of substation equipment, when a certain power monitoring data has reached the alarm threshold, another related power monitoring data has not reached the alarm threshold. At this time, the alarm action has been triggered, but in fact the substation equipment will not have any abnormal phenomenon, which increases the probability of false alarm. Summary of the Invention

[0005] In order to reduce the probability of misjudgment in a substation operation and maintenance monitoring and early warning system, the present application provides a substation operation and maintenance monitoring and early warning system and an early warning method.

[0006] In the first aspect, the present application provides a substation operation and maintenance monitoring and early warning system, which adopts the following technical solutions:

[0007] A substation operation and maintenance monitoring and early warning system, comprising a partial discharge sensor, a temperature sensor, a vibration sensor, and a data processing unit;

[0008] The partial discharge sensor is used to monitor the partial discharge signal of the equipment;

[0009] The vibration sensor is used to monitor the vibration acceleration signal of the equipment;

[0010] The temperature sensor is used to detect the temperature sequence signal of the device;

[0011] The data processing unit is used to perform data conversion, feature extraction, threshold comparison, data prediction and generate alarm information on the above-mentioned various signals based on the correlation experimental data of the substation equipment.

[0012] The above technical solution simulates the operating status of substation equipment under different conditions and monitors various experimental data corresponding to alarms, which require processing and recording. These data include partial discharge signals (PD signals), vibration acceleration signals, and temperature series signals. Based on the correlation between the different data properties in these scenarios, the relevant data is classified and used as trigger conditions for response measures in different scenarios.

[0013] For example, in the simulation operation of the substation equipment, when an alarm occurs, the local discharge signal, vibration acceleration signal and temperature sequence signal corresponding to the moment the alarm occurs are recorded. Since the above three signals exist at the same time when the alarm occurs, the three signals at the moment the alarm occurs are the correlation experimental data when the alarm occurs in the substation equipment. Therefore, the three signals at the moment the alarm occurs are used as the early warning trigger conditions for the alarm of the substation equipment.

[0014] During the actual operation of substation equipment detected by this early warning system, by monitoring the above three types of signals and performing feature extraction and data prediction on the above three types of signals, the signal characteristics of the above three types of signals after the next time period can be known. If the predicted three types of signal characteristics are consistent with the preset experimental data related to the occurrence of alarms, it means that the alarm is very likely to occur. At this time, an alarm message is generated to alert the operator to take relevant treatment measures to prevent accidents before they happen.

[0015] Since this early warning system uses various types of correlated experimental data to determine whether an alarm is likely to occur, compared to the traditional prediction and judgment based on a single data, it effectively reduces the probability of misjudgment of the substation operation and maintenance monitoring and early warning system, thereby reducing the phenomenon of decreased power supply stability caused by misjudgment and early warning, and thus improving power supply stability.

[0016] In a second aspect, based on the above-mentioned substation operation and maintenance monitoring and early warning system, the present application also provides a substation operation and maintenance monitoring and early warning method.

[0017] A method for monitoring and early warning of substation operation and maintenance, comprising the following processing steps:

[0018] Obtain partial discharge signals, temperature sequence signals, and vibration acceleration signals of substation equipment;

[0019] generating a partial discharge time-frequency diagram based on the partial discharge signal, and extracting the partial discharge time-frequency signal from the partial discharge time-frequency diagram;

[0020] Inputting the partial discharge time-frequency graph into a convolutional neural network model, and outputting a discharge probability from the convolutional neural network model;

[0021] The temperature sequence signal is input into the memory network model, and the memory network model outputs the predicted temperature;

[0022] Calculating a root mean square value of vibration acceleration based on the vibration acceleration signal;

[0023] Set graded alarms based on discharge probability, predicted temperature, and RMS vibration acceleration.

