Power system fault diagnosis device based on big data

Abstract

The present invention relates to the technical field of fault diagnosis devices, specifically to a power system fault diagnosis device based on big data, including a database, a fault location module, a data acquisition and processing module, a feature extraction and selection module, a fault diagnosis module, a fault factor derivation module, an integration module and an execution terminal. The present invention is conducive to improving the accuracy of power system fault diagnosis, while avoiding over-reliance on the operation data of the power system. When the operation data of the power system is missing, there is a problem of difficulty in diagnosis, and it is conducive to predicting the potential risks of power system failures in advance, thereby achieving the effect of pre-maintenance of the power system before failure, which is conducive to ensuring a more stable operation of the power system.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis devices, and in particular to a power system fault diagnosis device based on big data. Background Art

[0002] Power system failures not only threaten power supply reliability and safety, but also cause serious economic losses. Therefore, the importance of power system fault detection and diagnosis analysis in the power industry is becoming increasingly prominent, and the development and application of effective power system fault diagnosis technology is particularly important.

[0003] The existing power system fault diagnosis covers many types of diagnostic faults, resulting in excessive calculations required in the diagnosis process and over-reliance on the operation data of the power system. When the operation data of the power system is missing, there is a problem of difficulty in diagnosis. In addition, the diagnostic results calculated based on the separate power system operation data are relatively rough, making it difficult to ensure the accuracy of the diagnosis and provide corresponding suggestions. To this end, we propose a power system fault diagnosis device based on big data. Summary of the invention

[0004] The purpose of the present invention is to provide a power system fault diagnosis device based on big data to solve the problems raised by the above background technology.

[0005] The purpose of the present invention can be achieved by the following technical solution: A power system fault diagnosis device based on big data, comprising:

[0006] The fault location module is used to divide the power system into regions and set different number marks for different regions. When the operation data collected in this numbered region contains pre-fault characteristic parameters, this numbered region is recorded as abnormal. Otherwise, this numbered region is recorded as normal. Based on this, the fault position is located and sent to the fault factor derivation module;

[0007] The data acquisition and processing module is used to obtain the operation data of the power system, process the obtained operation data, extract the characteristic parameters of the pre-fault, and send them to the characteristic extraction and selection module and the fault diagnosis module;

[0008] The feature extraction and selection module selects similar historical faults according to the feature parameters of the pre-fault, extracts feature parameters from the operating data of the fault, and selects appropriate feature parameters as reference for fault diagnosis and sends them to the fault diagnosis module;

[0009] A fault diagnosis module is used to match the reference characteristic parameters of the fault feature extraction operation data with the screened pre-fault operation data parameters, obtain the fault feature evaluation index and send it to the integration module, and perform fault diagnosis based on it to obtain a diagnosis result;

[0010] The fault factor derivation module is used to derive and analyze the abnormal factors at the fault location, including environmental factors, human factors and equipment failures, and then obtain the fault factor evaluation index and send it to the integration module, based on which the fault diagnosis is carried out to obtain accurate diagnosis results;

[0011] The integration module is used to integrate the received fault feature evaluation index and fault factor evaluation index, and perform fault diagnosis based on them to propose corresponding processing suggestions and preventive measures, and feed back the results to the execution terminal, which transmits the received results to the mobile terminal of the management duty personnel.

[0012] As a further improvement of the present invention: extracting the parameters of the pre-fault characteristics, specifically:

[0013] Image screening: Set the current, voltage and power in the operating data as the horizontal axis, set the operating parameter values ​​of the current, voltage and power as the vertical axis, and draw the operating parameter fluctuation lines of the current, voltage and power based on the horizontal and vertical axes;

[0014] At the same time, a fluctuation range is set on the coordinate graph according to the minimum to maximum values ​​of the operating parameters of current, voltage and power. When the fluctuation range of the operating parameter fluctuation line is stably within the fluctuation range, it is marked as a normal operating state. When the fluctuation range of the operating parameter fluctuation line exceeds the fluctuation range, it is marked as an abnormal operating state, and the operating parameters exceeding the fluctuation range are extracted as characteristic parameters of pre-fault.

