System and method for predicting failure of power equipment according to dangerous weather phenomena

Abstract

A system and method for predicting failures of power facilities due to hazardous weather phenomena are provided. According to embodiments of the present invention, a comprehensive failure prediction index and a failure risk grade for power facilities can be calculated in an optimized manner for each area within the analysis target area and for a set period. In particular, according to embodiments of the present invention, by calculating failure prediction indices for wind and rain damage, lightning damage, and ice and snow damage—which are the biggest factors causing failures in power facilities—and then comprehensively considering them to calculate a comprehensive failure prediction index and a failure risk grade, the probability of failure in power facilities can be predicted more accurately.

Description

Technology Field

[0001] Embodiments of the present invention relate to a technology for predicting failures of power facilities caused by hazardous weather phenomena. Background Technology

[0003] Generally, power facilities can be classified into distribution facilities, transmission facilities, etc. These power facilities may fail due to various weather phenomena (or weather elements), and the causes of such failures are also very diverse. In particular, while a failure in power facilities may be caused by a single weather phenomenon, it may also occur as a result of being simultaneously affected by multiple weather phenomena. Furthermore, even for the same power facility, the frequency of failure may vary depending on the installation location and time of day. Prior art literature

[0005] Korean Registered Patent Publication No. 10-2356263 (January 24, 2022) The problem to be solved

[0006] The embodiments of the present invention are intended to efficiently calculate a comprehensive failure prediction index and a failure risk grade for power facilities by utilizing meteorological elements such as wind speed and precipitation for each region and set period. means of solving the problem

[0008] According to one embodiment, a power facility failure prediction system due to hazardous weather phenomena is provided, comprising: a data collection unit that collects power facility failure data including the time of failure, location of failure, equipment failure, and cause of failure for power facilities in an analysis target area, and weather forecast data for said analysis target area; a data preprocessing unit that mutually maps said power facility failure data and said weather forecast data by the same time period and same location; a failure risk analysis unit that extracts power facility failure data corresponding to a plurality of set causes of failure among said power facility failure data, analyzes said weather forecast data corresponding to said power facility failure data by each cause of failure of said power facility failure data, calculates a failure prediction index for each cause of failure for each set area within said analysis target area, calculates a comprehensive failure prediction index for said area from said failure prediction index for each cause of failure, and calculates a failure risk grade for said area according to said comprehensive failure prediction index; and a verification unit that verifies said failure risk grade for said area using a set verification model.

[0009] The above-mentioned failure risk analysis unit can extract the first power equipment failure data corresponding to wind and rain damage, the second power equipment failure data corresponding to lightning damage, and the third power equipment failure data corresponding to ice and snow damage from the above-mentioned power equipment failure data, and then analyze the weather forecast data corresponding to the extracted first power equipment failure data, the second power equipment failure data, and the third power equipment failure data to calculate the first failure prediction index corresponding to wind and rain damage, the second failure prediction index corresponding to lightning damage, and the third failure prediction index corresponding to ice and snow damage for each area.

[0010] The above failure risk analysis unit can calculate the first failure prediction index by obtaining the wind distribution by wind speed grade set for each area and calculating the frequency of failure occurring when wind of the specific wind speed grade occurs relative to the frequency of wind of the specific wind speed grade, calculate the second failure prediction index by obtaining the precipitation amount distribution by precipitation grade set for each area and calculating the frequency of failure occurring in the specific precipitation grade relative to the frequency of failure occurring due to lightning damage, and calculate the third failure prediction index by obtaining the snowfall amount distribution by snowfall grade set for each area and calculating the frequency of failure occurring in the specific snowfall grade relative to the frequency of failure occurring due to ice and snow damage.

[0011] The above failure risk analysis unit can calculate the comprehensive failure prediction index for each of the above areas using the following mathematical formula.

[0012] [Mathematical Formula]

[0013] Comprehensive Failure Prediction Index = a * 1st Failure Prediction Index + b * 2nd Failure Prediction Index + c * 3rd Failure Prediction Index

[0014] (Here, a, b, and c represent the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient, respectively, and a + b + c = 1)

[0015] The above failure risk analysis unit can determine the above first weighting coefficient, the above second weighting coefficient, and the above third weighting coefficient according to the failure occurrence rate caused by each failure cause for each area.

[0016] The verification unit can apply a confusion matrix to the failure risk grade for each area to calculate Accuracy (ACC) and Critical Success Index (CSI), and verify the failure risk grade for each area based on the Accuracy and Critical Success Index.

[0017] According to another embodiment, in a data collection unit, a step of collecting power equipment failure data including the time of failure, location of failure, equipment failure, and cause of failure for power equipment in an analysis target area, and weather forecast data for said analysis target area; in a data preprocessing unit, a step of mutually mapping said power equipment failure data and said weather forecast data by the same time period and same location; in a failure risk analysis unit, a step of extracting power equipment failure data corresponding to a plurality of set causes of failure among said power equipment failure data, and then analyzing the weather forecast data corresponding to said power equipment failure data by the cause of failure of said power equipment failure data to calculate a failure prediction index for each cause of failure for each set area within said analysis target area; in the failure risk analysis unit, a step of calculating a comprehensive failure prediction index for said area from the failure prediction index for each cause of failure; and in the failure risk analysis unit, a step of calculating a failure risk grade for said area according to said comprehensive failure prediction index. A method for predicting failure of power equipment due to hazardous weather phenomena is provided, comprising the step of verifying the failure risk grade for each area using a set verification model in the verification section.

[0018] The step of calculating the failure prediction index for each of the above failure causes can be performed by extracting the first power equipment failure data corresponding to wind and rain damage, the second power equipment failure data corresponding to lightning damage, and the third power equipment failure data corresponding to ice and snow damage from the above power equipment failure data, and then analyzing the weather forecast data corresponding to the extracted first power equipment failure data, the second power equipment failure data, and the third power equipment failure data to calculate the first failure prediction index corresponding to wind and rain damage, the second failure prediction index corresponding to lightning damage, and the third failure prediction index corresponding to ice and snow damage for each area.

[0019] The step of calculating the failure prediction index for each of the above failure causes may be to obtain the wind distribution by wind speed grade set for each of the above areas and calculate the first failure prediction index through the frequency of failure occurring when wind of the specific wind speed grade occurs relative to the frequency of wind of the specific wind speed grade, obtain the precipitation amount distribution by precipitation grade set for each of the above areas and calculate the second failure prediction index through the frequency of failure occurring in the specific precipitation grade relative to the frequency of failure occurring due to lightning damage, and obtain the snowfall amount distribution by snowfall grade set for each of the above areas and calculate the third failure prediction index through the frequency of failure occurring in the specific snowfall grade relative to the frequency of failure occurring due to ice and snow damage.