[0024] Through the above technical solution, after obtaining the partial discharge signal, temperature sequence signal and vibration acceleration signal, the partial discharge time-frequency signal extracted based on the partial discharge signal is input into the convolutional neural network model to output the required predicted discharge probability; the temperature sequence signal is input into the memory network model to output the predicted temperature; and based on the vibration acceleration signal, the root mean square value of the vibration acceleration is calculated.

[0025] In the experimental phase of the substation equipment, the various types of monitored correlation experimental data are divided into levels according to the different measures that need to be processed, and the correlation experimental data of different levels are set as the trigger conditions for different levels of alarms of the substation equipment.

[0026] By setting up a graded alarm mechanism, operators can take different measures according to the actual situation based on different levels of alarm information, so that they can respond to possible alarm situations more quickly, reasonably and scientifically.

[0027] In a preferred example, the present application may be further configured such that the setting of graded alarms includes:

[0028] If the predicted temperature is greater than 120°C and the RMS value of the vibration acceleration is greater than 5m / S 2 , then trigger the first level alarm;

[0029] If 110℃<predicted temperature≤120℃, or the predicted temperature exceeds the temperature threshold, a level 2 alarm is triggered;

[0030] If 100℃<predicted temperature≤110℃, or the discharge probability is greater than the probability threshold, a level 3 alarm is triggered.

[0031] Through the above technical solution, the triggering conditions of different levels of alarms in this technical solution are formulated based on the actual alarm triggering conditions measured by the substation equipment in the experimental phase, which can fit the actual operating status of the corresponding substation equipment, thereby improving the accuracy of alarm prediction.

[0032] In a preferred example, the present application can be further configured as follows:

[0033] The triggering action of the first-level alarm is to cut off the power supply of the device and send a text message to the operation and maintenance terminal;

[0034] The triggering action of the secondary alarm is to generate a work order and push it to the mobile terminal;

[0035] The triggering action of the third level alarm is to mark the device status and record the log.

[0036] Through the above technical solution, if the predicted data reaches the triggering conditions of the first-level alarm, the triggering action of the first-level alarm will be triggered, that is, cutting off the power supply of the equipment and sending a text message to the operation and maintenance terminal. The power supply of the equipment is directly cut off before the alarm occurs to prevent the damage caused by the actual occurrence of the fault, and the operation and maintenance terminal is notified via text message to remind the operation and maintenance personnel of the occurrence of the first-level alarm.

[0037] If the predicted data reaches the triggering conditions of the second-level alarm, the triggering action of the second-level alarm will be triggered, that is, a work order will be generated and pushed to the mobile terminal. Although the second-level alarm does not require cutting off the power supply of the equipment, for safety reasons, operation and maintenance personnel are still required to conduct on-site inspections and checks of the equipment.

[0038] If the predicted data reaches the triggering conditions of the third-level alarm, the triggering action of the third-level alarm will be triggered, that is, marking the equipment status and recording the log. The third-level alarm is usually understood as a general alarm. It is only necessary to record the operating status and related data of the equipment in the book to provide a reference for inspection and inspection for operation and maintenance personnel.

[0039] In a preferred example, the present application can be further configured as follows:

[0040] Get the predicted alarm level to which the predicted temperature belongs. If the predicted alarm level is not a level 1 alarm, get the alarm occurrence result. If an alarm occurs, get the predicted temperature, actual alarm temperature, and predicted deviation threshold.

[0041] Based on the predicted temperature, the actual warning temperature, and the prediction deviation threshold, determining whether a predicted temperature difference between the predicted temperature and the actual warning temperature is less than the prediction deviation threshold;

[0042] If the predicted temperature difference is less than the predicted deviation threshold, the temperature value, vibration acceleration root mean square value and discharge probability corresponding to the actual alarm temperature are added to the trigger conditions of the previous level of predicted alarm level.

