Electric power operation fault monitoring and early warning system based on digital twinning

Abstract

The invention relates to the technical field of electric power monitoring, in particular to an electric power operation fault monitoring and early warning system based on digital twinning, which comprises a positioning acquisition module used for acquiring operation data of an electric power bus, and the operation data comprises current, voltage, temperature and vibration; generating a positioning data set by the operation data according to a topological structure of the power bus; the positioning data set comprises operation data, a topological structure where the power bus is located and an actual connection relation; the risk point positioning module is used for analyzing the state of the current power bus according to the positioning data set, acquiring a fluctuation proportionality coefficient and a topological structure judgment coefficient of the power bus, and identifying a risk point during operation of the power bus based on the fluctuation proportionality coefficient and the topological structure judgment coefficient; and the safety and the stability of the power system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power monitoring, and specifically to a power operation fault monitoring and early warning system based on digital twin. Background Art

[0002] During the operation of the power system, as a key link for power transmission and distribution, the operating state of the power bus has an important impact on the stability and safety of the power system. However, due to the complexity and diversity of the operating environment of the power bus, its operating data often exhibits characteristics such as non-linearity and time-variation, posing great challenges to the monitoring and early warning of risk points.

[0003] Traditional power bus monitoring methods mainly rely on manual inspections and simple threshold alarms. Manual inspections have problems such as low monitoring efficiency and lagged response, making it difficult to detect and handle risk points in a timely manner. Simple threshold alarms lack accuracy and flexibility and are easily affected by external interference and false alarms.

[0004] For example, Chinese Patent Publication No. CN112256922A discloses a method for quickly identifying fault power outages, including constructing a ledger tree topology model according to the topology of the distribution network; constructing a hierarchical data storage model of the ledger tree topology model based on the ledger tree topology model; determining the circuit breaker switches and constructing a circuit breaker switch data storage model in combination with the real-time on-off states of each circuit breaker switch; when a fault occurs, obtaining a set of circuit breaker switches according to the upper-level trace target in the circuit breaker switch data storage model, and obtaining a set of power supply paths in combination with the upper-level trace target in the hierarchical data storage model; obtaining the current values of each device located before and after the fault reporting in each power supply path in the set of power supply paths to calculate the device load rate of each power supply path, and screening out the optimal fault troubleshooting path according to the device load rate.

[0005] For example, Chinese Patent Publication No. CN116595693A discloses a method for analyzing the anti-destruction performance of a power optical cable network, which relates to the technical field of power optical cables and solves the problems that the existing methods for analyzing the anti-destruction performance of a power optical cable network fail to effectively utilize the node spatial position information, the analysis results are limited, and the algorithm complexity is high. The method of the present invention can accurately obtain the most vulnerable areas of the power optical cable network under different target damage degrees, and use the damage radius of the minimum damage circle under different target damage degrees to express the anti-destruction performance of the power optical cable network in the spatial range damage faced by this damage degree. By analyzing the nodes covered by the minimum damage circle under different damage degrees, it can guide the strategy design of the power optical cable network to cope with natural disasters or malicious attacks; by observing the occurrence positions of the minimum damage circles, it helps to locate network problems in special cases.

[0006] The prior art describes the processing method of how to track risks in a power system. However, when the power system monitors using a power bus, it is not only necessary to identify the damage and power paths existing at these risk points, but also to adaptively locate the current risk points in real time, verify whether the current location is accurate, and whether there is an association between the located risk points, so as to achieve the control of the overall power system and improve the accuracy of early warning. Summary of the Invention

[0007] To solve the above technical problems, the technical solution adopted by the present invention is: A power operation fault monitoring and early warning system based on digital twin, including: A positioning acquisition module, used to acquire the operation data of the power bus, and the operation data includes current, voltage, temperature, vibration; Generate a positioning data set according to the topological structure of the power bus; The positioning data set includes operation data, the topological structure where the power bus is located, and the actual connection relationship.

[0008] A risk point positioning module, used to analyze the state of the current power bus according to the positioning data set, obtain the fluctuation ratio coefficient and the topological structure judgment coefficient of the power bus, and identify the risk points during the operation of the power bus based on the fluctuation ratio coefficient and the topological structure judgment coefficient.

[0009] A fault positioning module, used to identify risk points, determine the logical association and fault degree between risk points, and obtain the node correlation coefficient between the logical association and the actual operation in the risk points.

[0010] A positioning compensation module, used to determine the positioning error generated during the positioning of risk points based on the logical association between risk points, and obtain the positioning compensation coefficient related to the positioning error.