[0015] As a further improvement of the present invention: feature parameter extraction is performed based on the operating data of the fault, and appropriate feature parameters are selected as references for fault diagnosis, specifically:

[0016] According to the characteristic parameters of the pre-fault, similar historical faults are selected to extract the operating data parameters, and the minimum to maximum values ​​of the data parameters of normal operation are set as the reference operating interval. The fault operating data parameters are compared with the reference operating interval to obtain the fault characteristic parameters of its operating data, and the fault characteristic parameters are sorted in order according to the similarity of the difference with the characteristic parameters of the pre-fault as reference characteristic parameters for fault diagnosis.

[0017] As a further improvement of the present invention: analysis is performed based on the fault feature evaluation index, specifically:

[0018] The fault reference characteristic parameters are matched with the screened pre-fault characteristic parameters, and the similarity between the characteristic parameters of the fault reference current, voltage and power and the characteristic parameters of the pre-fault current, voltage and power is used as the evaluation fault value. Each evaluation fault value is substituted into the summation formula to obtain the fault characteristic evaluation index, and the preliminary fault diagnosis result is obtained based on this.

[0019] As a further improvement of the present invention: based on the derivation and analysis of the abnormal factors at the fault location, a fault factor evaluation index is obtained, specifically:

[0020] Based on the abnormal number area sent by the fault location module, the environment of the fault location is obtained. The specific analysis is as follows:

[0021] Determine whether the power system in this area is located in an outdoor environment:

[0022] If yes, extract the recent weather, temperature and humidity of the fault location and conduct a comprehensive analysis to obtain the first evaluation index of environmental factors;

[0023] If not, then extract whether the fault location has rodent and insect infestation conditions, temperature and humidity, and conduct a comprehensive analysis to obtain the second evaluation index of environmental factors;

[0024] Determine whether the power system in the area has human influence conditions:

[0025] When the power system is located in an outdoor environment, the probability of human influence is high, so it is set as the first evaluation index of human factors;

[0026] When the environment where the power system is located is not an outdoor environment, the probability of human influence is low, and it is set as the second evaluation index of human factors;

[0027] Determine whether the power system in the area is affected by equipment failure:

[0028] Check the service life, used years and load status of the equipment. If it is a equipment failure, it is set as the first evaluation index of the equipment factor. Otherwise, it is set as the second evaluation index of the equipment factor.

[0029] Substitute the evaluation index of each fault factor into the summation formula and calculate the sum to obtain the fault factor evaluation index.

[0030] As a further improvement of the present invention: mutual integration is performed based on the fault feature evaluation index and the fault factor evaluation index, specifically:

[0031] The fault feature evaluation index and the fault factor evaluation index are analyzed to obtain the fault analysis result value, and the corresponding processing suggestions and preventive measures are matched according to the fault analysis result value.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] The present invention diagnoses the faults of the power system by setting a fault feature evaluation index and a fault factor evaluation index for comprehensive analysis, thereby obtaining the diagnosis results and the causing factors. The diagnosis results and the causing factors are used to verify each other to improve the accuracy of the fault diagnosis, and corresponding processing suggestions and preventive measures are provided accordingly, which is beneficial to ensuring the normal operation of the power system.

[0034] The present invention sets a fault factor evaluation index, and can perform fault pre-occurrence analysis based on it, thereby realizing potential risk estimation of power system failures. The power system can also be maintained in advance before a failure occurs through the estimation of potential risks, thereby reducing the failure probability of the power system and facilitating the normal operation of the power system.

[0035] In summary, the present invention is conducive to improving the accuracy of power system fault diagnosis, while avoiding over-reliance on the operating data of the power system. When the operating data of the power system is missing, there is a problem of difficulty in diagnosis. It is also conducive to predicting the potential risks of power system failures in advance, thereby achieving the effect of advance maintenance before the power system fails, which is conducive to ensuring more stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0037] Figure 1 is a system block diagram of the present invention;

[0038] Figure 2 is a flow chart of the present invention;

[0039] Figure 3 Screening data graphs for images of the present invention. DETAILED DESCRIPTION

[0040] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0041] See also Figure 1-3 As shown, the power system fault diagnosis device based on big data includes a database, a fault location module, a data acquisition and processing module, a feature extraction and selection module, a fault diagnosis module, a fault factor derivation module, an integration module and an execution terminal.