[0043] Through the above technical solution, if the predicted temperature belongs to a non-level one alarm (such as a level two alarm and a level three alarm), taking the level three alarm as an example, if the predicted temperature belongs to the level three alarm and an alarm occurs, then based on the predicted temperature, the actual alarm temperature and the predicted deviation threshold, it is determined whether the predicted temperature difference between the predicted temperature and the actual alarm temperature is less than the predicted deviation threshold, that is, the predicted temperature and the actual alarm temperature (the actual temperature of the substation where the alarm occurs) are subtracted, the absolute value is taken, and then this absolute value is compared with the predicted deviation threshold.

[0044] If the absolute value of the predicted temperature difference is less than the predicted deviation threshold, it means that the temperature prediction has good accuracy. The temperature prediction can still be used as one of the main warning data bases. The temperature value, root mean square value of vibration acceleration and discharge probability corresponding to the actual alarm temperature are added to the trigger conditions of the second-level prediction alarm level. By upgrading the trigger conditions that originally belonged to the third-level alarm to the trigger conditions of the second-level alarm, the importance of the predicted temperature of the above-mentioned alarm is increased, and more cautious handling measures are taken to reduce the probability of similar alarms occurring.

[0045] In addition, when the trigger conditions that originally belonged to the third-level alarm jump to the trigger conditions of the second-level alarm, the trigger conditions that originally belonged to the third-level alarm will be deleted from the trigger conditions of the third-level alarm to avoid system confusion.

[0046] In a preferred example, the present application can be further configured such that if the predicted temperature is less than 100°C and an alarm occurs, the temperature value, root mean square value of vibration acceleration and discharge probability corresponding to the alarm occurring when the predicted temperature is less than 100°C are obtained and added to the triggering conditions of the third-level alarm.

[0047] Through the above technical solution, since the corresponding level alarm is not set when the predicted temperature is less than 100°C, there is still a possibility that an alarm may occur in the substation equipment. In this case, the temperature value, root mean square value of vibration acceleration and discharge probability corresponding to the alarm when the predicted temperature is less than 100°C are added to the trigger conditions of the third-level alarm, thereby increasing the importance of the predicted temperature for the above-mentioned alarm. When such an alarm occurs, it is recorded in the equipment log for the operator to view and perform further related processing, such as manually performing a transition process of the trigger condition.

[0048] In a preferred example, the present application can be further configured to record the newly added triggering conditions for transitioning from the third-level alarm to the second-level alarm, corresponding to the interval duration of the alarm occurrence. If the interval duration is less than the duration interval threshold, the newly added triggering conditions for transitioning from the third-level alarm to the second-level alarm will be added to the triggering conditions of the first-level alarm.

[0049] Through the above technical solution, if the new trigger condition for the transition from the third-level alarm to the second-level alarm, the interval between the occurrence of the alarm is less than the time interval threshold (the interval threshold between two adjacent alarms), it means that the relevant data of this new trigger condition has appeared frequently and has triggered an alarm. Therefore, it is necessary to further increase the importance of this new trigger condition. At this time, the new trigger condition for the transition from the third-level alarm to the second-level alarm will be added to the trigger condition of the first-level alarm, which can also increase the importance of the new trigger condition. When this new trigger condition is met again, the first-level alarm will be directly triggered, that is, the power supply of the equipment will be cut off, and a text message will be sent to the operation and maintenance terminal.

[0050] In a preferred example, the present application can be further configured as follows:

[0051] Get the trigger frequency of the newly added trigger conditions of the first-level alarm and the trigger frequency of the initial trigger conditions of the first-level alarm;

[0052] If the trigger frequency of the newly added trigger condition of the first level alarm reaches the trigger frequency of the initial trigger condition of the first level alarm, maintenance information of the substation equipment is generated;

[0053] After the maintenance is completed based on the maintenance information, the newly added triggering condition of the first-level alarm is reduced to the triggering condition of the second-level alarm;

[0054] The triggering conditions for downgrading to the second-level alarm will be reduced to the triggering conditions of the third-level alarm if the corresponding non-triggering time is greater than the downgrade time threshold.