[0020] The step of calculating the comprehensive failure prediction index for each of the above areas can be performed by using the following mathematical formula to calculate the comprehensive failure prediction index for each of the above areas.

[0021] [Mathematical Formula]

[0022] Comprehensive Failure Prediction Index = a * 1st Failure Prediction Index + b * 2nd Failure Prediction Index + c * 3rd Failure Prediction Index

[0023] (Here, a, b, and c represent the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient, respectively, and a + b + c = 1)

[0024] The step of calculating the comprehensive failure prediction index for each of the above areas may determine the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient according to the failure occurrence rate caused by each failure cause for each of the above areas.

[0025] The step of verifying the failure risk level for each of the above areas may involve applying a confusion matrix to the failure risk level for each of the above areas to calculate the Accuracy (ACC) and Critical Success Index (CSI), and verifying the failure risk level for each of the above areas based on the Accuracy and Critical Success Index. Effects of the invention

[0027] According to embodiments of the present invention, a comprehensive failure prediction index and a failure risk grade for power facilities can be calculated in an optimized manner for each area within the analysis target area and for a set period. In particular, according to embodiments of the present invention, by calculating a failure prediction index for each of the wind and rain damage, lightning damage, and ice and snow damage, which are the biggest factors causing failures in power facilities, and then comprehensively considering them to calculate a comprehensive failure prediction index and a failure risk grade, the probability of failure in power facilities can be predicted more accurately. Brief explanation of the drawing

[0029] FIG. 1 is a block diagram showing the detailed configuration of a fault prediction system according to an embodiment of the present invention. FIG. 2 is an example illustrating the process of mutually mapping power facility failure data and weather forecast data according to the first embodiment of the present invention. FIG. 3 is an example illustrating the process of mutually mapping power facility failure data and weather forecast data according to the second embodiment of the present invention. FIG. 4 is an example showing the wind speed distribution at the time of failure of power equipment in each region according to an embodiment of the present invention. FIG. 5 is an example showing the wind distribution by wind speed grade for each region according to an embodiment of the present invention. FIG. 6 is an example showing the first failure prediction index for each wind speed grade in each region according to the first embodiment of the present invention. FIG. 7 is an example showing the first failure prediction index for each region and wind speed grade according to the second embodiment of the present invention. FIG. 8 is an example showing the distribution of precipitation at the time of failure of power facilities in each region according to an embodiment of the present invention. FIG. 9 is an example showing the second fault prediction index for each region and precipitation grade according to the first embodiment of the present invention. FIG. 10 is an example showing the second fault prediction index for each region and precipitation grade according to the second embodiment of the present invention. FIG. 11 is an example showing the third failure prediction index for each region and snow load grade according to the first embodiment. FIG. 12 is an example showing the third failure prediction index by snow load grade for each area according to the second embodiment. FIG. 13 is a flowchart illustrating a fault prediction method according to an embodiment of the present invention. FIG. 14 is a block diagram illustrating a computing environment including a computing device suitable for use in exemplary embodiments. Specific details for implementing the invention

[0030] Hereinafter, specific embodiments of the present invention will be described with reference to the drawings. The following detailed description is provided to facilitate a comprehensive understanding of the methods, apparatuses, and / or systems described herein. However, this is merely illustrative and the present invention is not limited thereto.

[0031] In describing the embodiments of the present invention, detailed descriptions of known technologies related to the present invention are omitted if it is determined that such detailed descriptions may unnecessarily obscure the essence of the present invention. Furthermore, the terms described below are defined in consideration of their functions within the present invention, and these may vary depending on the intentions or practices of the user or operator. Therefore, such definitions should be based on the content throughout this specification. Terms used in the detailed description are intended merely to describe the embodiments of the present invention and should not be limiting in any way. Unless explicitly stated otherwise, expressions in the singular form include the meaning of the plural form. In this description, expressions such as "include" or "comprise" are intended to refer to certain characteristics, numbers, steps, actions, elements, parts thereof, or combinations thereof, and should not be interpreted to exclude the existence or possibility of one or more other characteristics, numbers, steps, actions, elements, parts thereof, or combinations thereof other than those described.

[0033] FIG. 1 is a block diagram showing the detailed configuration of a power equipment failure prediction system (100) due to hazardous weather phenomena according to one embodiment of the present invention. As shown in FIG. 1, the failure prediction system (100) according to one embodiment of the present invention includes a data collection unit (102), a data preprocessing unit (104), a failure risk analysis unit (106), and a verification unit (108).

[0034] The data collection unit (102) collects power equipment failure data of the analysis target area and weather forecast data of the analysis target area. In these embodiments, the analysis target area refers to an area that is subject to the calculation of the comprehensive failure prediction index and failure risk grade described later through the analysis of power equipment failure data and weather forecast data. At this time, the analysis target area is divided into grids of a set resolution unit (e.g., a 5km resolution unit), and based on each divided grid, it can be grouped and classified into a set number of regions (e.g., a metropolitan area, a central inland area, a central coastal area, a central mountainous area, a southern inland area, a southern coastal area, a southern mountainous area, Jeju area, etc.). Subsequently, the analysis of power equipment failure data and weather forecast data can be performed on a region-by-region basis within the analysis target area. Here, the power equipment failure data may include the time of failure, location (or address), faulty equipment, cause of failure, etc., for the power equipment in the analysis target area. Power facilities are physically interconnected facilities for supplying electricity produced at a power plant to users, and may include, for example, power generation facilities, distribution facilities, transmission facilities, substation facilities, etc. The data collection unit (102) may, for example, collect power facility failure data for the analysis target area from a server (not shown) of Korea Electric Power Corporation. In addition, weather forecast data may include temperature, humidity, precipitation, wind speed, location information for each grid or area within the analysis target area for each time period. The data collection unit (102) may collect weather forecast data on an hourly basis from a server (not shown) of the Korea Meteorological Administration.

[0035] The data preprocessing unit (104) preprocesses power equipment failure data and weather forecast data collected from the data collection unit (102). The data preprocessing unit (104) can map power equipment failure data and weather forecast data to each other by the same time period and the same location. To this end, the data preprocessing unit (104) converts location information included in the power equipment failure data into coordinate information including latitude and longitude, and based on the converted coordinate information, maps the location information included in the power equipment failure data and the location information included in the weather forecast data to each other corresponding locations.