[0042] The database is used to store characteristic data of historical faults.

[0043] The fault location module is used to divide the power system into regions and set different number tags for different regions. When the operating data collected in this numbered area contains pre-fault characteristic parameters, this numbered area is recorded as abnormal. Otherwise, this numbered area is recorded as normal. Based on this, the fault position is located and sent to the fault factor derivation module.

[0044] The data acquisition and processing module is used to obtain the operating data of the power system, including current, voltage and power; filter the acquired operating data, extract the characteristic parameters of the pre-fault, and send them to the feature extraction selection module and the fault diagnosis module, specifically:

[0045] Image screening, setting current, voltage and power as horizontal coordinates, setting the operating parameter values ​​of current, voltage and power as vertical coordinates, and drawing the operating parameter fluctuation lines of current, voltage and power based on the horizontal and vertical coordinates;

[0046] At the same time, a fluctuation range is set on the coordinate graph according to the minimum to maximum values ​​of the operating parameters of current, voltage and power. When the fluctuation range of the operating parameter fluctuation line is stably within the fluctuation range, it is marked as a normal operating state. When the fluctuation range of the operating parameter fluctuation line exceeds the fluctuation range, it is marked as an abnormal operating state, and the operating parameters exceeding the fluctuation range are extracted as characteristic parameters of pre-fault.

[0047] The feature extraction and selection module selects similar historical faults according to the feature parameters of the pre-fault, extracts feature parameters from the operating data of the fault, and selects appropriate feature parameters as reference for fault diagnosis and sends them to the fault diagnosis module;

[0048] Extract characteristic parameters based on the operating data of the fault, and select appropriate characteristic parameters as reference for fault diagnosis, specifically:

[0049] According to the characteristic parameters of the pre-fault, similar historical faults are selected to extract the operating data parameters, and the minimum to maximum values ​​of the data parameters of normal operation are set as the reference operating interval. The fault operating data parameters are compared with the reference operating interval to obtain the fault characteristic parameters of its operating data, and the fault characteristic parameters are sorted in order according to the similarity of the difference with the characteristic parameters of the pre-fault as reference characteristic parameters for fault diagnosis.

[0050] The fault diagnosis module is used to match the reference characteristic parameters of the fault feature extraction operation data with the screened pre-fault operation data parameters to obtain the fault feature evaluation index, which is then sent to the integration module, and the fault diagnosis is performed accordingly to obtain the diagnosis result; specifically:

[0051] Match the fault reference characteristic parameters with the screened pre-fault characteristic parameters, and use the similarity between the characteristic parameters of the fault reference current, voltage and power and the characteristic parameters of the pre-fault current, voltage and power as the evaluation fault value, substitute each evaluation fault value into the summation formula to obtain the fault characteristic evaluation index, and obtain the preliminary fault diagnosis result based on it;

[0052] The current reference characteristic parameter is set to I1, I2, I3, ..., In according to different fault types, the voltage reference characteristic parameter is set to U1, U2, U3 ... Un according to different fault types, and the power reference characteristic parameter is set to P1, P2, P3 ..., Pn according to different fault types;

[0053] The current real-time characteristic parameter is set to I', the voltage real-time characteristic parameter is set to U', and the power real-time characteristic parameter is set to P'; the current real-time characteristic parameter, the voltage real-time characteristic parameter, and the power real-time characteristic parameter are respectively calculated to differ from the corresponding current reference characteristic parameter, voltage reference characteristic parameter, and power reference characteristic parameter, and the evaluation fault value corresponding to the current reference characteristic parameter, voltage reference characteristic parameter, and power reference characteristic parameter with the smallest difference is matched; for example:

[0054] When the difference between I' and I1 is the smallest, that is, I' is close to I1, the evaluation fault value corresponding to the matching current reference characteristic parameter I1 is YI1; when the difference between I' and I2 is the smallest, that is, I' is close to I2, the evaluation fault value corresponding to the matching current reference characteristic parameter I2 is YI2; and so on, when the difference between I' and In is the smallest, that is, I' is close to In, the current evaluation fault value corresponding to the matching current reference characteristic parameter In is YIn;

[0055] Similarly, when U' is close to U1 or U2 or Un, the corresponding voltage evaluation fault value YU1 or YU2 or YUn is matched; when P' is close to P1 or P2 or Pn, the corresponding power evaluation fault value YP1 or YP2 or YPn is matched;

[0056] The preset formula A=YIn×α1+YUn×α2+YPn×α3 is used to obtain the fault characteristic evaluation index A, where α1, α2, and α3 are set weight factors respectively, and their sizes are customized by technical personnel in this field when used.

[0057] The fault factor derivation module is used to derive and analyze the abnormal factors at the fault location, including environmental factors, human factors and equipment failures, obtain the fault factor evaluation index, send it to the integration module, and perform fault diagnosis based on it to further refine the diagnosis results; specifically:

[0058] Based on the abnormal number area sent by the fault location module, the environment of the fault location is obtained. The environment of the power system in the abnormal number area is whether it is an outdoor environment:

[0059] If yes, extract the weather, temperature and humidity data for the preset time period corresponding to the fault location.

[0060] Extreme weather is identified on weather data, including extreme weather such as thunder, strong wind and heavy rain, the number of occurrences and duration of extreme weather are counted, the preset weather impact value corresponding to each extreme weather is obtained, and the value of the duration is multiplied by the corresponding weather impact value to obtain the sky shadow value corresponding to the extreme weather; all sky shadow values ​​corresponding to all extreme weather are summed up to obtain the total sky shadow value; the number of occurrences of extreme weather is divided by the total duration of the preset time period to obtain the occurrence frequency; the occurrence frequency and the total sky shadow value are normalized and the values ​​of the two are taken, and the occurrence frequency and the total sky shadow value are marked as PL1 and TY2; the weather factor value X is obtained using the formula X=PL1×fs1+TY2×fs2, where fs1 and fs2 are both preset weight factors, and their sizes are custom settings.

[0061] The temperature data is judged, and the temperature exceeding the high temperature threshold of the power system load limit or below the low temperature threshold of the power system load limit is marked as an over-threshold temperature or a low-threshold temperature; the number of over-threshold temperatures and low-threshold temperatures corresponding to the data is counted, and the corresponding numbers of the two are multiplied by the corresponding preset weight coefficients, and then the sum is obtained to obtain the corresponding temperature factor value X;

[0062] Process the humidity data to determine whether the humidity is greater than the set humidity threshold, and calculate the duration corresponding to the humidity being greater than the set humidity threshold to obtain the humidity duration, and multiply the humidity duration by the corresponding weight coefficient to obtain the humidity factor value Z;

[0063] The weather factor value, temperature factor value and humidity factor value are normalized, and the values ​​of the three are input into the computer, and the values ​​of the three are converted into straight line lengths according to a certain ratio by the computer; an ellipse is constructed with the straight line lengths of the temperature factor value and the humidity factor value as the major axis and the minor axis, the center of the ellipse is selected, and a straight line with a length equal to the weather factor value is drawn with the center of the ellipse as the starting point, and then the ellipse and the straight line cone are used to extract the volume of the cone and mark the value of the volume to obtain the first evaluation index B of the environmental factor. 1 .

[0064] If not, the environmental factors such as whether the fault location has rodent and insect infestation conditions, temperature and humidity are extracted, and a comprehensive analysis is performed to obtain the second evaluation index of environmental factors; including the damage of rodents and insects to the power system, high and low temperature environments that exceed the carrying limit of the power system, excessive humidity and high moisture, etc.