[0055] Through the above technical solution, if the trigger frequency of the newly added trigger conditions of the first-level alarm reaches the trigger frequency of the initial trigger conditions of the first-level alarm, it means that the newly added trigger conditions have been triggered frequently and need to be repaired in time. At this time, maintenance information of the substation equipment is generated to alert the operator to inspect and maintain the substation equipment.

[0056] After the operator completes the operation and maintenance of the substation equipment, the newly added trigger conditions of the first-level alarm are downgraded to the trigger conditions of the second-level alarm. After the trigger conditions of the second-level alarm are downgraded, if the corresponding non-triggering time is greater than the downgrade time threshold (the time required for the newly added trigger conditions to be downgraded to the next level of alarm), the trigger conditions of the second-level alarm will be downgraded to the trigger conditions of the third-level alarm. By downgrading the newly added trigger conditions step by step, rather than directly removing the newly added trigger conditions from the multi-level alarm, even after the maintenance of the substation equipment has been completed, a certain level of vigilance against the newly added trigger conditions can be retained.

[0057] In summary, this application has the following beneficial technical effects:

[0058] 1. This early warning system uses various types of correlation experimental data to determine whether an alarm situation is likely to occur. Compared with the traditional prediction and judgment based on a single data, it effectively reduces the probability of misjudgment of the substation operation and maintenance monitoring and early warning system, thereby reducing the phenomenon of power supply stability decline caused by misjudgment and early warning, and thus improving power supply stability;

[0059] 2. This early warning system sets different alarm levels according to the different measures required by the substation equipment, refines and improves the operating logic of the early warning system for substation equipment, and improves the accuracy of the response strategies adopted by operators;

[0060] 3. By setting up a transition mechanism for the newly added trigger conditions, the newly added trigger conditions will transition to a higher level of alarm level, thereby increasing the importance of the predicted temperature for the above-mentioned alarm situations, and then taking more cautious handling measures to reduce the probability of similar alarm situations occurring. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a structural diagram of the substation operation and maintenance monitoring and early warning system in an embodiment of the present application.

[0062] Figure 2 It is a flow chart of the substation operation and maintenance monitoring and early warning method in the embodiment of the present application.

[0063] Figure 3 It is a flowchart of the newly added trigger condition alarm level transition in the embodiment of the present application.

[0064] Figure 4 This is a flowchart of adding a trigger condition for degradation after the substation equipment maintenance is completed in an embodiment of the present application. DETAILED DESCRIPTION

[0065] The following is combined with Figure 1 -Attached Figure 4 This application is described in further detail.

[0066] The embodiment of the present application discloses a substation operation and maintenance monitoring and early warning system. Based on the monitoring and early warning system, the present application also discloses a substation operation and maintenance monitoring and early warning method.

[0067] Refer to the attached Figure 1 As shown, a substation operation and maintenance monitoring and early warning system includes a partial discharge sensor, a temperature sensor, a vibration sensor, and a data processing unit. The partial discharge sensor, temperature sensor, and vibration sensor are all installed on the substation equipment to be monitored.

[0068] Partial discharge sensors monitor equipment's partial discharge signals. Vibration sensors monitor equipment's vibration acceleration signals. Temperature sensors detect equipment temperature data, including temperature sequence signals. The data processing unit, based on relevant experimental data from substation equipment, performs data conversion, feature extraction, threshold comparison, data prediction, and alarm generation on these signals.

[0069] By simulating the operating status of substation equipment under different conditions and monitoring various experimental data corresponding to alarms, processing, and recording, including partial discharge signals (PD signals), vibration acceleration signals, and temperature series signals, the system categorizes the correlations between these data types in different scenarios, using them as triggers for response measures in different scenarios.

