CN118014405B - Power failure early warning system based on data processing - Google Patents
Power failure early warning system based on data processing Download PDFInfo
- Publication number
- CN118014405B CN118014405B CN202410417279.7A CN202410417279A CN118014405B CN 118014405 B CN118014405 B CN 118014405B CN 202410417279 A CN202410417279 A CN 202410417279A CN 118014405 B CN118014405 B CN 118014405B
- Authority
- CN
- China
- Prior art keywords
- fault
- prediction value
- power failure
- natural
- power
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012545 processing Methods 0.000 title claims abstract description 37
- 238000004458 analytical method Methods 0.000 claims description 66
- 238000000034 method Methods 0.000 claims description 25
- 238000012937 correction Methods 0.000 claims description 15
- 238000003491 array Methods 0.000 claims description 9
- 238000012544 monitoring process Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 4
- 241000764238 Isis Species 0.000 claims description 3
- 229910010293 ceramic material Inorganic materials 0.000 claims description 3
- 239000002131 composite material Substances 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 6
- 238000007405 data analysis Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Abstract
The invention relates to the technical field of fault early warning, in particular to a power fault early warning system based on data processing. The system comprises a power data acquisition module, an index data processing module, a fault prediction module and a fault early warning module. According to the invention, the historical power failure data, the historical natural weather data and the real-time natural weather data corresponding to all areas are obtained and are respectively preprocessed; according to the historical power failure data and the historical natural weather data corresponding to all areas, the natural failure correlation coefficient, the power failure occurrence index and the comprehensive occurrence index of the power failure are respectively analyzed, then the first failure prediction value, the second failure prediction value and the third failure prediction value are analyzed, finally the first failure prediction value, the second failure prediction value and the third failure prediction value are comprehensively analyzed, and failure early warning is sent out, so that the accurate prediction of the power failure is improved, and the problem that the prediction of the power failure in the prior art is inaccurate is solved.
Description
Technical Field
The invention relates to the technical field of fault early warning, in particular to a power fault early warning system based on data processing.
Background
The power failure early warning is a technology for detecting potential problems in advance and giving an alarm by monitoring the operation state of a power system. By using sensors, monitoring equipment and advanced data analysis techniques, the power failure warning system can identify abnormal conditions, such as equipment failure, voltage fluctuation or load abnormality, so as to prevent possible system failure and improve the reliability and stability of the power system.
Implementation of power failure warning systems typically involves data acquisition, sensor technology, real-time monitoring, and big data analysis. The sensors acquire real-time data of the power system operating conditions and the data analysis algorithms evaluate the data to identify potential signs of failure. By setting the threshold and the rule, the system triggers an alarm once an abnormality is detected, so that operation and maintenance personnel can take measures in time to prevent or alleviate potential power faults.
The invention discloses a power system fault early warning method, a device, a terminal device and a storage medium of China patent application CN113988325A, which comprises the following steps: extracting measurement point data of each measurement point in the power system in the current period to obtain each operation state parameter of the power system, and dividing all operation state parameters into two types to obtain a characteristic row operation state parameter and a tag row operation state parameter; inputting the characteristic string running state parameters into a trained prediction model so that the prediction model generates corresponding prediction tag string running state parameters according to the characteristic string running state parameters; and calculating a parameter difference value between the running state parameter of the tag array and the running state parameter of the predicted tag array, and then carrying out fault early warning according to the parameter difference value.
The invention patent application CN113554526A discloses a fault early warning method, a device, a storage medium and a processor of power equipment, comprising the following steps: acquiring real-time data of the power equipment in a preset working time period, wherein the real-time data comprises: working data of various working parameters; generating a real-time operating profile of the electrical device based on real-time data within a predetermined operating time period; comparing a real-time working curve graph of the power equipment with a standard curve graph to obtain a difference value change, wherein the standard curve graph is used for representing a working trend graph of the power equipment in a fault working state; and generating early warning information according to the difference value change and the set early warning threshold value.
However, in the process of implementing the technical scheme of the embodiment of the application, the application discovers that the above technology has at least the following technical problems:
In the prior art, the selected characteristics are insufficient to comprehensively reflect the working state of the power system, so that the problem of inaccurate prediction of power faults exists.
Disclosure of Invention
The embodiment of the application solves the problem that the prediction of the power failure is inaccurate in the prior art by providing the power failure early warning system based on data processing, and realizes the improvement of the accurate prediction of the power failure.
The embodiment of the application provides a power failure early warning system based on data processing, which comprises the following steps: the system comprises a power data acquisition module, an index data processing module, a fault prediction module and a fault early warning module; the power data acquisition module is used for dividing a plurality of power data corresponding areas, acquiring historical power fault data, historical natural weather data, current observed power time information and real-time natural weather data corresponding to all the areas, and respectively preprocessing the historical power fault data, the historical natural weather data, the current observed power time information and the real-time natural weather data; the index data processing module is used for respectively analyzing natural fault correlation coefficients, power fault occurrence indexes and power fault comprehensive occurrence indexes according to historical power fault data and historical natural meteorological data corresponding to all areas; the fault prediction module is used for analyzing and obtaining a first fault prediction value, a second fault prediction value and a third fault prediction value according to the current observed power time information, the real-time natural meteorological data, the natural fault correlation coefficient, the power fault occurrence index and the power fault comprehensive occurrence index corresponding to all the areas; the fault early warning module is used for comprehensively analyzing the first fault prediction value, the second fault prediction value and the third fault prediction value and sending out fault early warning.