[0011] A loss assessment module, used to obtain the spatial position and damage data corresponding to the power bus during the positioning of risk points, and identify the early warning loss degree of risk points.

[0012] The beneficial effects of the present invention are as follows: First, through the positioning acquisition module of the present invention, the operation data such as current, voltage, temperature, and vibration of the power bus are acquired in real time, and a positioning data set is generated based on the topological structure of the power bus, providing data support for the accurate positioning of risk points.

[0013] Second, the present invention uses the risk point positioning module to analyze the state of the current power bus according to the positioning data set, obtain the fluctuation ratio coefficient and the topological structure judgment coefficient, so as to accurately identify the risk points during the operation of the power bus and send out an early warning signal.

[0014] Third, the present invention uses the fault positioning module to identify risk points, determine the logical association and fault degree between risk points, and obtain the node correlation coefficient, providing a scientific basis for the quantitative assessment of the fault degree.

[0015] IV. The present invention uses a positioning compensation module to determine the positioning error generated by a risk point during positioning based on the logical association between risk points, calculate the positioning compensation coefficient, and correct the positioning result, thereby improving the accuracy of positioning.

[0016] V. The present invention uses a loss assessment module to obtain the spatial position and damage data of the power bus corresponding to the risk point during positioning, identify the early warning loss degree, and provide an effective reference basis for decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be further described below in conjunction with the drawings and embodiments.

[0018] Figure 1 is a system framework diagram of a power operation fault monitoring and early warning system based on digital twin.

[0019] Figure 2 is a system schematic diagram of a power operation fault monitoring and early warning system based on digital twin.

[0020] Figure 3 is a process schematic diagram of a risk point positioning module of a power operation fault monitoring and early warning system based on digital twin.

[0021] Figure 4 is a process schematic diagram of a positioning compensation module of a power operation fault monitoring and early warning system based on digital twin. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] The embodiments of the present invention will be described in detail below. The described embodiments are exemplary and are only used to explain the present invention, and should not be construed as a limitation of the present invention. For those not specified in the embodiments, the techniques or conditions described in the literature in the field or according to the product specifications are followed.

[0023] Refer to Figure 1 , a power operation fault monitoring and early warning system based on digital twin, comprising: a positioning acquisition module, a risk point positioning module, a fault positioning module, a positioning compensation module, and a loss assessment module.

[0024] The positioning acquisition module continuously acquires the operation data of the power bus, including current, voltage, temperature, vibration, etc., and generates a positioning data set according to the topological structure of the power bus; the positioning data set is transmitted to the risk point positioning module.

[0025] The risk point positioning module receives the positioning dataset transmitted by the positioning acquisition module, analyzes the state of the current power bus, obtains the fluctuation ratio coefficient and the topology structure judgment coefficient, and identifies the risk points during the operation of the power bus; transmits the risk point information including the position of the risk point, the fluctuation ratio coefficient, the topology structure judgment coefficient, etc. to the fault location module and the positioning compensation module.

[0026] The fault location module receives the risk point information transmitted by the risk point positioning module, identifies the risk points, determines the logical relationship and the degree of fault between the risk points, and obtains the node correlation coefficient between the logical relationship and the actual operation; transmits the fault point information including the position, the degree of fault, the logical relationship, etc. to the positioning compensation module and the loss assessment module.

[0027] The positioning compensation module receives the fault point information transmitted by the fault location module, determines the positioning error based on the logical relationship between the risk points, and calculates the positioning compensation coefficient; transmits the positioning compensation coefficient to the fault location module and the loss assessment module.

[0028] The loss assessment module receives the information transmitted by the fault location module and the positioning compensation module, and identifies the early warning loss degree of the risk points; outputs the loss assessment result to the system administrator or the operation and maintenance personnel for taking corresponding treatment measures.

[0029] As Figure 2 shown, the positioning acquisition module is used to acquire the operation data of the power bus, and the operation data includes current, voltage, temperature, and vibration; generates a positioning dataset according to the topology structure of the power bus; the positioning dataset includes operation data, the topology structure where the power bus is located, and the actual connection relationship, and at this time, the actual connection relationship represents the actual position of the power bus and the corresponding connected devices.

[0030] The risk point positioning module is used to analyze the state of the current power bus according to the positioning dataset, obtain the fluctuation ratio coefficient and the topology structure judgment coefficient of the power bus, and identify the risk points during the operation of the power bus based on the fluctuation ratio coefficient and the topology structure judgment coefficient.

[0031] The fault location module is used to identify the risk points, determine the logical relationship and the degree of fault between the risk points, and obtain the node correlation coefficient between the logical relationship and the actual operation in the risk points.

