An abnormal recognition system for gateway electric energy metering devices

Abstract

The present invention relates to an abnormal identification system for a gateway electric energy metering device, comprising: a photovoltaic panel, an AC / DC module, a DC / DC module, a power loss detection module, a switching output module, a battery, a power quantity detection module, a metering transformer and a control module; the switching output module includes a charging path and two discharging paths, and the two discharging paths can ensure that they are not conducted simultaneously or are both closed simultaneously; the control module can predict the fault risk of the electric energy metering device within a future period of time, and give an early warning to the high-voltage metering transformer with a high over-tolerance risk, so as to realize the rapid positioning of the fault of the electric energy metering device and the dynamic early warning of the state.

Description

Technical Field

[0001] The present invention belongs to the technical field of power equipment, and particularly relates to an abnormal recognition system for a gateway electric energy metering device. Background Art

[0002] With the continuous upgrading of power grid equipment, the stability and accuracy of electric energy metering are also continuously improved. When judging the abnormality of the metering device, it is more inclined to develop in the direction of intelligence and convenience. However, the abnormality of the metering device still occurs from time to time due to the complex working environment. For example, direct lightning strikes, induced lightning strikes, short circuits and grounding faults on the low-voltage side line, overvoltage caused by short circuits and grounding faults on the high-voltage line, and faults in the secondary circuit of the voltage transformer, etc., will all cause damage to the voltage transformer or fuse blowing of the high-voltage fuse. Many operating conditions will cause the metering device to be abnormal. Sometimes, even artificial reverse wiring, etc., will cause abnormal electric energy metering. How to improve the accurate recognition of the abnormality of electric energy metering and promptly eliminate abnormal situations is an important task that inevitably needs to be solved in the power grid upgrade. Summary of the Invention

[0003] The present invention discloses an abnormal recognition system for a gateway electric energy metering device, including: a photovoltaic panel, an AC / DC module, a DC / DC module, a power loss detection module, a switching output module, a battery, a power quantity detection module, a metering transformer, and a control module;

[0004] The switching output module includes a charging path and two discharging paths, and the two discharging paths can ensure that they are not conducted simultaneously or are both closed simultaneously;

[0005] The metering transformer receives electric energy provided by the commercial power, the photovoltaic panel or the battery for stable power supply. The power loss detection module is used to detect that neither the commercial power nor the photovoltaic panel can provide stable electric energy, determine a power loss signal, and automatically switch the battery to supply power to the metering transformer. The switching output module realizes automatic shutdown or remote shutdown control of the battery when the battery power is too low; the power quantity detection module is used to detect the power quantity of the battery to provide battery power quantity parameters to the control module. The control module determines whether to provide abnormal recognition for the gateway electric energy metering device according to the detected battery power quantity parameters. When the battery supplies power, the control module suspends abnormal recognition. At this time, the control module is only used to control the switching output module. When the power supply is not from the battery, the abnormal recognition function of the control module is enabled;

[0006] The control module receives the maintenance data and operation data within the gateway range, quantifies the deviation abnormality of the output data of the metering potential transformer, determines the operation status of the in-service high-voltage metering potential transformer based on the maintenance data, operation data, and the output data after deviation abnormality quantification, conducts prediction and risk assessment based on the operation status, displays and further analyzes the status data and calculation data, establishes a sample set using the historical status evaluation results of the electric energy metering device and the historical operation data of the electric energy metering device, trains the model to obtain the degradation curve of the metering performance of the high-voltage metering potential transformer, predicts the fault risk of the electric energy metering device in a future period of time, and gives an early warning for the high-voltage metering potential transformer with a high over-tolerance risk, realizing the rapid positioning of the electric energy metering device fault and the dynamic early warning of the status.

[0007] The described abnormal identification system for the gateway electric energy metering device, the switching output module includes: a first diode (D1), a second diode (D2), a third resistor (R3), a fourth resistor (R4), a fifth resistor (R5), a capacitor (C1), a controllable switch (K1), and a thyristor (D4);

[0008] One charging path is the charging path formed by the first diode (D1), the third resistor (R3), and the capacitor (C1) to charge the capacitor (C1), and the capacitor (C1) is charged by the electric energy provided by the mains or the photovoltaic panel.