Further, the index data processing module comprises a natural fault correlation coefficient analysis unit, a power fault occurrence index analysis unit and a power fault comprehensive occurrence index analysis unit; the natural fault correlation coefficient analysis unit: the correlation between the power failure type data and the natural weather type data is analyzed according to the historical power failure data and the historical natural weather data, and a natural failure correlation coefficient is obtained; the power failure occurrence index analysis unit: the power failure generation index is used for analyzing the power failure generation index according to the historical power failure data; the comprehensive power failure occurrence index analysis unit comprises: the method comprises the steps of comprehensively analyzing the relevance between power failure type data, power failure occurrence time information and natural weather type data according to historical power failure data and historical natural weather data to obtain a power failure comprehensive occurrence index; the historical power failure data includes: power failure type data, power failure occurrence time information; the historical natural weather data includes: natural weather type data, natural weather occurrence time information.
Further, the fault prediction module comprises a first fault prediction value analysis unit, a second fault prediction value analysis unit and a third fault prediction value analysis unit; the first failure prediction value analysis unit: the method comprises the steps of obtaining a first fault prediction value according to real-time natural meteorological data and natural fault correlation coefficients, outputting a result if the first fault prediction value exceeds a first threshold value, and outputting a fault-free result if the first fault prediction value exceeds the first threshold value; the second failure prediction value analysis unit: the method comprises the steps of obtaining a second fault prediction value according to current observation power time information and a fault occurrence prediction model, outputting a result if the second fault prediction value exceeds a second threshold value, and outputting a fault-free result if the second fault prediction value exceeds the second threshold value; the third failure prediction value analysis unit: and the method is used for obtaining a third fault prediction value according to the real-time natural meteorological data of the area and the current observed power time information, outputting a result if the third fault prediction value exceeds a third threshold value, and outputting a fault-free result if the third fault prediction value exceeds the third threshold value.
Further, the comprehensive analysis method of the fault early warning module comprises the following steps: sending out fault early warning according to the power fault prediction result: if the first fault prediction value, the second fault prediction value and the third fault prediction value are all output to have faults, determining to send out fault early warning; if and only if one output has no fault, sending out fault early warning according to the result of the third fault prediction value; if the third predicted value result is faulty, determining that fault early warning is sent out, and if the third predicted value result is non-faulty, determining that no early warning is sent out, namely: if the first fault prediction value and the second fault prediction value are both output with faults and the third fault prediction value is output without faults, determining that no early warning is performed; if the first fault prediction value and the third fault prediction value are both output with faults and the second fault prediction value is output without faults, determining to send out fault early warning; if the second fault prediction value and the third fault prediction value are both output with faults and the first fault prediction value is output without faults, determining to send out fault early warning; if no fault exists in two or more outputs among the three, determining that no early warning exists, namely: if the first fault prediction value, the second fault prediction value and the third fault prediction value are all output to be fault-free, determining that no early warning is performed; if the first fault prediction value and the second fault prediction value are output without faults and the third fault prediction value is output with faults, determining that early warning is not performed; if the first fault prediction value and the third fault prediction value are output without faults and the second fault prediction value is output with faults, determining that early warning is not performed; if the second fault prediction value and the third fault prediction value are output without faults and the first fault prediction value is output with faults, the early warning is determined not to be performed.
Further, the specific analysis method of the natural fault correlation coefficient comprises the following steps: extracting power failure type data and power failure occurrence time information from the historical power failure data; extracting natural weather type data and natural weather occurrence time information from historical natural weather data; according to the power failure occurrence time information and the natural weather occurrence time information, natural weather type data matched with the power failure occurrence time are found out; acquiring power failure type data corresponding to the power failure occurrence time information, constructing a natural failure correlation coefficient formula, and calculating natural correlation coefficients between the power failure types and the natural meteorological types in all areas according to the natural failure correlation coefficient formula; numbering the regions asRecording，Is the total number of regions; the power failure type is numberedRecording，Is the total number of power failure types, and will beThe power failure type data corresponding to each region is recorded as a power failure array; Numbering natural weather types asRecording，Is the total number of natural weather types and will beThe natural weather type data corresponding to each region is recorded as a natural weather array; And calculating a natural fault correlation coefficient of the total area according to a natural fault correlation coefficient formula, wherein the natural fault correlation coefficient formula is as follows: Wherein, the method comprises the steps of, wherein, For natural fault correlation coefficients between the power fault types and the natural weather types in all areas,Is the standard deviation of the power failure array,Is the standard deviation of the natural weather array,Is a correction factor for the natural fault correlation coefficient.
Further, the specific analysis method of the first fault prediction value is as follows: acquiring the real-time natural weather type with the largest proportion in all the current areas, and recording the real-time natural weather type as; Extracting a natural fault correlation coefficient, constructing a first fault prediction value formula, and calculating a first fault prediction value according to the first fault prediction value formula; the first failure prediction value formula is: Wherein, the method comprises the steps of, wherein, As a result of the first failure prediction value,Is the firstThe overall mean value of the natural weather array of each region,Is the firstThe overall average of the individual regional power failure arrays,Is a correction factor for the first failure prediction value.
