Intelligent electric energy meter burning prediction method, device and equipment

Abstract

The invention provides an intelligent electric energy meter burning prediction method, device and equipment, and relates to the technical field of power systems. The method comprises the following steps: collecting topological data and intelligent electric energy meter data in a target area; wherein the intelligent electric energy meter data comprises binding post data of the intelligent electric energy meter and operation data of the intelligent electric energy meter; constructing an impedance equation according to the topological data, the binding post data and the operation data; solving the impedance equation to obtain impedance data of each intelligent electric energy meter; and evaluating the impedance data of each intelligent electric energy meter so as to determine whether each intelligent electric energy meter in the target area has a meter burning risk. According to the method, the binding post data of the electric energy meter is introduced innovatively when the impedance equation is constructed, and scientificity is improved. Besides, an impedance method is easy to be interfered by other factors during solving, so that a solving result is inaccurate, and therefore, a step of evaluating the solved impedance data is additionally added, so that the prediction accuracy is improved.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a method, apparatus and equipment for predicting the burnout of smart energy meters. Background Technology

[0002] Smart meters are susceptible to burnout due to various factors, including installation techniques, operating environment, service life, and electrical load. A burnt-out meter can damage wiring, causing power outages, meter box damage, and even fires, posing significant risks to electrical safety and convenience. Furthermore, meter burnout can impair the meter's data collection function, negatively impacting the economic and management capabilities of power supply companies. A major cause of burnout is high contact resistance and sustained high current, leading to a continuous rise in temperature. This can cause deformation, charring, or even fire of the external insulation of the wires. For example, in one province, 70% to 80% of meter burnout incidents occur in rural areas. This may be due to the high current and prolonged duration of electric heating during winter, coupled with the use of easily oxidized aluminum wires, resulting in increased contact resistance.

[0003] With the widespread deployment of smart meters in my country and the continuous improvement of the supporting electricity consumption information collection system, a large amount of operational data has been accumulated. Using this operational data, line impedance models can be created, establishing a model capable of identifying and predicting the risk of meter burnout, thereby preventing high-risk meters from malfunctioning.

[0004] Current methods for predicting and analyzing meter burnout primarily involve extracting features from multiple dimensions, including basic information about the meter manufacturer, daily electricity consumption, and environmental information such as temperature, and then using machine learning methods for modeling and prediction. However, this method relies on numerous data features, and feature extraction is quite difficult. If the extracted features are inaccurate or of poor quality, it will affect the model's predictive performance. Summary of the Invention

[0005] This invention provides a method, apparatus, and device for predicting the burnout of smart energy meters, in order to solve the problem that traditional burnout analysis methods have limitations in data feature extraction, leading to inaccurate prediction results.

[0006] In a first aspect, embodiments of the present invention provide a method for predicting the burnout of a smart energy meter, comprising: Collect topology data and smart meter data within the target transformer area; the smart meter data includes the terminal data and operating data of the smart meters. Based on the topology data, terminal data, and operational data, construct the impedance equation; The impedance equation is solved to obtain the impedance data of each smart energy meter; The impedance data of each smart meter is evaluated to determine whether there is a risk of meter burnout in the target area.

[0007] In one possible implementation, the terminal data includes the neutral and live currents of the smart meter terminals; based on the topology data, terminal data, and operational data, an impedance equation is constructed, including: Based on the topology data, each smart meter is paired to obtain multiple pairs of smart meter pairings; each pair of smart meter pairings includes two smart meters. For any pair of smart meters, determine whether the terminal data of the two smart meters meet the preset conditions. If the preset conditions are met, the first impedance equation is constructed based on the operating data of the two smart meters. If the preset conditions are not met, a second impedance equation is constructed based on the operating data of the two smart meters. The first impedance equation includes the neutral wire resistance parameter and the live wire resistance parameter; the second impedance equation does not include the neutral wire resistance parameter and the live wire resistance parameter.

[0008] In one possible implementation, smart meters are paired based on topology data to obtain multiple pairs of smart meter combinations, including: Based on the topology data, identify each smart energy meter under the same final-level branch; If the number of smart meters under any given terminal branch is equal to the preset number, then these two smart meters will be paired as a single smart meter combination. If the number of smart meters under any given end branch is greater than the preset number, then the smart meters under the same end branch will be randomly paired to obtain multiple pairs of smart meter pairs. Based on the smart meter pairings obtained from the same final-level branch, multiple pairs of smart meter pairings are obtained.

[0009] In one possible implementation, the first impedance equation is:

[0010] in, The voltage measured by the first smart energy meter; The voltage measured by the second smart energy meter; This refers to the voltage measurement error of smart energy meters. The live wire current measured by the second smart energy meter; The live wire input resistor for the second smart energy meter; The contact resistance of the live wire input of the second smart energy meter; The neutral current measured by the second smart energy meter; The neutral wire input resistance of the second smart energy meter; The neutral contact resistance of the second smart energy meter; The live wire current measured by the first smart energy meter; The live wire input resistor for the first smart energy meter; The contact resistance of the live wire input of the first smart energy meter; The neutral current measured by the first smart energy meter; The neutral wire input resistance of the first smart energy meter; The neutral contact resistance of the first smart energy meter; The second impedance equation is:

[0011] in, The resistance of the circuit where the second smart energy meter is located; The measured current of the second smart energy meter; The resistance of the circuit where the first smart energy meter is located; This is the measured current of the first smart energy meter.