[0009] The two discharging paths include a first discharging path and a second discharging path. The first discharging path is the first path formed by the battery, the MOS transistor, and the thyristor (D4) to discharge the metering potential transformer; the second discharging path is the second path formed by the battery, the MOS transistor, and the controllable switch (K1) to discharge the metering potential transformer; when the first discharging path and the second discharging path discharge, due to the unidirectional conduction function of the second diode (D2) set, neither can charge the capacitor (C1).

[0010] The described abnormal recognition system for the gateway power metering device. The anode of the first diode (D1) is connected to the output terminals of the AC / DC module and the DC / DC module. The cathode of the first diode (D1) is respectively connected to one end of the third resistor (R3) and the anode of the second diode (D2). The second end of the third resistor (R3) is connected to the first end of the capacitor (C1), the first end of the fourth resistor (R4), and the first end of the fifth resistor (R5). The second end of the capacitor (C1) and the second end of the fourth resistor (R4) are both grounded. The second end of the fifth resistor (R5) is connected to the control terminal (G2) of the thyristor (D4). The cathode of the second diode (D2) is connected to the first end of the controllable switch (K1), the cathode (K) of the thyristor (D4), and the power supply input terminal of the gateway watt-hour meter. The second end of the controllable switch (K1) is connected to the drain (D) of the MOS transistor and the anode (A) of the thyristor (D4). The control terminal of the controllable switch (K1) is connected to the output terminal of the control module;

[0011] The described abnormal recognition system for the gateway power metering device. The input terminal of the AC / DC module is connected to the AC mains, and the output terminal outputs direct current, and the output terminal is connected to the charging circuit, the power loss detection module, and the switching output module;

[0012] DC / DC module. The input terminal of the DC / DC is connected to the photovoltaic panel, and the output terminal is connected to the charging circuit, the power loss detection module, and the switching output module;

[0013] The power loss detection module includes: a first resistor (R1), a second resistor (R2), a voltage regulator diode (D3), and a MOS transistor. The first end of the first resistor (R1) is connected to the output terminals of the AC / DC module and the DC / DC module. The second end of the first resistor (11) is connected to the first end of the second resistor (R2), the cathode of the voltage regulator diode (D3), and the gate (G1) of the MOS transistor. The anode of the voltage regulator diode (D3) is grounded. The second end of the second resistor (R2) and the source (S) of the MOS transistor are connected to the connection point of the charging circuit and the battery; when the power loss detection module detects a power loss in the mains and the photovoltaic panel, it controls the MOS transistor to conduct.

[0014] The described abnormal recognition system for the gateway power metering device. The control module includes a processing unit, a switching control unit, and an abnormal recognition unit. The processing unit can control the opening and closing of the switching control unit and the abnormal recognition unit.

[0015] The described abnormal identification system for the gateway electric energy metering device. The control module determines whether to provide abnormal identification for the gateway electric energy metering device according to the detected battery power parameter. When powered by the battery, the control module suspends the abnormal identification. At this time, the control module is only used to control the switching output module. When not powered by the battery, the specific steps to enable the abnormal identification function of the control module are as follows:

[0016] When the power detection module detects that the power of the battery decreases, it is determined that the external power supply loses power at this time and is powered by the battery. The first electric energy data recorded by the gateway electric energy metering device at this time is recorded into the abnormal identification unit. Then the processing unit closes the abnormal identification unit of the control module and enables the switching control unit. When the power of the battery detected by the power detection unit is too low, the switching control unit issues a driving signal to control the controllable switch (K1) to conduct first and then turn off, so as to control the battery to stop supplying power externally. After the controllable switch (K1) conducts and before it turns off, the processing unit briefly enables the abnormal identification unit to record the second electric energy data at this time and store it;

[0017] When the external power supply is normal, the abnormal identification unit is enabled and the battery is charged. At the same time, the abnormal identification unit analyzes the first electric energy data and the second electric energy data to judge whether they are the same. If they are the same, it means that the gateway electric energy metering device is normal after the external power loss. If they are not the same, it means that there is an abnormality and an abnormal warning is given.

