Electric power information system fault diagnosis method

Abstract

The invention discloses a fault diagnosis method for an electric power information system, relates to the technical field of fault diagnosis, and solves the technical problems that a statistical error of the electric power information system is difficult to find, and an error area is difficult to accurately locate when an electric power statistical error is found. The business area is planned into a plurality of detection areas through the artificial intelligence model, the target data of each detection area is acquired from the electric power information system, the standard target data is acquired from the database based on the environmental condition, and the target data and the standard target data are analyzed to obtain the unreasonable area. The unreasonable area is analyzed to obtain the fault instruction, corresponding alarm information is sent based on the fault instruction, and the reliability and safety of the electric power information system are improved.

Description

Technical Field

[0001] The present invention belongs to the field of smart grids and relates to fault diagnosis technology, in particular to a method for diagnosing faults in an electric power information system. Background Art

[0002] The power information system is an indispensable part of modern society. It is responsible for managing and allocating power resources and ensuring the stability and security of the power system. As the scale and complexity of the power information system continue to increase, the interactions and dependencies between the components of the system, such as equipment, networks, and software, become more complex, resulting in a corresponding increase in the possibility of failures. The power information system may encounter various failures during operation, such as equipment failures, network failures, and statistical error failures. Equipment failures and network failures are relatively easy to detect system failures, but statistical errors are difficult to detect in the power information system. When such failures are discovered, they have already had a great impact on the management of power resources.

[0003] At present, most fault diagnosis methods used in power information systems only diagnose surface faults, and it is difficult to detect statistical errors in the power information system in real time. If such errors are not found, it will bring certain safety hazards to the operation of the power information system. At the same time, when power statistical errors are found, it is difficult to accurately locate the sub-area or subsystem where the error occurs through the abnormal values ​​in the power information system, which increases the difficulty for relevant personnel to troubleshoot errors.

[0004] Therefore, the present invention discloses a fault diagnosis method for an electric power information system, which is used to solve the above technical problems. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a fault diagnosis method for an electric power information system, which is used to solve the situation that it is difficult to discover statistical errors in the electric power information system, and when electric power statistical errors are discovered, it is difficult to accurately locate the technical problem of the error area. The present invention plans the business area into several detection areas through an artificial intelligence model, obtains target data of each detection area from the electric power information system, obtains standard target data from a database based on environmental conditions, analyzes the target data and the standard target data to obtain unreasonable areas, analyzes the unreasonable areas to obtain fault instructions, and issues corresponding alarm information based on the fault instructions to solve the above problems.

[0006] To achieve the above object, a first aspect of the present invention provides a method for diagnosing faults in a power information system, comprising:

[0007] S1: The business area is divided into several detection areas through the artificial intelligence model, and the target data of each detection area is obtained from the power information system; the environmental conditions of each detection area on the day are obtained; the target data includes power generation and power consumption, and the environmental conditions include: maximum temperature, weather conditions, and date conditions;

[0008] S2: Obtain standard target data from the database based on environmental conditions, analyze the target data and standard target data to obtain unreasonable areas, and analyze the unreasonable areas to obtain fault instructions; wherein the standard target data includes standard power generation and standard power consumption;

[0009] S3: Issue corresponding alarm information based on the fault instruction.

[0010] In the present invention, the business area is planned into several detection areas through the artificial intelligence model in order to plan a larger business area into several smaller detection areas, because a larger area needs to consider more influences of different situations, while a smaller area may need to consider only a few influences, which can reduce the computational complexity of the system; standard target data is obtained from the database based on environmental conditions in order to take into account the influence of temperature on power generation and power consumption, which can improve the accuracy of data analysis.