[0012] In one possible implementation, the impedance equation is solved to obtain the impedance data of each smart energy meter, including: The impedance equation is solved using a non-negative linear regression algorithm to obtain the fitted voltage residual data. Determine whether there are outlier data points in the fitted voltage residual data; If outlier data points exist, delete the outlier data points in the fitted voltage residual data, and return to the step of using the non-negative linear regression algorithm to solve the impedance equation and obtain the fitted voltage residual data, until there are no outlier data points in the fitted voltage residual data. If no outlier data points exist, the impedance data corresponding to the current fitted voltage residual data will be used as the impedance data for each smart energy meter.

[0013] In one possible implementation, determining whether outlier data points exist in the fitted voltage residual data includes: An outlier detection method is used to detect outlier data points in the fitted voltage residual data.

[0014] In one possible implementation, the impedance data of each smart meter is evaluated to determine whether there is a risk of meter burnout in the target distribution area, including: Extract sample points from the smart energy meter that exceed a preset current threshold, and calculate the residual signal-to-noise ratio of the smart energy meter based on its operating data. Determine the fitting voltage residual data corresponding to the impedance data of the smart energy meter, and determine the standard deviation of the fitting voltage residual based on the fitting voltage residual data; If the residual signal-to-noise ratio is greater than the preset first threshold and the standard deviation of the fitted voltage residual is less than the preset standard deviation, then the impedance data of the smart energy meter is considered reliable. Based on the impedance data, it can be determined whether there is a risk of meter burnout in each smart meter within the target distribution area.

[0015] In one possible implementation, the impedance data of each smart meter is evaluated to determine whether there is a risk of meter burnout in the target distribution area, including: For any given smart meter, if the number of smart meters under the same final-level branch exceeds a preset number, then the following steps are executed: Determine the multiple pairs of smart meters to which the smart meter belongs, and determine the impedance data of the smart meter based on each pair of smart meters. Extract reliable impedance data from each impedance data set, and obtain average impedance data based on the reliable impedance data. Based on the average impedance data, determine whether the smart energy meter is at risk of burning out.

[0016] Secondly, embodiments of the present invention provide a smart energy meter burnout prediction device, comprising: The data acquisition module is used to collect topology data and smart meter data within the target transformer area; the smart meter data includes the terminal data of the smart meters and the operating data of the smart meters. The module is used to construct impedance equations based on topology data, terminal data, and operational data. The solver module is used to solve the impedance equation and obtain the impedance data of each smart energy meter. The prediction module is used to evaluate the impedance data of each smart meter to determine whether there is a risk of meter burnout in each smart meter within the target area.

[0017] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0018] In this embodiment of the invention, the method provided only requires the data measured by the electricity meter itself, and does not require multi-dimensional data features such as environmental data and manufacturer data. Furthermore, feature extraction is not required in the prediction and judgment process, which can solve the problem of inaccurate prediction caused by feature extraction in traditional methods.

[0019] Furthermore, the method provided in this embodiment of the invention takes into account that when an electricity meter is burnt out, the terminal data on the terminals corresponding to the neutral and live wires will change. Therefore, this embodiment of the invention innovatively incorporates the terminal data of the electricity meter when constructing the impedance equation, thereby improving its scientific rigor. In addition, the impedance method is easily affected by other factors during solution, leading to inaccurate results. Therefore, this embodiment adds an additional step of evaluating the obtained impedance data to improve the accuracy of the prediction. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the implementation of the smart energy meter burn-out prediction method provided in this embodiment of the invention. Figure 2 This is a wiring diagram for a smart energy meter; Figure 3 This is a schematic diagram showing the voltage drop relationship between two smart energy meters under the same final branch; Figure 4 This is a flowchart illustrating the implementation of a smart energy meter burn-out prediction method according to another embodiment of the present invention. Figure 5 This is a schematic diagram of the structure of the smart energy meter burn-out prediction device provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0021] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0022] Figure 1 This is a flowchart illustrating the implementation of the smart energy meter burnout prediction method provided in this embodiment of the invention. Figure 1 As shown, the method may include: Step 110: Collect topology data and smart meter data within the target transformer area; the smart meter data includes the terminal data of the smart meters and the operating data of the smart meters.

[0023] In this embodiment, the topology data includes the topology structure of each smart meter within the target distribution area. The smart meter terminal data may include the neutral and live currents at the neutral and / or live wire terminals of each smart meter; the smart meter operating data may include the current and voltage data collected by each smart meter from each user, i.e., measured current and measured voltage. In addition, the smart meter data may also include user profiles, distribution area profiles, etc., corresponding to each smart meter.

[0024] Step 120: Construct the impedance equation based on the topology data, terminal data, and operational data.

[0025] In this embodiment, the branch to which each smart meter belongs can be determined based on topology data. When constructing the impedance equation, it needs to be based on the operating data of the smart meters under each final-level branch.