[0070] For example, in the simulation operation of the substation equipment, when an alarm occurs, the local discharge signal, vibration acceleration signal and temperature sequence signal corresponding to the moment the alarm occurs are recorded. Since the above three signals exist at the same time when the alarm occurs, the three signals at the moment the alarm occurs are the correlation experimental data when the alarm occurs in the substation equipment. Therefore, the three signals at the moment the alarm occurs are used as the early warning trigger conditions for the alarm of the substation equipment.

[0071] During the actual operation of substation equipment detected by this early warning system, by monitoring the above three types of signals and performing feature extraction and data prediction on the above three types of signals, the signal characteristics of the above three types of signals after the next time period can be known. If the predicted three types of signal characteristics are consistent with the preset experimental data related to the occurrence of alarms, it means that the alarm is very likely to occur. At this time, an alarm message is generated to alert the operator to take relevant treatment measures to prevent accidents before they happen.

[0072] Since this early warning system uses various types of correlated experimental data to determine whether an alarm is likely to occur, compared to the traditional prediction and judgment based on a single data, it effectively reduces the probability of misjudgment of the substation operation and maintenance monitoring and early warning system, thereby reducing the phenomenon of decreased power supply stability caused by misjudgment and early warning, and thus improving power supply stability.

[0073] Based on the above-mentioned substation operation and maintenance monitoring and early warning system, the present application also discloses a substation operation and maintenance monitoring and early warning method.

[0074] Refer to the attached Figure 2 As shown, a substation operation and maintenance monitoring and early warning method includes the following processing steps:

[0075] S101. Obtain partial discharge signals, temperature sequence signals, and vibration acceleration signals of substation equipment.

[0076] In implementation, the partial discharge signal is obtained through a partial discharge sensor, the temperature series signal is obtained through a stability sensor, and the vibration acceleration signal is obtained through a vibration sensor.

[0077] S102 : generating a partial discharge time-frequency diagram based on the partial discharge signal, and extracting the partial discharge time-frequency signal from the partial discharge time-frequency diagram.

[0078] In implementation, after obtaining the partial discharge signal, a partial discharge time-frequency diagram is generated based on the partial discharge signal, and the partial discharge time-frequency signal in the partial discharge time-frequency diagram is extracted.

[0079] S103: Input the partial discharge time-frequency diagram into a convolutional neural network model, and output the discharge probability from the convolutional neural network model.

[0080] In implementation, the convolutional neural network model outputs the required predicted discharge probability of the substation equipment.

[0081] S104: Input the temperature sequence signal into the memory network model, and the memory network model outputs the predicted temperature.

[0082] In implementation, the memory network model outputs the predicted temperature of the substation equipment required for prediction.

[0083] S105: Calculate the root mean square value of the vibration acceleration based on the vibration acceleration signal.

[0084] S106: Setting a graded alarm based on the discharge probability, the predicted temperature, and the root mean square value of the vibration acceleration.

[0085] During implementation, in the experimental phase of the substation equipment, the various types of correlated experimental data monitored are classified into levels according to the different measures that need to be processed, and the correlated experimental data of different levels are set as trigger conditions for different levels of alarms of the substation equipment.

[0086] By setting up a graded alarm mechanism, operators can take different measures according to the actual situation based on different levels of alarm information, so that they can respond to possible alarm situations more quickly, reasonably and scientifically.

[0087] Refer to the attached Figure 2 As shown, in step S106, setting a graded alarm includes:

[0088] If the predicted temperature is greater than 120°C and the RMS value of the vibration acceleration is greater than 5m / S 2 , a level one alarm is triggered, the power to the device is cut off and a text message is sent to the operation and maintenance terminal.

[0089] If the predicted data reaches the triggering conditions of the first-level alarm, the triggering action of the first-level alarm will be triggered, that is, cutting off the power supply of the equipment and sending a text message to the operation and maintenance terminal. The power supply of the equipment will be directly cut off before the alarm occurs to prevent the damage caused by the actual occurrence of the fault. The operation and maintenance terminal will be notified via text message to remind the operation and maintenance personnel of the occurrence of the first-level alarm.

