Power grid hidden fault diagnosis system based on multi-source measurement switch state collaborative analysis

Abstract

The invention relates to the technical field of power grid fault diagnosis, and particularly discloses a power grid hidden fault diagnosis system based on multi-source measurement switch state collaborative analysis, and the system comprises an acquisition module which obtains a plurality of groups of electrical measurement data of a target power transmission line under a plurality of continuous different environment temperature conditions; determining real-time equivalent total impedance and real-time equivalent total admittance of the target power transmission line based on the electrical measurement data; the analysis module is used for establishing a first function relationship taking the environment temperature as an independent variable and the real-time equivalent total impedance as a dependent variable, and establishing a second function relationship taking the environment temperature as the independent variable and the real-time equivalent total admittance as a dependent variable; extracting a parameter constant component based on the first function relation and the second function relation; and the judgment module is used for acquiring the actually measured equivalent total impedance X1 and the actually measured equivalent total admittance X2 at the current moment, and judging whether the target power transmission line has a hidden fault or not by combining the parameter constant component. The accuracy of power grid fault diagnosis is improved.

Description

Technical Field

[0001] This invention relates to the field of power grid fault diagnosis technology, and more specifically to a power grid latent fault diagnosis system based on collaborative analysis of multi-source measurement switch states. Background Technology

[0002] In extreme cold weather, the ice that forms on the surface of transmission lines is not a static insulator. When the ice layer contains impurities or partially melts, its conductivity increases significantly, thereby reducing the overall insulation resistance of the line and causing an increase in leakage current. This change in electrical parameters caused by environmental factors often manifests as small anomalies such as current fluctuations and voltage deviations in monitoring data.

[0003] However, latent faults within the line itself, such as insulation aging or localized damage, can also produce nearly identical electrical characteristics: increased leakage current and decreased insulation performance. Existing fault diagnosis methods typically establish a baseline based on the line's rated parameters, but icing causes actual parameters to continuously deviate from this baseline. This results in discrepancies between monitoring data and standard values, which are simultaneously mixed with environmental interference and genuine fault signals. This makes it difficult for operators to accurately identify the root cause of anomalies: they may misjudge normal parameter drift caused by icing as a fault, unnecessarily increasing maintenance costs; or they may miss potential hazards because icing masks the true characteristics of the fault, thereby increasing the risk of line outages or even accidents. Summary of the Invention

[0004] The purpose of this invention is to provide a power grid latent fault diagnosis system based on multi-source measurement switch state collaborative analysis, thereby solving the above-mentioned technical problems.

[0005] The objective of this invention can be achieved through the following technical solutions: A power grid latent fault diagnosis system based on multi-source measurement switch state collaborative analysis includes: Acquisition module: Acquires multiple sets of electrical measurement data of the target transmission line under various ambient temperature conditions. The real-time equivalent total impedance and real-time equivalent total admittance of the target transmission line are determined based on electrical measurement data. Analysis module: Establish a first functional relationship with ambient temperature as the independent variable and real-time equivalent total impedance as the dependent variable, and establish a second functional relationship with ambient temperature as the independent variable and real-time equivalent total admittance as the dependent variable. Extract constant components of parameters based on the first and second functional relationships. Judgment module: Obtain the measured equivalent total impedance X1 and measured equivalent total admittance X2 at the current moment, and determine whether there is a hidden fault in the target transmission line by combining the constant component of the parameter.