[0018] The described abnormal identification system for the gateway electric energy metering device. Using the historical state evaluation results of the electric energy metering device and the historical operation data of the electric energy metering device to establish a sample set, train a model, and obtain the degradation curve of the metering performance of the high-voltage metering transformer, and predict the fault risk of the electric energy metering device in the next period of time. The specific steps are as follows:

[0019] Obtain historical operation data to establish a historical sample set;

[0020] Input the feature set data of the normalized historical sample set into the neural network, use the neural network to perform feature analysis of the historical sample set, and combine the historical state evaluation results to obtain the normal event boundary threshold for training the gateway metering device;

[0021] Determine the basic training model through the boundary threshold;

[0022] Input the current metering data into the basic training model to predict whether the operation curve of the electric energy metering device exceeds the boundary threshold. If it exceeds, there is a fault risk.

[0023] The present invention provides an abnormal identification system for a gateway power metering device, which can identify various types of faults, avoid metering faults during external power failure, and can perform orderly control of abnormal identification and switching control through a control module to ensure the stable operation of the metering device for a period of time during external power failure. One of the main improvements of the present invention is to set a switching output module. Through a separate charging path and two discharging paths, it can automatically control the shutdown of the battery according to the power, set the charging path of the capacitor between the first diode and the second diode, and ensure the timely shutdown of the second discharging path by setting the second diode. And after the controllable switch of the second discharging path is turned on and then turned off, and when there is external power, the capacitor can be charged through the first diode. With the one-way conduction function of the set second diode (D2), neither of the two discharging paths can charge the capacitor (C1), without considering the conduction time of the controllable switch, and the first path will not be turned on again, ensuring that the switching process can be more stable. Another improvement of the present invention lies in that the control module includes a processing unit, a switching control unit, and an abnormal identification unit. The processing unit can control the opening and closing of the switching control unit and the abnormal identification unit, and can perform separate control of the two functions, saving the electric energy consumed by the control module. At the same time, according to the battery power supply state, when the battery supplies power, the control module pauses abnormal identification. At this time, the control module is only used to control the switching output module. When it is not battery-powered, the abnormal identification function of the control module is turned on to identify abnormalities in stages and states. After the controllable switch (K1) is turned on and before it is turned off, the processing unit briefly turns on the abnormal identification unit to record the second power data at this time and store it, and determines the abnormal data of the metering device after external power failure by comparing the first data and the second data. Another improvement of the present invention is to use the historical state evaluation results of the power metering device and the historical operation data of the power metering device to establish a sample set, train a model, obtain a degradation curve of the metering performance of the high-voltage metering transformer, and predict the fault risk of the power metering device in the next period of time. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a system block diagram of the abnormal identification system for the gateway power metering device of the present invention.

[0025] Figure 2 It is a specific connection schematic diagram of the abnormal identification system for the gateway power metering device of the present invention.

[0026] Figure 3 It is a schematic diagram of the abnormal identification method for the gateway power metering device of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0027] The present application will be further described in detail below in conjunction with the accompanying drawings. It is necessary to point out here that the following specific embodiments are only used to further illustrate the present application and should not be construed as limiting the protection scope of the present application. Those skilled in the art can make some non-essential improvements and adjustments to the present application based on the above application content.

[0028] According to the prior art, between the anode and cathode of a thyristor, as long as a trigger voltage with different positive polarities is applied between the control electrode G and the anode, it can be triggered to conduct and present a low-resistance state. At this time, the voltage drop between the anode and cathode is also about 1V. Once the thyristor conducts, even if the trigger voltage G is lost, it can continue to maintain the conducting state. Only when the anode current decreases, is less than the holding current, or when the voltage polarity between the anode and cathode changes and there is no trigger voltage, the thyristor will turn off. At this time, only by reapplying the trigger voltage can it conduct. That is, the thyristor requires a trigger voltage after turning off, and after conducting, it can maintain conduction without a trigger voltage or with a trigger voltage.