Further, the specific analysis method of the power failure occurrence index comprises the following steps: setting the cycle period of power failure monitoring asAnd numbering the cycle period as，，Dividing the cycle period equally intoThe number of observation periods isIs marked as，To at the firstA total number of observation periods in each cycle period; acquiring power failure occurrence time information in historical failure data corresponding to each observation period in each cycle period; constructing a power failure occurrence index formula, and calculating a power failure occurrence index; the power failure occurrence index formula is: Wherein, the method comprises the steps of, wherein, As an index of the occurrence of a power failure,Is the firstArea numberThe number of power failures occurring within a single observation period,Is a natural constant which is used for the production of the high-temperature-resistant ceramic material,Is a correction factor for the power failure occurrence index.
Further, the specific analysis method of the second fault prediction value is as follows: constructing a fault occurrence prediction model based on a model of a time sequence according to the power fault occurrence index, the cycle period, the observation period and the power fault occurrence time information; inputting the current observed power time information into a fault occurrence prediction model to obtain a second fault prediction value。
Further, the specific analysis method of the comprehensive occurrence index of the power failure comprises the following steps: acquiring historical power failure data, historical natural weather data and power failure occurrence time information corresponding to the area; the power failure occurrence time information is numbered asRecording，Is the total amount of the power failure occurrence time information and will beThe power failure occurrence time information corresponding to each area is recorded as a failure time array; According to the power failure arrayAnd natural weather array; Constructing a comprehensive power failure occurrence index formula, and calculating a comprehensive power failure occurrence index; the comprehensive occurrence index formula of the power failure is as follows: In which, in the process, Is the firstThe power failure composite occurrence index of each region,And the correction factor of the comprehensive occurrence index of the power failure.
Further, the specific analysis method of the third fault prediction value is as follows: acquiring real-time natural weather type data in a current area, and recording the real-time natural weather type data as; Acquiring current observed power time information in a current area, and recording as; Extracting an integrated occurrence index of the power failure, constructing a third failure prediction value formula, and calculating a third failure prediction value according to the third failure prediction value formula; the third failure prediction value formula is: Wherein, the method comprises the steps of, wherein, As a result of the third failure prediction value,Is the firstThe overall mean value of the natural weather array of each region,Is the firstThe overall average of the individual regional power failure arrays,Is the firstThe overall mean of the individual zone time to failure arrays,Is a correction factor for the third failure prediction value.
One or more technical solutions provided in the embodiments of the present application at least have the following technical effects or advantages:
1. The method comprises the steps of respectively preprocessing historical power failure data, historical natural weather data and real-time natural weather data corresponding to all areas by acquiring the historical power failure data, the historical natural weather data and the real-time natural weather data; according to the historical power failure data and the historical natural weather data corresponding to all areas, the natural failure correlation coefficient, the power failure occurrence index and the comprehensive occurrence index of the power failure are respectively analyzed, and then the first failure prediction value, the second failure prediction value and the third failure prediction value are analyzed, so that the first failure prediction value, the second failure prediction value and the third failure prediction value are comprehensively analyzed, failure early warning is sent out, further the accurate prediction of the power failure is improved, and the problem that the prediction of the power failure in the prior art is inaccurate is effectively solved.
2. By analyzing the occurrence rules of the power faults at different times in one year, time factors are fully considered, so that the rules of seasonality, periodicity and the like can be found, and the prediction capability of faults is improved.
3. By adopting multi-level fault prediction and early warning strategies, the first, second and third fault prediction values are analyzed, and different early warning strategies are implemented, so that the accuracy of judging the fault occurrence probability is improved.
Drawings
Fig. 1 is a schematic structural diagram of a power failure early warning system based on data processing according to an embodiment of the present application;
Fig. 2 is a schematic structural diagram of a finger data processing module in the power failure early warning system based on data processing according to an embodiment of the present application;
Fig. 3 is a schematic structural diagram of a fault prediction module in a power fault early warning system based on data processing according to an embodiment of the present application.
Detailed Description
The embodiment of the application solves the problem that the prediction of the power failure is inaccurate in the prior art by providing the power failure early warning system based on data processing, and improves the accurate prediction of the power failure by comprehensively analyzing the first failure prediction value, the second failure prediction value and the third failure prediction value and sending out failure early warning.
The technical scheme in the embodiment of the application aims to solve the problem that the prediction of the power failure is not accurate enough, and the overall thought is as follows: reading the jurisdiction range of the power failure early warning system, dividing the jurisdiction range into a plurality of areas, and numbering each area; acquiring all historical power failure data in a range and historical natural meteorological data at the same time, and respectively preprocessing the two types of data; according to the historical power failure data and the historical natural weather data at the same time, analyzing a failure correlation coefficient between the power failure and the natural weather; acquiring and preprocessing real-time natural weather data, obtaining a first fault prediction value according to the real-time natural weather data and a fault correlation coefficient, outputting a result if the first fault prediction value exceeds a first threshold value, otherwise outputting a fault-free result; according to the historical fault data and the corresponding time, analyzing the power fault occurrence indexes at different times in one year, and constructing a fault occurrence prediction model; acquiring the current time and preprocessing, obtaining a second fault prediction value according to a fault occurrence prediction model, outputting a result if the second fault prediction value exceeds a second threshold value, and outputting no fault if the second fault prediction value exceeds the second threshold value; analyzing the comprehensive occurrence index of the power faults under different natural weather in different time in one year according to the relation among the historical power fault data, the historical natural weather data and the fault occurrence time information corresponding to the region, obtaining a third fault prediction value according to the real-time natural weather data and the current observed power time information of the region, outputting a result to have faults if the third fault prediction value exceeds a third threshold value, otherwise outputting no faults; if the third fault prediction value and the other two fault prediction values are out of order, determining to send out fault early warning; if one of the three fault prediction values is fault-free, a fault early warning is sent out according to the result of the third fault prediction value: if the third predicted value result is faulty, determining that fault early warning is sent out, and if the third predicted value result is non-faulty, determining that no early warning is sent out; if two or more than two of the three fault prediction values are fault-free, the early warning is determined not to be performed, and the accuracy of predicting the power faults is improved.
