Method and system for detecting abnormal power state of electrical equipment

Abstract

The invention belongs to the technical field of power state detection and abnormity diagnosis of electrical equipment in a high-voltage switch cabinet, and particularly relates to an electrical equipment abnormal power state detection method and system. The invention provides a brand new method for monitoring the power state of electrical equipment based on temperature information, and the method is suitable for a high-voltage switch cabinet and other scenes where the internal space is narrow, the environment is severe, and power is difficult to measure directly. According to the method, temperature sensors are installed at key parts of equipment, temperature data are collected and preprocessed, heating points, heating sections and heating areas are identified, temperature characteristics of the areas are extracted, and a model is established by using historical data to predict the influence of each area on the power consumption of the equipment. And finally, carrying out weighted summation on the influence values of all the regions to obtain a corrected power value, comparing the corrected power value with a power range of normal operation of the equipment, and evaluating whether the power state of the equipment is abnormal or not, so that an alarm is triggered or a protection measure is taken, and potential safety risks are avoided.

Description

Technical Field

[0001] The present invention belongs to the technical field of power state detection and abnormal diagnosis of electrical equipment inside a high-voltage switch cabinet, and in particular relates to a method and system for detecting abnormal power state of electrical equipment. Background Art

[0002] As an important part of the power system, the normal operation of the electrical equipment inside the high-voltage switchgear is crucial to the safety and stability of the power system. However, the internal space of the high-voltage switchgear is small and closed, and the operating environment of the electrical equipment is harsh. There are many difficulties in directly measuring the voltage, current and power inside the equipment. Traditional power status detection and abnormal diagnosis technology of electrical equipment inside high-voltage switchgear mainly relies on the following methods:

[0003] 1. Direct measurement method: By installing high-voltage voltage transformers and current transformers outside the equipment, the voltage and current inside the equipment are indirectly measured, and the power is calculated. However, the space of the high-voltage switch cabinet is small, and there are space restrictions for installing measurement equipment, which may affect the operation of the equipment; high-voltage insulation requirements are high, and there are safety risks in the installation and maintenance of measurement equipment; the measurement accuracy is greatly affected by the external environment, and it is difficult to guarantee accuracy.

[0004] 2. Relay protection method: When the equipment current exceeds the set value, the relay will operate, cut off the fault circuit and protect the equipment. However, this method can only detect current anomalies and cannot identify other types of power anomalies, such as power factor anomalies, harmonic distortion, etc. The sensitivity of relay protection action is limited and it may not be able to detect slight power anomalies in time.

[0005] 3. Acoustic monitoring method: Use acoustic sensors to monitor the noise generated by equipment operation and analyze the noise characteristics to determine the equipment status. However, the internal environment of the high-voltage switchgear is complex and the noise interference is serious, making it difficult to effectively extract equipment operation noise information. The relationship between acoustic characteristics and equipment status is complex, and it is difficult to establish an accurate correspondence.

[0006] Traditional detection methods have the following shortcomings:

[0007] 1) Space limitation: The internal space of the high-voltage switch cabinet is small, making it difficult to install measurement equipment.

[0008] 2) Safety risks: High voltage insulation requirements are high, and the installation and maintenance of measuring equipment pose safety risks.

[0009] 3) Insufficient accuracy: The measurement accuracy is greatly affected by the external environment, and it is difficult to establish an accurate correspondence.

[0010] Therefore, traditional methods are difficult to meet the needs of power status detection and abnormal diagnosis of electrical equipment inside high-voltage switchgear, and new detection methods are urgently needed to solve these problems. Summary of the invention

[0011] The present invention overcomes the shortcomings of the prior art and provides a method and system for detecting abnormal power status of electrical equipment, which solves the problem that traditional methods are difficult to accurately detect power status in environments such as high-voltage switch cabinets, and realizes accurate detection of the power status of electrical equipment and safety warning.

[0012] To achieve the above object, the present invention provides a method for detecting abnormal power state of electrical equipment, comprising the following steps:

[0013] S1. Collect real-time temperature data of electrical equipment and pre-process the collected temperature data.