[0036] FIG. 2 is an example illustrating the process of mutually mapping power equipment failure data and weather forecast data according to the first embodiment of the present invention. FIG. 2(a) is a diagram showing each coordinate information within the power equipment failure data for a distribution facility, and FIG. 2(b) is a diagram showing grid points within the weather forecast data corresponding to each coordinate information within the power equipment failure data shown in FIG. 2(a).

[0037] FIG. 3 is an example illustrating the process of mutually mapping power equipment failure data and weather forecast data according to a second embodiment of the present invention. FIG. 3(a) is a diagram showing each coordinate information within power equipment failure data for transmission equipment, and FIG. 3(b) is a diagram showing grid points within weather forecast data corresponding to each coordinate information within power equipment failure data shown in FIG. 3(a).

[0038] Referring to FIGS. 2 and 3, the data preprocessing unit (104) can map the location information included in the power equipment failure data and the location information included in the weather forecast data to corresponding locations based on the converted coordinate information. Subsequently, the preprocessing unit (104) can assign the power equipment failure data and weather forecast data by time period to each grid point. As described above, the power equipment failure data may include the failure time, failure location, faulted equipment, and cause of failure for the power equipment in the analysis target area. At this time, the failure time may be expressed in the format, for example, yyyy (year)-mm (month)-dd (day)-HH:MM (hour). In addition, the failure location may be expressed as the unique grid number of the grid point within the weather forecast data. Furthermore, the faulted equipment may include information regarding the type of equipment where the failure occurred (e.g., transformer, wire, etc.) and identification number. In addition, the cause of failure may include information regarding set causes of failure such as wind and rain damage, lightning damage, and ice and snow damage.

[0039] Returning to Fig. 1, the failure risk analysis unit (106) calculates a failure prediction index for each failure cause for each area set within the target area, calculates a comprehensive failure prediction index for each area from the failure prediction index for each failure cause, and calculates a failure risk grade for each area based on the comprehensive failure prediction index.

[0040] To this end, the failure risk analysis unit (106) can extract power equipment failure data corresponding to a set number of failure causes from the power equipment failure data, and then analyze weather forecast data corresponding to the power equipment failure data for each failure cause of the extracted power equipment failure data. Here, the set number of failure causes may include wind and rain damage, lightning damage, and ice and snow damage. The failure causes for power equipment may be broadly classified into three types: wind and rain damage, lightning damage, and ice and snow damage. Accordingly, the failure risk analysis unit (106) can extract the first power equipment failure data corresponding to wind and rain, the second power equipment failure data corresponding to lightning, and the third power equipment failure data corresponding to ice and snow among the power equipment failure data, and then analyze the weather forecast data corresponding to the extracted first power equipment failure data, second power equipment failure data, and third power equipment failure data to calculate the first failure prediction index corresponding to wind and rain, the second failure prediction index corresponding to lightning, and the third failure prediction index corresponding to ice and snow for each area.

[0041] First, the failure risk analysis unit (106) can calculate a first failure prediction index by obtaining the wind distribution for each wind speed grade set for each area and then calculating the frequency of failures occurring when wind of the specific wind speed grade occurs relative to the frequency of wind of the specific wind speed grade. Generally, failures of power equipment due to wind and rain damage are mainly caused by tree collapse, wire breakage, and contact caused by strong winds. The most influential meteorological element in wind and rain damage is wind, and in particular, wind intensity, that is, wind speed, is important. Furthermore, wind is a meteorological phenomenon that occurs continuously, not a temporary one. Accordingly, the present invention enables the calculation of a first failure prediction index by calculating the frequency of failures occurring when wind of the specific wind speed grade occurs relative to the frequency of wind of the specific wind speed grade.

[0042] FIG. 4 is an example showing the wind speed distribution at the time of failure of power equipment in each region according to an embodiment of the present invention. FIG. 4(a) is an example showing the wind speed distribution at the time of failure of distribution equipment in each region, and FIG. 4(b) is an example showing the wind speed distribution at the time of failure of transmission equipment in each region.

[0043] Referring to Figure 4, it can be seen that the wind speed distribution at the time of failure of the power distribution and transmission facilities differs from each other in each area.

[0044] FIG. 5 is an example showing wind distribution by wind speed grade for each region according to an embodiment of the present invention. The eight distribution graphs shown in FIG. 5 represent wind distribution by wind speed grade for each region, for example, a metropolitan area (a), a central inland area (b), a central coastal area (c), a central mountainous area (d), a southern inland area (e), a southern coastal area (f), a southern mountainous area (g), and a Jeju area (h). Here, the wind speed grade can be expressed as a grade according to the Beaufort wind scale, for example, from grade 0 to grade 12.

[0045] Referring to Figure 5, it can be seen that in the metropolitan area (a), wind speed grades 1 and 2 each account for the largest proportion at approximately 35% to 40%, while in the central inland area (b), wind speed grade 1 accounts for the largest proportion at approximately 45%. In other words, it can be confirmed that the wind speed grades of normal winds differ for each area.

[0046] As described above, the failure risk analysis unit (106) can calculate a first failure prediction index by obtaining the wind distribution for each wind speed grade set for each area and then calculating the frequency of failures occurring when wind of the specific wind speed grade occurs relative to the frequency of wind of the specific wind speed grade. Here, the frequency of wind of the specific wind speed grade can be counted as once when wind of the specific wind speed grade occurs during a set time unit. In addition, the frequency of failures occurring when wind of the specific wind speed grade occurs may be the number of times a failure occurs when wind of the specific wind speed grade occurs during a set time unit. That is, the first failure prediction index can be calculated based on the premise that the probability of a failure in power equipment increases when wind of the wind speed at the time of failure occurs, through the frequency of wind of the wind speed at the time of failure relative to the frequency of wind blowing normally (i.e., background wind).

[0047] Specifically, the failure risk analysis unit (106) can calculate the first failure prediction index using the following mathematical formula 1. The frequency of failures occurring when wind of a specific wind speed class occurs and the frequency of wind of a specific wind speed class can be derived from the power equipment failure data and weather forecast data described above.

[0049] [Mathematical Formula 1]

[0050] 1st Fault Prediction Index (Wind and Rain Index)

[0051] = Frequency of failures when wind of a specific speed class occurs / Frequency of wind of a specific speed class

[0053] Here, the first fault prediction index can be expressed as 0 to 100. As the first fault prediction index approaches 100, it means that the probability of failure of power equipment due to wind and rain damage increases. This first fault prediction index can be calculated for each wind speed class.