In order to better understand the above technical solutions, the following detailed description will refer to the accompanying drawings and specific embodiments.
As shown in fig. 1, a schematic structural diagram of a power failure early warning system based on data processing according to an embodiment of the present application is shown, where the power failure early warning system based on data processing according to the embodiment of the present application includes: the system comprises a power data acquisition module, an index data processing module, a fault prediction module and a fault early warning module; the power data acquisition module is used for reading the range governed by the power failure early warning system, dividing the range into a plurality of power data corresponding areas, setting area numbers, acquiring all historical power failure data in the range and historical natural meteorological data under the same time from a block chain, acquiring real-time natural meteorological data and current observed power time information, and respectively preprocessing the real-time natural meteorological data and the current observed power time information; the system comprises an index data processing module, a fault occurrence prediction module and a power failure analysis module, wherein the index data processing module is used for analyzing a natural fault correlation coefficient between a power fault and natural weather according to historical power fault data and historical natural weather data under the same time, analyzing power fault occurrence indexes under different times in one year according to the historical fault data and corresponding occurrence time information, constructing a fault occurrence prediction model, and analyzing power fault comprehensive occurrence indexes under different natural weather in different times in one year according to the relations among the historical power fault data, the historical natural weather data and the fault occurrence time information corresponding to the region; the fault prediction module is used for analyzing and obtaining a first fault prediction value, a second fault prediction value and a third fault prediction value according to the current observed power time information, the real-time natural meteorological data, the natural fault correlation coefficient, the power fault occurrence index and the power fault comprehensive occurrence index corresponding to all the areas: obtaining a first fault prediction value according to the real-time natural weather data and the fault correlation coefficient, outputting a result if the first fault prediction value exceeds a first threshold value, otherwise outputting a fault-free value, obtaining a second fault prediction value according to the current observed power time information and a fault occurrence prediction model, outputting a result if the second fault prediction value exceeds a second threshold value, otherwise outputting a fault-free value, obtaining a third fault prediction value according to the regional real-time natural weather data and the current observed power time information, outputting a result if the third fault prediction value exceeds a third threshold value, otherwise outputting a fault-free value; the fault early warning module is used for comprehensively analyzing the first fault predicted value, the second fault predicted value and the third fault predicted value and sending out fault early warning: if the third fault prediction value and the other two fault prediction values are out of order, determining to send out fault early warning; if one of the three fault prediction values is fault-free, a fault early warning is sent out according to the result of the third fault prediction value: if the third predicted value result is faulty, determining that fault early warning is sent out, and if the third predicted value result is non-faulty, determining that no early warning is sent out; if two or more of the three fault prediction values obtain no faults, determining that no early warning is performed.
Further, as shown in fig. 2, the structure diagram of the index data processing module in the data processing-based power failure early warning system according to the embodiment of the present application is shown, where the index data processing module includes a natural failure correlation coefficient analysis unit, a power failure occurrence index analysis unit, and a power failure comprehensive occurrence index analysis unit; the natural fault correlation coefficient analysis unit is used for analyzing the correlation between the power fault type data and the natural weather type data according to the historical power fault data and the historical natural weather data to obtain a natural fault correlation coefficient; the power failure occurrence index analysis unit is used for analyzing the power failure occurrence index according to the historical power failure data; the power failure comprehensive occurrence index analysis unit is used for comprehensively analyzing the relevance between the power failure type data, the power failure occurrence time information and the natural weather type data according to the historical power failure data and the historical natural weather data to obtain a power failure comprehensive occurrence index; the historical power failure data comprises power failure type data and power failure occurrence time information; the historical natural weather data includes natural weather type data, natural weather occurrence time information.
In the embodiment, quantitative evaluation of the relationship between the power failure and the natural weather is helpful for understanding the influence of environmental factors on the stability of the power system. Faults that may occur in a power system under certain natural weather conditions can be more accurately predicted, helping to take preventive measures to reduce the risk of power faults. According to the historical power failure data, the power failure occurrence index is analyzed, the trend and probability of power system failure under different time are reflected, weak links possibly existing in the power system can be recognized, potential power failure risks can be recognized more effectively in real-time monitoring, and the real-time grasping capability of the power system stability is improved. And analyzing the comprehensive occurrence index of the power faults according to the historical power fault data and the historical natural weather data, and comprehensively considering the influence of natural weather factors.