[0026] In this embodiment, considering that in actual application, when there is a risk of the smart energy meter burning out, the contact resistance at its terminals will change, which will correspondingly cause changes in the thermal current and neutral current at the terminals.

[0027] However, due to measurement errors, this criterion is not entirely accurate. To improve prediction accuracy and reduce computational load, this embodiment uses the terminal data of each smart meter as the basis for judgment. Based on the terminal data under different conditions, different impedance equations are constructed for each smart meter to improve prediction accuracy.

[0028] Step 130: Solve the impedance equation to obtain the impedance data of each smart energy meter.

[0029] In this embodiment, solving the impedance equation can yield the wire resistance at the neutral wire connection point and the live wire connection point of each smart energy meter, the impedance at the neutral wire exit point and the live wire exit point, or the total impedance of the corresponding connection line and exit line of each smart energy meter, and other related impedance data.

[0030] Step 140: Evaluate the impedance data of each smart meter to determine whether there is a risk of meter burnout in each smart meter within the target area.

[0031] In this embodiment, since the calculation method used in this application is the impedance method, which is easily affected by interference, these interference factors may lead to inaccurate solution results. In this case, the prediction result of smart meter burnout will be inaccurate. Therefore, before predicting whether there is a risk of meter burnout for each smart meter in the target area, this embodiment will also evaluate and test the impedance data of each smart meter to verify whether the impedance data obtained according to the impedance equation is accurate. If it is accurate, the meter burnout risk prediction is made based on the accurate impedance data; if it is inaccurate, it needs to be re-solved to avoid misjudgment due to inaccurate data.

[0032] In summary, the method provided in this embodiment only requires the data measured by the electricity meter itself, and does not require multi-dimensional data features such as environmental data and manufacturer data. Furthermore, feature extraction is not required in the prediction and judgment process, which can solve the problem of inaccurate prediction caused by feature extraction in traditional methods.

[0033] Furthermore, the method provided in this embodiment of the invention takes into account that when an electricity meter is burnt out, the terminal data on the terminals corresponding to the neutral and live wires will change. Therefore, this embodiment of the invention innovatively incorporates the terminal data of the electricity meter when constructing the impedance equation, thereby improving its scientific rigor. In addition, the impedance method is easily affected by other factors during solution, leading to inaccurate results. Therefore, this embodiment adds an additional step of evaluating the obtained impedance data to improve the accuracy of the prediction.

[0034] In an optional embodiment, step 120, which involves constructing the impedance equation based on the topology data, terminal data, and operational data, may include: Step 121: Based on the topology data, pair each smart meter to obtain multiple pairs of smart meter pairings; wherein each pair of smart meter pairings includes two smart meters.

[0035] Step 122: For any pair of smart meters, determine whether the terminal data of the two smart meters meet the preset conditions.

[0036] Step 123: If the preset conditions are met, construct the first impedance equation based on the operating data of the two smart energy meters.

[0037] Step 124: If the preset conditions are not met, construct the second impedance equation based on the operating data of the two smart meters.

[0038] The first impedance equation includes the neutral wire resistance parameter and the live wire resistance parameter; the second impedance equation does not include the neutral wire resistance parameter and the live wire resistance parameter.

[0039] In this embodiment, before constructing the impedance equation, the collected data can be preprocessed to obtain effective data.

[0040] The preprocessing process may include: removing outliers, for example, determining the normal voltage range and removing measured voltage data that does not belong to the normal voltage range; where the normal voltage range can be between 200V and 240V; for current data, removing data that exceeds a preset measured current value; and for missing data, completing the missing data.

[0041] In this embodiment, the smart meters under each terminal branch can be determined based on the topology data. If the number of smart meters under any given terminal branch is equal to two, these two smart meters can be directly paired as a group, and the impedance equations for these two smart meters can be established using the impedance method.

[0042] If the number of smart meters under any given terminal branch is greater than 2, a random matching method can be used to randomly pair the smart meters under the terminal branch to obtain multiple pairs of smart meter pairs. Based on the operating data of each pair, the impedance equation of that pair can be established.

[0043] Figure 2 This is a wiring diagram for a smart energy meter; such as Figure 2 As shown, the wiring structure of a smart energy meter is explained using the voltage measurement circuit of a low-voltage single-phase smart energy meter as an example.

[0044] exist Figure 2 In a smart meter, when the terminals are connected to the power line, there is a corresponding contact resistance. Under these circumstances, the voltage measured by the smart meter will change. Figure 2 In the middle, the potential on the live wire side of the power supply is The potential of the neutral wire is 0V. The resistances of the live wire and the neutral wire are respectively... , The current flowing into the live wire and the neutral wire, that is, the live wire current and the neutral wire current measured by the smart energy meter, are respectively , The contact resistances of the four terminals of the smart energy meter, and the contact resistances of the live wire and neutral wire are respectively... , , , The live wire voltage and neutral wire voltage measured by the smart energy meter at the live wire end and neutral wire end are respectively , .

[0045] Accordingly Figure 2 The structure can be expressed by the following relationship: (1) In this case, the voltage measured by the smart energy meter is: (2) according to Figure 2 The relationship between the potential on the live wire side of the power supply and other parameters can be expressed as: (3) Correspondingly, the voltage measured by the smart energy meter can be expressed as: (4) Figure 3 This is a schematic diagram showing the voltage drop relationship between two smart meters under the same final-level branch, as shown below. Figure 3 As shown.