[0032] The positioning compensation module is used to determine the positioning error generated during the positioning of the risk points based on the logical relationship between the risk points, and obtain the positioning compensation coefficient related to the positioning error.

[0033] The loss assessment module is used to obtain the spatial position and damage data corresponding to the power bus during the positioning of the risk points, and identify the early warning loss degree of the risk points.

[0034] In one embodiment of the present invention, a positioning and acquisition module is configured to acquire the operation data of a power bus. The operation data includes current, voltage, temperature, and vibration. The operation data is generated into a positioning data set according to the topological structure of the power bus. The positioning data set includes operation data, the topological structure where the power bus is located, and the actual connection relationship.

[0035] In this module, the topological structure of the power bus shows the Internet of Things connection between the power bus and other electrical equipment, as well as the logical relationship of the bus in the power system, such as power distribution, load connection, etc., to represent how the current power bus is connected to the equipment.

[0036] In this topological structure, the location where the power bus is located will be described in detail, and sensors are used to acquire the operation data of the power bus at each location. For example, voltage and current sensors are used to acquire current and voltage data, a temperature sensor is used to acquire the temperature of the power bus during operation, and a vibration sensor is used to acquire the vibration on the physical structure where the power bus is located to determine whether the structure supporting the power bus is stable.

[0037] In one embodiment of the present invention, a risk point positioning module is configured to analyze the state of the current power bus according to the positioning data set, obtain the fluctuation ratio coefficient and the topological structure judgment coefficient of the power bus, and identify the risk points during the operation of the power bus based on the fluctuation ratio coefficient and the topological structure judgment coefficient.

[0038] During risk point positioning, the current, voltage, temperature, and vibration are sequentially identified to identify the points where the fluctuations in the operation data exceed the preset fluctuation ratio, and verify whether the topological structure position of the power bus is normal. The points with abnormal topological structure positions and exceeding the preset fluctuation ratio are regarded as the risk points at this time.

[0039] Such as Figure 3 As shown, the implementation method for analyzing the state of the power bus includes obtaining the fluctuation ratio coefficient of the power bus based on the operation data.

[0040] Based on the topological structure of the power bus, obtain the topological structure judgment coefficient of the power bus.

[0041] Based on the fluctuation ratio coefficient and the topological structure judgment coefficient, output the risk points of the power bus.

[0042] After identifying current, voltage, temperature, and vibration using the fluctuation ratio coefficient, a comprehensive coefficient is obtained. This coefficient contains risk points in current, voltage, temperature, and vibration. These risk points can be verified in the form of average values, maximum values, and fluctuation ranges during identification. At this time, to further describe the possible fluctuations, different situations of these four values will be processed. For example, if the current and voltage tend to operate stably and the overall data needs to be stable, the average value and fluctuation range can be used for identification. If the temperature tends to identify abnormal temperatures, the maximum value can be used for identification. Vibration is easily identified when the structure is unstable, and at this time, it is represented according to the occurrence frequency and frequency interval to screen out the risk points in the fluctuation ratio.

[0043] The fluctuation ratio coefficient is expressed as follows: successively obtain the first ratio coefficient related to current and voltage, the second ratio coefficient related to temperature, and the third ratio coefficient related to vibration, and combine the first ratio coefficient, the second ratio coefficient, and the third ratio coefficient into the fluctuation ratio coefficient.

[0044] Therefore, the processing method for the first ratio coefficient related to current and voltage in the fluctuation ratio coefficient is expressed as: obtain the standard deviation, average value, and fluctuation range index value of current and voltage, and calculate to obtain the first ratio coefficient.

[0045] ; where represents the first ratio coefficient, represents the standard deviation of current, represents the average value of current, represents the index value of the fluctuation range of current. This index value represents the difference between the upper limit value and the lower limit value of the fluctuation range; represents the standard value of the fluctuation range index value of current. At this time, the standard value of the fluctuation range index value of current represents the average value corresponding to the fluctuation range index value in historical data; represents the standard deviation of voltage, represents the average value of voltage, represents the index value of the fluctuation range of voltage, represents the standard value of the fluctuation range index value of voltage. At this time, the index value of the fluctuation range of voltage is the same as that of current, and the average value of the fluctuation range in historical data is selected as the standard value, and the difference between the upper limit value and the lower limit value is used as the index value; represents the weight coefficient of current, represents the weight coefficient of voltage. The weight coefficients are sequentially set to 0.5 and 0.5 in the order of current and voltage, represents the exponential constant.