Further, as shown in fig. 3, a schematic structural diagram of a fault prediction module in a power fault early warning system based on data processing according to an embodiment of the present application is shown, where the fault prediction module includes a first fault prediction value analysis unit, a second fault prediction value analysis unit, and a third fault prediction value analysis unit; the first fault prediction value analysis unit is used for obtaining a first fault prediction value according to real-time natural meteorological data and natural fault correlation coefficients, outputting a result if the first fault prediction value exceeds a first threshold value, and outputting no fault if the first fault prediction value exceeds the first threshold value; the second fault prediction value analysis unit is used for obtaining a second fault prediction value according to the current observed power time information and the fault occurrence prediction model, outputting a result if the second fault prediction value exceeds a second threshold value, and outputting a fault-free result if the second fault prediction value exceeds the second threshold value; the third fault prediction value analysis unit is used for obtaining a third fault prediction value according to the real-time natural meteorological data of the area and the current observed power time information, outputting a result if the third fault prediction value exceeds a third threshold value, and outputting no fault if the third fault prediction value exceeds the third threshold value.
In this embodiment, the first fault prediction value analysis unit rapidly generates the first fault prediction value through real-time natural weather data and the fault correlation coefficient, so as to immediately determine whether a potential fault risk exists in the current environment. The second fault prediction value analysis unit is used for providing fault probability assessment for specific time by combining the current observed power time information and the fault occurrence prediction model, and providing more detailed fault prediction information of time dimension for system operators. The third fault prediction value analysis unit is used for generating a third fault prediction value by combining real-time natural meteorological data and current observed power time information in consideration of regional differences and providing customized assessment for fault risks of different regions.
Further, the comprehensive analysis method of the fault early warning module comprises the following steps: sending out fault early warning according to the power fault prediction result: if the first fault prediction value, the second fault prediction value and the third fault prediction value are all output to have faults, determining to send out fault early warning; if and only if one output has no fault, sending out fault early warning according to the result of the third fault prediction value; if the third predicted value result is faulty, determining that fault early warning is sent out, and if the third predicted value result is non-faulty, determining that no early warning is sent out, namely: if the first fault prediction value and the second fault prediction value are both output with faults and the third fault prediction value is output without faults, determining that no early warning is performed; if the first fault prediction value and the third fault prediction value are both output with faults and the second fault prediction value is output without faults, determining to send out fault early warning; if the second fault prediction value and the third fault prediction value are both output with faults and the first fault prediction value is output without faults, determining to send out fault early warning; if no fault exists in two or more outputs among the three, determining that no early warning exists, namely: if the first fault prediction value, the second fault prediction value and the third fault prediction value are all output to be fault-free, determining that no early warning is performed; if the first fault prediction value and the second fault prediction value are output without faults and the third fault prediction value is output with faults, determining that early warning is not performed; if the first fault prediction value and the third fault prediction value are output without faults and the second fault prediction value is output with faults, determining that early warning is not performed; if the second fault prediction value and the third fault prediction value are output without faults and the first fault prediction value is output with faults, the early warning is determined not to be performed.
Further, the specific analysis method of the natural fault correlation coefficient comprises the following steps: extracting power failure type data and power failure occurrence time information from the historical power failure data; extracting natural weather type data and natural weather occurrence time information from historical natural weather data; according to the power failure occurrence time information and the natural weather occurrence time information, natural weather type data matched with the power failure occurrence time are found out; acquiring power failure type data corresponding to the power failure occurrence time information, constructing a natural failure correlation coefficient formula, and calculating natural correlation coefficients between the power failure types and the natural meteorological types in all areas according to the natural failure correlation coefficient formula; numbering the regions asRecording，Is the total number of regions; the power failure type is numberedRecording，Is the total number of power failure types, and will beThe power failure type data corresponding to each region is recorded as a power failure array; Numbering natural weather types asRecording，Is the total number of natural weather types and will beThe natural weather type data corresponding to each region is recorded as a natural weather array; And according to the natural fault correlation coefficient formula, calculating the natural fault correlation coefficient of the total area, wherein the natural fault correlation coefficient formula is as follows: Wherein, the method comprises the steps of, wherein, For natural fault correlation coefficients between the power fault types and the natural weather types in all areas,Is the standard deviation of the power failure array,Is the standard deviation of the natural weather array,Is a correction factor for the natural fault correlation coefficient.
In the embodiment, comprehensive monitoring and timely response to the power system are realized by combining a fault prediction value analysis result of the fault prediction module and a fault early warning comprehensive analysis method of the fault early warning module. The failure prediction value analysis of the failure prediction module is responsible for comprehensively judging failure prediction results according to the output of the first, second and third failure prediction values, if any one of the prediction values exceeds a set threshold value, outputting that the failure exists, otherwise, outputting that the failure does not exist. And the comprehensive fault early warning analysis method of the fault early warning module is used for making a strategy for sending out fault early warning based on the prediction results. If all three output faults, the fault early warning is sent out clearly, and the operation and maintenance personnel are informed to take necessary measures. If only one predicted value is output without faults, judging whether fault early warning is sent out according to the result of the third fault predicted value, and helping to avoid excessive alarming. If two or more outputs have no faults, no early warning is sent out clearly, and the alarm system is ensured to be more accurate and reliable.
Further, the specific analysis method of the first fault prediction value is as follows: acquiring the real-time natural weather type with the largest proportion in all the current areas, and recording the real-time natural weather type as; Extracting a natural fault correlation coefficient, constructing a first fault prediction value formula, and calculating a first fault prediction value according to the first fault prediction value formula; the first failure prediction value formula is: Wherein, the method comprises the steps of, wherein, As a result of the first failure prediction value,Is the firstThe overall mean value of the natural weather array of each region,Is the firstThe overall average of the individual regional power failure arrays,Is a correction factor for the first failure prediction value.