[0054] FIG. 6 is an example showing the first fault prediction index for each wind speed class in each region according to the first embodiment of the present invention. The graphs of the eight first fault prediction indices for each wind speed class shown in FIG. 6 represent the first fault prediction index for power distribution facilities for each region, for example, a metropolitan area (a), a central inland area (b), a central coastal area (c), a central mountainous area (d), a southern inland area (e), a southern coastal area (f), a southern mountainous area (g), and a Jeju area (h).

[0055] Referring to Figure 6, it can be seen that in the metropolitan area (a), the first failure prediction index is 100 when the wind speed grade is 6 or higher, and in the central inland area (b), the first failure prediction index is 100 when the wind speed grade is 5 or higher. That is, in the metropolitan area (a), the probability of power equipment failure due to wind and rain damage is 100% when the wind speed grade is 6 or higher, and in the central inland area (b), the probability of power equipment failure due to wind and rain damage is 100% when the wind speed grade is 5 or higher.

[0056] FIG. 7 is an example showing the first fault prediction index for each wind speed class in each region according to the second embodiment of the present invention. The graphs of the eight first fault prediction indices for each wind speed class shown in FIG. 7 represent the first fault prediction index for transmission facilities for each region, for example, a metropolitan area (a), a central inland area (b), a central coastal area (c), a central mountainous area (d), a southern inland area (e), a southern coastal area (f), a southern mountainous area (g), and a Jeju area (h).

[0057] Referring to Figure 7, it can be seen that in the metropolitan area (a), the first failure prediction index is 100 when the wind speed grade is 6 or higher, and in the central inland area (b), the first failure prediction index is 100 when the wind speed grade is 5 or higher. In the metropolitan area (a), the probability of power equipment failure due to wind and rain damage is 100% when the wind speed grade is 6 or higher, and in the central inland area (b), the probability of power equipment failure due to wind and rain damage is 100% when the wind speed grade is 5 or higher.

[0058] As such, it can be confirmed that the first failure prediction index varies by region and by wind speed class.

[0059] Next, the failure risk analysis unit (106) can obtain the precipitation distribution by precipitation grade set for each area and calculate a second failure prediction index by comparing the frequency of failures occurring at a specific precipitation grade with the frequency of failures occurring due to lightning damage. Generally, failures of power facilities caused by lightning damage are mainly caused by damage to power facilities due to lightning strikes. Although weather forecast data generally does not include information on lightning strikes, precipitation phenomena are detected in more than 70% of accident data caused by lightning damage. In addition, a mutually linear relationship exists between lightning frequency and precipitation amount (Sheridan et al. (1997)), and lightning density and precipitation intensity have a high correlation with a correlation coefficient of 0.89 to 0.98 (Petrova et al. (2009)). Lightning frequency and precipitation intensity also have a strong positive correlation, and lightning can serve as an indicator of convective rainfall (Oh Seok-geun et al. (2010)). Accordingly, the present invention enables the calculation of a second failure prediction index based on the frequency of failures occurring at a specific precipitation grade relative to the frequency of failures occurring due to lightning damage. Here, precipitation grades can be classified according to precipitation amount as shown in Table 1 below. However, Table 1 is merely an example, and the numerical range of precipitation amount for each precipitation grade, the number of each precipitation grade, etc., are not specifically limited.

[0061] Precipitation grade Grade 1 (light rain) Grade 2 (moderate rain) Grade 3 (heavy rain) Grade 4 (very heavy rain) Hourly precipitation 1mm or more and less than 3mm 3mm or more and less than 15mm 15mm or more and less than 30mm 30mm or more

[0063] FIG. 8 is an example showing the precipitation distribution at the time of failure of power facilities in each region according to an embodiment of the present invention. FIG. 8 (a) is an example showing the precipitation distribution at the time of failure of distribution facilities in each region, and FIG. 8 (b) is an example showing the precipitation distribution at the time of failure of transmission facilities in each region.

[0064] Referring to Figure 8, it can be seen that the distribution of precipitation at the time of failure differs for each area and for each type of power equipment.

[0065] The failure risk analysis unit (106) can calculate a second failure prediction index through a classification of precipitation grades based on the linear relationship between the precipitation intensity and lightning frequency described above. Specifically, the failure risk analysis unit (106) can calculate the second failure prediction index using the following mathematical formula 2. The frequency of failures occurring at a specific precipitation grade and the frequency of failures occurring due to lightning damage can be derived from the power equipment failure data and weather forecast data described above.

[0067] [Mathematical Formula 2]

[0068] Second fault prediction index (lightning damage index)

[0069] = Frequency of failures at a specific precipitation grade / Frequency of failures caused by lightning strikes

[0071] Here, the second fault prediction index can be expressed as 0 to 100. As the second fault prediction index approaches 100, it means that the probability of failure of power facilities due to lightning damage increases. This second fault prediction index can be calculated for each precipitation grade.

[0072] FIG. 9 is an example showing the second fault prediction index for each region and precipitation grade according to the first embodiment of the present invention. The eight graphs of the second fault prediction index for each precipitation grade shown in FIG. 9 represent the second fault prediction index for power distribution facilities for each region, for example, a metropolitan area (a), a central inland area (b), a central coastal area (c), a central mountainous area (d), a southern inland area (e), a southern coastal area (f), a southern mountainous area (g), and a Jeju area (h).

[0073] Referring to Figure 9, it can be seen that in the metropolitan area (a), the central inland area (b), and the central coastal area (c), the second failure prediction index is 80 or higher when the precipitation grade is 2 or higher, and in the central mountainous area (d), the second failure prediction index is 80 or higher when the precipitation grade is 1 or higher. That is, in the metropolitan area (a), the central inland area (b), and the central coastal area (c), the probability of power equipment failure due to lightning strikes is 80% or higher when the precipitation grade is 2 or higher, and in the central mountainous area (d), the probability of power equipment failure due to lightning strikes is 80% or higher when the precipitation grade is 1 or higher.

[0074] FIG. 10 is an example showing the second fault prediction index for each region and precipitation grade according to the second embodiment of the present invention. The eight graphs of the second fault prediction index for each precipitation grade shown in FIG. 10 represent the second fault prediction index for transmission facilities for each region, for example, a metropolitan area (a), a central inland area (b), a central coastal area (c), a central mountainous area (d), a southern inland area (e), a southern coastal area (f), a southern mountainous area (g), and a Jeju area (h).