[0011] Preferably, the business area is planned into several detection areas by using an artificial intelligence model, including:

[0012] Divide the business area into several assessment areas by manual setting, obtain the area, power generation and power consumption of each assessment area, and the distance between the centers of each assessment area;

[0013] Extract the historical area, historical power generation and historical power consumption of each historical assessment area with different division times, and the distance between the centers of each historical assessment area from the database; obtain the historical detection area range corresponding to the historical assessment area with different division times from the database; use the historical area, historical power generation and historical power consumption of each historical assessment area, and the distance between the centers of each historical assessment area as training data, and use the historical detection area range as test data; use the training data to train the artificial intelligence model, use the test data to test the trained artificial intelligence model, and adjust the parameters of the artificial intelligence model according to the test results to obtain a regional matching model with the area, power generation, power consumption and distance between the centers of each assessment area as input, and the range of each detection area as output; wherein the artificial intelligence model includes a K-Means clustering algorithm and a deep belief network;

[0014] The area, power generation, power consumption, and distance of the center of each assessment area are transmitted to the regional matching model to obtain the scope of each detection area; based on the scope of each detection area, the business area is planned into several detection areas.

[0015] It should be noted that extracting historical assessment areas with different numbers of divisions can be understood as: the database records the division of the assessment areas H times, and the historical area, historical power production and historical power consumption of each assessment area corresponding to each division, as well as the distance between the centers of each historical assessment area are recorded in the data packet of that time. When constructing the regional matching model, the H data packets are used as training data to train the artificial intelligence model; the test data corresponding to the training data is obtained by manual identification.

[0016] Preferably, the acquiring target data of each detection area from the power information system includes:

[0017] At the set time, the power generation and consumption of each detection area on that day are extracted from the power information system.

[0018] Preferably, the obtaining of the environmental conditions of each detection area on that day includes:

[0019] Extract the highest temperature of each detection area on that day from the computer;

[0020] Obtain the weather conditions of each detection area on the day; determine whether the weather conditions of each detection area on the day are extreme weather; if yes, obtain the type of extreme weather and extract the extreme weather factor according to the type; if no, mark the extreme weather factor as 1; wherein the types of extreme weather include heavy rain, heavy snow, and strong wind; the extreme weather factor is obtained by manual setting;

[0021] Get the date of the day and determine whether the period of the day is a holiday; if yes, get the type of holiday and extract the holiday factor according to the type; if no, mark the holiday factor as 1; wherein the holiday factor is set manually.

[0022] Preferably, the acquiring of standard target data from a database based on environmental conditions comprises:

[0023] Extract the highest temperature, weather conditions, and date conditions from the environmental conditions of each detection area, and mark the historical days with the same highest temperature as the current day as candidate data from the database;

[0024] The historical power generation in the selected data is marked as group A, and the historical power consumption is marked as group B;

[0025] Obtain the variance of each group in turn, and determine whether the variance of each group is less than the corresponding set value; if yes, calculate the average value of the group to obtain the eigenvalue; if no, extract the mode of the group, remove the number with the largest absolute value of difference from the mode in the group, and recalculate the variance of the remaining data of the group, and judge the variance of the remaining data of the group with the set value until the variance of the remaining data of the group meets the set value, stop removing data, and calculate the average value of the remaining data of the group to obtain the eigenvalue; wherein the set value is set manually;

[0026] Obtain the target data of the previous n days, and calculate the average value of each data in the target data of the previous n days to obtain the corresponding average power generation and average power consumption; where n is set manually;

[0027] The characteristic values ​​of each group are averaged with the average data corresponding to the previous n days to obtain the reference target data corresponding to each group; the average data includes the average power generation and the average power consumption, and the reference target data includes the reference power generation and the reference power consumption;

[0028] Based on the weather conditions of the current date and the date conditions, extreme weather factors and holiday factors are extracted, and based on the extreme weather factors and holiday factors, the reference target data is updated to obtain the standard target data.

[0029] It should be noted that when obtaining the eigenvalues ​​of each group, the present invention first performs analysis through variance. When the variance of a group is less than the corresponding set value, it means that the average value of the data of the group can be used as a stable eigenvalue. When the variance of a group is not less than the corresponding set value, it means that the data of the group has a maximum or minimum value that affects volatility. At this time, the maximum or minimum value that affects volatility is removed, and then the obtained average value can also be used as a stable eigenvalue, which provides data support for the subsequent acquisition of standard target data.