[0046] Figure 3 middle This represents the voltage of the upper-level common node. and These represent the measured voltages of the first and second smart energy meters corresponding to the upper-level common node, respectively. The resistance of the circuit where the second smart energy meter is located; The measured current of the second smart energy meter; The resistance of the circuit where the first smart energy meter is located; This is the measured current of the first smart energy meter.

[0047] based on Figure 3 It can be seen that the duck-talk relationship from the upper-level common node to the first smart energy meter and the second smart energy meter can be represented as: (5) (6) Subtracting the two equations, we get: (7) The above formula can represent the impedance equation obtained by the traditional impedance method.

[0048] In this embodiment, the terminal data of the energy meter is innovatively introduced when constructing the impedance equation, which improves the scientific nature. Therefore, this embodiment improves the traditional impedance equation by judging each pair of smart energy meters to see if they meet the preset conditions, so as to construct different impedance equations for smart energy meter pairings under different conditions.

[0049] Specifically, it is determined whether both smart meters in each pair of smart meters have a neutral current, or whether the neutral current of the first smart meter is consistent with the neutral current of the second smart meter, or whether the live current of the first smart meter is consistent with the live current of the second smart meter. If both have a neutral current, or if the neutral current and the live current are consistent, then a first impedance equation is constructed based on the operating data of that pair. The first impedance equation can be expressed as:

[0050] in, The voltage measured by the first smart energy meter; The voltage measured by the second smart energy meter; This refers to the voltage measurement error of smart energy meters. The live wire current measured by the second smart energy meter; The live wire input resistor for the second smart energy meter; The contact resistance of the live wire input of the second smart energy meter; The neutral current measured by the second smart energy meter; The neutral wire input resistance of the second smart energy meter; The neutral contact resistance of the second smart energy meter; The live wire current measured by the first smart energy meter; The live wire input resistor for the first smart energy meter; The contact resistance of the live wire input of the first smart energy meter; The neutral current measured by the first smart energy meter; The neutral wire input resistance of the first smart energy meter; The neutral contact resistance of the first smart energy meter; The process of constructing the first impedance equation is as follows: Substituting the measured voltage of the smart energy meter in formula (3) into formulas (5) and (6), we get:

[0051]

[0052] Subtracting formulas (9) and (10), we get: (11) In this embodiment, considering that the measured voltage measured by the smart energy meter itself also has errors, the measurement error needs to be taken into account when constructing the impedance equation. Therefore, it is necessary to adjust the formula (11).

[0053] However, since we usually apply the impedance method to solve the problem at the final branch, the measured voltages of the two smart meters are... The correlation is very high. If the two voltage terms and Adding error terms to all solutions can lead to trivial solutions and may also cause instability in the solution.

[0054] Therefore, in this embodiment, only the error term is added for the first smart energy meter. After sorting and optimization, the first impedance equation shown in formula (8) can be obtained. In this embodiment, since the smart energy meter targeted in this embodiment is a single-phase meter, the values ​​of the live and neutral wire currents are completely consistent and inseparable when there are no current measurement errors, series connections, or common zeros. However, when the proportion of neutral wire sampling is lower than that of live wire sampling in actual situations, and when the neutral wire current is missing or the two are consistent, only the live wire current can be used. That is, when there is no neutral wire current, or when the neutral wire current and the live wire current are inconsistent, formula (8) can be converted into the following format to obtain the second impedance equation: (12) in, ; . , All currents are triggered by the live wire of the common node, and flow through the energy meter and the neutral wire back to the total impedance of the common node.

[0055] By solving the impedance equations constructed above, the impedance data corresponding to each smart energy meter can be obtained.

[0056] In an optional embodiment, solving the impedance equation in step 130 to obtain the impedance data of each smart energy meter may include: Step 131: Use a non-negative linear regression algorithm to solve the impedance equation and obtain the fitted voltage residual data.

[0057] Step 132: Determine whether there are outlier data points in the fitted voltage residual data.

[0058] Step 133: If outlier data points exist, delete the outlier data points in the fitted voltage residual data and return to the step of using the non-negative linear regression algorithm to solve the impedance equation and obtain the fitted voltage residual data, until there are no outlier data points in the fitted voltage residual data.

[0059] Step 134: If there are no outlier data points, then use the impedance data corresponding to the current fitted voltage residual data as the impedance data of each smart energy meter.

[0060] In this embodiment, for the impedance equation, each coefficient to be solved is an impedance value, which should theoretically be non-negative. Therefore, non-negative linear regression is used to solve the equation, such as using the least squares method with constraints or quadratic programming to solve the equation, so that the solved resistance value is non-negative, thus satisfying the actual physical phenomenon meaning of the coefficient.

[0061] Considering that there may be abnormal noise points during the solution process, this embodiment verifies and optimizes the obtained solution to remove abnormal noise points, making the solution of the impedance equation more accurate.