[0006] As a further aspect of the present invention: acquiring electrical measurement data includes: Electrical measurement data is periodically collected within a preset time span. The electrical measurement data includes the instantaneous values ​​of the three-phase voltage at the beginning, the instantaneous values ​​of the three-phase current at the beginning, the instantaneous values ​​of the three-phase voltage at the end, and the instantaneous values ​​of the three-phase current at the end. The ambient temperature at the time of sampling is also recorded. Convert the instantaneous values ​​of the three-phase voltages at the beginning to positive sequence voltage components, denoted as the positive sequence voltage at the beginning; convert the instantaneous values ​​of the three-phase voltages at the end to positive sequence voltage components, denoted as the positive sequence voltage at the end; convert the instantaneous values ​​of the three-phase currents at the beginning to positive sequence current components, denoted as the positive sequence current at the beginning; convert the instantaneous values ​​of the three-phase currents at the end to positive sequence current components, denoted as the positive sequence current at the end. The first-end positive sequence voltage, first-end positive sequence current, last-end positive sequence voltage and last-end positive sequence current, and the corresponding ambient temperature at each sampling time are stored as a set of electrical measurement data.

[0007] As a further aspect of the present invention: the calculation of real-time equivalent total impedance and real-time equivalent total admittance includes: Obtain the electrical measurement data at the i-th sampling time, including the first-end positive sequence voltage Us, the first-end positive sequence current Is, the last-end positive sequence voltage Ur, and the last-end positive sequence current Ir. Establish computational relationships based on the equations of a two-port network of transmission lines. These relationships include: ; ; Among them, transmission parameters L represents the length of the target transmission line, and the propagation constant of the target transmission line is... Z and Y represent the series impedance per unit length and the parallel admittance per unit length of the target transmission line, respectively. Transmission parameters Characteristic impedance ; Transmission parameters ; Transmission parameters ; Solve the above equations simultaneously to obtain the series impedance Z per unit length and the parallel admittance Y per unit length of the target transmission line; The real-time equivalent total impedance is obtained by multiplying the series impedance Z per unit length by the length L of the target transmission line; the real-time equivalent total admittance is obtained by multiplying the parallel admittance per unit length by the length L of the target transmission line.

[0008] As a further aspect of the present invention: extracting constant components of parameters includes: The first data point is generated in the coordinate system with ambient temperature as the abscissa and the corresponding real-time equivalent total impedance as the ordinate. The first function relationship is obtained by curve fitting of the first data point. If the first functional relationship is monotonically increasing or monotonically decreasing, then the real-time equivalent total impedance at an ambient temperature of 0 is taken as the constant component of the first parameter; if the first functional relationship is neither monotonically increasing nor monotonically decreasing, then the mean of the real-time equivalent total impedance among all the first data points is calculated as the constant component of the first parameter. A second data point is generated in the coordinate system with ambient temperature as the abscissa and the corresponding real-time equivalent total admittance as the ordinate. The second function relationship is obtained by curve fitting of the second data point. If the second functional relationship is monotonically increasing or monotonically decreasing, then the real-time equivalent total admittance at an ambient temperature of 0 is taken as the constant component of the second parameter; if the second functional relationship is neither monotonically increasing nor monotonically decreasing, then the mean of the real-time equivalent total admittance among all the second data points is calculated as the constant component of the second parameter. The first constant component and the second constant component are referred to as the constant component of the parameter.

[0009] As a further aspect of the present invention: determining whether a latent fault exists includes: Calculate the absolute value ΔX1 of the difference between the measured equivalent total impedance X1 and the first constant parameter component x1 in the constant parameter components; calculate the absolute value ΔX2 of the difference between the measured equivalent total admittance X2 and the second constant parameter component x2 in the constant parameter components. Calculate the first offset ΔX1' = ΔX1 / x1, and calculate the second offset ΔX2' = ΔX2 / x2; If max(ΔX1', ΔX2') > YZ, then the target transmission line is determined to have a hidden fault. YZ represents the preset judgment threshold, and max(ΔX1', ΔX2') represents obtaining the larger value between ΔX1' and ΔX2'.

[0010] As a further aspect of the present invention: after determining that a latent fault exists, the method further includes: Substituting the current ambient temperature into the first and second functional relationships respectively, we obtain the theoretical impedance P1 and theoretical admittance P2. Calculate the difference P1' = X1 - P1, and calculate the difference P2' = X2 - P2; like If the type of latent fault is determined to be abnormal series impedance, K is a preset proportional coefficient and K>1; like If so, the type of latent fault is determined to be parallel admittance anomaly; If neither of the above two conditions is met, then the latent fault is determined to affect both the series impedance and the parallel admittance.