[0046] The second proportionality coefficient related to temperature in the fluctuation proportionality coefficient is expressed as obtaining the maximum value and the average value of the temperature, and calculating the second proportionality coefficient.

[0047] ; where represents the second proportionality coefficient, represents the maximum value of the temperature, represents the average value of the temperature; at this time, comparing the maximum value with the average value can identify the proportion by which the current temperature exceeds when the temperature shows corresponding abnormalities.

[0048] The third proportionality coefficient related to vibration in the fluctuation proportionality coefficient is expressed as obtaining the occurrence frequency and the frequency interval of the vibration, and obtaining the third proportionality coefficient.

[0049] ; where represents the third proportionality coefficient, represents the occurrence frequency of the vibration, represents the standard value of the occurrence frequency of the vibration, represents the frequency interval of the vibration, represents the standard value of the frequency interval of the vibration; represents the weight coefficient of the occurrence frequency of the vibration, represents the weight coefficient of the frequency interval of the vibration. At this time, the weight coefficients are set to 0.6 and 0.4 in the order of the occurrence frequency and the frequency interval; the above standard value of the occurrence frequency of the vibration is the average value of the occurrence frequency of the vibration in the historical data; the standard value of the frequency interval of the vibration will select the average value of the frequency intervals in the historical data.

[0050] Then at this time, the fluctuation proportionality coefficient can be expressed as directly outputting after combining the three coefficients without comprehensively calculating the three coefficients, or choosing to comprehensively calculate the first proportionality coefficient, the second proportionality coefficient, and the third proportionality coefficient to obtain the current required fluctuation proportionality coefficient.

[0051] For example, the fluctuation proportionality coefficient can be expressed as ; when comprehensively considering these coefficients for the fluctuation proportionality coefficient, it is necessary to consider the comprehensive situation of the three coefficients to obtain the coefficient of the corresponding average situation as much as possible to identify the risk points existing in the current situation.

[0052] When obtaining the topological structure judgment coefficient of the power bus, it mainly judges whether the position of the power bus is abnormal and processes it according to the relative load of the equipment connected to the power bus to judge whether there are risk points in the current power bus in the corresponding topological structure.

[0053] Therefore, the topological structure judgment coefficient is expressed as follows: according to the actual connection relationship of the power busbars, a topological graph is constructed; the topological graph can be represented by a graph theory model, where the nodes represent electrical equipment and the edges represent the connection relationships between the equipment.

[0054] Calculate the position deviation between the actual position and the expected position of the nodes in the topological graph, as well as the load deviation between the actual load and the expected load of the nodes in the topology. Combine the load deviation and the position deviation to form the topological structure judgment coefficient.

[0055] The position at this time can be obtained from the images captured by the drone or the corresponding patrol robot. The load represents the electrical energy value consumed by the nodes in the topological graph at this time. The electrical energy value consumed by the corresponding nodes can be measured using current and voltage sensors, and combined with the expected load to determine the risk point situation existing in the topological structure at this time.

[0056] Since the position of the corresponding power equipment basically does not change after installation, when obtaining the position deviation, in addition to the deviation adjusted according to the actual appearance position of the power equipment, it can also be identified according to the position where the load deviation occurs, that is, identify the position where the load exceeds the average value as the actual position identified by the node at this time to determine the risk point where the corresponding deviation occurs.

[0057] The topological structure judgment coefficient is expressed as follows: obtain the Pearson correlation coefficient corresponding to the load deviation and the position deviation, and combine the Pearson correlation coefficient corresponding to the load deviation and the position deviation, the load deviation and the position deviation to obtain the topological structure judgment coefficient.

[0058] ; where represents the topological structure judgment coefficient, represents the Pearson correlation coefficient corresponding to the load deviation and the position deviation, represents the weight coefficient of the Pearson correlation coefficient, represents the load deviation, represents the maximum value of the load deviation, represents the weight of the load deviation, represents the position deviation, represents the maximum value of the position deviation, represents the weight of the position deviation.

[0059] For the Pearson correlation coefficient corresponding to the above load deviation and position deviation, it is calculated by taking the load deviation and the position deviation as inputs and comparing them with the average values of the load deviation and position deviation at the corresponding positions in the historical data. At the same time, the position deviation is expressed as the difference distance value at this time.

[0060] When obtaining the risk points of the power bus based on the fluctuation ratio coefficient and the topology structure judgment coefficient, the components of the fluctuation ratio coefficient and the topology structure judgment coefficient, as well as the magnitudes of the corresponding coefficient values, will be judged in sequence. When a single coefficient value exceeds the standard threshold, the corresponding data will be regarded as a risk point. When the combined coefficient value exceeds the standard threshold, the data corresponding to the combined coefficient value will also be regarded as a risk point, so as to obtain all the currently identified risk points. This standard coefficient will use the average value of the corresponding data in the historical data to represent the standard threshold of the coefficient.