[0014] S2. Based on the historical operating data of the equipment and the current temperature distribution, the initial predicted power is obtained through regression analysis.

[0015] S3. Use temperature data to perform image processing and feature extraction to identify abnormal areas, including heating points, heating sections and heating areas.

[0016] S4. Extract the features of the heating point, heating section and heating area.

[0017] S5. Based on historical data, a mathematical model is established to describe the impact of heating points, heating sections and heating areas on power consumption.

[0018] S6. Input the extracted features into the mathematical model, calculate the influence value of each feature on power consumption, combine the calculated influence value with the initial predicted power value, and obtain a revised power value. The revised power value reflects the actual power consumption.

[0019] S7. Compare the corrected power value with the power consumption range of the equipment during normal operation to evaluate the equipment operation status.

[0020] Furthermore, S1 includes the following steps:

[0021] S11. Use sensors to directly collect temperature data and obtain current temperature distribution information of the device.

[0022] The formula used to define the real-time temperature distribution is T(x,y,t)=T sensor (x,y,t).

[0023] Where: T(x,y,t) is the temperature at coordinate (x,y) at time t. sensor The temperature value measured by the sensor.

[0024] S12. Use a filter to perform noise filtering on the temperature data collected in S11 to improve data quality and reduce the impact of errors.

[0025] S13. Use the moving average method to smooth the data obtained in S12 to eliminate sudden changes in the temperature data, enhance data stability, and facilitate subsequent analysis.

[0026] The formula used for smoothing is

[0027] Where: T smooth is the temperature value after preprocessing. N is the number of neighboring points involved in smoothing. (x i ,y i ) are the coordinates of the neighboring points.

[0028] Furthermore, S2 refers to the preprocessed data of S1 and uses the historical operating data to establish a linear relationship between temperature and power, establishes a linear regression model, and obtains the initial predicted power through regression analysis based on the current temperature, and quickly estimates the current power of the equipment as a reference value for subsequent analysis.

[0029] The initial prediction formula is

[0030] Where: T initial is the initial predicted power. is the regression coefficient. avg is the average temperature of the device. ∈ is the error term.

[0031] Further, S3 includes the following steps:

[0032] S31. Set a temperature threshold according to the specifications and historical operation data of the equipment.

[0033] S32. Scan the temperature distribution data to find out all abnormal areas where the temperature exceeds the threshold.

[0034] S33, classifying the identified abnormal areas into heating points, heating sections and heating areas, and defining a feature set.

[0035] A hot spot is a point where the temperature exceeds a maximum threshold.

[0036] The heating section is a continuous area where the temperature exceeds the intermediate threshold and the temperature gradient exceeds the set threshold.

[0037] The heating area is a continuous area where the temperature exceeds the intermediate threshold and the temperature gradient exceeds the set threshold.

[0038] The formula used to define the abnormal area is H region ={(x, y)|T smooth (x,y)>T threshold}.

[0039] Among them, H regionis the set of abnormal regions. threshold is the preset temperature threshold.

[0040] Further, S4 includes the following steps:

[0041] S41. In each heating point, extract the highest local temperature T spotrmax , temperature gradient and heating point radius R.

[0042] S42, in each heating section, extract the heating section length L, the heating section average temperature T segmentavg And the standard deviation of the heating section temperature S td .

[0043] S43, in each heating area, extract the heating area A, the heating area average temperature T areavg and the maximum temperature of the heating area T aremax .

[0044] Further, S5 includes the following steps:

[0045] S51. Analyze historical data to identify the influence of the characteristics of heating points, heating sections, and heating areas on power consumption.

[0046] S52. Based on the historical data analysis results, three nonlinear regression models are established to describe the mathematical relationship between the characteristics of the heating point, heating section, and heating area and the power consumption respectively.

[0047] Further, S6 includes the following steps:

[0048] S61, input the characteristic data extracted in S4 into the three models in S5 respectively, and predict the impact value P of each area on power consumption spot , P segment and P are .

[0049] S62. Weight the impact value according to the importance of the feature.