[0011] As a further aspect of the present invention, it also includes: Record the time points at which a latent fault is identified, calculate the time interval between two adjacent time points, and visualize the time intervals using methods such as line charts and bar charts.

[0012] The beneficial effects of this invention compared to the prior art are as follows: This invention collects multi-source measurement data of transmission lines under different ambient temperature conditions and establishes the relationship between electrical parameters and temperature variations. This effectively separates the variability caused by low-temperature environmental factors such as icing, allowing for the identification and compensation of normal drift in monitoring parameters due to environmental factors. The changes in the equivalent impedance and equivalent admittance of the line after temperature-related characteristic stripping reflect the line's inherent condition, thus highlighting the additional offset caused by latent faults in monitoring. By comprehensively judging the offset, the existence of latent faults can be accurately identified even under significant environmental interference, and the emphasis of fault impact can be further differentiated based on the differences in impedance and admittance directions. Recording the fault occurrence time and its time interval provides continuous status information for operation and maintenance decisions. In summary, this invention maintains the stability and reliability of diagnostic results under strong environmental disturbances, distinguishes the superimposed effects of environmental factors and inherent equipment degradation on electrical quantities, and thus improves the accuracy and specificity of latent fault identification in transmission lines. Attached Figure Description

[0013] The invention will now be further described with reference to the accompanying drawings.

[0014] Figure 1 This is a flowchart illustrating the power grid latent fault diagnosis system based on multi-source measurement switch state collaborative analysis of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Please see Figure 1 As shown, this invention is a power grid latent fault diagnosis system based on multi-source measurement switch state collaborative analysis, comprising: Acquisition module: Acquires multiple sets of electrical measurement data of the target transmission line under various ambient temperature conditions. In a preferred embodiment of the present invention, acquiring electrical measurement data includes: Electrical measurement data is periodically collected within a preset time span. The electrical measurement data includes the instantaneous values ​​of the three-phase voltage at the beginning, the instantaneous values ​​of the three-phase current at the beginning, the instantaneous values ​​of the three-phase voltage at the end, and the instantaneous values ​​of the three-phase current at the end. The ambient temperature at the time of sampling is also recorded. Convert the instantaneous values ​​of the three-phase voltages at the beginning to positive sequence voltage components, denoted as the positive sequence voltage at the beginning; convert the instantaneous values ​​of the three-phase voltages at the end to positive sequence voltage components, denoted as the positive sequence voltage at the end; convert the instantaneous values ​​of the three-phase currents at the beginning to positive sequence current components, denoted as the positive sequence current at the beginning; convert the instantaneous values ​​of the three-phase currents at the end to positive sequence current components, denoted as the positive sequence current at the end. The first-end positive sequence voltage, first-end positive sequence current, last-end positive sequence voltage and last-end positive sequence current, and the corresponding ambient temperature at each sampling time are stored as a set of electrical measurement data.

[0017] It should be noted that periodically acquiring instantaneous measurements of the voltage and current at the beginning and end points under different ambient temperatures and uniformly converting them into positive-sequence components essentially constructs a consistent and electrically comparable state characterization for the line. Three-phase quantities are significantly affected by phase sequence imbalance, noise, and transient disturbances, while positive-sequence quantities can extract the core components reflecting the parameter characteristics of the line under steady-state conditions at power frequency, providing a clear physical correspondence when subsequently solving for the line's equivalent impedance and equivalent admittance based on port relationships. By recording the ambient temperature simultaneously with the data acquisition, a correlated sample is formed between electrical quantities and environmental factors, allowing parameter changes caused by low-temperature conditions such as icing to be characterized in the temperature dimension. The construction of positive-sequence quantities ensures the comparability of the same physical quantity across different sampling times, while temperature recording provides the necessary input for establishing temperature correlations. The combination of these two aspects makes it possible to extract the patterns of environmental influences and separate them from the changes in the line's own state, contributing to the formation of stable functional relationships in parameter inversion and temperature fitting. This ensures that the final offset value used for fault diagnosis has a true state indication meaning, thereby supporting the accuracy of latent fault identification.