[0090] If 110℃<predicted temperature≤120℃, or the predicted temperature exceeds the temperature threshold, a secondary alarm is triggered, a work order is generated and pushed to the mobile terminal.

[0091] If the predicted data reaches the triggering conditions of the second-level alarm, the triggering action of the second-level alarm will be triggered, that is, a work order will be generated and pushed to the mobile terminal. Although the second-level alarm does not require cutting off the power supply of the equipment, for safety reasons, operation and maintenance personnel are still required to conduct on-site inspections and checks of the equipment.

[0092] If 100℃<predicted temperature≤110℃, or the discharge probability is greater than the probability threshold, a level 3 alarm is triggered, the device status is marked, and a log is recorded.

[0093] If the predicted data reaches the triggering conditions of the third-level alarm, the triggering action of the third-level alarm will be triggered, that is, marking the equipment status and recording the log. The third-level alarm is usually understood as a general alarm. It is only necessary to record the operating status and related data of the equipment in the book to provide a reference for inspection and inspection for operation and maintenance personnel.

[0094] Refer to the attached Figure 3 As shown, this substation operation and maintenance monitoring and early warning method can also include the following processing steps.

[0095] S201. Obtain the predicted alarm level to which the predicted temperature belongs. If the predicted alarm level is not a level 1 alarm, obtain the alarm occurrence result. If an alarm occurs, obtain the predicted temperature, the actual alarm temperature, and the predicted deviation threshold.

[0096] In implementation, if the predicted temperature belongs to a non-level 1 alarm (such as a level 2 alarm and a level 3 alarm), taking the level 3 alarm as an example, if the predicted temperature belongs to the level 3 alarm and an alarm occurs, the predicted temperature, the actual alarm temperature and the predicted deviation threshold are obtained.

[0097] S202: Based on the predicted temperature, the actual warning temperature, and the predicted deviation threshold, determine whether the predicted temperature difference between the predicted temperature and the actual warning temperature is less than the predicted deviation threshold.

[0098] In implementation, the predicted temperature is subtracted from the actual alarm temperature (the actual temperature of the substation where the alarm occurs), and the absolute value is taken, which is then compared with the predicted deviation threshold.

[0099] S203: If the predicted temperature difference is less than the predicted deviation threshold, the temperature value, vibration acceleration root mean square value, and discharge probability corresponding to the actual alarm temperature are added to the triggering conditions of the previous predicted alarm level.

[0100] In implementation, if the absolute value of the predicted temperature difference is less than the predicted deviation threshold, it means that the temperature prediction has good accuracy. The temperature prediction can still be used as one of the main warning data bases. The temperature value, root mean square value of vibration acceleration and discharge probability corresponding to the actual alarm temperature are added to the trigger conditions of the second-level prediction alarm level. By upgrading the trigger conditions that originally belonged to the third-level alarm to the trigger conditions of the second-level alarm, the importance of the predicted temperature of the above-mentioned alarm is increased, and more cautious handling measures are taken to reduce the probability of similar alarms occurring.

[0101] In addition, when the trigger conditions that originally belonged to the third-level alarm jump to the trigger conditions of the second-level alarm, the trigger conditions that originally belonged to the third-level alarm will be deleted from the trigger conditions of the third-level alarm to avoid system confusion.

[0102] If the predicted temperature is less than 100°C and an alarm occurs, the temperature value, vibration acceleration root mean square value, and discharge probability corresponding to the alarm when the predicted temperature is less than 100°C are obtained and added to the triggering conditions of the third-level alarm.

[0103] Since no corresponding level alarm is set when the predicted temperature is less than 100°C, there is still a possibility that an alarm may occur in the substation equipment. Therefore, the temperature value, root mean square value of vibration acceleration and discharge probability corresponding to the alarm when the predicted temperature is less than 100°C are added to the trigger conditions of the third-level alarm to increase the importance of the predicted temperature for the above-mentioned alarm. When such an alarm occurs, it is recorded in the equipment log for the operator to view and perform further related processing, such as manually performing a transition process on the trigger condition.