[0061] For example, taking the first ratio coefficient, the second ratio coefficient, the third ratio coefficient, the fluctuation ratio coefficient, and the topology structure judgment coefficient as the basic conditions, compare them with the average values of the corresponding historical data of the first ratio coefficient, the second ratio coefficient, the third ratio coefficient, the fluctuation ratio coefficient, and the topology structure judgment coefficient under the normal operation of the power bus in sequence. When there is a coefficient that exceeds the 10% deviation range compared to the average value of its corresponding historical data, the position where the corresponding data is located will be regarded as a risk point.

[0062] In an embodiment of the present invention, a fault location module is used to identify risk points, determine the logical association and fault degree between risk points, and obtain the node correlation coefficient between the logical association and the actual operation in the risk points;

[0063] At this time, the logical association is used to identify whether there is a logical association between risk points, and the association of the current risk point with the identified risk problems, and to quantify these risk problems, so as to express the fault degree existing in the risk point; Quantify the node association of the risk point through the logical association and the fault degree, so as to obtain a comprehensive node correlation coefficient to represent the process of monitoring the power system using the power bus.

[0064] When performing fault location, the logical association between risk points can be obtained first, and a logical association matrix between risk points is set. The logical association matrix is represented as an n×n matrix, where n represents the number of risk points. Each element in the logical association matrix represents whether there is a logical association between risk points. If there is a logical association, it is 1; if there is no logical association, it is set to 0.

[0065] The node correlation coefficient can be expressed as the product of the logical association matrix and the fault degree to represent the corresponding node correlation coefficient between the current risk points, so as to obtain the situation of whether the nodes between the risk points are associated.

[0066] The fault degree can be expressed as the product of the topology structure judgment coefficient and the fluctuation ratio coefficient corresponding to the risk point.

[0067] Therefore, the node correlation coefficient is expressed as follows: obtain the logical correlation matrix corresponding to the logical correlation, the fluctuation ratio coefficient corresponding to the degree of failure, and the topological structure judgment coefficient, and calculate the node correlation coefficient.

[0068] ; where represents the node correlation coefficient, represents the number of risk points, and the value ranges of i and j are both from 1 to n. represents the index value of the i-th risk point and the j-th risk point in the logical correlation matrix. represents the fluctuation ratio coefficient of the i-th risk point. represents the topological structure judgment coefficient of the i-th risk point. and and represent the weight coefficients; where , , .

[0069] At this time, the possible degree of failure can be quantified through the fluctuation ratio coefficient and the topological structure judgment coefficient, and the relationship between these risk points can be further described according to the relevant logical relationship matrix, so that when verifying node correlation, more attention is paid to the fluctuations in the current operating data and the deviations in the current operating data; making the overall consideration more complete and improving the early warning and monitoring effects on the system.

[0070] In an embodiment of the present invention, the positioning compensation module determines the positioning error generated by the risk point during positioning based on the logical correlation between the risk points, and obtains the positioning compensation coefficient related to the positioning error.

[0071] The positioning compensation coefficient is used to adjust the positioning error generated by the risk point. The risk point is obtained through the data measured on the power bus. If the currently measured risk point is easily affected by the surrounding power buses, it will cause the risk point to not be directly located at the problem position during identification; at this time, it is necessary to verify the probability of the corresponding risk classification of the current risk point under the corresponding logical relationship to identify the positioning error, so as to obtain the positioning compensation coefficient.

[0072] The positioning error can be identified by clustering. The obtained risk points are classified in the form of clustering. After the classification is completed, for each classification, the positioning error of the risk point is calculated. The positioning error measures the difference between the position where the risk point is located and its actual position, so as to more accurately identify the current error and improve the accuracy of identifying risk points.

[0073] For example Figure 4As shown, the method for obtaining the positioning compensation coefficient is as follows: obtain the positioning error of the risk point, determine the average recognition period, maximum positioning error, and average positioning error ratio of the risk point under the corresponding positioning error; based on the average recognition period, maximum positioning error, and average positioning error ratio of the risk point, decompose and sort the risk points, calculate the distribution degree of the risk points; according to the distribution degree of the risk points, output the risk classification probability of the current risk point, and calculate the positioning compensation coefficient according to the risk classification probability and the positioning error.