[0050] S63, the weighted influence value and the initial predicted power P initial Combined, calculate the corrected power P corrected , the modified formula is P corrected =P initial +αP spot +βP segment +γP are .

[0051] Among them: α, β, γ are weight coefficients, which are used to adjust the contribution of the influence value and can be adjusted according to the importance and uncertainty of the feature.

[0052] Further, S7 includes the following steps:

[0053] S71. Record the power range P of the equipment during normal operation normal , as the judgment criterion.

[0054] S72, compare the power data corrected in step S6 with the normal power range P in S71 normal For comparison, the judgment formula is

[0055] Where: A is the abnormal evaluation result, 1 represents abnormality, and 0 represents normality. normal is the power of the equipment when it is operating normally. δ is the tolerance deviation of the normal power.

[0056] S73. According to the abnormality assessment result in S72, when an abnormality is indicated, an alarm system is triggered according to the abnormality assessment result, and protective measures are taken according to the severity of the abnormality to avoid potential safety risks.

[0057] The present invention also provides an abnormal power state detection system for electrical equipment, comprising the following modules:

[0058] Data acquisition and preprocessing module: The data acquisition and preprocessing module is responsible for collecting real-time temperature data of key parts of the equipment, performing noise filtering and smoothing on the collected data to improve data quality.

[0059] Initial power prediction module: The initial power prediction module uses historical operating data and current temperature distribution to establish a linear regression model, and makes an initial power prediction based on the model as a reference value for subsequent analysis.

[0060] Abnormal area identification module. A temperature threshold is set in the abnormal area identification module to distinguish between normal and abnormal temperature ranges. The abnormal area identification module scans the temperature distribution data, identifies areas where the temperature exceeds the threshold, and classifies the identified abnormal areas into heating points, heating sections, and heating areas.

[0061] Abnormal feature extraction module. The abnormal feature module can extract the features of heating points, heating sections and heating areas, and use the extracted feature data as input for subsequent analysis.

[0062] Power correction model module. The power correction module can analyze historical data, identify the correlation of abnormal area features, establish a nonlinear regression model, establish a mathematical relationship between features and power consumption, and use the model to calculate the impact value of features on power consumption, combine the calculated impact value with the initial predicted power, and obtain the corrected power.

[0063] Abnormal detection and alarm module. The abnormal detection and alarm module compares the corrected power with the normal power range to determine whether there is an abnormality. Based on the abnormal evaluation results, the alarm system is triggered and protective measures are taken according to the severity of the abnormality.

[0064] The present invention solves the defects existing in the background technology and has the following beneficial effects:

[0065] (1) The present invention effectively solves the problem that the traditional method is difficult to accurately detect the power status in an environment such as a high-voltage switch cabinet through seven steps, and realizes accurate detection of the power status of electrical equipment and safety warning.

[0066] (2) The present invention utilizes temperature characteristic information (heating point, heating section, heating area) and characteristic data (local maximum temperature T spotmax , temperature gradient and the heating point radius R, etc.) to analyze the power state of the device. By establishing a nonlinear regression model, the temperature characteristics are linked to the power consumption, the impact value of the characteristics on the power is calculated, and the initial predicted power is corrected, thereby improving the accuracy of the power state assessment, being able to identify abnormal situations that are difficult to detect with traditional methods, and avoiding potential equipment failures. Compared with the traditional direct measurement method, the present invention can indirectly assess the power state through temperature information when it is difficult to directly measure the voltage and current, thereby improving the accuracy and reliability of the detection. In addition, the present invention can further improve the accuracy and reliability of the power state assessment by continuously accumulating historical data and continuously optimizing the model.

[0067] (3) The present invention adopts non-contact temperature measurement technology, which can identify abnormal power status without direct contact with high-voltage equipment, timely warn, and take protective measures according to the degree of abnormality, effectively avoiding safety risks caused by equipment failure, improving the safety of electrical equipment, and avoiding equipment damage or casualties caused by power abnormalities. Compared with traditional direct measurement methods, the present invention avoids the risk of high-voltage electric shock and is safer and more reliable. In addition, the present invention can be combined with other safety protection measures to form a more complete equipment safety protection system.