[0104] Record the newly added triggering conditions for transitioning from the third-level alarm to the second-level alarm, corresponding to the interval duration of the alarm occurrence. If the interval duration is less than the duration interval threshold, the newly added triggering conditions for transitioning from the third-level alarm to the second-level alarm will be added to the triggering conditions of the first-level alarm.

[0105] If the new trigger condition for transitioning from level 3 alarm to level 2 alarm has a time interval between alarms that is less than the time interval threshold (the time interval threshold between two adjacent alarms), it means that the relevant data of this new trigger condition has appeared frequently and has triggered an alarm. Therefore, it is necessary to further increase the importance attached to this new trigger condition. At this time, the new trigger condition for transitioning from level 3 alarm to level 2 alarm will be added to the trigger condition of level 1 alarm, which will also increase the importance attached to the new trigger condition. When this new trigger condition is met again, the level 1 alarm will be directly triggered, that is, the power supply of the equipment will be cut off, and a text message will be sent to the operation and maintenance terminal.

[0106] Refer to the attached Figure 4 As shown, this substation operation and maintenance monitoring and early warning method can also include the following processing steps.

[0107] S301: Acquire the trigger frequency of the newly added trigger condition of the first level alarm and the trigger frequency of the initial trigger condition of the first level alarm.

[0108] In implementation, the new triggering conditions for the first level alarm are the triggering conditions for the transition to the first level alarm; the initial triggering conditions for the first level alarm are that the predicted temperature is greater than 120°C and the root mean square value of the vibration acceleration is greater than 5m / S 2 .

[0109] S302: If the trigger frequency of the newly added trigger condition of the first-level alarm reaches the trigger frequency of the initial trigger condition of the first-level alarm, then generate maintenance information of the substation equipment.

[0110] During implementation, if the trigger frequency of the newly added trigger conditions of the first-level alarm reaches the trigger frequency of the initial trigger conditions of the first-level alarm, it means that the newly added trigger conditions have been triggered frequently and need to be repaired in time. At this time, maintenance information of the substation equipment is generated to alert the operator to inspect and maintain the substation equipment.

[0111] S303: After the maintenance is completed based on the maintenance information, the newly added triggering condition of the first-level alarm is reduced to the triggering condition of the second-level alarm.

[0112] In implementation, since the corresponding fault of the newly added triggering condition of the first-level alarm has been handled, the newly added triggering condition of the first-level alarm is reduced to the triggering condition of the second-level alarm.

[0113] S304: When the triggering condition for downgrading to the second-level alarm is not triggered for a period longer than the downgraded period threshold, the triggering condition for downgrading to the second-level alarm is downgraded to the triggering condition for the third-level alarm.

[0114] In practice, if a trigger condition that has been downgraded to a Level 2 alarm remains untriggered for longer than the downgrade threshold (the time required for a newly added trigger condition to be downgraded to the next level of alarm), it will be downgraded to a Level 2 alarm, or even a Level 3 alarm. By gradually downgrading new trigger conditions rather than simply removing them from the multi-level alarm system, a certain level of vigilance against newly added trigger conditions can be maintained even after substation maintenance is complete.

[0115] The embodiments of this specific implementation method are all preferred embodiments of the present application and are not intended to limit the scope of protection of the present application in turn. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A substation operation and maintenance monitoring and early warning system, characterized in that: It includes a partial discharge sensor, a temperature sensor, a vibration sensor and a data processing unit; The partial discharge sensor is used to monitor the partial discharge signal of the equipment; The vibration sensor is used to monitor the vibration acceleration signal of the equipment; The temperature sensor is used to detect the temperature sequence signal of the device; The data processing unit is used to perform data conversion, feature extraction, threshold comparison, data prediction and generate alarm information on the above-mentioned various signals based on the correlation experimental data of the substation equipment.