[0074] The positioning error can be obtained by combining the position deviation calculated when screening the risk point with the actual position of the risk point. The average recognition period represents the average time from the appearance of each risk point to its recognition, which is used to describe the number of risk points existing in the corresponding time period. The average positioning error ratio represents the ratio of the average value of the current positioning error to the average value of the positioning error in historical data. At this time, the risk points can be divided into multiple regions, sorted for multiple regions, and continuously iterated for this region to calculate the positioning compensation coefficient.

[0075] The method for decomposing and sorting the risk points is as follows: calculate the evaluation value of the risk point according to the average recognition period, maximum positioning error, and average positioning error ratio of the risk point, sort the risk points according to the evaluation value, and obtain the distribution degree of the risk points according to the cumulative distribution value of the risk points.

[0076] The evaluation value of the risk point can be expressed as ; where represents the evaluation value of the risk point, represents the average recognition period, represents the maximum value of the average recognition period, represents the weight coefficient of the average recognition period, represents the maximum positioning error, represents the standard value of the maximum positioning error, and the standard value of the maximum positioning error is represented by the average value of the maximum positioning error in historical data; represents the weight coefficient of the maximum positioning error, represents the average positioning error ratio, represents the weight coefficient of the average positioning error ratio.

[0077] At this time, the risk points can be evaluated through the average recognition period, maximum positioning error, and average positioning error ratio of the risk points. At the same time, the weight coefficients of the average recognition period, maximum positioning error, and average positioning error ratio can be set in the form of 0.3, 0.4, and 0.3 in sequence to describe the positioning error of the current risk point.

[0078] The distribution degree of risk points is expressed as the ratio of the number of risk points determined according to the evaluation values of risk points to the total number of risk points under the corresponding positioning error values. The obtained distribution degree value is used to represent the risk classification probability of the current risk points; the evaluation value of risk points can represent the risk classification during classification.

[0079] The positioning compensation coefficient is expressed as follows: based on the distribution degree of risk points, obtain the risk classification probability of risk points, and calculate the positioning compensation coefficient.

[0080] ; where represents the positioning compensation coefficient, represents the risk classification, represents the positioning error, represents the risk classification probability of the i-th risk point under the corresponding positioning error, represents the positioning error of the i-th risk point, represents the basic value of positioning compensation, represents the number of risk points, and the value range of i is from 1 to n.

[0081] The obtained positioning compensation coefficient at this time can represent the corresponding distance value that should be compensated during the positioning analysis of risk points to reduce the error generated during the positioning of risk points.

[0082] In an embodiment of the present invention, a loss assessment module is used to obtain the spatial position and damage data corresponding to the power bus during the positioning of risk points, and determine the relative loss degree of risk points; based on the relative loss degree of risk points, identify the early warning loss degree of risk points.

[0083] The selected spatial position at this time represents the distribution of risk points according to the actual position, and combines this actual position with the damage data to verify the specific situation of the damage data at different spatial positions.

[0084] The damage data is expressed as the wire damage and damage radius generated by the power bus.

[0085] Wire damage: The wire may be damaged for various reasons (such as mechanical stress, environmental corrosion, high temperature, current impact, etc.) during use. These damages may be manifested as wear on the wire surface, breakage of internal metal wires, damage to the insulation layer, etc. The wire damage represents the current type of damage generated.

[0086] Damage radius: The damage radius refers to the range or influence range of the damaged area on the wire. This range can be linear (along the wire length direction) or radial (along the wire diameter direction). At this time, the damage radius will include the measurement method and measurement length of the loss radius to represent the physical damage of the power bus when there are risk points.

[0087] The measurement method of the damage radius can be the following methods. Linear damage: Damage along the length direction of the wire, usually represented by the damage length. For example, the surface wear length of a certain section of the wire is 1 meter; Radial damage: Damage along the diameter direction of the wire, usually represented by the damage depth or width. For example, the surface wear depth of a certain point on the wire is 1 millimeter, or the width of the insulation layer breakage is 2 millimeters.

[0088] The monitoring method of the damage radius can be measured using multiple sensors. For example, strain sensors: Monitor the strain of the wire and detect the tensile or compressive deformation of the wire; Corrosion sensors: Monitor the corrosion of the wire and detect the chemical changes on the wire surface; Visual inspection: Use cameras or drones for visual inspection to observe the surface damage of the wire; Ultrasonic inspection: Use ultrasonic flaw detectors to detect the internal damage of the wire.

[0089] Wire damage will affect its electrical performance, such as increasing resistance, reducing conductivity, causing local overheating, etc.; Wire damage will reduce its mechanical strength, which may lead to wire breakage or detachment, affecting the physical structure of the power system; Severe wire damage may lead to safety accidents such as short circuits and fires, threatening the safety of personnel and equipment.