[0018] The real-time equivalent total impedance and real-time equivalent total admittance of the target transmission line are determined based on electrical measurement data. In another preferred embodiment of the present invention, the calculation of the real-time equivalent total impedance and the real-time equivalent total admittance includes: Obtain the electrical measurement data at the i-th sampling time, including the first-end positive sequence voltage Us, the first-end positive sequence current Is, the last-end positive sequence voltage Ur, and the last-end positive sequence current Ir. Establish computational relationships based on the equations of a two-port network of transmission lines. These relationships include: ; ; Among them, transmission parameters L represents the length of the target transmission line, and the propagation constant of the target transmission line is... Z and Y represent the series impedance per unit length and the parallel admittance per unit length of the target transmission line, respectively. Transmission parameters Characteristic impedance ; Transmission parameters ; Transmission parameters ; Solve the above equations simultaneously to obtain the series impedance Z per unit length and the parallel admittance Y per unit length of the target transmission line; The real-time equivalent total impedance is obtained by multiplying the series impedance Z per unit length by the length L of the target transmission line; the real-time equivalent total admittance is obtained by multiplying the parallel admittance per unit length by the length L of the target transmission line.

[0019] It is understandable that constructing the two-port relationship of a transmission line based on the positive-sequence quantities of voltage and current at both ends, and then using this relationship to inversely derive the series impedance and parallel admittance per unit length of the line, relies on the physical characteristic that the transmission line can be considered a distributed parameter line under steady-state power frequency conditions. In this model, voltage and current propagate along the line according to specific transmission equations, and the correspondence between their port quantities can be expressed by the propagation constant and characteristic impedance. Therefore, the port quantities contain all the information about the line parameters. With a known line length, a nonlinear mapping can be established between the port measurements in the two-port equations and the parameters to be determined. Since the positive-sequence quantities can represent the dominant behavior of the line under symmetrical power frequency conditions, this mapping is unique and solvable. The mathematical form of the transmission parameters ensures that as long as the measurements at both ends are sufficiently accurate, the actual state of the line impedance and admittance can be recovered from the inverse process of the equations. Through this mechanism, the physical changes of the line under different ambient temperatures can be directly presented in the form of parameter changes, allowing subsequent temperature fitting and offset judgments to be based on real physical quantities rather than apparent measurements, thus establishing a reliable foundation for distinguishing between environmental influences and line anomalies.

[0020] In steady-state operation of transmission lines, the series impedance and parallel admittance distributed along the line jointly determine the propagation mode of voltage and current. Although this propagation is continuous in physical space, it can be mathematically compressed into a linear mapping relationship between the beginning and end points, that is, the entire line is regarded as a network element with two ports. The reason for this treatment is that steady-state positive-sequence voltage and current satisfy linear differential equations. When the parameters along the line remain uniform over the entire length, the solutions to these differential equations must be exponential, which leads to the quantities at the beginning and end points being related only in a linear combination. The resulting two-port expression represents the voltage at the beginning point as a linear combination of the voltage and current at the end point, and the current at the beginning point as a linear combination of the voltage and current at the end point. The coefficients in these linear combinations are completely determined by the propagation constant and characteristic impedance, which are in turn determined by the impedance and admittance per unit length. Therefore, the physical parameters of the line are embedded in the port relationship, allowing internal parameters that cannot be directly measured to be inversely derived from measurable port quantities. The fundamental reason for adopting this structure is that it is the only way to completely characterize the physical process between the quantities at the beginning and end of a uniform transmission line under steady-state conditions. It does not lose the influence of line impedance and admittance on port quantities, nor does it introduce degrees of freedom that do not conform to actual behavior. This allows a solvable correspondence to be formed between port measurements and internal electrical characteristics, laying the necessary foundation for subsequent recovery of line electrical parameters based on voltage and current measurements.