[0062] For example, by using a nonnegative linear regression algorithm to solve the impedance equation, impedance data and fitted voltage residual data can be obtained. The box plot method is used to detect outliers in the fitted voltage residual data. If outlier points exist, they are treated as data noise points and deleted. The solution is then repeated until no outliers are found.

[0063] In an optional embodiment, the process may include: Step 1: Select any data range as a sample, and use the current and voltage data in this range as the data to solve the impedance equation; Step 2: Apply the impedance solution model to calculate the impedance, and fit the solution impedance to obtain the fitted voltage residual; Step 3: Apply the box-plot outlier detection method to the residuals to detect outlier data points and delete them. The specific method for removing outliers using box plots is as follows: Find the minimum and maximum values ​​of the sequence; the lower quartile (Q1, at the 25th percentile), the upper quartile (Q3, at the 75th percentile), and the median (at the 50th percentile). Box plots can display outliers, which can be calculated using the IQR (Interquartile Range, i.e., Q3-Q1). Outliers are defined as values ​​less than Q1 - 1.5IQR or greater than Q3 + 1.5IQR.

[0064] Step 4: Repeat Step 1-Step 3 until no outlier noise data can be detected in the residual data.

[0065] In an optional embodiment, step 140, which evaluates the impedance data of each smart meter to determine whether there is a risk of meter burnout in each smart meter within the target area, may include: Extract sample points from the smart energy meter that exceed a preset current threshold, and calculate the residual signal-to-noise ratio of the smart energy meter based on its operating data.

[0066] Determine the fitting voltage residual data corresponding to the impedance data of the smart energy meter, and determine the standard deviation of the fitting voltage residual based on the fitting voltage residual data.

[0067] If the residual signal-to-noise ratio is greater than the preset first threshold and the standard deviation of the fitted voltage residual is less than the preset standard deviation, then the impedance data of the smart energy meter is considered reliable.

[0068] Based on the impedance data, it can be determined whether there is a risk of meter burnout in each smart meter within the target distribution area.

[0069] In this embodiment, impedance calculations are performed only by pairing meters from the same final branch. However, the impedance method is susceptible to interference from factors such as clock deviations, which can lead to unreliable results. Therefore, the impedance calculation results need to be evaluated; only reliable results can accurately predict the risk of meter damage.

[0070] Credibility evaluation can be divided into two dimensions: smart meter dimension and overall solution fitting.

[0071] The credibility evaluation of smart energy meters can be achieved through the following steps: In this embodiment, the credibility evaluation of the smart energy meter dimension is essentially an evaluation of the regression coefficient impedance value of the smart energy meter, which can be performed using the T-test method. Accordingly, p-value thresholds and t-value thresholds are set. When the p-value of the regression coefficient is less than the p-value threshold or the t-value is greater than the t-value threshold, the regression result is considered significant and credible. For example, a p-value threshold of 0.01 or a t-value threshold of 1.97 can be used.

[0072] Specifically, multiple sample points exceeding a preset current threshold are extracted from the smart energy meter, and the residual signal-to-noise ratio (SNR) of the fitted voltage data corresponding to these sample points is calculated. The reliability of the obtained impedance data is evaluated based on the SNR. The formula for calculating the residual SNR is as follows: (13) In the formula, This represents the fitted impedance of the x-th smart energy meter. Let x represent the current of the x-th meter at time t. Let be the fitted voltage residual at time t, and n represent the number of sample points.

[0073] For example, at time t, a certain current point of the first smart energy meter is... The current of the second smart energy meter is At this time, the pressure difference is Using the impedance obtained from the solution That is, the impedance at this time The residual is 0.3V, and the signal-to-noise ratio is: .

[0074] According to the preset first threshold, when the calculated residual signal-to-noise ratio is greater than the threshold, the signal-to-noise ratio is considered to be significant and the impedance data is considered reliable.

[0075] The residual signal-to-noise ratio of the smart energy meter is calculated based on its operating data.

[0076] The reliability evaluation of the overall solution fitting dimension can be achieved through the following steps: In this embodiment, the credibility evaluation of the overall solution fitting dimension is essentially to determine the quantization noise of the branch voltage difference. The closer the overall fitted voltage residual is to the white noise of the quantization noise, the higher the credibility of the overall solution.

[0077] Specifically, to evaluate whether the fitted voltage residuals are close to white noise, the Ljung-Box test algorithm can be used to determine whether the residuals have significant autocorrelation. If the p-value is high (usually greater than 0.05), it is considered close to white noise.

[0078] To evaluate the relationship between the voltage residual amplitude and quantization noise, the standard deviation of the fitted voltage residual is considered reliable when it is within a set threshold (which can be 0.2). That is, when the standard deviation of the fitted residual is within 0.2V, the overall fitting evaluation result is considered reliable; otherwise, it is not reliable.

[0079] In this embodiment, if the residual signal-to-noise ratio is greater than a preset first threshold and the standard deviation of the fitted voltage residual is less than a preset standard deviation, then the impedance data of the smart energy meter is considered reliable.

[0080] Based on reliable impedance data, determine whether there is a risk of meter burnout in each smart meter within the target distribution area.