In this embodiment, the pearson correlation coefficient provides a quantitative method by measuring the linear correlation between the type of power failure and the type of natural weather, reflecting the direction and strength of the trend between them. The relation between the power system faults and natural meteorological factors is effectively revealed, and scientific basis is provided for fault prediction of the system.
Further, the specific analysis method of the power failure occurrence index is as follows: setting the cycle period of power failure monitoring asAnd numbering the cycle period as，，Dividing the cycle period equally intoThe number of observation periods isIs marked as，To at the firstA total number of observation periods in each cycle period; acquiring power failure occurrence time information in historical failure data corresponding to each observation period in each cycle period; constructing a power failure occurrence index formula, and calculating a power failure occurrence index; the power failure occurrence index formula is: Wherein, the method comprises the steps of, wherein, As an index of the occurrence of a power failure,Is the firstArea numberThe number of power failures occurring within a single observation period,Is a natural constant which is used for the production of the high-temperature-resistant ceramic material,Is a correction factor for the power failure occurrence index.
In the present embodiment, the power failure occurrence index represents a ratio of the frequency of occurrence of power failure to the total number of observations within a specific time period.The ratio indicates the relative frequency of occurrence of faults during the observation period during the cycle. If this value is close to 1, this indicates that the frequency of occurrence during the observation period is relatively high. If approaching 0, this indicates that relatively little occurs.
Further, the specific analysis method of the second fault prediction value is as follows: constructing a fault occurrence prediction model based on a model of a time sequence according to the power fault occurrence index, the cycle period, the observation period and the power fault occurrence time information; inputting the current observed power time information into a fault occurrence prediction model to obtain a second fault prediction value。
In this embodiment, a time series based model, such as ARIMA, prophet, is used to predict future power failure occurrence index based on time information.
Further, the specific analysis method of the comprehensive occurrence index of the power failure comprises the following steps: acquiring historical power failure data, historical natural weather data and power failure occurrence time information corresponding to the area; the power failure occurrence time information is numbered asRecording，Is the total amount of the power failure occurrence time information and will beThe power failure occurrence time information corresponding to each area is recorded as a failure time array; According to the power failure arrayAnd natural weather array; Constructing a comprehensive power failure occurrence index formula, and calculating a comprehensive power failure occurrence index; the comprehensive occurrence index formula of the power failure is as follows: In which, in the process, Is the firstThe power failure composite occurrence index of each region,And the correction factor of the comprehensive occurrence index of the power failure.
In the embodiment, the comprehensive power failure occurrence index analysis method adopting the pearson correlation coefficient can comprehensively evaluate the relation among the historical power failure data, the historical natural weather data and the power failure occurrence time information, provide scientific decision support for system operation and maintenance, and is beneficial to improving the reliability of a power system and the capability of coping with potential failures.
Further, the specific analysis method of the third fault prediction value is as follows: acquiring real-time natural weather type data in a current area, and recording the real-time natural weather type data as; Acquiring current observed power time information in a current area, and recording as; Extracting an integrated occurrence index of the power failure, constructing a third failure prediction value formula, and calculating a third failure prediction value according to the third failure prediction value formula; the third failure prediction value formula is: Wherein, the method comprises the steps of, wherein, As a result of the third failure prediction value,Is the firstThe overall mean value of the natural weather array of each region,Is the firstThe overall average of the individual regional power failure arrays,Is the firstThe overall mean of the individual zone time to failure arrays,Is a correction factor for the third failure prediction value.
Claims (10)
1. The power failure early warning system based on data processing is characterized by comprising a power data acquisition module, an index data processing module, a failure prediction module and a failure early warning module;
The power data acquisition module is used for dividing a plurality of power data corresponding areas, acquiring historical power fault data, historical natural weather data, current observed power time information and real-time natural weather data corresponding to all the areas, and respectively preprocessing the historical power fault data, the historical natural weather data, the current observed power time information and the real-time natural weather data;
The index data processing module is used for respectively analyzing natural fault correlation coefficients, power fault occurrence indexes and power fault comprehensive occurrence indexes according to historical power fault data and historical natural meteorological data corresponding to all areas;
The fault prediction module is used for analyzing and obtaining a first fault prediction value, a second fault prediction value and a third fault prediction value according to the current observed power time information, the real-time natural meteorological data, the natural fault correlation coefficient, the power fault occurrence index and the power fault comprehensive occurrence index corresponding to all the areas;
The fault early warning module is used for comprehensively analyzing the first fault prediction value, the second fault prediction value and the third fault prediction value and sending out fault early warning.
2. The data processing-based power failure warning system of claim 1, wherein: the index data processing module comprises a natural fault correlation coefficient analysis unit, a power fault occurrence index analysis unit and a power fault comprehensive occurrence index analysis unit;
The natural fault correlation coefficient analysis unit: the correlation between the power failure type data and the natural weather type data is analyzed according to the historical power failure data and the historical natural weather data, and a natural failure correlation coefficient is obtained;
the power failure occurrence index analysis unit: the power failure generation index is used for analyzing the power failure generation index according to the historical power failure data;
The comprehensive power failure occurrence index analysis unit comprises: the method comprises the steps of comprehensively analyzing the relevance between power failure type data, power failure occurrence time information and natural weather type data according to historical power failure data and historical natural weather data to obtain a power failure comprehensive occurrence index;
the historical power failure data includes: power failure type data, power failure occurrence time information;
The historical natural weather data includes: natural weather type data, natural weather occurrence time information.