[0021] Analysis module: Establish a first functional relationship with ambient temperature as the independent variable and real-time equivalent total impedance as the dependent variable, and establish a second functional relationship with ambient temperature as the independent variable and real-time equivalent total admittance as the dependent variable. Extract constant components of parameters based on the first and second functional relationships. In another preferred embodiment of the present invention, the extraction of constant parameter components includes: The first data point is generated in the coordinate system with ambient temperature as the abscissa and the corresponding real-time equivalent total impedance as the ordinate. The first function relationship is obtained by curve fitting of the first data point. If the first functional relationship is monotonically increasing or monotonically decreasing, then the real-time equivalent total impedance at an ambient temperature of 0 is taken as the constant component of the first parameter; if the first functional relationship is neither monotonically increasing nor monotonically decreasing, then the mean of the real-time equivalent total impedance among all the first data points is calculated as the constant component of the first parameter. A second data point is generated in the coordinate system with ambient temperature as the abscissa and the corresponding real-time equivalent total admittance as the ordinate. The second function relationship is obtained by curve fitting of the second data point. If the second functional relationship is monotonically increasing or monotonically decreasing, then the real-time equivalent total admittance at an ambient temperature of 0 is taken as the constant component of the second parameter; if the second functional relationship is neither monotonically increasing nor monotonically decreasing, then the mean of the real-time equivalent total admittance among all the second data points is calculated as the constant component of the second parameter. The first constant component and the second constant component are referred to as the constant component of the parameter.

[0022] It should be noted that the real-time equivalent total impedance and equivalent total admittance obtained under different ambient temperatures can be understood as the sum of the relatively stable inherent part of the line itself and the part that changes with temperature. By fitting these data points with temperature as the independent variable, the "temperature-related change law" is essentially described explicitly using a function curve, thus laying the foundation for extracting a stable reference quantity from it.

[0023] Determining whether a function is monotonic is to confirm that the direction of parameter change with temperature remains consistent and the trend is continuous within the range under investigation, without a situation where the same temperature corresponds to multiple possible values ​​due to alternating periods of increase and decrease. Under the premise of stable trend and no repetition, taking the function value at an agreed reference temperature (such as zero degrees) can yield a constant component of the parameter with a unique calculation method that remains unchanged under the current overall state of the circuit. This makes it convenient to use as a unified benchmark under any subsequent temperature conditions.

[0024] When the fitted relationship does not have monotonicity, it indicates that the influence of temperature on the parameters is mixed with local fluctuations or outliers. Simply selecting the function value at a certain temperature is easily amplified by local changes. In this case, the constant component is extracted by the mean of all data points, which is equivalent to selecting a statistically significant central level in the overall data distribution to represent the average parameter state of the line in the current operating cycle.

[0025] By processing the constant components separately according to their monotonicity, the constant components can maintain the determinism and repeatability of the calculation, and reflect the overall level of the line itself under the collected operating conditions. When the real-time parameters are compared with this fixed benchmark, the normal changes caused by temperature are mainly reflected in the fitted curve or mean. The extra offsets that exceed this pattern can be more clearly reflected, which can be used to indicate abnormal changes related to latent faults, thereby serving the final fault identification.

[0026] Judgment module: Obtain the measured equivalent total impedance X1 and measured equivalent total admittance X2 at the current moment, and determine whether there is a hidden fault in the target transmission line by combining the constant component of the parameter.

[0027] In another preferred embodiment of the present invention, determining whether a latent fault exists includes: Calculate the absolute value ΔX1 of the difference between the measured equivalent total impedance X1 and the first constant parameter component x1 in the constant parameter components; calculate the absolute value ΔX2 of the difference between the measured equivalent total admittance X2 and the second constant parameter component x2 in the constant parameter components. Calculate the first offset ΔX1' = ΔX1 / x1, and calculate the second offset ΔX2' = ΔX2 / x2; If max(ΔX1', ΔX2') > YZ, then the target transmission line is determined to have a hidden fault. YZ represents the preset judgment threshold, and max(ΔX1', ΔX2') represents obtaining the larger value between ΔX1' and ΔX2'.

[0028] Understandably, after obtaining the constant component, calculating the difference between the real-time equivalent parameters and this fixed benchmark essentially locks the temperature-dependent portion within the baseline, allowing the real-time measured parameters to be compared with the same reference level at any temperature. When the line is in normal condition, the difference between the real-time parameters and the constant component is mainly determined by the previously fitted temperature-dependent portion, and its variation range is usually within a predictable range. However, once the line itself experiences latent degradation, the equivalent parameters will exhibit additional shifts under all temperature conditions. This shift is independent of ambient temperature, thus significantly increasing the difference.

[0029] By relativizing the differences, the disparities caused by the magnitude of the line's rated parameters can be further mitigated, allowing for evaluation of variations under different operating conditions on a unified scale. Using preset thresholds to determine whether relative offset exceeds the normal range essentially leverages statistical laws to identify anomalies deviating from normal change patterns, clearly highlighting any additional increments in the line's inherent characteristics against the backdrop of environmental influences. In this way, the degree of offset between real-time quantities and a constant benchmark can reflect the degree of anomaly in the line's condition, providing quantifiable evidence for identifying latent faults and thus supporting the final diagnostic decision.

[0030] For example, during a winter operation, a transmission line collected multiple sets of real-time equivalent total impedance and real-time equivalent total admittance under different temperature conditions over a time span. As the temperature gradually decreased, the equivalent total impedance showed a continuous decreasing trend, and the equivalent total admittance also showed a relatively stable unidirectional change with temperature. Therefore, when these data were fitted with temperature as the horizontal axis, a monotonic functional relationship was formed.

[0031] In this case, the fitted curve reveals the pattern of temperature change in the line under the current overall health condition. Zero degrees is selected as the reference temperature on the coordinate axis corresponding to the line temperature, and the corresponding value of the fitted curve at this fixed temperature point is extracted as the constant component of the parameter. This constant component has a benchmark significance for subsequent state measurement.

[0032] As the weather changes further, new operating data points appear on the line under low temperature conditions. At a certain moment, the real-time equivalent total impedance is significantly higher than the previous decreasing trend with temperature changes. The change range of the real-time equivalent total admittance is also inconsistent with the normal trend. At this time, the real-time equivalent total impedance at this moment is compared with the previously determined constant component. The difference obtained is more prominent than under normal fluctuation background. The difference between the real-time equivalent total admittance and its constant component also shows the same characteristics.

[0033] Because the constant component is determined by a unified reference temperature point during the fitting process and remains unchanged with real-time temperature variations, the real-time parameters can establish a stable comparison relationship with this benchmark at any temperature, allowing the line's condition at the current temperature to be reflected through this difference. Under normal circumstances, the difference between the real-time equivalent total impedance and the constant component exhibits a smooth change consistent with the fitting trend as temperature changes. However, when the line itself begins to show latent degradation, additional physical changes will be superimposed on the temperature-varying part, manifesting as an overall shift in the difference. This shift exceeds the normal range formed by previous temperature changes, thus being identified as an anomaly in the difference criterion.

[0034] The difference between the real-time equivalent total admittance and the constant component can also reflect changes in the line insulation performance in the same way. After relativization, the changes in the two types of differences can be further normalized, making them unaffected by dimensions and rated values. This allows us to use preset thresholds to determine whether the value has exceeded the normal range. When the relative magnitude of the difference is so large that it cannot be explained by temperature changes, it can be determined that there is an additional deviation in the line beyond the temperature pattern. This deviation comes from changes in the line's own characteristics. Therefore, by comparing the difference with the threshold, a latent fault can be determined.