[0081] In this embodiment, the method is a prediction for the case where the number of smart meters under the same terminal branch is equal to 2. When the number of smart meters under the same terminal branch is equal to 2, it is possible to directly determine whether there is a risk of meter burnout in each smart meter in the target area based on reliable impedance data.

[0082] If the number of smart meters under the same final branch is greater than 2, the following method is used to determine whether there is a risk of meter burnout in each smart meter in the target area.

[0083] In an optional embodiment, step 140, which evaluates the impedance data of each smart meter to determine whether there is a risk of meter burnout in each smart meter within the target area, may include: For any given smart meter, if the number of smart meters under the same final-level branch exceeds a preset number, then the following steps are executed: Determine the multiple pairs of smart meters to which the smart meter belongs, and determine the impedance data of the smart meter based on each pair of smart meters. Extract reliable impedance data from each impedance data set, and obtain average impedance data based on the reliable impedance data. Based on the average impedance data, determine whether the smart energy meter is at risk of burning out.

[0084] In this embodiment, firstly, the impedance data corresponding to each smart energy meter has been evaluated and determined to be reliable impedance data.

[0085] In this embodiment, if there are multiple smart meters under the same final branch, taking any one smart meter as an example, the impedance data of that smart meter is determined based on the pairing and combination of each pair of smart meters.

[0086] Using the smart energy meter as a granular unit, the average value of multiple reliable results is calculated to obtain the average impedance data. This average impedance data is then used as the final circuit impedance of the meter to determine whether any smart energy meter is at risk of burnout.

[0087] The formula for calculating the average impedance data is as follows: (14) in, For the first The impedance value obtained by the group, For the first j The number of combinations and pairs corresponding to each smart energy meter.

[0088] In this embodiment, if the total impedance of any smart energy meter is greater than a preset impedance threshold, it can be determined that the smart energy meter is at risk of burning out, and an early warning will be issued.

[0089] For example, the preset impedance threshold can be determined as follows: if preliminary analysis shows that there is a risk of meter burnout when the loop impedance is 100 milliohms, then the preset impedance threshold can be set to 100 milliohms. Accordingly, when the loop impedance of a smart meter is detected to be 100 milliohms, an early warning is issued, and personnel need to be dispatched to the site for handling.

[0090] Figure 4 This is a flowchart illustrating the implementation of a smart energy meter burnout prediction method according to another embodiment of the present invention, as shown below. Figure 4 As shown, the method may include: First, obtain the data from the transformer substation archives and the voltage and current data measured by each smart meter.

[0091] Then, the voltage and current data measured by each smart energy meter are initially cleaned.

[0092] Group the smart energy meters under the same final branch, and determine whether each smart energy meter in each group has neutral current data, or whether the neutral current data and live current data are consistent.

[0093] If they are consistent, then construct the first impedance equation with zero-wire resistance parameter based on the operating data of each group of smart energy meters. If they are inconsistent, a second impedance equation without zero-wire and live-wire resistance parameters is constructed based on the operating data of each group of smart energy meters.

[0094] The impedance equations are solved using a constrained nonnegative linear regression method.

[0095] During the solution process, residual iteration of voltage and current data is performed to remove noise points, and the corresponding impedance data is determined based on the current and voltage data when there are no noise points.

[0096] To ensure the reliability of the obtained impedance data, each impedance data point is evaluated to obtain impedance data with high reliability.

[0097] If the number of smart meters in the same terminal branch is equal to two, the prediction result is determined directly based on their corresponding impedance data. If the number of smart meters in the same terminal branch is greater than two, a smart meter burn-out warning will be issued based on the average impedance corresponding to that terminal branch.

[0098] In summary, compared to machine learning algorithms, which require extracting features from multiple dimensions and face significant challenges in extracting effective features, this method generally lacks interpretability. This invention, through modeling the burning of electricity meters based on the heating principle of electrical wires, utilizes actual physical phenomena to perform more accurate modeling and analysis. Furthermore, it provides explanations for model solving, outlier removal, and result reliability evaluation, thus exhibiting broader applicability. Ultimately, it can more accurately achieve online analysis and early warning of meter burning phenomena, enabling early detection of meters at risk of burning. For risks not caused by meter problems, it allows for investigation and replacement, reducing the risk of meter damage. Timely replacement of meters already at risk of burning reduces economic losses for power companies and users, improving the safety of electricity use for users. It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0099] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0100] Figure 5 A schematic diagram of the structure of the smart energy meter burn-out prediction device provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below: like Figure 5 As shown, the smart energy meter burnout prediction device 5 includes: The acquisition module 51 is used to acquire topology data and smart meter data within the target transformer area; the smart meter data includes the terminal data of the smart meter and the operating data of the smart meter. Module 52 is used to construct impedance equations based on topology data, terminal data, and operational data. The solver module 53 is used to solve the impedance equation and obtain the impedance data of each smart energy meter. The prediction module 54 is used to evaluate the impedance data of each smart energy meter to determine whether there is a risk of meter burnout in each smart energy meter in the target area.

[0101] In one possible implementation, the terminal data includes the neutral current and live current of the smart meter terminals. Module 52 is used specifically for: Based on the topology data, each smart meter is paired to obtain multiple pairs of smart meter pairings; each pair of smart meter pairings includes two smart meters. For any pair of smart meters, determine whether the terminal data of the two smart meters meet the preset conditions. If the preset conditions are met, the first impedance equation is constructed based on the operating data of the two smart meters. If the preset conditions are not met, a second impedance equation is constructed based on the operating data of the two smart meters. The first impedance equation includes the neutral wire resistance parameter and the live wire resistance parameter; the second impedance equation does not include the neutral wire resistance parameter and the live wire resistance parameter.

[0102] In one possible implementation, module 52 is specifically used for: Based on the topology data, identify each smart energy meter under the same final-level branch; If the number of smart meters under any given terminal branch is equal to the preset number, then these two smart meters will be paired as a single smart meter combination. If the number of smart meters under any given end branch is greater than the preset number, then the smart meters under the same end branch will be randomly paired to obtain multiple pairs of smart meter pairs. Based on the smart meter pairings obtained from the same final-level branch, multiple pairs of smart meter pairings are obtained.

[0103] In one possible implementation, the first impedance equation is:

[0104] in, The voltage measured by the first smart energy meter; The voltage measured by the second smart energy meter; This refers to the voltage measurement error of smart energy meters. The live wire current measured by the second smart energy meter; The live wire input resistor for the second smart energy meter; The contact resistance of the live wire input of the second smart energy meter; The neutral current measured by the second smart energy meter; The neutral wire input resistance of the second smart energy meter; The neutral contact resistance of the second smart energy meter; The live wire current measured by the first smart energy meter; The live wire input resistor for the first smart energy meter; The contact resistance of the live wire input of the first smart energy meter; The neutral current measured by the first smart energy meter; The neutral wire input resistance of the first smart energy meter; The neutral contact resistance of the first smart energy meter; The second impedance equation is:

[0105] in, The resistance of the circuit where the second smart energy meter is located; The measured current of the second smart energy meter; The resistance of the circuit where the first smart energy meter is located; This is the measured current of the first smart energy meter.

[0106] In one possible implementation, the solver module 53 is specifically used for: The impedance equation is solved using a non-negative linear regression algorithm to obtain the fitted voltage residual data. Determine whether there are outlier data points in the fitted voltage residual data; If outlier data points exist, delete the outlier data points in the fitted voltage residual data, and return to the step of using the non-negative linear regression algorithm to solve the impedance equation and obtain the fitted voltage residual data, until there are no outlier data points in the fitted voltage residual data. If no outlier data points exist, the impedance data corresponding to the current fitted voltage residual data will be used as the impedance data for each smart energy meter.

[0107] In one possible implementation, the solver module 53 is specifically used for: An outlier detection method is used to detect outlier data points in the fitted voltage residual data.

[0108] In one possible implementation, the prediction module 54 is specifically used for: Extract sample points from the smart energy meter that exceed a preset current threshold, and calculate the residual signal-to-noise ratio of the smart energy meter based on its operating data. Determine the fitting voltage residual data corresponding to the impedance data of the smart energy meter, and determine the standard deviation of the fitting voltage residual based on the fitting voltage residual data; If the residual signal-to-noise ratio is greater than the preset first threshold and the standard deviation of the fitted voltage residual is less than the preset standard deviation, then the impedance data of the smart energy meter is considered reliable. Based on the impedance data, it can be determined whether there is a risk of meter burnout in each smart meter within the target distribution area.

[0109] In one possible implementation, the prediction module 54 is specifically used for: For any given smart meter, if the number of smart meters under the same final-level branch exceeds a preset number, then the following steps are executed: Determine the multiple pairs of smart meters to which the smart meter belongs, and determine the impedance data of the smart meter based on each pair of smart meters. Extract reliable impedance data from each impedance data set, and obtain average impedance data based on the reliable impedance data. Based on the average impedance data, determine whether the smart energy meter is at risk of burning out.

[0110] Figure 6 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 6 As shown, the electronic device 6 of this embodiment includes a processor 60 and a memory 61. The memory 61 stores a computer program 62. When the processor 60 executes the computer program 62, it implements the steps in the various method embodiments described above. Alternatively, when the processor 60 executes the computer program 62, it implements the functions of each module / unit in the various device embodiments described above.

[0111] For example, computer program 62 may be divided into one or more modules / units, which are stored in memory 61 and executed by processor 60 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 62 in electronic device 6.

[0112] Electronic device 6 may include, but is not limited to, processor 60 and memory 61. Those skilled in the art will understand that... Figure 6 This is merely an example of electronic device 6 and does not constitute a limitation on electronic device 6. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 6 may also include input / output devices, network access devices, buses, etc.

[0113] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0114] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0115] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for predicting the burnout of a smart energy meter, characterized in that, include: Collect topology data and smart meter data within the target transformer area; wherein, the smart meter data includes smart meter terminal data and smart meter operating data; Based on the topology data, the terminal data, and the operational data, an impedance equation is constructed. The impedance equation is solved to obtain the impedance data of each smart energy meter; The impedance data of each smart energy meter is evaluated to determine whether there is a risk of meter burnout in each smart energy meter within the target distribution area.

2. The method for predicting the burnout of a smart energy meter according to claim 1, characterized in that, The terminal data includes the neutral current and live current of the smart energy meter terminals; the construction of the impedance equation based on the topology data, the terminal data, and the operating data includes: Based on the topology data, each smart meter is paired to obtain multiple pairs of smart meter pairings; wherein each pair of smart meter pairings includes two smart meters. For any pair of smart meters, determine whether the terminal data of the two smart meters meet the preset conditions. If the preset conditions are met, the first impedance equation is constructed based on the operating data of the two smart meters. If the preset conditions are not met, a second impedance equation is constructed based on the operating data of the two smart meters. The first impedance equation includes neutral wire resistance parameters and live wire resistance parameters; the second impedance equation does not include neutral wire resistance parameters and live wire resistance parameters.

3. The method for predicting the burnout of a smart energy meter according to claim 2, characterized in that, The process of pairing smart meters based on the topology data to obtain multiple pairs of smart meter combinations includes: Based on the topology data, identify each smart energy meter under the same final-level branch; If the number of smart meters under any given terminal branch is equal to the preset number, then these two smart meters will be paired as a single smart meter combination. If the number of smart meters under any given end branch is greater than the preset number, then the smart meters under the same end branch will be randomly paired to obtain multiple pairs of smart meter pairs. Based on the smart meter pairings obtained from the same final-level branch, multiple pairs of smart meter pairings are obtained.

4. The method for predicting the burnout of a smart energy meter according to claim 2, characterized in that, The first impedance equation is: in, The voltage measured by the first smart energy meter; The voltage measured by the second smart energy meter; This refers to the voltage measurement error of smart energy meters. The live wire current measured by the second smart energy meter; The live wire input resistor for the second smart energy meter; The contact resistance of the live wire input of the second smart energy meter; The neutral current measured by the second smart energy meter; The neutral wire input resistance of the second smart energy meter; The neutral contact resistance of the second smart energy meter; The live wire current measured by the first smart energy meter; The live wire input resistor for the first smart energy meter; The contact resistance of the live wire input of the first smart energy meter; The neutral current measured by the first smart energy meter; The neutral wire input resistance of the first smart energy meter; The neutral contact resistance of the first smart energy meter; The second impedance equation is: in, The resistance of the circuit where the second smart energy meter is located; The measured current of the second smart energy meter; The resistance of the circuit where the first smart energy meter is located; This is the measured current of the first smart energy meter.

5. The method for predicting the burnout of a smart energy meter according to claim 1, characterized in that, Solving the impedance equation to obtain the impedance data of each smart energy meter includes: The impedance equation is solved using a non-negative linear regression algorithm to obtain the fitted voltage residual data. Determine whether there are outlier data points in the fitted voltage residual data; If outlier data points exist, delete the outlier data points in the fitted voltage residual data, and return to the step of using the non-negative linear regression algorithm to solve the impedance equation and obtain the fitted voltage residual data, until there are no outlier data points in the fitted voltage residual data. If no outlier data points exist, the impedance data corresponding to the current fitted voltage residual data will be used as the impedance data for each smart energy meter.

6. The method for predicting the burnout of a smart energy meter according to claim 5, characterized in that, The step of determining whether outlier data points exist in the fitted voltage residual data includes: An outlier detection method is used to detect whether there are outlier data points in the fitted voltage residual data.

7. The method for predicting the burnout of a smart energy meter according to claim 1, characterized in that, The evaluation of the impedance data of each smart meter to determine whether there is a risk of meter burnout in each smart meter within the target distribution area includes: Extract sample points from the smart energy meter that exceed a preset current threshold, and calculate the residual signal-to-noise ratio of the smart energy meter based on its operating data. Determine the fitting voltage residual data corresponding to the impedance data of the smart energy meter, and determine the standard deviation of the fitting voltage residual based on the fitting voltage residual data; If the residual signal-to-noise ratio is greater than a preset first threshold and the standard deviation of the fitted voltage residual is less than a preset standard deviation, then the impedance data of the smart energy meter is considered reliable. Based on the impedance data, it is determined whether there is a risk of meter burnout in each smart energy meter within the target distribution area.

8. The method for predicting the burnout of a smart energy meter according to claim 7, characterized in that, The evaluation of the impedance data of each smart meter to determine whether there is a risk of meter burnout in each smart meter within the target distribution area includes: For any given smart meter, if the number of smart meters under the same final-level branch exceeds a preset number, then the following steps are executed: Determine the multiple pairs of smart meters to which the smart meter belongs, and determine the impedance data of the smart meter based on each pair of smart meters. Extract reliable impedance data from each impedance data set, and obtain average impedance data based on the reliable impedance data. Based on the average impedance data, determine whether the smart energy meter is at risk of burning out.

9. A smart energy meter burnout prediction device, characterized in that, include: The data acquisition module is used to acquire topology data and smart meter data within the target transformer area; wherein, the smart meter data includes the terminal data of the smart meter and the operating data of the smart meter. The construction module is used to construct impedance equations based on the topology data, the terminal data, and the operating data; The solution module is used to solve the impedance equation to obtain the impedance data of each smart energy meter. The prediction module is used to evaluate the impedance data of each smart energy meter to determine whether there is a risk of meter burnout in each smart energy meter within the target distribution area.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 8.

Images