[0029] As Figure 1 shown, it is a system block diagram of the abnormal identification system for the gateway electric energy metering device of the present invention. The present invention discloses an abnormal identification system for a gateway electric energy metering device, including: a photovoltaic panel, an AC / DC module, a DC / DC module, a power loss detection module, a switching output module, a battery, a power detection module, a metering current transformer, and a control module.

[0030] The switching output module includes a charging path and two discharging paths, and the two discharging paths can ensure that they do not conduct simultaneously or are both closed simultaneously;

[0031] The metering current transformer receives the electric energy provided by the mains power, the photovoltaic panel, or the battery for stable power supply. The power loss detection module is used to detect that neither the mains power nor the photovoltaic panel can provide stable electric energy, determine a power loss signal, and automatically switch the battery to supply power to the metering current transformer. The switching output module realizes automatic shutdown or remote shutdown control of the battery when the battery power is too low. The power detection module is used to detect the power of the battery to provide battery power parameters to the control module. The control module determines whether to provide abnormal identification for the gateway electric energy metering device according to the detected battery power parameters. When the battery supplies power, the control module suspends the abnormal identification. At this time, the control module is only used to control the switching output module. When the power supply is not from the battery, the abnormal identification function of the control module is enabled;

[0032] The control module receives the maintenance data and operation data within the gateway range, quantifies the deviation abnormality of the output data of the metering potential transformer, determines the operation status of the in-service high-voltage metering potential transformer based on the maintenance data, operation data, and the output data after deviation abnormality quantification, conducts prediction and risk assessment based on the operation status, displays and further analyzes the status data and calculation data, establishes a sample set using the historical status evaluation results of the electric energy metering device and the historical operation data of the electric energy metering device, trains the model to obtain the degradation curve of the metering performance of the high-voltage metering potential transformer, predicts the fault risk of the electric energy metering device in a future period of time, and gives an early warning to the high-voltage metering potential transformer with a high over-tolerance risk, realizing the rapid positioning of the electric energy metering device fault and the dynamic early warning of the status.

[0033] As Figure 2 shown, it is a specific connection schematic diagram of the abnormal identification system for the gateway electric energy metering device of the present invention. The described abnormal identification system for the gateway electric energy metering device, the switching output module includes: a first diode (D1), a second diode (D2), a third resistor (R3), a fourth resistor (R4), a fifth resistor (R5), a capacitor (C1), a controllable switch (K1), and a thyristor (D4);

[0034] One charging path is the charging path formed by the first diode (D1), the third resistor (R3), and the capacitor (C1) to charge the capacitor (C1), and the capacitor (C1) is charged by the electric energy provided by the mains or the photovoltaic panel;

[0035] The two discharging paths include a first discharging path and a second discharging path. The first discharging path is the first path formed by the battery, the MOS transistor, and the thyristor (D4) to discharge the metering potential transformer; the second discharging path is the second path formed by the battery, the MOS transistor, and the controllable switch (K1) to discharge the metering potential transformer; when the first discharging path and the second discharging path discharge, due to the unidirectional conduction function of the second diode (D2) set, neither can charge the capacitor (C1).

[0036] The described abnormal recognition system for the gateway electric energy metering device. The anode of the first diode (D1) is connected to the output terminals of the AC / DC module and the DC / DC module. The cathode of the first diode (D1) is respectively connected to one end of the third resistor (R3) and the anode of the second diode (D2). The second end of the third resistor (R3) is connected to the first end of the capacitor (C1), the first end of the fourth resistor (R4), and the first end of the fifth resistor (R5). The second end of the capacitor (C1) and the second end of the fourth resistor (R4) are both grounded. The second end of the fifth resistor (R5) is connected to the control terminal (G2) of the thyristor (D4). The cathode of the second diode (D2) is connected to the first end of the controllable switch (K1), the cathode (K) of the thyristor (D4), and the power supply input terminal of the gateway electric energy meter. The second end of the controllable switch (K1) is connected to the drain (D) of the MOS transistor and the anode (A) of the thyristor (D4). The control terminal of the controllable switch (K1) is connected to the output terminal of the control module;