[0090] By comprehensively processing the physical damage existing in the power busbar and the risk points identified from the operation data, the specific types of the currently identified risk points can be further shortened, making the expression of the risk points more specific. At the same time, the effect of risk point positioning and early warning can be improved, the risk points can be discovered in time, and the service life of the equipment can be extended.

[0091] For the obtained damage data, the damage caused to the current power busbar can be quantified according to the identified wire damage and damage radius. For example, obtain the damage degree value from the database according to the wire damage at this time, and calculate the damage difference amount of the corresponding damage radius relative to the average value of the damage radius in the historical data. Finally, combine the values of the node correlation coefficient and the positioning compensation coefficient to obtain the early warning degree of the risk point.

[0092] The damage degree value at this time can be obtained according to the method of image recognition to obtain the damage type at the corresponding risk point of the current power busbar during monitoring, and a score is set for each damage type.

[0093] Therefore, the warning loss degree is expressed as obtaining the damage degree value and the damage difference amount corresponding to the damage data, taking the damage degree value, the damage difference amount, the node correlation coefficient and the positioning compensation coefficient as inputs, and calculating to obtain the warning loss degree.

[0094] ; where represents the warning loss degree, Represents the node correlation coefficient, Represents the positioning compensation coefficient, Represents the damage degree value, Represents the damage difference amount, Represents the standard value of the damage difference amount. At this time, the standard value of the damage difference amount is selected as the average value of the damage difference amount in historical data. At this time, the damage difference amount is equivalent to a distance value, and at the same time, the positioning compensation coefficient is also equivalent to a distance value. The damage difference amount and the positioning compensation coefficient at this time are used to determine whether the size of the damage radius can affect the compensation value during risk point positioning under the corresponding position conditions. The node correlation coefficient is simply used to describe the overall value-taking situation of the current risk point when there is a corresponding correlation coefficient. The damage degree value supplements this finally calculated value to make the early warning degree more in line with the overall requirements.

[0095] The above formulas are all dimensionless and take their numerical calculations. The formula is obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0096] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0097] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0098] In addition, in each embodiment of this application, the functional modules can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0099] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

[0100] Finally, the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A power operation fault monitoring and early warning system based on digital twin, characterized in that, Including: A positioning and acquisition module, used to acquire the operation data of the power bus, where the operation data includes current, voltage, temperature, and vibration; Generate a positioning data set according to the topological structure of the power bus for the operation data; the positioning data set includes operation data, the topological structure where the power bus is located, and the actual connection relationship; A risk point positioning module, used to analyze the state of the current power bus according to the positioning data set, obtain the fluctuation ratio coefficient and the topological structure judgment coefficient of the power bus, and identify the risk points during the operation of the power bus based on the fluctuation ratio coefficient and the topological structure judgment coefficient; A fault positioning module, used to identify the risk points, determine the logical association and fault degree between the risk points, and obtain the node correlation coefficient between the logical association and the actual operation in the risk points; A positioning compensation module, used to determine the positioning error generated during the positioning of the risk points based on the logical association between the risk points, and obtain the positioning compensation coefficient related to the positioning error; A loss assessment module, used to acquire the spatial position and damage data corresponding to the power bus during the positioning of the risk points, and identify the early warning loss degree of the risk points.

2. The power operation fault monitoring and early warning system based on digital twin according to claim 1, characterized in that, The implementation method for analyzing the state of the power bus includes: Based on the operation data, obtain the fluctuation ratio coefficient of the power bus; Based on the topological structure of the power bus, obtain the topological structure judgment coefficient of the power bus; Based on the fluctuation ratio coefficient and the topological structure judgment coefficient, output the risk points of the power bus.

3. A power operation fault monitoring and early warning system based on digital twin according to claim 2, characterized in that, The fluctuation ratio coefficient is expressed as follows: successively obtain the first ratio coefficient related to current and voltage, the second ratio coefficient related to temperature, and the third ratio coefficient related to vibration, and combine the first ratio coefficient, the second ratio coefficient, and the third ratio coefficient into the fluctuation ratio coefficient; The processing method of the first ratio coefficient related to current and voltage in the fluctuation ratio coefficient is expressed as: obtain the standard deviation, average value, and index value of the fluctuation range of current and voltage, and calculate to obtain the first ratio coefficient; The second ratio coefficient related to temperature in the fluctuation ratio coefficient is expressed as: obtain the maximum value and average value of the temperature, and calculate to obtain the second ratio coefficient; The third ratio coefficient related to vibration in the fluctuation ratio coefficient is expressed as: obtain the occurrence frequency and frequency interval of the vibration, and obtain the third ratio coefficient.