[0068] (4) The present invention combines temperature information analysis with model building to achieve accurate detection of the power status of electrical equipment and safety warning, overcomes the limitations of traditional methods, effectively improves the safety, reliability and operating efficiency of electrical equipment, and has important application value in the monitoring and diagnosis of electrical equipment such as high-voltage switchgear. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

[0070] Figure 1 It is a flow chart of a method for detecting abnormal power status of electrical equipment;

[0071] Figure 2 The present invention is a framework diagram of an abnormal power state detection system for electrical equipment. DETAILED DESCRIPTION

[0072] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0073] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0074] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the scope of protection of the present application. In addition, the terms "first", "second", etc. are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, features defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of the invention, unless otherwise specified, "multiple" means two or more.

[0075] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection. It can be a mechanical connection or an electrical connection. It can be directly connected or indirectly connected through an intermediate medium, and it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0076] The present invention provides a new method for monitoring the power status of electrical equipment based on temperature information, which is suitable for scenarios such as high-voltage switchgear where the internal space is small, the environment is harsh, and it is difficult to directly measure power. This embodiment takes the high-voltage switchgear as an example. The temperature sensors should be set in key parts such as busbars, circuit breakers, transformers, current transformers, lightning arresters, etc., and the appropriate installation location should be selected according to the type of other equipment and the heat concentration area. By reasonably setting the temperature sensor, the temperature changes of the equipment can be effectively monitored, abnormal areas can be identified, the power status can be predicted, and timely measures can be taken to avoid potential safety risks.

[0077] like Figure 1 As shown, the present invention provides a method for detecting abnormal power state of electrical equipment, comprising the following steps:

[0078] S1. Temperature sensors are installed on the busbar, circuit breaker, transformer, current transformer and lightning arrester of the high-voltage switchgear to collect real-time temperature data and pre-process the collected temperature data.

[0079] S2. Based on the historical operating data of the equipment and the current temperature distribution, the initial predicted power is obtained through regression analysis.

[0080] S3. Use temperature data for image processing and feature extraction to identify abnormal areas, including heating points, heating sections and heating areas.

[0081] S4. Extract the features of the heating point, heating section and heating area.

[0082] S5. Based on historical data, a mathematical model is established to describe the impact of heating points, heating sections and heating areas on power consumption.

[0083] S6. Input the extracted features into the mathematical model, calculate the influence value of each feature on power consumption, combine the calculated influence value with the initial predicted power value, and obtain a revised power value. The revised power value reflects the actual power consumption.

[0084] S7. Compare the corrected power value with the power consumption range of the equipment during normal operation to evaluate the equipment operation status.

[0085] Furthermore, S1 includes the following steps:

[0086] S11. Use sensors to directly collect temperature data and obtain current temperature distribution information of the device.

[0087] The formula used to define the real-time temperature distribution is T(x,y,t)=T sensor (x,y,t).

[0088] Where: T(x,y,t) is the temperature at coordinate (x,y) at time t. sensor The temperature value measured by the sensor.

[0089] S12. Use a filter to perform noise filtering on the temperature data collected in S11 to improve data quality and reduce the impact of errors.

[0090] S13. Use the moving average method to smooth the data obtained in S12 to eliminate sudden changes in the temperature data, enhance data stability, and facilitate subsequent analysis.

[0091] The formula used for smoothing is

[0092] Where: T smooth is the temperature value after preprocessing. N is the number of neighboring points involved in smoothing. (x i ,y i ) are the coordinates of the neighboring points.

[0093] Furthermore, S2 refers to the preprocessed data of S1 and uses the historical operating data to establish a linear relationship between temperature and power, establishes a linear regression model, and obtains the initial predicted power through regression analysis based on the current temperature, and quickly estimates the current power of the equipment as a reference value for subsequent analysis.

[0094] The initial prediction formula is

[0095] Where: T initial is the initial predicted power. is the regression coefficient. avg is the average temperature of the equipment. ∈ is the error term.

[0096] Further, S3 includes the following steps:

[0097] S31. Set a temperature threshold according to the specifications and historical operation data of the equipment.