[0030] It is worth noting that in the present invention, the characteristic values ​​of each group and the average data corresponding to the previous n days are averaged to obtain the reference target data corresponding to each group. The reference target data obtained in this step includes the influence of temperature and electricity price. The temperature influence is expressed through the characteristic value, and the electricity price influence is expressed through the average data. The reference target data obtained by combining the two is more scientific.

[0031] Preferably, the updating of the reference target data based on the extreme weather factors and the holiday factors to obtain the standard target data includes:

[0032] The extreme weather factor is marked as JTY, the holiday factor is marked as JRY, the reference target data CKMi is extracted, and the standard target data BZMi of each detection area is obtained based on the formula BZMi=(α×JTY+β×JRY)×CKMi; wherein α and β are both proportional adjustment coefficients greater than 0, and α+β=1; i is the target data type serial number, i=1 indicates power generation, and i=2 indicates power consumption.

[0033] It is worth noting that due to the differences in the location of the detection area, the population living there, and the enterprises where they are located, the same extreme weather and holidays will have different impacts on the target data of different detection areas on the same day. Therefore, the target data of each detection area is calculated to obtain the corresponding standard target data. This is conducive to obtaining standard data suitable for each detection area.

[0034] Preferably, the analysis of the target data and the standard target data to obtain the unreasonable area includes:

[0035] A1: Extract the standard power generation, standard power consumption, power generation and power consumption of each detection area on that day;

[0036] A2: Determine whether the ratio of power generation to standard power generation exceeds the corresponding setting range; if yes, mark the current detection area as the power generation change area; if no, do nothing; the setting range is obtained through experience;

[0037] A3: Determine whether the ratio of power consumption to standard power consumption exceeds the corresponding set range; if yes, mark the current detection area as a power consumption change area; if no, do nothing; among them, unreasonable areas include power generation change areas and power consumption change areas.

[0038] Preferably, the analyzing the unreasonable area to obtain the fault instruction includes:

[0039] Extract the power consumption change area, and determine whether the ratio of the number of the power consumption change areas to the number of the detection areas exceeds the ratio threshold value 1; if yes, issue instruction 1; if no, issue instruction 2; wherein the ratio threshold value 1 is obtained through experience;

[0040] Extract the area where power generation changes, and determine whether the ratio of the number of power generation change areas to the number of detection areas exceeds ratio threshold 2; if yes, issue instruction 3; if not, issue instruction 4; wherein, ratio threshold 2 is obtained through experience, and fault instructions include instruction 1, instruction 2, instruction 3 and instruction 4.

[0041] Preferably, the issuing of corresponding alarm information based on the fault instruction includes:

[0042] When instruction one is received, an alarm of systematic power consumption statistics error is issued, and manual investigation is arranged; when instruction two is received, an alarm of small-area power consumption statistics error is issued, and the location of the power consumption change area and the ratio of power consumption to standard power consumption are sent to the management personnel;

[0043] When instruction three is received, an alarm of systematic power generation statistics error is issued, and manual investigation is arranged; when instruction four is received, an alarm of small area power generation statistics error is issued, and the location of the power generation change area and the ratio of power generation to standard power generation are sent to the management personnel.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] 1. The present invention plans the business area into several detection areas through an artificial intelligence model, obtains target data of each detection area from the power information system, obtains standard target data from the database based on environmental conditions, analyzes the target data and the standard target data to obtain unreasonable areas, analyzes the unreasonable areas to obtain fault instructions, and issues corresponding alarm information based on the fault instructions, thereby solving the problem that it is difficult to discover statistical errors in the power information system, and the technical problem that it is difficult to accurately locate the error area when power statistical errors are discovered, thereby improving the reliability and security of the power information system.

[0046] 2. In the present invention, the characteristic values ​​of each group and the average data corresponding to the previous n days are averaged to obtain the reference target data corresponding to each group. The reference target data obtained in this step includes the influence of temperature and electricity price. The temperature influence is expressed through the characteristic values, and the electricity price influence is expressed through the average data. The reference target data obtained by combining the two is more scientific. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] 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 will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 A schematic diagram of the operation steps in an embodiment of the present invention;

[0049] Figure 2 A schematic diagram of the operation steps for obtaining standard target data of the present invention;

[0050] Figure 3 The figure is a schematic diagram of the operation steps for obtaining the unreasonable area of ​​the present invention. DETAILED DESCRIPTION

[0051] 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.