[0075] Referring to Figure 10, it can be seen that in the metropolitan area (a), the second failure prediction index is 80 or higher when the precipitation level is 3 or higher, and in the central inland area (b), the second failure prediction index is 80 or higher when the precipitation level is 2 or higher. That is, in the metropolitan area (a), the probability of power equipment failure due to lightning strikes is 80% or higher when the precipitation level is 3 or higher, and in the central inland area (b), the probability of power equipment failure due to lightning strikes is 80% or higher when the precipitation level is 2 or higher.

[0076] Next, the failure risk analysis unit (106) can obtain the distribution of snowfall amounts by snowfall grade set for each area and calculate a third failure prediction index by comparing the frequency of failures occurring in a specific snowfall grade to the frequency of failures occurring due to ice and snow damage. Generally, failures of power equipment due to ice and snow damage are mainly caused by contact and disconnection due to tree collapse, etc. However, weather forecast data does not include information on snowfall. Accordingly, in the present invention, the amount of snowfall is estimated according to the precipitation-snowfall conversion method (Baxter et al. (2005)), and then snowfall grades are classified according to the estimated amount of snowfall as shown in Table 2 below, and a third failure prediction index can be calculated by comparing the frequency of failures occurring in a specific snowfall grade to the frequency of failures occurring due to ice and snow damage. However, Table 2 is merely an example, and the numerical range of precipitation for each snowfall grade, the number of each snowfall grade, etc. are not specifically limited.

[0078] Snow depth grade Grade 0 (Eye) Grade 1 (Caution) Level 2 (Alert) Grade 3 (Alert (Mountain Area)) Hourly precipitation Less than 5cm 5cm or more and less than 20cm 20cm or more and less than 30cm 30cm or more

[0080] Specifically, the failure risk analysis unit (106) can calculate the third failure prediction index using the following mathematical formula 3. The frequency of failures occurring at a specific snow depth level and the frequency of failures occurring due to ice and snow damage can be derived from the aforementioned power equipment failure data and weather forecast data.

[0082] [Mathematical Formula 3]

[0083] 3rd Fault Prediction Index (Ice and Snow Index)

[0084] = Frequency of failures at a specific snow load grade / Frequency of failures caused by ice and snow damage

[0086] Here, the third fault prediction index can be expressed as 0 to 100. As the third fault prediction index approaches 100, it means that the probability of failure of power equipment due to ice and snow damage increases. This third fault prediction index can be calculated for each snow depth grade.

[0087] FIG. 11 is an example showing the third failure prediction index for each region and snow depth grade according to the first embodiment. The graph of the eight third failure prediction indices for each precipitation grade shown in FIG. 11 represents the second failure prediction index for power distribution facilities for each region, for example, metropolitan area (a), central inland area (b), central coastal area (c), central mountainous area (d), southern inland area (e), southern coastal area (f), southern mountainous area (g), and Jeju area (h).

[0088] Referring to Figure 11, it can be seen that in the metropolitan area (a), the third failure prediction index is approximately 65 when the snow level is 0, the third failure prediction index is approximately 72 when the snow level is 1, and the third failure prediction index is approximately 98 when the snow level is 2. This means that when the snow level is 0, the probability of power equipment failure due to ice and snow is approximately 65%, when the snow level is 1, the probability of power equipment failure due to ice and snow is approximately 72%, and when the snow level is 2, the probability of power equipment failure due to ice and snow is approximately 98%. In addition, it can be seen that in the central inland region (b), the third failure prediction index is approximately 50 when the snow depth is 0, the third failure prediction index is approximately 62 when the snow depth is 1, and the third failure prediction index is approximately 100 when the snow depth is 2. This means that when the snow depth is 0, the probability of power equipment failure due to ice and snow is approximately 50%, when the snow depth is 1, the probability of power equipment failure due to ice and snow is approximately 62%, and when the snow depth is 2, the probability of power equipment failure due to ice and snow is 100%.

[0089] FIG. 12 is an example showing the third failure prediction index for each region and snow depth grade according to the second embodiment. The graph of the eight third failure prediction indices for each precipitation grade shown in FIG. 12 represents the second failure prediction index for transmission facilities for each region, for example, metropolitan area (a), central inland area (b), central coastal area (c), central mountainous area (d), southern inland area (e), southern coastal area (f), and southern mountainous area (g), respectively.

[0090] Referring to Fig. 12, it can be seen that in the metropolitan area (a), the third failure prediction index is approximately 3 when the snow depth is 0, the third failure prediction index is approximately 20 when the snow depth is 1, and the third failure prediction index is approximately 100 when the snow depth is 2. Additionally, in the central inland area (b), the third failure prediction index is approximately 33 when the snow depth is 0, the third failure prediction index is approximately 45 when the snow depth is 1, and the third failure prediction index is approximately 100 when the snow depth is 2.

[0091] In this way, the failure risk analysis unit (106) can calculate the first failure prediction index, the second failure prediction index, and the third failure prediction index for each area set within the analysis target area through the method described above.

[0092] Subsequently, the failure risk analysis unit (106) can calculate a comprehensive failure prediction index for each area using the first failure prediction index, the second failure prediction index, and the third failure prediction index. Specifically, the failure risk analysis unit (106) can calculate a comprehensive failure prediction index for each area by assigning weighting coefficients set for the first failure prediction index, the second failure prediction index, and the third failure prediction index, and then summing the first failure prediction index, the second failure prediction index, and the third failure prediction index to which the weighting coefficients have been assigned.

[0093] As an example, the failure risk analysis unit (106) can calculate the overall failure prediction index for each area using the following mathematical formula 4.

[0095] [Mathematical Formula 4]

[0096] Comprehensive Failure Prediction Index = a * 1st Failure Prediction Index + b * 2nd Failure Prediction Index + c * 3rd Failure Prediction Index

[0097] (Here, a, b, and c represent the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient, respectively, and a + b + c = 1)

[0099] At this time, the failure risk analysis unit (106) can determine a first weighting coefficient, a second weighting coefficient, and a third weighting coefficient according to the failure occurrence rate of each failure cause for each area. As an example, the failure risk analysis unit (106) can determine the occurrence rate of each failure cause, namely the failure occurrence rate due to wind and rain, the failure occurrence rate due to lightning, and the failure occurrence rate due to ice and snow, for each period (e.g., monthly) set as shown in Table 3 below for the power distribution equipment, as the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient, respectively. The failure occurrence rate due to wind and rain, the failure occurrence rate due to lightning, and the failure occurrence rate due to ice and snow can be obtained from power equipment failure data. These first weighting coefficient, second weighting coefficient, and third weighting coefficient may differ for each set period and each area.

[0101] month Wind and rain sea Brain sea Ice and snow sea 1 0.55 0.07 0.38 2 0.43 0.16 0.41 3 0.50 0.26 0.23 4 0.74 0.25 0.00 5 0.46 0.54 0.00 6 0.29 0.71 0.00 7 0.41 0.59 0.00 8 0.75 0.25 0.00 9 0.91 0.09 0.00 10 0.89 0.11 0.00 11 0.53 0.39 0.08 12 0.63 0.09 0.28

[0103] In the above example, the failure risk analysis unit (106) can determine the failure rate due to wind and rain damage (0.55), the failure rate due to lightning damage (0.07), and the failure rate due to ice and snow damage (0.38) of the power distribution equipment during the month as the first weighting coefficient (0.55), the second weighting coefficient (0.07), and the third weighting coefficient (0.38), respectively.

[0104] Additionally, the failure risk analysis unit (106) can determine the failure rate (0.43) caused by wind and rain damage, the failure rate (0.16) caused by lightning damage, and the failure rate (0.41) caused by ice and snow damage during the two months of the power distribution facility as the first weighting coefficient (0.43), the second weighting coefficient (0.16), and the third weighting coefficient (0.41), respectively.

[0105] As another example, the failure risk analysis unit (106) can determine the occurrence rate of each cause of failure, namely the failure rate due to wind and rain, the failure rate due to lightning, and the failure rate due to ice and snow, for each period (e.g., monthly) set as shown in Table 4 below for the transmission equipment, as a first weighting coefficient, a second weighting coefficient, and a third weighting coefficient, respectively. These first weighting coefficient, second weighting coefficient, and third weighting coefficient may differ for each set period and each area.

[0107] month Wind and rain sea Brain sea Ice and snow sea 1 0.00 0.10 0.90 2 0.00 0.17 0.83 3 0.02 0.24 0.74 4 0.08 0.55 0.37 5 0.07 0.93 0.00 6 0.00 1.00 0.00 7 0.00 1.00 0.00 8 0.15 0.85 0.00 9 0.52 0.48 0.00 10 0.32 0.68 0.00 11 0.00 0.89 0.11 12 0.00 0.04 0.96

[0109] In the above example, the failure risk analysis unit (106) can determine the failure rate due to wind and rain damage (0.00), the failure rate due to lightning damage (0.10), and the failure rate due to ice and snow damage (0.90) of the transmission equipment during the month as the first weighting coefficient (0.00), the second weighting coefficient (0.10), and the third weighting coefficient (0.90), respectively.

[0110] Additionally, the failure risk analysis unit (106) can determine the failure rate due to wind and rain damage (0.00), the failure rate due to lightning damage (0.10), and the failure rate due to ice and snow damage (0.90) of the power distribution equipment during the two months as the first weighting coefficient (0.00), the second weighting coefficient (0.10), and the third weighting coefficient (0.90), respectively.

[0111] Afterwards, the failure risk analysis unit (106) can calculate the overall failure prediction index for each area using the mathematical formula 4 described above.

[0112] Additionally, the failure risk analysis unit (106) can calculate failure risk grades for each area according to the comprehensive failure prediction index. The failure risk analysis unit (106) can calculate multiple different failure risk grades (e.g., 4) according to the numerical range of the comprehensive failure prediction index as shown in Table 5 below. However, the numerical range of the comprehensive failure prediction index, the number of failure risk grades, and the classification criteria listed in Table 5 are merely examples and are not limited to those listed in Table 5.

[0114] Failure risk rating Grade 1 (Interest) Level 2 (Caution) Grade 3 (Border) Level 4 (Severe) Numerical range of the comprehensive failure prediction index 0 or more and less than 22 22 or more and less than 44 44 or more and less than 75 75 or more and 100 or less

[0116] Returning to FIG. 1, the verification unit (108) verifies the failure risk level for each area using a set verification model. In these embodiments, the verification model may be, for example, a confusion matrix. A confusion matrix is ​​a table that analyzes the number of actual values ​​and the number of predicted values ​​of a specific class, and can be expressed as shown in Table 6 below.

[0118] Predicted value Positive Negative Actual value Positive True Positive(TP) False Negative(FN) Accuracy(TP + TN) / (TP + TN + FP + FN) Negative False Positive (FP) True Negative(TN) CSITP / (TP + TN + FP)

[0120] Here, True Positive (TP) refers to cases where both the actual and predicted values ​​are positive, and True Negative (TN) refers to cases where both the actual and predicted values ​​are negative. Additionally, False Positive (FP) refers to cases where the predicted value is positive but the actual value is negative, and False Negative (FN) refers to cases where the predicted value is negative but the actual value is positive.

[0121] In addition, Accuracy (ACC) is calculated using the formula (TP + TN) / (TP + TN + FP + FN) and represents the probability that a prediction corresponds to reality.

[0122] In addition, the Critical Success Index (CSI) is calculated using the formula TP / (TP + TN + FP), and is defined as an index that excludes the category of a state where the prediction was negative but the actual result was positive (FN), representing the degree to which the event occurrence was correctly estimated among the estimated and observed data.

[0123] In the present invention, a verification was performed on the failure risk grade, i.e., the predicted value, calculated by applying a confusion matrix to the predicted value and the actual value of the failure risk grade of a specific facility (e.g., power distribution facility) during a specific period (e.g., 2021).

[0124] The results of verifying the confusion matrix for the interest grades of power distribution facilities are as shown in Table 7 below.

[0126] Predicted value Positive Negative Actual value Positive 136 49 Accuracy 0.75 Negative 14 63 CSI 0.68

[0128] In addition, the results of verifying the confusion matrix for the caution levels of power distribution facilities are as shown in Table 8 below.

[0130] Predicted value Positive Negative Actual value Positive 30 31 Accuracy 0.74 Negative 36 165 CSI 0.31

[0132] In addition, the results of verifying the confusion matrix for the boundary grades of the distribution facilities are as shown in Table 9 below.

[0134] Predicted value Positive Negative Actual value Positive 12 4 Accuracy 0.85 Negative 34 212 CSI 0.24

[0136] The higher the accuracy (ACC) and critical success index (CSI) of the confusion matrix, the higher the reliability of the calculated failure risk grade can be determined. Accordingly, the verification unit (108) can determine the reliability of each failure risk grade based on whether the sum of the accuracy (ACC) and critical success index (CSI) exceeds a reference value.