[0035] In a preferred embodiment of the present invention, after determining that a latent fault exists, the method further includes: Substituting the current ambient temperature into the first and second functional relationships respectively, we obtain the theoretical impedance P1 and theoretical admittance P2. Calculate the difference P1' = X1 - P1, and calculate the difference P2' = X2 - P2; like If the type of latent fault is determined to be abnormal series impedance, K is a preset proportional coefficient and K>1; like If so, the type of latent fault is determined to be parallel admittance anomaly; If neither of the above two conditions is met, then the latent fault is determined to affect both the series impedance and the parallel admittance.

[0036] It is worth noting that the time points at which the latent faults are identified are recorded, the time intervals between two adjacent time points are calculated, and the time intervals are visualized, including line charts and bar charts.

[0037] This invention utilizes the operational relationship between temperature and parameters for structured modeling, thereby decomposing the actual monitored parameter changes into two categories: environmentally driven terms and line-inherent terms. Parameter changes caused by icing and low temperatures are essentially continuous processes that change with temperature, while changes caused by latent line faults typically manifest as temperature-independent overall shifts. By obtaining a constant component from the fitted relationship and using it as a reference, the deviation of real-time parameters relative to this reference can reflect whether there are additional line-specific changes. In this way, environmental factors cause parameters to move smoothly along the fitted trend, while changes caused by latent faults cause parameters to deviate from the trend, forming abnormally large shifts. Comparing the shift with a threshold can clearly distinguish between the two types of phenomena, solving the problem of misjudgment caused by the overlap of environmental changes and fault characteristics.

[0038] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation. The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A power grid latent fault diagnosis system based on multi-source measurement switch state collaborative analysis, characterized in that, include: Acquisition module: Acquires multiple sets of electrical measurement data of the target transmission line under various ambient temperature conditions. The real-time equivalent total impedance and real-time equivalent total admittance of the target transmission line are determined based on electrical measurement data. Analysis module: Establish a first functional relationship with ambient temperature as the independent variable and real-time equivalent total impedance as the dependent variable, and establish a second functional relationship with ambient temperature as the independent variable and real-time equivalent total admittance as the dependent variable. Extract constant components of parameters based on the first and second functional relationships. Judgment module: Obtain the measured equivalent total impedance X1 and measured equivalent total admittance X2 at the current moment, and determine whether there is a hidden fault in the target transmission line by combining the constant component of the parameter.

2. The power grid latent fault diagnosis system based on multi-source measurement switch state collaborative analysis according to claim 1, characterized in that, Obtaining electrical measurement data includes: Electrical measurement data is periodically collected within a preset time span. The electrical measurement data includes the instantaneous values ​​of the three-phase voltage at the beginning, the instantaneous values ​​of the three-phase current at the beginning, the instantaneous values ​​of the three-phase voltage at the end, and the instantaneous values ​​of the three-phase current at the end. The ambient temperature at the time of sampling is also recorded. Convert the instantaneous values ​​of the three-phase voltages at the beginning to positive sequence voltage components, denoted as the positive sequence voltage at the beginning; convert the instantaneous values ​​of the three-phase voltages at the end to positive sequence voltage components, denoted as the positive sequence voltage at the end; convert the instantaneous values ​​of the three-phase currents at the beginning to positive sequence current components, denoted as the positive sequence current at the beginning; convert the instantaneous values ​​of the three-phase currents at the end to positive sequence current components, denoted as the positive sequence current at the end. The first-end positive sequence voltage, first-end positive sequence current, last-end positive sequence voltage and last-end positive sequence current, and the corresponding ambient temperature at each sampling time are stored as a set of electrical measurement data.