[0037] When there is external power, the capacitor is always charged through the first diode, maintaining sufficient electrical energy. This sufficient electrical energy enables the capacitor to discharge through the resistor to the control electrode G2 of the thyristor at the initial stage of external power loss, forming a trigger voltage;

[0038] When the external power is lost, since the capacitor can trigger the thyristor to conduct at the initial stage of external power loss, at this time, the battery can supply power to the load through the thyristor. After the external power is lost, the capacitor cannot be charged through the first diode at this time. Then, after discharging the electrical energy to the control electrode of the thyristor, the electrical energy is exhausted. According to the working principle of the thyristor, even if there is no electricity at the control electrode of the thyristor, due to the fact that conduction has been formed, it can continue to conduct;

[0039] When the external power supply is lost during the working process, after being judged by the second resistor R2 and the voltage regulator diode, the MOS transistor (PMOS) is turned on.

[0040] When the controllable switch K1 is turned on, the battery current will discharge through the second discharge path, that is, supply power to the load through the controllable switch K1. At this time, since the controllable switch K1 is turned on and is equivalent to a short - circuit state, the electromotive forces of the anode and cathode of the thyristor D4 are equal, and the forward current is instantly zero, making it impossible to maintain the conduction between A and K. The thyristor D4 is turned off. The capacitor connected to the control stage G2 also cannot be charged due to the one - way conduction of the second diode and has no voltage. The discharge path of the battery can only discharge through the second path. When the controllable switch is turned off, the battery discharge path is completely closed. The controllable switch does not need to set the time length. As long as the controllable switch is turned off at any time after being turned on, it can stably control the closing of the battery power supply path. There is a control advantage compared with the prior art that also needs to consider the conduction time of the controllable switch.

[0041] The described abnormal identification system for a gateway power metering device, the input end of the AC / DC module is connected to the AC mains, the output end outputs direct current, and the output end is connected to a charging circuit, a power loss detection module, and a switching output module;

[0042] A DC / DC module, the input end of the DC / DC is connected to a photovoltaic panel, and the output end is connected to a charging circuit, a power loss detection module, and a switching output module;

[0043] The power loss detection module includes: a first resistor (R1), a second resistor (R2), a voltage stabilizing diode (D3), and a MOS transistor, such as a PMOS transistor. The first end of the first resistor (R1) is connected to the output ends of the AC / DC module and the DC / DC module. The second end of the first resistor (11) is connected to the first end of the second resistor (R2), the cathode of the voltage stabilizing diode (D3), and the gate (G1) of the MOS transistor. The anode of the voltage stabilizing diode (D3) is grounded. The second end of the second resistor (R2) and the source (S) of the MOS transistor are connected to the connection point of the charging circuit and the battery; when the power loss detection module detects a power loss of the mains and the photovoltaic panel, it controls the MOS transistor to conduct.

[0044] The described abnormal identification system for a gateway power metering device, the control module includes a processing unit, a switching control unit, and an abnormal identification unit, and the processing unit can control the opening and closing of the switching control unit and the abnormal identification unit.

[0045] The described abnormal identification system for a gateway power metering device, the control module determines whether to provide abnormal identification for the gateway power metering device according to the detected battery power parameter. When powered by the battery, the control module suspends abnormal identification. At this time, the control module is only used to control the switching output module. When not powered by the battery, the specific steps for enabling the abnormal identification function of the control module are as follows:

[0046] When the power detection module detects that the power of the battery decreases, it is determined that the external power supply has a power loss at this time and is powered by the battery. The first power data recorded by the gateway power metering device at this time is recorded into the abnormal identification unit. Then the processing unit closes the abnormal identification unit of the control module and opens the switching control unit. When the power of the battery detected by the power detection unit is too low, the switching control unit issues a drive signal to control the controllable switch (K1) to conduct first and then turn off, so as to control the battery to stop supplying power externally. After the controllable switch (K1) conducts and before it turns off, the processing unit briefly opens the abnormal identification unit, records the second power data at this time and stores it;

[0047] When the external power supply is normal, the anomaly recognition unit is turned on and the battery is charged. At the same time, the anomaly recognition unit analyzes the first electrical energy data and the second electrical energy data to determine whether they are the same. If they are the same, it means that there is no anomaly in the gateway electrical energy metering device after a power outage. If they are different, it means that there is an anomaly and an anomaly warning is issued.