[0052] See also Figure 1-2 The first embodiment of the present invention provides a method for diagnosing faults in a power information system, comprising:

[0053] S1: The business area is divided into several detection areas through the artificial intelligence model, and the target data of each detection area is obtained from the power information system; the environmental conditions of each detection area on the day are obtained; the target data includes power generation and power consumption, and the environmental conditions include: maximum temperature, weather conditions, and date conditions;

[0054] S2: Obtain standard target data from the database based on environmental conditions, analyze the target data and standard target data to obtain unreasonable areas, and analyze the unreasonable areas to obtain fault instructions; wherein the standard target data includes standard power generation and standard power consumption;

[0055] S3: Issue corresponding alarm information based on the fault instruction.

[0056] For example, in this embodiment, the highest temperature of the entire business area is 30° C., the weather condition is heavy rain, and the date condition is not a holiday;

[0057] The business area is planned into 100 assessment areas. The area, power generation and power consumption of each assessment area, as well as the distance to the center of each assessment area are transmitted to the regional matching model to obtain the range of each detection area. Based on the range of each detection area, the business area is planned into 25 detection areas.

[0058] At 23:59:59 every day, the power generation and power consumption of each detection area are obtained; the extreme weather factor of heavy rain is obtained and the holiday factor is marked as 1;

[0059] Obtain the characteristic values ​​of the two groups of 25 detection areas A and B, obtain the target data of the previous 10 days, and calculate the average value of each data in the target data of the previous 10 days to obtain the corresponding average power generation and average power consumption; calculate the average value of the characteristic value of group A and the average power generation to obtain the reference power generation CKM1, and calculate the average value of the characteristic value of group B and the average power consumption to obtain the reference power consumption CKM2; obtain the standard target data BZMi of each detection area based on the formula BZMi=(α×JTY+β×JRY)×CKMi; where group A represents the historical power generation and group B represents the historical power consumption;

[0060] The target data and the standard target data are analyzed to obtain unreasonable areas, the unreasonable areas are analyzed to obtain fault instructions, and then corresponding alarm information is issued based on the fault instructions.

[0061] In this application, the business area is divided into several evaluation areas by artificial setting through an artificial intelligence model, and the area, power generation and power consumption of each evaluation area, as well as the distance of the center of each evaluation area are obtained;

[0062] Extract the historical area, historical power generation and historical power consumption of each historical assessment area with different division times, and the distance between the centers of each historical assessment area from the database; obtain the historical detection area range corresponding to the historical assessment area with different division times from the database; use the historical area, historical power generation and historical power consumption of each historical assessment area, and the distance between the centers of each historical assessment area as training data, and use the historical detection area range as test data; use the training data to train the artificial intelligence model, use the test data to test the trained artificial intelligence model, and adjust the parameters of the artificial intelligence model according to the test results to obtain a regional matching model with the area, power generation, power consumption and distance between the centers of each assessment area as input, and the range of each detection area as output; wherein the artificial intelligence model includes a K-Means clustering algorithm and a deep belief network;

[0063] The area, power generation, power consumption, and distance of the center of each assessment area are transmitted to the regional matching model to obtain the scope of each detection area; based on the scope of each detection area, the business area is planned into several detection areas.

[0064] It should be noted that the area of ​​each evaluation area is obtained because: the area of ​​the evaluation area has a non-negligible impact on the planning of the detection area. When the area of ​​the evaluation area of ​​the fixed business area is small, the corresponding number of evaluation areas will increase accordingly, and the number of planned detection areas will also increase accordingly; and when the area of ​​the evaluation area of ​​the fixed business area is large, the corresponding number of evaluation areas will decrease accordingly, and the restrictions on planning the detection area will increase.

[0065] In this application, target data of each detection area is obtained from the power information system, including:

[0066] At the set time, the power generation and consumption of each detection area on that day are extracted from the power information system.

[0067] It should be noted that the present invention obtains the power generation and power consumption of the day at a fixed time set manually.