4. The power operation fault monitoring and early warning system based on digital twin according to claim 3, characterized in that, Fluctuation ratio coefficient It can be expressed as: ； Among them, represents the first proportionality coefficient, represents the second proportionality coefficient, represents the third proportionality coefficient.

5. A power operation fault monitoring and early warning system based on digital twin according to claim 2, characterized in that, The topological structure judgment coefficient is expressed as: construct a topological graph according to the actual connection relationship of the power bus; Calculate the position deviation between the actual position and the expected position of the nodes in the topological graph, and the load deviation between the actual load and the expected load of the nodes in the topology; Obtain the Pearson correlation coefficient corresponding to the load deviation and the position deviation, and combine the Pearson correlation coefficient corresponding to the load deviation and the position deviation, the load deviation, and the position deviation to obtain the topological structure judgment coefficient; ； Among them, represents the topological structure judgment coefficient, represents the Pearson correlation coefficient corresponding to the load deviation and the position deviation, represents the weight coefficient of the Pearson correlation coefficient, represents the load deviation, represents the maximum value of the load deviation, represents the weight of the load deviation, represents the position deviation, represents the maximum value of the position deviation, represents the weight of the position deviation.

6. A power operation fault monitoring and early warning system based on digital twin according to claim 1, characterized in that, The node correlation coefficient is expressed as: obtain the logical association matrix corresponding to the logical association, the fluctuation ratio coefficient and the topological structure judgment coefficient corresponding to the fault degree, and calculate to obtain the node correlation coefficient; ； Among them, represents the node correlation coefficient, represents the number of risk points, and the value ranges of i and j are both from 1 to n, represents the index value of the i-th risk point and the j-th risk point in the logical correlation matrix, represents the fluctuation ratio coefficient of the i-th risk point, represents the topological structure judgment coefficient of the i-th risk point, , , represent the weight coefficients.

7. A power operation fault monitoring and early warning system based on digital twin according to claim 1, characterized in that The method for obtaining the positioning compensation coefficient is as follows: obtain the positioning error of the risk point, determine the average recognition period, maximum positioning error, and average positioning error ratio of the risk point under the corresponding positioning error; based on the average recognition period, maximum positioning error, and average positioning error ratio of the risk point, decompose and sort the risk point, calculate the distribution degree of the risk point; according to the distribution degree of the risk point, output the risk classification probability of the current risk point, and calculate the positioning compensation coefficient according to the risk classification probability and the positioning error.

8. The power operation fault monitoring and early warning system based on digital twin according to claim 7, characterized in that, The method for decomposing and sorting the risk point is as follows: calculate the evaluation value of the risk point according to the average recognition period, maximum positioning error, and average positioning error ratio of the risk point, sort the risk points according to the evaluation value, and obtain the distribution degree of the risk point according to the cumulative distribution value of the risk point: ； Among them, represents the evaluation value of the risk point, represents the average recognition period, represents the maximum value of the average recognition period, represents the weight coefficient of the average recognition period, represents the maximum positioning error, represents the standard value of the maximum positioning error; represents the weight coefficient of the maximum positioning error, represents the average positioning error ratio, represents the weight coefficient of the average positioning error ratio; The distribution degree of the risk point is expressed as the ratio of the number of risk points determined according to the evaluation value of the risk point to the total number of risk points under the corresponding positioning error value.

9. A power operation fault monitoring and early warning system based on digital twin according to claim 7, characterized in that, The positioning compensation coefficient is expressed as: based on the distribution degree of the risk point, obtain the risk classification probability of the risk point, and calculate the positioning compensation coefficient: ； Among them, represents the positioning compensation coefficient, represents the risk classification, represents the positioning error, represents the risk classification probability of the i-th risk point under the corresponding positioning error, represents the positioning error of the i-th risk point, represents the base value of the positioning compensation, represents the number of risk points, and the value range of i is from 1 to n.

10. A power operation fault monitoring and early warning system based on digital twin according to claim 1, characterized in that, The early warning loss degree is expressed as: obtain the damage degree value and damage difference amount corresponding to the damage data, take the damage degree value, damage difference amount, node correlation coefficient, and positioning compensation coefficient as inputs, and calculate the early warning loss degree; ； Among them, represents the early warning loss degree, represents the node correlation coefficient, represents the positioning compensation coefficient, represents the damage degree value, represents the damage difference amount, represents the standard value of the damage difference amount.

Images