[0048] For the described anomaly recognition system of the gateway electrical energy metering device, using the historical state evaluation results of the electrical energy metering device and the historical operation data of the electrical energy metering device to establish a sample set, training a model, and obtaining a degradation curve of the metering performance of the high-voltage metering transformer, predicting the fault risk of the electrical energy metering device in a future period specifically includes:

[0049] Obtain historical operation data to establish a historical sample set;

[0050] Input the feature set data of the normalized historical sample set into a neural network, use the neural network to perform feature analysis of the historical sample set, and combine the historical state evaluation results to obtain the normal event boundary threshold for training the gateway metering device;

[0051] Determine the basic training model through the boundary threshold;

[0052] Input the current metering data into the basic training model, and predict whether the operation curve of the electrical energy metering device exceeds the boundary threshold. If it exceeds, there is a fault risk.

[0053] As Figure 3 shown, a method for anomaly recognition and control of a gateway electrical energy metering device according to the present invention includes the following steps:

[0054] Step S1: The metering transformer receives electrical energy provided by the mains power, photovoltaic panels or batteries for stable power supply. The power outage detection module is used to detect that neither the mains power nor the photovoltaic panels can provide stable electrical energy, determine the power outage signal, and automatically switch the battery to supply power to the metering transformer;

[0055] Step S2: The battery power detection module is used to detect the battery power to provide battery power parameters to the control module. The control module determines whether to provide anomaly recognition for the gateway electrical energy metering device according to the detected battery power parameters. When the battery supplies power, it enters step S3. When the battery does not supply power, it enters step S4;

[0056] Step S3: When the power detection module detects that the battery power decreases, it is determined that the external power supply has lost power at this time and the battery supplies power. Record the first power data recorded by the gateway power metering device at this time into the anomaly recognition unit. Then, the processing unit closes the anomaly recognition unit of the control module and activates the switching control unit. When the battery power detected by the power detection unit is too low, the switching control unit issues a drive signal to control the pilot conduction and then turn-off of the controllable switch (K1), so as to control the battery to stop supplying power externally. After the controllable switch (K1) conducts and before it turns off, the processing unit briefly activates the anomaly recognition unit to record and store the second power data at this time;

[0057] Step S4: When the external power supply is normal, activate the anomaly recognition unit and charge the battery. At the same time, the anomaly recognition unit analyzes the first power data and the second power data to determine whether they are the same. If they are the same, it means that the gateway power metering device is normal after the external power loss. If they are not the same, it means that there is an anomaly and an anomaly warning is issued;

[0058] The control module receives the maintenance data and operation data within the gateway range, and quantifies the deviation anomaly of the output data of the metering transformer. Based on the maintenance data, operation data and the output data after deviation anomaly quantification, determine the operation trend of the in-service high-voltage metering transformer. Based on the operation trend, conduct prediction and risk assessment, display and further analyze the status data and calculation data. Use the historical status evaluation results of the power metering device and the historical operation data of the power metering device to establish a sample set, train the model, obtain the degradation curve of the metering performance of the high-voltage metering transformer, predict the power metering device failure risk in the next period of time, and issue a warning for the high-voltage metering transformer with a high over-tolerance risk to achieve rapid positioning of power metering device failures and dynamic status warnings.