[0068] This application obtains the environmental conditions of each detection area on the day, including:

[0069] Extract the highest temperature of each detection area on that day from the computer;

[0070] Obtain the weather conditions of each detection area on the day; determine whether the weather conditions of each detection area on the day are extreme weather; if yes, obtain the type of extreme weather and extract the extreme weather factor according to the type; if no, mark the extreme weather factor as 1; wherein the types of extreme weather include heavy rain, heavy snow, and strong wind; the extreme weather factor is obtained by manual setting;

[0071] Get the date of the day and determine whether the period of the day is a holiday; if yes, get the type of holiday and extract the holiday factor according to the type; if no, mark the holiday factor as 1; wherein the holiday factor is set manually.

[0072] It should be noted that the extreme weather factors and holiday factors established in this step are not all greater than 1, and may be less than 1. This is because some extreme weather or holidays may have an impact on the target data growth, while some extreme weather or holidays may have an impact on the target data reduction.

[0073] It should be noted that marking the extreme weather factor as 1 indicates that the weather conditions on that day are not extreme, and the impact of the weather conditions on the target data is 0; marking the holiday factor as 1 indicates that the day is not a holiday, and the impact of the date conditions on the target data is 0.

[0074] In this application, standard target data is obtained from the database based on the environmental conditions, including:

[0075] Extract the highest temperature, weather conditions, and date conditions from the environmental conditions of each detection area, and mark the historical days with the same highest temperature as the current day as candidate data from the database;

[0076] The historical power generation in the selected data is marked as group A, and the historical power consumption is marked as group B;

[0077] Obtain the variance of each group in turn, and determine whether the variance of each group is less than the corresponding set value; if yes, calculate the average value of the group to obtain the eigenvalue; if no, extract the mode of the group, remove the number with the largest absolute value of difference from the mode in the group, and recalculate the variance of the remaining data of the group, and judge the variance of the remaining data of the group with the set value until the variance of the remaining data of the group meets the set value, stop removing data, and calculate the average value of the remaining data of the group to obtain the eigenvalue; wherein the set value is set manually;

[0078] Obtain the target data of the previous n days, and calculate the average value of each data in the target data of the previous n days to obtain the corresponding average power generation and average power consumption; where n is set manually;

[0079] The characteristic values ​​of each group are averaged with the average data corresponding to the previous n days to obtain the reference target data corresponding to each group; the average data includes the average power generation and the average power consumption, and the reference target data includes the reference power generation and the reference power consumption;

[0080] Based on the weather conditions of the current date and the date conditions, extreme weather factors and holiday factors are extracted, and based on the extreme weather factors and holiday factors, the reference target data is updated to obtain the standard target data.

[0081] It should be noted that the reference target data corresponding to each group is obtained by averaging the characteristic values ​​of each group with the average data corresponding to the previous n days: the characteristic values ​​of group A and the average power generation, and the characteristic values ​​of group B and the average power consumption.

[0082] It is worth noting that the purpose of obtaining the average value of each data in the target data of the previous n days is to obtain the impact of the recent electricity price on each data. Since electricity prices may fluctuate to a certain extent, this fluctuation will have a certain impact on the target data. Therefore, adding the impact of electricity prices to the analysis of standard target data can better obtain the corresponding standard situation for the day.

[0083] In this application, the reference target data is updated based on extreme weather factors and holiday factors to obtain standard target data, including:

[0084] The extreme weather factor is marked as JTY, the holiday factor is marked as JRY, the reference target data CKMi is extracted, and the standard target data BZMi of each detection area is obtained based on the formula BZMi=(α×JTY+β×JRY)×CKMi; wherein α and β are both proportional adjustment coefficients greater than 0, and α+β=1; i is the target data type serial number, i=1 indicates power generation, and i=2 indicates power consumption.

[0085] See also Figure 3 In this application, the target data and standard target data are analyzed to obtain unreasonable areas, including:

[0086] A1: Extract the standard power generation, standard power consumption, power generation and power consumption of each detection area on that day;

[0087] A2: Determine whether the ratio of power generation to standard power generation exceeds the corresponding set range; if yes, mark the current detection area as a power generation change area; if no, do nothing;

[0088] A3: Determine whether the ratio of power consumption to standard power consumption exceeds the corresponding set range; if yes, mark the current detection area as a power consumption change area; if no, do nothing; among them, unreasonable areas include power generation change areas and power consumption change areas.

[0089] In this application, the unreasonable area is analyzed to obtain fault instructions, including:

[0090] Extract the power consumption change area, and determine whether the ratio of the number of the power consumption change areas to the number of the detection areas exceeds the ratio threshold value 1; if yes, issue instruction 1; if no, issue instruction 2; wherein the ratio threshold value 1 is obtained through experience;

[0091] Extract the area where power generation changes, and determine whether the ratio of the number of power generation change areas to the number of detection areas exceeds ratio threshold 2; if yes, issue instruction 3; if not, issue instruction 4; wherein, ratio threshold 2 is obtained through experience, and fault instructions include instruction 1, instruction 2, instruction 3 and instruction 4.

[0092] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0093] Working principle of the present invention:

[0094] The present invention first plans the business area into several detection areas through an artificial intelligence model, obtains target data of each detection area from the power information system, and obtains the environmental conditions of each detection area on the day; obtains standard target data from the database based on the environmental conditions, extracts the standard power generation, standard power consumption, power generation and power consumption of each detection area on the day, marks the detection area where the ratio of power generation to standard power generation exceeds the corresponding set range as a power generation change area, marks the detection area where the ratio of power consumption to standard power consumption exceeds the corresponding set range as a power consumption change area, marks the power generation change area and the power consumption change area as unreasonable areas, analyzes the unreasonable areas to obtain fault instructions; finally, issues corresponding alarm information based on the fault instructions.

[0095] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for diagnosing faults in a power information system, characterized in that: S1: The business area is divided into several detection areas through the artificial intelligence model, and the target data of each detection area is obtained from the power information system; Obtain the environmental conditions of each detection area on the day; the target data includes power generation and power consumption, and the environmental conditions include: maximum temperature, weather conditions, and date conditions; S2: Obtain standard target data from the database based on environmental conditions, analyze the target data and standard target data to obtain unreasonable areas, and analyze the unreasonable areas to obtain fault instructions; wherein the standard target data includes standard power generation and standard power consumption; S3: Issue corresponding alarm information based on the fault instruction.

2. A method for diagnosing faults in a power information system according to claim 1, characterized in that: The business area is planned into several detection areas through the artificial intelligence model, including: Divide the business area into several assessment areas by manual setting, and obtain the area, power generation and power consumption of each assessment area, as well as the distance between the centers of each assessment area; Extract the historical area, historical power generation and historical power consumption of each historical assessment area with different division times, and the distance between the centers of each historical assessment area from the database; obtain the historical detection area range corresponding to the historical assessment area with different division times from the database; use the historical area, historical power generation and historical power consumption of each historical assessment area, and the distance between the centers of each historical assessment area as training data, and use the historical detection area range as test data; use the training data to train the artificial intelligence model, use the test data to test the trained artificial intelligence model, and adjust the parameters of the artificial intelligence model according to the test results to obtain a regional matching model with the area, power generation, power consumption and distance between the centers of each assessment area as input, and the range of each detection area as output; wherein the artificial intelligence model includes a K-Means clustering algorithm and a deep belief network; The area, power generation, power consumption, and distance of the center of each assessment area are transmitted to the regional matching model to obtain the scope of each detection area; based on the scope of each detection area, the business area is planned into several detection areas.

3. A method for diagnosing faults in a power information system according to claim 2, characterized in that: The step of acquiring target data of each detection area from the power information system includes: At the set time, the power generation and consumption of each detection area on that day are extracted from the power information system.

4. A method for diagnosing faults in a power information system according to claim 3, characterized in that: The obtaining of the environmental conditions of each detection area on the day includes: Extract the highest temperature of each detection area on that day from the computer; Obtain the weather conditions of each detection area on the day; determine whether the weather conditions of each detection area on the day are extreme weather; if yes, obtain the type of extreme weather and extract the extreme weather factor according to the type; if no, mark the extreme weather factor as 1; wherein the types of extreme weather include heavy rain, heavy snow, and strong wind; Get the date of the day and determine whether the period of the day is a holiday; if yes, get the type of holiday and extract the holiday factor according to the type; if no, mark the holiday factor as 1.