[0098] S32. Scan the temperature distribution data to find out all abnormal areas where the temperature exceeds the threshold.

[0099] S33, classifying the identified abnormal areas into heating points, heating sections and heating areas, and defining a feature set.

[0100] A hot spot is a point where the temperature exceeds a maximum threshold.

[0101] The heating section is a continuous area where the temperature exceeds the intermediate threshold and the temperature gradient exceeds the set threshold.

[0102] The heating area is a continuous area where the temperature exceeds the intermediate threshold and the temperature gradient exceeds the set threshold.

[0103] The formula used to define the abnormal area is H region ={(x, y)|T smooth (x,y)>T threshold}.

[0104] Among them, H region is the set of abnormal regions. threshold is the preset temperature threshold.

[0105] Further, S4 includes the following steps:

[0106] S41. In each heating point, extract the highest local temperature T spotmax , temperature gradient and the heating point radius R.

[0107] S42, in each heating section, extract the heating section length L, the heating section average temperature T segmentavg And the standard deviation of the heating section temperature S td .

[0108] S43, in each heating area, extract the heating area A, the heating area average temperature T areavg and the maximum temperature of the heating area T aremax .

[0109] Further, S5 includes the following steps:

[0110] S51. Analyze historical data to identify the influence of the characteristics of heating points, heating sections, and heating areas on power consumption.

[0111] S52. According to the historical data analysis results, three nonlinear regression models are established to describe the mathematical relationship between the characteristics of the heating point, heating section, and heating area and the power consumption respectively:

[0112] Where: α1 is the local maximum temperature T spotmax The weight coefficient; α2 is the temperature gradient ; α3 is the weight coefficient of the heating point radius R; in the present invention: α1 is 0.5, α2 is 0.3, and α3 is 0.2.

[0113] P segment =β1*2.5L 1.2 +β2*0.8T segmentavg +β3*S td +g; where: β1 is the weight coefficient of the heating section length L; β2 is the average temperature of the heating section T segmentavg The weight coefficient; β3 is the standard deviation of the heating section temperature S td The weight coefficient is g; g is a constant term; in the present invention: β1 is 0.6, β2 is 0.2, β3 is 0.2, and g is 10.

[0114] P are =γ1*A+γ2*T areavg +γ3*T aremax ; Where: γ1 is the weight coefficient of the heating area A; γ2 is the average temperature T of the heating area areavg The weight coefficient; γ3 is the maximum temperature T of the heating area aremax Weight coefficient; in the present invention: γ1 is 0.6, γ2 is 0.3, and γ3 is 0.1.

[0115] Further, S6 includes the following steps:

[0116] S61, input the characteristic data extracted in S4 into the three models in S5 respectively, and predict the impact value P of each area on power consumption spot , P segment and P are .

[0117] S62. Weight the impact value according to the importance of the feature.

[0118] S63, the weighted influence value and the initial predicted power P initial Combined, calculate the corrected power P corrected , the modified formula is P corrected =P initial +αP spot +βP segment +γP are .

[0119] Wherein: α, β, γ are weight coefficients, which are used to adjust the contribution of the influence value, and can be adjusted according to the importance and uncertainty of the feature; in the present invention: α is 0.1, β is 0.5, and γ is 0.2.

[0120] Further, S7 includes the following steps:

[0121] S71. Record the power range P of the equipment during normal operation normal , as the judgment criterion.

[0122] S72, compare the power data corrected in step S6 with the normal power range P in S71 normal For comparison, the judgment formula is

[0123] Where: A is the abnormal evaluation result, 1 represents abnormality, and 0 represents normality. normal is the power of the equipment when it is operating normally. δ is the tolerance deviation of the normal power.

[0124] S73. According to the abnormality assessment result in S72, when an abnormality is indicated, an alarm system is triggered according to the abnormality assessment result, and protective measures are taken according to the severity of the abnormality to avoid potential safety risks.

[0125] like Figure 2 As shown, the present invention also provides an abnormal power state detection system for electrical equipment, comprising the following modules:

[0126] Data acquisition and preprocessing module: The data acquisition and preprocessing module is responsible for collecting real-time temperature data of key parts of the equipment, performing noise filtering and smoothing on the collected data to improve data quality.

[0127] Initial power prediction module: The initial power prediction module uses historical operating data and current temperature distribution to establish a linear regression model, and makes an initial power prediction based on the model as a reference value for subsequent analysis.

[0128] Abnormal area identification module. A temperature threshold is set in the abnormal area identification module to distinguish between normal and abnormal temperature ranges. The abnormal area identification module scans the temperature distribution data, identifies areas where the temperature exceeds the threshold, and classifies the identified abnormal areas into heating points, heating sections, and heating areas.

[0129] Abnormal feature extraction module. The abnormal feature module can extract the features of heating points, heating sections and heating areas, and use the extracted feature data as input for subsequent analysis.

[0130] Power correction model module. The power correction module can analyze historical data, identify the correlation of abnormal area features, establish a nonlinear regression model, establish a mathematical relationship between features and power consumption, and use the model to calculate the impact value of features on power consumption, combine the calculated impact value with the initial predicted power, and obtain the corrected power.

[0131] Abnormal detection and alarm module. The abnormal detection and alarm module compares the corrected power with the normal power range to determine whether there is an abnormality. Based on the abnormal evaluation results, the alarm system is triggered and protective measures are taken according to the severity of the abnormality.

[0132] The above is based on the ideal embodiment of the present invention. Through the above description, relevant personnel can make various changes and modifications without departing from the technical concept of the present invention. The technical scope of the present invention is not limited to the content in the specification, and the technical scope must be determined according to the scope of the claims.

Claims

1. A method for detecting abnormal power state of electrical equipment, characterized in that: The following steps are involved: S1. Collect real-time temperature data of electrical equipment and pre-process the collected temperature data; S2. Based on the historical operation data of the equipment and the current temperature distribution, the initial predicted power is obtained through regression analysis; S3, using temperature data to perform image processing and feature extraction to identify abnormal areas, including heating points, heating sections and heating areas; S4, extracting the features of the heating point, heating section and heating area; S5. Based on historical data, a mathematical model is established to describe the influence of heating points, heating sections and heating areas on power consumption; S6, inputting the extracted features into a mathematical model, calculating the influence value of each feature on power consumption, combining the calculated influence value with the initial predicted power value, and obtaining a corrected power value; the corrected power value reflects the actual power consumption; S7. Compare the corrected power value with the power consumption range of the equipment during normal operation to evaluate the equipment operation status.

2. The method for detecting abnormal power state of electrical equipment according to claim 1, characterized in that: S1 includes the following steps: S11. Use sensors to directly collect temperature data and obtain current temperature distribution information of the device; The formula used to define the real-time temperature distribution is T(x,y,t)=T sensor (x,y,t); Where: T(x,y,t) is the temperature at coordinate (x,y) at time t; T sensor The temperature value measured by the sensor; S12, using a filter to perform noise filtering on the temperature data collected in S11 to improve data quality and reduce the impact of errors; S13, using the moving average method to smooth the data obtained in S12, eliminate the sudden change in the temperature data, enhance the data stability, and facilitate subsequent analysis; The formula used for smoothing is Where: T smooth is the temperature value after preprocessing; N is the number of neighboring points involved in smoothing; (x i ,y i ) are the coordinates of the neighboring points.

3. The method for detecting abnormal power state of electrical equipment according to claim 2, characterized in that: S2 refers to the pre-processed data of S1 and uses historical operation data to establish a linear relationship between temperature and power, and establishes a linear regression model. According to the current temperature, regression analysis obtains the initial predicted power, and quickly estimates the current power of the equipment as a reference value for subsequent analysis; The initial prediction formula is Where: T initial is the initial predicted power; is the regression coefficient; T avg is the average temperature of the equipment; ∈ is the error term.

4. The method for detecting abnormal power state of electrical equipment according to claim 3, characterized in that: S3 includes the following steps: S31. Setting a temperature threshold according to the equipment specifications and historical operation data; S32, scanning the temperature distribution data to find all abnormal areas where the temperature exceeds the threshold; S33, classifying the identified abnormal area into heating points, heating sections and heating areas, and defining a feature set; A heating point is a point where the temperature exceeds the maximum threshold; The heating section is a continuous area where the temperature exceeds the intermediate threshold and the temperature gradient exceeds the set threshold; The heating area is a continuous area where the temperature exceeds the intermediate threshold and the temperature gradient exceeds the set threshold; The formula used to define the abnormal area is H region ={(x,y)|T smooth (x,y)>T threshold }; Among them, H region is the set of abnormal regions; T thredhold is the preset temperature threshold.

5. The method for detecting abnormal power state of electrical equipment according to claim 4, characterized in that: S4 includes the following steps: S41. In each heating point, extract the highest local temperature T spotmax , temperature gradient and the heating point radius R; S42, in each heating section, extract the heating section length L, the heating section average temperature T segmentavg And the standard deviation of the heating section temperature S td ; S43, in each heating area, extract the heating area A, the heating area average temperature T areavg and the maximum temperature of the heating area T aremax .

6. The method for detecting abnormal power state of electrical equipment according to claim 5, characterized in that: S5 includes the following steps: S51, analyzing historical data to identify the influence of characteristics of heating points, heating sections, and heating areas on power consumption; S52. Based on the historical data analysis results, three nonlinear regression models are established to describe the mathematical relationship between the characteristics of the heating point, heating section, and heating area and the power consumption respectively.

7. The method for detecting abnormal power state of electrical equipment according to claim 6, characterized in that: S6 includes the following steps: S61, input the characteristic data extracted in S4 into the three models in S5 respectively, and predict the impact value P of each area on power consumption spot , P segment and P are ; S62. Weighting the impact value according to the importance of the feature; S63, the weighted influence value and the initial predicted power P initial Combined, calculate the corrected power P corrected , the modified formula is P corrected =P initial +αP spot +βP segment +γP are ; Among them: α, β, γ are weight coefficients, which are used to adjust the contribution of the influence value and can be adjusted according to the importance and uncertainty of the feature.

8. The method for detecting abnormal power state of electrical equipment according to claim 7, characterized in that: S7 includes the following steps: S71. Record the power range P of the equipment during normal operation nromal , as the criterion for judgment; S72, compare the power data corrected in step S6 with the normal power range P in S71 normal For comparison, the judgment formula is Where: A is the abnormal assessment result, 1 represents abnormality, and 0 represents normality; P normal is the power of the equipment during normal operation; δ is the tolerance deviation of normal power; S73. According to the abnormality assessment result in S72, when an abnormality is indicated, an alarm system is triggered according to the abnormality assessment result, and protective measures are taken according to the severity of the abnormality to avoid potential safety risks.

9. An abnormal power state detection system for electrical equipment, characterized in that: include: Data acquisition and preprocessing module; The data acquisition and preprocessing module is responsible for collecting real-time temperature data of key parts of the equipment, performing noise filtering and smoothing on the collected data to improve data quality; Initial power prediction module; the initial power prediction module uses historical operating data and current temperature distribution to establish a linear regression model, and performs initial power prediction based on the model as a reference value for subsequent analysis; An abnormal area identification module; a temperature threshold is set in the abnormal area identification module to distinguish between normal and abnormal temperature ranges; the abnormal area identification module scans the temperature distribution data, identifies the area where the temperature exceeds the threshold, and classifies the identified abnormal area into heating points, heating sections and heating areas; Abnormal feature extraction module; the abnormal feature module can extract the features of the heating point, heating section and heating area, and use the extracted feature data as input for subsequent analysis; Power correction model module; the power correction module can analyze historical data, identify the correlation of abnormal area features, establish a nonlinear regression model, establish a mathematical relationship between features and power consumption, and use the model to calculate the impact value of the features on power consumption, combine the calculated impact value with the initial predicted power, and obtain the corrected power; Abnormal detection and alarm module: The abnormal detection and alarm module compares the corrected power with the normal power range to determine whether there is an abnormality, and triggers the alarm system based on the abnormality evaluation result, and takes protective measures based on the severity of the abnormality.

Images