2. A method for monitoring and early warning of substation operation and maintenance, according to the substation operation and maintenance monitoring and early warning system of claim 1, characterized in that: The method comprises: Obtain partial discharge signals, temperature sequence signals, and vibration acceleration signals of substation equipment; generating a partial discharge time-frequency diagram based on the partial discharge signal, and extracting the partial discharge time-frequency signal from the partial discharge time-frequency diagram; Inputting the partial discharge time-frequency graph into a convolutional neural network model, and outputting a discharge probability from the convolutional neural network model; The temperature sequence signal is input into the memory network model, and the memory network model outputs the predicted temperature; Calculating a root mean square value of vibration acceleration based on the vibration acceleration signal; Set graded alarms based on discharge probability, predicted temperature, and RMS vibration acceleration.

3. A method for monitoring and early warning of substation operation and maintenance according to claim 2, characterized in that: The step of setting a graded alarm includes: If the predicted temperature is greater than 120°C and the RMS value of the vibration acceleration is greater than 5m / S 2 , then trigger the first level alarm; If 110℃<predicted temperature≤120℃, or the predicted temperature exceeds the temperature threshold, a level 2 alarm is triggered; If 100℃<predicted temperature≤110℃, or the discharge probability is greater than the probability threshold, a level 3 alarm is triggered.

4. A method for monitoring and early warning of substation operation and maintenance according to claim 3, characterized in that: The triggering action of the first-level alarm is to cut off the power supply of the device and send a text message to the operation and maintenance terminal; The triggering action of the secondary alarm is to generate a work order and push it to the mobile terminal; The triggering action of the third level alarm is to mark the device status and record the log.

5. A method for monitoring and early warning of substation operation and maintenance according to claim 3, characterized in that: Also includes: Get the predicted alarm level to which the predicted temperature belongs. If the predicted alarm level is not a level 1 alarm, get the alarm occurrence result. If an alarm occurs, get the predicted temperature, actual alarm temperature, and predicted deviation threshold. Based on the predicted temperature, the actual warning temperature, and the prediction deviation threshold, determining whether a predicted temperature difference between the predicted temperature and the actual warning temperature is less than the prediction deviation threshold; If the predicted temperature difference is less than the predicted deviation threshold, the temperature value, vibration acceleration root mean square value and discharge probability corresponding to the actual alarm temperature are added to the trigger conditions of the previous level of predicted alarm level.

6. A method for monitoring and early warning of substation operation and maintenance according to claim 5, characterized in that: If the predicted temperature is less than 100°C and an alarm occurs, the temperature value, vibration acceleration root mean square value, and discharge probability corresponding to the alarm when the predicted temperature is less than 100°C are obtained and added to the triggering conditions of the third-level alarm.

7. A method for monitoring and early warning of substation operation and maintenance according to claim 5, characterized in that: Record the newly added triggering conditions for transitioning from the third-level alarm to the second-level alarm, corresponding to the interval duration of the alarm occurrence. If the interval duration is less than the duration interval threshold, the newly added triggering conditions for transitioning from the third-level alarm to the second-level alarm will be added to the triggering conditions of the first-level alarm.

8. A method for monitoring and early warning of substation operation and maintenance according to claim 7, characterized in that: Get the trigger frequency of the newly added trigger conditions of the first-level alarm and the trigger frequency of the initial trigger conditions of the first-level alarm; If the trigger frequency of the newly added trigger condition of the first level alarm reaches the trigger frequency of the initial trigger condition of the first level alarm, maintenance information of the substation equipment is generated; After the maintenance is completed based on the maintenance information, the newly added triggering condition of the first-level alarm is reduced to the triggering condition of the second-level alarm; The triggering conditions for downgrading to the second-level alarm will be reduced to the triggering conditions of the third-level alarm if the corresponding non-triggering time is greater than the downgrade time threshold.

Images