5. A method for diagnosing faults in a power information system according to claim 4, characterized in that: The step of acquiring standard target data from a database based on environmental conditions includes: Extract the highest temperature, weather conditions, and date conditions from the environmental conditions of each detection area, and mark the historical days with the same highest temperature as the current day as candidate data from the database; The historical power generation in the selected data is marked as group A, and the historical power consumption is marked as group B; Obtain the variance of each group in turn, and determine whether the variance of each group is less than the corresponding set value; if yes, calculate the average value of the group to obtain the eigenvalue; if no, extract the mode of the group, remove the number with the largest absolute value of the difference with the mode in the group, and recalculate the variance of the remaining data of the group, and judge the variance of the remaining data of the group with the set value until the variance of the remaining data of the group meets the set value, and stop removing data, and calculate the average value of the remaining data of the group to obtain the eigenvalue; Obtain the target data of the previous n days, and calculate the average value of each data in the target data of the previous n days to obtain the corresponding average power generation and average power consumption; where n is set manually; The characteristic values ​​of each group are averaged with the average data corresponding to the previous n days to obtain the reference target data corresponding to each group; the average data includes the average power generation and the average power consumption, and the reference target data includes the reference power generation and the reference power consumption; Based on the weather conditions of the current date and the date conditions, extreme weather factors and holiday factors are extracted, and based on the extreme weather factors and holiday factors, the reference target data is updated to obtain the standard target data.

6. A method for diagnosing faults in a power information system according to claim 5, characterized in that: The updating of the reference target data based on the extreme weather factors and holiday factors to obtain the standard target data includes: The extreme weather factor is marked as JTY, the holiday factor is marked as JRY, the reference target data CKMi is extracted, and the standard target data BZMi of each detection area is obtained based on the formula BZMi=(α×JTY+β×JRY)×CKMi; wherein α and β are both proportional adjustment coefficients greater than 0, and α+β=1; i is the target data type serial number, i=1 indicates power generation, and i=2 indicates power consumption.

7. A method for diagnosing faults in a power information system according to claim 5, characterized in that: The unreasonable areas obtained by analyzing the target data and the standard target data include: A1: Extract the standard power generation, standard power consumption, power generation and power consumption of each detection area on that day; A2: Determine whether the ratio of power generation to standard power generation exceeds the corresponding set range; if yes, mark the current detection area as a power generation change area; if no, do nothing; A3: Determine whether the ratio of power consumption to standard power consumption exceeds the corresponding set range; if yes, mark the current detection area as a power consumption change area; if no, do nothing; among them, unreasonable areas include power generation change areas and power consumption change areas.

8. A method for diagnosing faults in a power information system according to claim 7, characterized in that: The analyzing the unreasonable area to obtain the fault instruction includes: Extract the power consumption change area, and determine whether the ratio of the number of the power consumption change areas to the number of the detection areas exceeds the ratio threshold 1; if yes, issue instruction 1; if no, issue instruction 2; Extract the power generation change area, and determine whether the ratio of the number of the power generation change areas to the number of detection areas exceeds the ratio threshold 2; if yes, issue instruction 3; if no, issue instruction 4; wherein the fault instructions include instruction 1, instruction 2, instruction 3 and instruction 4.

9. A method for diagnosing faults in a power information system according to claim 8, characterized in that: The issuing of corresponding alarm information based on the fault instruction includes: When instruction one is received, an alarm of systematic power consumption statistics error is issued, and manual investigation is arranged; when instruction two is received, an alarm of small-area power consumption statistics error is issued, and the location of the power consumption change area and the ratio of power consumption to standard power consumption are sent to the management personnel; When instruction three is received, an alarm of systematic power generation statistics error is issued, and manual investigation is arranged; when instruction four is received, an alarm of small area power generation statistics error is issued, and the location of the power generation change area and the ratio of power generation to standard power generation are sent to the management personnel.

Images