[0059] The present invention provides an abnormal recognition system for gateway power metering devices, which can identify various types of faults, avoid metering faults during external power failure, and can perform orderly control of abnormal recognition and switching control through a control module to ensure the stable operation of the metering device for a period of time during external power failure. One of the main improvements of the present invention is to set a switching output module. Through a separate charging path and two discharging paths, it can automatically control the cut-off control of the battery according to the electric quantity. The charging path of the capacitor is set between the first diode and the second diode. By setting the second diode, it can ensure the timely cut-off of the control of the second discharging path. And after the controllable switch of the second discharging path is turned on and then turned off, and when there is external power, the capacitor can be charged through the first diode. Due to the one-way conduction function of the set second diode (D2), neither of the two discharging paths can charge the capacitor (C1), without considering the conduction time of the controllable switch, and the first path will not be turned on again, ensuring that the switching process can be more stable. Another improvement of the present invention lies in that the control module includes a processing unit, a switching control unit, and an abnormal recognition unit. The processing unit can control the opening and closing of the switching control unit and the abnormal recognition unit, and can perform separate control of the two functions, saving the electric energy consumed by the control module. At the same time, according to the battery power supply state, when the battery supplies power, the control module pauses abnormal recognition. At this time, the control module is only used to control the switching output module. When it is not powered by the battery, the abnormal recognition function of the control module is turned on to perform phased and state abnormal recognition. After the controllable switch (K1) is turned on and before it is turned off, the processing unit briefly turns on the abnormal recognition unit to record the second electric energy data at this time and store it. The abnormal data of the metering device after external power failure is determined by comparing the first data and the second data. Another improvement of the present invention is to establish a sample set by using the historical state evaluation results of the power metering device and the historical operation data of the power metering device, train the model, obtain the degradation curve of the metering performance of the high-voltage metering transformer, and predict the fault risk of the power metering device in the future for a period of time.

Claims

1. An abnormal recognition method for a gateway electric energy metering device, characterized in that The method is based on an abnormal identification system for a gateway electric energy metering device, and the system includes: a photovoltaic panel, an AC / DC module, a DC / DC module, a power loss detection module, a switching output module, a battery, a power detection module, a metering mutual inductor, and a control module; The input end of the AC / DC module is connected to the AC mains, and the output end outputs direct current, and the output end is connected to a charging circuit and a power loss detection module; the input end of the DC / DC module is connected to the photovoltaic panel, and the output end is connected to the charging circuit and the power loss detection module; the output end of the power loss detection module is connected to the switching output module; The control module includes a processing unit, a switching control unit, and an abnormal identification unit; The switching output module includes a charging path and two discharging paths, and the two discharging paths can ensure that they are not conducted simultaneously or are both closed simultaneously; The method includes: The metering mutual inductor receives the electric energy provided by the mains, the photovoltaic panel or the battery for stable power supply. The power loss detection module is used to detect that neither the mains nor the photovoltaic panel can provide stable electric energy, determine a power loss signal, so as to automatically switch the battery to supply power to the metering mutual inductor. When the battery power is too low, the switching output module realizes automatic shutdown or remote shutdown control of the battery; the power detection module is used to detect the battery power to provide battery power parameters to the control module, and the control module determines whether to provide abnormal identification for the gateway electric energy metering device according to the detected battery power parameters; When the power detection module detects that the battery power decreases, it is determined that the external power supply has a power loss at this time, and the battery supplies power. The first electric energy data recorded by the gateway electric energy metering device at this time is recorded into the abnormal identification unit. Then the processing unit turns off the abnormal identification unit of the control module and turns on the switching control unit. When the battery power detected by the power detection module is too low, the switching control unit issues a driving signal to control the controllable switch of the switching output module to conduct first and then turn off, so as to control the battery to stop supplying power externally. After the controllable switch conducts and before it turns off, the processing unit briefly turns on the abnormal identification unit, records the second electric energy data at this time and stores it; When the external power supply is normal, the abnormal identification unit is turned on, and the battery is charged. At the same time, the abnormal identification unit analyzes the first electric energy data and the second electric energy data to judge whether they are the same. If they are the same, it means that there is no abnormality in the gateway electric energy metering device after the external power loss. If they are not the same, it means that there is an abnormality, and an abnormal warning is given; The control module receives the maintenance data and operation data within the scope of the switchyard, quantifies the deviation abnormality of the output data of the metering transformer, determines the operation status of the in-service high-voltage metering transformer based on the maintenance data, operation data, and the output data after deviation abnormality quantification, conducts prediction and risk assessment based on the operation status, displays and further analyzes the status data and calculation data, establishes a sample set by using the historical status evaluation results of the electric energy metering device and the historical operation data of the electric energy metering device, trains the model, obtains the deterioration curve of the metering performance of the high-voltage metering transformer, predicts the fault risk of the electric energy metering device in a future period of time, and gives an early warning to the high-voltage metering transformer with a high over-tolerance risk, so as to achieve rapid positioning of the electric energy metering device fault and dynamic status early warning.