[0137] Specifically, the verification unit (108) applies a confusion matrix to each failure risk grade for a specific area, and if the number of failure risk grades for which the sum of Accuracy (ACC) and Critical Success Index (CSI) among each failure risk grade exceeds a threshold value exceeds a set number, the verification result of each failure risk grade for the specific area may be determined to be normal. At this time, the verification unit (108) may perform verification by applying a confusion matrix to each failure risk grade for each set period (e.g., monthly) for the specific area. As an example, the verification unit (108) may perform verification on the January failure risk grade of a metropolitan area and perform verification on the February failure risk grade of a metropolitan area. If the number of failure risk grades for which the sum of Accuracy (ACC) and Critical Success Index (CSI) exceeds a threshold value is less than or equal to a set number, the verification unit (108) may determine that the verification result of each failure risk grade for a specific period of the specific area is abnormal. In this case, the failure risk analysis unit (106) can recalculate the failure risk grade for the specific area by adjusting the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient described above.

[0138] At this time, the failure risk analysis unit (106) can select the area and period that are most similar to the wind speed pattern and precipitation pattern of the specific area that is judged to be abnormal among the areas judged to be normal by the verification result, and can recalculate the failure risk grade for the specific area based on the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient corresponding to the selected area and period. As an example, if the verification result for the failure risk grade for January in a metropolitan area is judged to be abnormal, the failure risk analysis unit (106) can select February in the southern inland area as the area and period that shows the pattern most similar to the wind speed pattern and precipitation pattern for January in the metropolitan area among the areas judged to be normal by the verification result. Subsequently, the failure risk analysis unit (106) can recalculate the failure risk grade for the specific area based on the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient corresponding to February in the southern inland area. For example, the failure risk analysis unit (106) can recalculate the failure risk grade for the specific area by changing the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient by 0.01, starting from the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient corresponding to February in the southern inland region. The verification unit (108) can verify the failure risk grade for the specific area recalculated in the same way as described above. Since verification may not be performed or the verification results may appear abnormal if there is insufficient case data, i.e., actual values, for verifying each failure risk grade, the present invention recalculates the failure risk grade for the specific area based on the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient of other areas and periods that exhibit similar wind speed and precipitation patterns.

[0139] In this way, the calculation and verification of failure risk grades for each area and set period are repeated, and accordingly, the reliability of the failure risk grades calculated by the failure risk analysis unit (106) can be improved.

[0140] FIG. 13 is a flowchart illustrating a fault prediction method according to an embodiment of the present invention. In the illustrated flowchart, the method is described by dividing it into a plurality of steps, but at least some of the steps may be performed in a different order, combined with other steps and performed together, omitted, divided into detailed steps, or performed with one or more steps not illustrated added.

[0141] In step S102, the data collection unit (102) collects power equipment failure data and weather forecast data for the area to be analyzed.

[0142] In step S104, the data preprocessing unit (104) maps power equipment failure data and weather forecast data to each other by time zone and location.

[0143] In step S106, the failure risk analysis unit (106) extracts power equipment failure data corresponding to multiple set failure causes from the power equipment failure data, and then analyzes the weather forecast data corresponding to the power equipment failure data for each failure cause of the extracted power equipment failure data, and calculates a failure prediction index for each failure cause for each set area within the analysis target area.

[0144] In step S108, the failure risk analysis unit (106) calculates a comprehensive failure prediction index for each area from the failure prediction index for each cause of failure.

[0145] In step S110, the failure risk analysis unit (106) calculates the failure risk grade for each area based on the comprehensive failure prediction index.

[0146] In step S112, the verification unit (108) verifies the failure risk level for each area using the established verification model.

[0147] FIG. 14 is a block diagram illustrating a computing environment including a computing device suitable for use in exemplary embodiments. In the illustrated embodiments, each component may have different functions and capabilities in addition to those described below, and may include additional components in addition to those not described below.

[0148] The illustrated computing environment (10) includes a computing device (12). In one embodiment, the computing device (12) may be a fault prediction system (100) or one or more components included in the fault prediction system (100).

[0149] The computing device (12) includes at least one processor (14), a computer-readable storage medium (16), and a communication bus (18). The processor (14) can cause the computing device (12) to operate according to the exemplary embodiment described above. For example, the processor (14) can execute one or more programs stored in the computer-readable storage medium (16). The one or more programs may include one or more computer-executable instructions, and the computer-executable instructions may be configured to cause the computing device (12) to perform operations according to the exemplary embodiment when executed by the processor (14).

[0150] A computer-readable storage medium (16) is configured to store computer-executable instructions or program code, program data and / or other suitable forms of information. A program (20) stored in the computer-readable storage medium (16) includes a set of instructions executable by a processor (14). In one embodiment, the computer-readable storage medium (16) may be memory (volatile memory such as random access memory, non-volatile memory, or a suitable combination thereof), one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, or other forms of storage media that are accessed by a computing device (12) and capable of storing desired information, or a suitable combination thereof.

[0151] The communication bus (18) interconnects various other components of the computing device (12), including the processor (14) and the computer-readable storage medium (16).

[0152] The computing device (12) may also include one or more input / output interfaces (22) and one or more network communication interfaces (26) that provide interfaces for one or more input / output devices (24). The input / output interfaces (22) and network communication interfaces (26) are connected to a communication bus (18). The input / output devices (24) may be connected to other components of the computing device (12) through the input / output interfaces (22). An exemplary input / output device (24) may include an input device such as a pointing device (such as a mouse or trackpad), a keyboard, a touch input device (such as a touchpad or touchscreen), a voice or sound input device, various types of sensor devices and / or imaging devices, and / or an output device such as a display device, a printer, a speaker and / or a network card. An exemplary input / output device (24) may be included inside the computing device (12) as a component constituting the computing device (12), or it may be connected to the computing device (12) as a separate device distinct from the computing device (12).

[0154] Although the present invention has been described in detail above through representative embodiments, those skilled in the art will understand that various modifications can be made to the aforementioned embodiments without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined by the claims set forth below as well as equivalents thereof. Explanation of the symbols

[0156] 100 : Fault Prediction System 102 : Data Collection Department 104 : Data Preprocessing Section 106 : Failure Risk Analysis Department 108 : Verification Department

Claims

Claim 1 A system for predicting failures of power facilities due to hazardous weather phenomena, comprising: a data collection unit that collects power facility failure data including failure time, failure location, failure facility, and failure cause for power facilities in an analysis target area, and weather forecast data for said analysis target area; a data preprocessing unit that mutually maps said power facility failure data and said weather forecast data by the same time period and same location; a failure risk analysis unit that extracts power facility failure data corresponding to a plurality of set failure causes from said power facility failure data, analyzes said weather forecast data corresponding to said power facility failure data by failure cause of said power facility failure data for each set area within said analysis target area, calculates a failure prediction index for each failure cause for said area from said failure prediction index for each failure cause, calculates a comprehensive failure prediction index for said area from said failure prediction index for said failure cause, and calculates a failure risk grade for said area according to said comprehensive failure prediction index; and a verification unit that verifies said failure risk grade for said area using a set verification model. Claim 2 A system for predicting failure of power equipment due to hazardous weather phenomena according to claim 1, wherein the failure risk analysis unit extracts, respectively, first power equipment failure data corresponding to wind and rain damage, second power equipment failure data corresponding to lightning damage, and third power equipment failure data corresponding to ice and snow damage from among the power equipment failure data, and then analyzes weather forecast data corresponding to the extracted first power equipment failure data, second power equipment failure data, and third power equipment failure data to calculate a first failure prediction index corresponding to wind and rain damage, a second failure prediction index corresponding to lightning damage, and a third failure prediction index corresponding to ice and snow damage for each area. Claim 3 A system for predicting failures of power equipment due to hazardous weather phenomena according to claim 2, wherein the failure risk analysis unit calculates the first failure prediction index by obtaining a wind distribution by wind speed grade set for each area and calculating the first failure prediction index by the frequency of failures occurring when wind of the specific wind speed grade occurs relative to the frequency of wind of the specific wind speed grade; calculates the second failure prediction index by obtaining a precipitation distribution by precipitation grade set for each area and calculating the second failure prediction index by the frequency of failures occurring in the specific precipitation grade relative to the frequency of failures occurring due to lightning damage; and calculates the third failure prediction index by obtaining a snowfall distribution by snowfall grade set for each area and calculating the third failure prediction index by the frequency of failures occurring in the specific snowfall grade relative to the frequency of failures occurring due to ice and snow damage. Claim 4 In claim 3, the fault risk analysis unit calculates a comprehensive fault prediction index for each area using the following mathematical formula, in a system for predicting failures of power equipment due to hazardous weather phenomena. [Mathematical Formula] Comprehensive fault prediction index = a * 1st fault prediction index + b * 2nd fault prediction index + c * 3rd fault prediction index (wherein a, b, and c represent the 1st weighting coefficient, 2nd weighting coefficient, and 3rd weighting coefficient, respectively, and a + b + c = 1) Claim 5 A system for predicting failure of power equipment due to hazardous weather phenomena according to claim 4, wherein the failure risk analysis unit determines the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient according to the failure occurrence rate of each failure cause for each area. Claim 6 A system for predicting failure of power equipment due to hazardous weather phenomena according to claim 1, wherein the verification unit calculates Accuracy (ACC) and Critical Success Index (CSI) by applying a confusion matrix to the failure risk grade for each area, and verifies the failure risk grade for each area based on the Accuracy and Critical Success Index. Claim 7 A step of collecting power equipment failure data including the time of failure, location of failure, equipment failure, and cause of failure for power equipment in the analysis target area, and weather forecast data for the analysis target area in the data collection unit; a step of mutually mapping the power equipment failure data and the weather forecast data by the same time period and same location in the data preprocessing unit; a step of calculating a failure prediction index for each cause of failure by each set area within the analysis target area by the failure risk analysis unit by each extracting power equipment failure data corresponding to a plurality of set causes of failure from the power equipment failure data, and then analyzing the weather forecast data corresponding to the power equipment failure data for each cause of failure of the extracted power equipment failure data; a step of calculating a comprehensive failure prediction index for each area from the failure prediction index for each cause of failure in the failure risk analysis unit; a step of calculating a failure risk grade for each area according to the comprehensive failure prediction index in the failure risk analysis unit. A method for predicting failure of power equipment due to hazardous weather phenomena, comprising the step of verifying the failure risk grade for each area using a set verification model in the verification section. Claim 8 A method for predicting failure of power equipment due to hazardous weather phenomena according to claim 7, wherein the step of calculating a failure prediction index for each cause of failure comprises extracting, respectively, a first power equipment failure data corresponding to wind and rain damage, a second power equipment failure data corresponding to lightning damage, and a third power equipment failure data corresponding to ice and snow damage from among the power equipment failure data, and then analyzing weather forecast data corresponding to the extracted first power equipment failure data, the second power equipment failure data, and the third power equipment failure data to calculate the first failure prediction index corresponding to wind and rain damage, the second failure prediction index corresponding to lightning damage, and the third failure prediction index corresponding to ice and snow damage for each area. Claim 9 A method for predicting failure of power equipment due to hazardous weather phenomena according to claim 8, wherein the step of calculating a failure prediction index for each cause of failure comprises: obtaining a wind distribution by wind speed grade set for each area and calculating the first failure prediction index through the frequency of failure occurring when wind of a specific wind speed grade occurs relative to the frequency of wind of a specific wind speed grade; obtaining a precipitation distribution by precipitation grade set for each area and calculating the second failure prediction index through the frequency of failure occurring in a specific precipitation grade relative to the frequency of failure occurring due to lightning damage; and obtaining a snowfall distribution by snowfall grade set for each area and calculating the third failure prediction index through the frequency of failure occurring in a specific snowfall grade relative to the frequency of failure occurring due to ice and snow damage. Claim 10 In claim 9, the step of calculating the comprehensive failure prediction index for each region is a method for predicting failure of power equipment due to hazardous weather phenomena, wherein the comprehensive failure prediction index for each region is calculated using the following mathematical formula. [Mathematical Formula] Comprehensive failure prediction index = a * 1st failure prediction index + b * 2nd failure prediction index + c * 3rd failure prediction index (wherein a, b, and c represent the 1st weighting coefficient, 2nd weighting coefficient, and 3rd weighting coefficient, respectively, and a + b + c = 1) Claim 11 A method for predicting failure of power equipment due to hazardous weather phenomena according to claim 10, wherein the step of calculating the comprehensive failure prediction index for each area determines the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient according to the failure occurrence rate of each failure cause for each area. Claim 12 A method for predicting failure of power equipment due to hazardous weather phenomena according to claim 7, wherein the step of verifying the failure risk grade for each area comprises applying a confusion matrix to the failure risk grade for each area to calculate Accuracy (ACC) and Critical Success Index (CSI), and verifying the failure risk grade for each area based on the Accuracy and Critical Success Index.

Images