3. The power grid latent fault diagnosis system based on multi-source measurement switch state collaborative analysis according to claim 1, characterized in that, The calculation of real-time equivalent total impedance and real-time equivalent total admittance includes: Obtain the electrical measurement data at the i-th sampling time, including the first-end positive sequence voltage Us, the first-end positive sequence current Is, the last-end positive sequence voltage Ur, and the last-end positive sequence current Ir. Establish computational relationships based on the equations of a two-port network of transmission lines. These relationships include: ； ； Among them, transmission parameters L represents the length of the target transmission line, and the propagation constant of the target transmission line is... Z and Y represent the series impedance per unit length and the parallel admittance per unit length of the target transmission line, respectively. Transmission parameters Characteristic impedance ; Transmission parameters ; Transmission parameters ; Solve the above equations simultaneously to obtain the series impedance Z per unit length and the parallel admittance Y per unit length of the target transmission line; The real-time equivalent total impedance is obtained by multiplying the series impedance Z per unit length by the length L of the target transmission line; the real-time equivalent total admittance is obtained by multiplying the parallel admittance per unit length by the length L of the target transmission line.

4. The power grid latent fault diagnosis system based on multi-source measurement switch state collaborative analysis according to claim 1, characterized in that, Extracting constant components of parameters includes: The first data point is generated in the coordinate system with ambient temperature as the abscissa and the corresponding real-time equivalent total impedance as the ordinate. The first function relationship is obtained by curve fitting of the first data point. If the first functional relationship is monotonically increasing or monotonically decreasing, then the real-time equivalent total impedance at an ambient temperature of 0 is taken as the constant component of the first parameter; if the first functional relationship is neither monotonically increasing nor monotonically decreasing, then the mean of the real-time equivalent total impedance among all the first data points is calculated as the constant component of the first parameter. A second data point is generated in the coordinate system with ambient temperature as the abscissa and the corresponding real-time equivalent total admittance as the ordinate. The second function relationship is obtained by curve fitting of the second data point. If the second functional relationship is monotonically increasing or monotonically decreasing, then the real-time equivalent total admittance at an ambient temperature of 0 is taken as the constant component of the second parameter; if the second functional relationship is neither monotonically increasing nor monotonically decreasing, then the mean of the real-time equivalent total admittance among all the second data points is calculated as the constant component of the second parameter. The first constant component and the second constant component are referred to as the constant component of the parameter.

5. The power grid latent fault diagnosis system based on multi-source measurement switch state collaborative analysis according to claim 4, characterized in that, Determining whether a latent fault exists includes: Calculate the absolute value ΔX1 of the difference between the measured equivalent total impedance X1 and the first constant parameter component x1 in the constant parameter components; calculate the absolute value ΔX2 of the difference between the measured equivalent total admittance X2 and the second constant parameter component x2 in the constant parameter components. Calculate the first offset ΔX1' = ΔX1 / x1, and calculate the second offset ΔX2' = ΔX2 / x2; If max(ΔX1', ΔX2') > YZ, then the target transmission line is determined to have a hidden fault. YZ represents the preset judgment threshold, and max(ΔX1', ΔX2') represents obtaining the larger value between ΔX1' and ΔX2'.

6. The power grid latent fault diagnosis system based on multi-source measurement switch state collaborative analysis according to claim 1, characterized in that, After determining that a latent fault exists, the following steps are also included: Substituting the current ambient temperature into the first and second functional relationships respectively, we obtain the theoretical impedance P1 and theoretical admittance P2. Calculate the difference P1' = X1 - P1, and calculate the difference P2' = X2 - P2; like If the type of latent fault is determined to be abnormal series impedance, K is a preset proportional coefficient and K>1; like If so, the type of latent fault is determined to be parallel admittance anomaly; If neither of the above two conditions is met, then the latent fault is determined to affect both the series impedance and the parallel admittance.

7. The power grid latent fault diagnosis system based on multi-source measurement switch state collaborative analysis according to claim 1, characterized in that, Also includes: Record the time points at which a latent fault is identified, calculate the time interval between two adjacent time points, and visualize the time intervals using methods such as line charts and bar charts.

Images