2. The abnormal recognition method of a gateway power metering device according to claim 1, wherein The switching output module includes: a first diode (D1), a second diode (D2), a third resistor (R3), a fourth resistor (R4), a fifth resistor (R5), a capacitor (C1), a controllable switch (K1), and a thyristor (D4); One charging path is the charging path formed by the first diode (D1), the third resistor (R3), and the capacitor (C1) for charging the capacitor (C1), and the capacitor (C1) is charged by the electric energy provided by the mains or the photovoltaic panel. The two discharging paths include a first discharging path and a second discharging path. The first discharging path is the first path formed by the battery, the MOS transistor, and the thyristor (D4) for discharging to the metering transformer; the second discharging path is the second path formed by the battery, the MOS transistor, and the controllable switch (K1) for discharging to the metering transformer; when the first discharging path and the second discharging path discharge, due to the unidirectional conduction function of the second diode (D2) provided, neither of them can charge the capacitor (C1).

3. The abnormal recognition method for a gateway electric energy metering device according to claim 2, characterized in that, The anode of the first diode (D1) is connected to the output terminals of the AC / DC module and the DC / DC module. The cathode of the first diode (D1) is respectively connected to one end of the third resistor (R3) and the anode of the second diode (D2). The second end of the third resistor (R3) is connected to the first end of the capacitor (C1), the first end of the fourth resistor (R4), and the first end of the fifth resistor (R5). The second end of the capacitor (C1) and the second end of the fourth resistor (R4) are both grounded. The second end of the fifth resistor (R5) is connected to the control terminal (G2) of the thyristor (D4). The cathode of the second diode (D2) is connected to the first end of the controllable switch (K1), the cathode (K) of the thyristor (D4), and the power supply input terminal of the switchyard watt-hour meter. The second end of the controllable switch (K1) is connected to the drain (D) of the MOS transistor and the anode (A) of the thyristor (D4). The control terminal of the controllable switch (K1) is connected to the output terminal of the control module.

4. The abnormal recognition method for a gateway electric energy metering device according to claim 1, characterized in that; The power loss detection module includes: a first resistor (R1), a second resistor (R2), a voltage regulator diode (D3), and a MOS transistor. The first end of the first resistor (R1) is connected to the output ends of the AC / DC module and the DC / DC module. The second end of the first resistor (11) is connected to the first end of the second resistor (R2), the cathode of the voltage regulator diode (D3), and the gate (G1) of the MOS transistor. The anode of the voltage regulator diode (D3) is grounded. The second end of the second resistor (R2) and the source (S) of the MOS transistor are connected to the connection point of the charging circuit and the battery. When the power loss detection module detects a power loss in the mains power and the photovoltaic panel, it controls the MOS transistor to conduct.

5. The abnormal recognition method of a gateway power metering device according to claim 1, characterized in that, The processing unit can control the activation and deactivation of the switching control unit and the anomaly recognition unit.

6. The abnormal recognition method for a gateway electric energy metering device according to claim 1, characterized in that Establishing a sample set using the historical state evaluation results of the power metering device and the historical operation data of the power metering device, training a model, obtaining a degradation curve of the metering performance of the voltage transformer for high-voltage metering, and predicting the fault risk of the power metering device in a future period specifically includes: Obtaining historical operation data to establish a historical sample set; Inputting the feature set data of the normalized historical sample set into a neural network, using the neural network to perform feature analysis of the historical sample set, and combining the historical state evaluation results to obtain the normal event boundary threshold for training the gateway metering device; Determining a basic training model through the boundary threshold; Inputting the current metering data into the basic training model, and predicting whether the operation curve of the power metering device exceeds the boundary threshold. If it exceeds, there is a fault risk.

Images