3. The data processing-based power failure warning system of claim 2, wherein: the fault prediction module comprises a first fault prediction value analysis unit, a second fault prediction value analysis unit and a third fault prediction value analysis unit;
The first failure prediction value analysis unit: the method comprises the steps of obtaining a first fault prediction value according to real-time natural meteorological data and natural fault correlation coefficients, outputting a result if the first fault prediction value exceeds a first threshold value, and outputting a fault-free result if the first fault prediction value exceeds the first threshold value;
The second failure prediction value analysis unit: the method comprises the steps of obtaining a second fault prediction value according to current observation power time information and a fault occurrence prediction model, outputting a result if the second fault prediction value exceeds a second threshold value, and outputting a fault-free result if the second fault prediction value exceeds the second threshold value;
The third failure prediction value analysis unit: and the method is used for obtaining a third fault prediction value according to the real-time natural meteorological data of the area and the current observed power time information, outputting a result if the third fault prediction value exceeds a third threshold value, and outputting a fault-free result if the third fault prediction value exceeds the third threshold value.
4. The power failure early warning system based on data processing as claimed in claim 3, wherein the comprehensive analysis method of the failure early warning module is as follows:
Sending out fault early warning according to the power fault prediction result:
If the first fault prediction value, the second fault prediction value and the third fault prediction value are all output to have faults, determining to send out fault early warning;
If and only if one output has no fault, sending out fault early warning according to the result of the third fault prediction value;
if the third predicted value result is faulty, determining that fault early warning is sent out, and if the third predicted value result is non-faulty, determining that no early warning is sent out, namely:
If the first fault prediction value and the second fault prediction value are both output with faults and the third fault prediction value is output without faults, determining that no early warning is performed;
if the first fault prediction value and the third fault prediction value are both output with faults and the second fault prediction value is output without faults, determining to send out fault early warning;
if the second fault prediction value and the third fault prediction value are both output with faults and the first fault prediction value is output without faults, determining to send out fault early warning;
if no fault exists in two or more outputs among the three, determining that no early warning exists, namely:
If the first fault prediction value, the second fault prediction value and the third fault prediction value are all output to be fault-free, determining that no early warning is performed;
if the first fault prediction value and the second fault prediction value are output without faults and the third fault prediction value is output with faults, determining that early warning is not performed;
If the first fault prediction value and the third fault prediction value are output without faults and the second fault prediction value is output with faults, determining that early warning is not performed;
If the second fault prediction value and the third fault prediction value are output without faults and the first fault prediction value is output with faults, the early warning is determined not to be performed.
5. The data processing-based power failure warning system of claim 3, wherein: the specific analysis method of the natural fault correlation coefficient comprises the following steps:
according to the power failure occurrence time information and the natural weather occurrence time information, natural weather type data matched with the power failure occurrence time are found out;
Acquiring power failure type data corresponding to the power failure occurrence time information, constructing a natural failure correlation coefficient formula, and calculating natural correlation coefficients between the power failure types and the natural meteorological types in all areas according to the natural failure correlation coefficient formula;
numbering the regions as Recording，Is the total number of regions;
The power failure type is numbered Recording As a total number of types of power failures,
And will be the firstThe power failure type data corresponding to each region is recorded as a power failure array；
Numbering natural weather types asRecording，Is the total number of natural weather types and will beThe natural weather type data corresponding to each region is recorded as a natural weather array；
And calculating a natural fault correlation coefficient of the total area according to a natural fault correlation coefficient formula, wherein the natural fault correlation coefficient formula is as follows:
，
Wherein, For natural fault correlation coefficients between the power fault types and the natural weather types in all areas,Is the standard deviation of the power failure array,Is the standard deviation of the natural weather array,Is a correction factor for the natural fault correlation coefficient.
6. The data processing-based power failure warning system of claim 5, wherein: the specific analysis method of the first fault prediction value comprises the following steps:
acquiring the real-time natural weather type with the largest proportion in all the current areas, and recording the real-time natural weather type as ；
Extracting a natural fault correlation coefficient, constructing a first fault prediction value formula, and calculating a first fault prediction value according to the first fault prediction value formula;
The first failure prediction value formula is:
，
Wherein, As a result of the first failure prediction value,Is the firstThe overall mean value of the natural weather array of each region,Is the firstThe overall average of the individual regional power failure arrays,Is a correction factor for the first failure prediction value.
7. The data processing-based power failure warning system of claim 5, wherein: the specific analysis method of the power failure occurrence index comprises the following steps:
Setting the cycle period of power failure monitoring as And numbering the cycle period as，，Dividing the cycle period equally intoThe number of observation periods isIs marked as，To at the firstA total number of observation periods in each cycle period;
acquiring power failure occurrence time information in historical failure data corresponding to each observation period in each cycle period;
constructing a power failure occurrence index formula, and calculating a power failure occurrence index;
the power failure occurrence index formula is:
，
Wherein, As an index of the occurrence of a power failure,Is the firstArea numberThe number of power failures occurring within a single observation period,Is a natural constant which is used for the production of the high-temperature-resistant ceramic material,Is a correction factor for the power failure occurrence index.