[0065] Specifically, the video data corresponding to the fault location is obtained, and the preset rodent and insect identification is performed in the video data to obtain the number of rodent and insect occurrences and marked as S; the number of times the temperature exceeds the maximum high temperature of the power system and the number of times the temperature is lower than the maximum low temperature of the power system, the two numbers are summed to obtain the number of temperature differences and marked as Y; all humidity values ​​are summed and averaged to obtain the average humidity value Z; the number of rodent and insect occurrences, the number of temperature differences and the average humidity value are normalized and the values ​​of the three are substituted into formula B 2 =S×δ1+Y×δ2+Z×δ3 to obtain the second evaluation index B of environmental factors 2 , δ1, δ2, and δ3 represent the set proportional coefficient factors respectively.

[0066] Determine whether the power system in the area has human influence conditions:

[0067] When the power system is located in an outdoor environment, the number of cable thefts and malicious damage to power equipment in the area is collected and marked as L, the number of misoperations of switches or controllers is collected, and the number of times maintenance is ignored is collected M and other factors; the number of times maintenance is ignored is the number of times maintenance instructions are sent to the corresponding personnel and the corresponding maintenance is not performed within the preset time range. Substitute into formula B 3 =L×φ1+M×φ2+N×φ3 to obtain the first evaluation index B of human factors 3 , φ1, φ2, φ3 represent the set weight factors respectively;

[0068] When the power system is located in a non-outdoor environment, the formula Get the second evaluation index B of human factors 4 , They are respectively represented as the set weight factors, and φ2>φ4, φ3>φ5.

[0069] Is the power system in this area affected by equipment failure?

[0070] Check the total service life, service life and equipment load rate of the equipment, mark the corresponding values ​​of the total service life, service life and equipment load rate as D, E and F respectively, and substitute them into the formula Get the first evaluation index B of the equipment factor 5 , γ1, γ2, and γ3 represent the set weight factors respectively.

[0071] Substitute the evaluation index of each fault factor into the summation formula B = B i ×λ1+B j ×λ2+B 5×λ3 are summed up to obtain the fault factor evaluation index B, λ1, λ2, λ3 represent the set proportional coefficient factors respectively, i=1 or 2, j=3 or 4.

[0072] The integration module is used to integrate the received fault feature evaluation index and fault factor evaluation index, and perform fault diagnosis based on them, and then propose corresponding processing suggestions and preventive measures, and feed back the results to the execution terminal;

[0073] Based on the mutual integration of fault feature evaluation index and fault factor evaluation index, the specific ones are:

[0074] Substitute the fault feature evaluation index and the fault factor evaluation index into the preset formula C=A×0.4+B×0.6 for sum calculation to obtain the fault analysis result value C, and provide corresponding processing suggestions and preventive measures based on the fault analysis result value; pre-set a number of processing suggestions and preventive measures, each preventive measure corresponds to a result range; match the fault analysis result value C with all result ranges, and if the fault analysis result value C falls within the result range, match the preventive measures corresponding to the result range.

[0075] The execution terminal transmits the received results to the mobile terminal of the management duty personnel.

[0076] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A power system fault diagnosis device based on big data, characterized in that: include; The fault location module is used to divide the power system into regions and set different number marks for different regions. When the operation data collected in this numbered region contains pre-fault characteristic parameters, this numbered region is recorded as abnormal. Otherwise, this numbered region is recorded as normal. Based on this, the fault position is located and sent to the fault factor derivation module; The data acquisition and processing module is used to obtain the operation data of the power system, process the obtained operation data, extract the characteristic parameters of the pre-fault, and send them to the characteristic extraction and selection module and the fault diagnosis module; The feature extraction and selection module selects similar historical faults according to the feature parameters of the pre-fault, extracts feature parameters from the operating data of the fault, and selects appropriate feature parameters as reference for fault diagnosis and sends them to the fault diagnosis module; A fault diagnosis module is used to match the reference characteristic parameters of the fault feature extraction operation data with the screened pre-fault operation data parameters, obtain the fault feature evaluation index and send it to the integration module, and perform fault diagnosis based on it to obtain a diagnosis result; The fault factor derivation module is used to derive and analyze the abnormal factors at the fault location, including environmental factors, human factors and equipment failures, and then obtain the fault factor evaluation index and send it to the integration module, based on which the fault diagnosis is carried out to obtain accurate diagnosis results; Based on the derivation and analysis of abnormal factors at the fault location, the fault factor evaluation index is obtained, which is: Based on the abnormal number area sent by the fault location module, the environment of the fault location is obtained. The specific analysis is as follows: Determine whether the power system in this area is located in an outdoor environment: If yes, the weather, temperature and humidity of the fault location in recent period are extracted and comprehensively analyzed to obtain the first evaluation index of environmental factors; if no, whether the fault location has rodent and insect infestation conditions, temperature and humidity are extracted and comprehensively analyzed to obtain the second evaluation index of environmental factors; Determine whether the power system in the area has human influence conditions: When the power system is located in an outdoor environment, the number of cable thefts, malicious damage to power equipment, etc., the number of misoperations of switches or controllers, and the number of neglected maintenance are collected for comprehensive analysis to obtain the first evaluation index of human factors. When the power system is located in a non-outdoor environment, the number of misoperations of switches or controllers in the area, as well as the number of neglected maintenance are collected for comprehensive analysis to obtain the second evaluation index of human factors. Determine whether the power system in the area is affected by equipment failure: Check and conduct a comprehensive analysis of the service life, used years and load conditions of the equipment. If it is an equipment failure, it is set as the first evaluation index of the equipment factor. Otherwise, it is set as the second evaluation index of the equipment factor. Substituting the evaluation index of each fault factor into the summation formula for summation calculation to obtain the fault factor evaluation index; The integration module is used to integrate the received fault feature evaluation index and fault factor evaluation index, and perform fault diagnosis based on them to propose corresponding processing suggestions and preventive measures, and feed back the results to the execution terminal, which transmits the received results to the mobile terminal of the management duty personnel.

2. The power system fault diagnosis device based on big data according to claim 1 is characterized in that: The parameters of the pre-fault characteristics are extracted, specifically: Image screening: Set the current, voltage and power in the operating data as the horizontal axis, set the operating parameter values ​​of the current, voltage and power as the vertical axis, and draw the operating parameter fluctuation lines of the current, voltage and power based on the horizontal and vertical axes; At the same time, a fluctuation range is set on the coordinate graph according to the minimum to maximum values ​​of the operating parameters of current, voltage and power. When the fluctuation range of the operating parameter fluctuation line is stably within the fluctuation range, it is marked as a normal operating state. When the fluctuation range of the operating parameter fluctuation line exceeds the fluctuation range, it is marked as an abnormal operating state, and the operating parameters exceeding the fluctuation range are extracted as characteristic parameters of pre-fault.

3. The power system fault diagnosis device based on big data according to claim 1 is characterized in that: Extract characteristic parameters based on the operating data of the fault, and select appropriate characteristic parameters as reference for fault diagnosis, specifically: According to the characteristic parameters of the pre-fault, similar historical faults are selected to extract the operating data parameters, and the minimum to maximum values ​​of the data parameters of normal operation are set as the reference operating interval. The fault operating data parameters are compared with the reference operating interval to obtain the fault characteristic parameters of its operating data, and the fault characteristic parameters are sorted in order according to the similarity of the difference with the characteristic parameters of the pre-fault as reference characteristic parameters for fault diagnosis.

4. The power system fault diagnosis device based on big data according to claim 1, characterized in that: The analysis is based on the fault feature evaluation index, specifically: The fault reference characteristic parameters are matched with the screened pre-fault characteristic parameters, and the similarity between the characteristic parameters of the fault reference current, voltage and power and the characteristic parameters of the pre-fault current, voltage and power is used as the evaluation fault value. Each evaluation fault value is substituted into the summation formula to obtain the fault characteristic evaluation index, and the preliminary fault diagnosis result is obtained based on this.

5. The power system fault diagnosis device based on big data according to claim 1, characterized in that: Based on the mutual integration of fault feature evaluation index and fault factor evaluation index, the specific ones are: The fault feature evaluation index and the fault factor evaluation index are analyzed to obtain the fault analysis result value, and the corresponding processing suggestions and preventive measures are matched according to the fault analysis result value.

Images