8. The data processing-based power failure warning system of claim 7, wherein: the specific analysis method of the second fault prediction value comprises the following steps:
Constructing a fault occurrence prediction model based on a model of a time sequence according to the power fault occurrence index, the cycle period, the observation period and the power fault occurrence time information;
inputting the current observed power time information into a fault occurrence prediction model to obtain a second fault prediction value 。
9. The data processing-based power failure warning system of claim 7, wherein: the specific analysis method of the comprehensive occurrence index of the power failure comprises the following steps:
Acquiring historical power failure data, historical natural weather data and power failure occurrence time information corresponding to the area;
the power failure occurrence time information is numbered as Recording，Is the total amount of the power failure occurrence time information and will beThe power failure occurrence time information corresponding to each area is recorded as a failure time array；
According to the power failure arrayAnd natural weather array；
Constructing a comprehensive power failure occurrence index formula, and calculating a comprehensive power failure occurrence index;
The comprehensive occurrence index formula of the power failure is as follows:
，
in the method, in the process of the invention, Is the firstThe power failure composite occurrence index of each region,And the correction factor of the comprehensive occurrence index of the power failure.
10. The data processing-based power failure warning system of claim 9, wherein: the specific analysis method of the third fault prediction value comprises the following steps:
acquiring real-time natural weather type data in a current area, and recording the real-time natural weather type data as ；
Acquiring current observed power time information in a current area, and recording as；
Extracting an integrated occurrence index of the power failure, constructing a third failure prediction value formula, and calculating a third failure prediction value according to the third failure prediction value formula;
the third failure prediction value formula is:
，
Wherein, As a result of the third failure prediction value,Is the firstThe overall mean value of the natural weather array of each region,Is the firstThe overall average of the individual regional power failure arrays,Is the firstThe overall mean of the individual zone time to failure arrays,Is a correction factor for the third failure prediction value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410417279.7A CN118014405B (en) | 2024-04-09 | Power failure early warning system based on data processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410417279.7A CN118014405B (en) | 2024-04-09 | Power failure early warning system based on data processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118014405A CN118014405A (en) | 2024-05-10 |
| CN118014405B true CN118014405B (en) | 2024-07-12 |
Family
ID=
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110175745A (en) * | 2019-04-26 | 2019-08-27 | 中国电力科学研究院有限公司 | A kind of electric power telecommunication network risk assessment method and system based on fault modeling |
| CN115936448A (en) * | 2023-02-13 | 2023-04-07 | 南京深科博业电气股份有限公司 | Urban distribution network power evaluation system and method based on big data |
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110175745A (en) * | 2019-04-26 | 2019-08-27 | 中国电力科学研究院有限公司 | A kind of electric power telecommunication network risk assessment method and system based on fault modeling |
| CN115936448A (en) * | 2023-02-13 | 2023-04-07 | 南京深科博业电气股份有限公司 | Urban distribution network power evaluation system and method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109524139B (en) | Real-time equipment performance monitoring method based on equipment working condition change | |
| CN109186813B (en) | Temperature sensor self-checking device and method | |
| CN110647133B (en) | Rail transit equipment state detection maintenance method and system | |
| CN115329812B (en) | Bridge infrastructure anomaly monitoring method based on artificial intelligence | |
| CN107844067B (en) | A kind of gate of hydropower station on-line condition monitoring control method and monitoring system | |
| CN114997313B (en) | Abnormality detection method for ocean on-line monitoring data | |
| CN113177646A (en) | Power distribution equipment online monitoring method and system based on self-adaptive edge proxy | |
| CN104317778A (en) | Massive monitoring data based substation equipment fault diagnosis method | |
| CN114487705A (en) | Power grid equipment fault positioning detection method | |
| CN111314110B (en) | Fault early warning method for distributed system | |
| CN118014405B (en) | Power failure early warning system based on data processing | |
| CN118014405A (en) | Power failure early warning system based on data processing | |
| CN115034094A (en) | Prediction method and system for operation state of metal processing machine tool | |
| CN113269266A (en) | Knowledge distillation-based wafer manufacturing process fault detection method under multi-fault mode | |
| CN104731955A (en) | Methods and systems for diagnostic standard establishment and intelligent diagnosis of wind generation set oil monitoring | |
| CN116633026B (en) | Clean energy perception monitoring method and device, computer equipment and storage medium | |
| CN117978628B (en) | Communication control method and system based on intelligent park | |
| KR102667861B1 (en) | heavy electrical equipment treatment apparatus using deep learning-based prediction model and method therefor | |
| CN114151291B (en) | Early fault monitoring method for wind turbine generator | |
| CN118118319B (en) | Intelligent diagnosis method and system for network equipment based on big data | |
| CN117808456B (en) | Equipment fault early warning method and device based on intelligent operation management | |
| CN117783792B (en) | Valve side sleeve insulation state detection method and system based on multiparameter real-time monitoring | |
| Li et al. | Health Monitoring Framework for Weather Radar Based on Long Short‐Term Memory Network with a Real Case Study | |
| CN115860123A (en) | Fault diagnosis reasoning and checking method for water turbine | |
| Abdubrani et al. | Achieving Unprecedented Precision in Multiphase Flow Meter Monitoring with Deep Learning-Driven Abnormality Detection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |