Power equipment fault data processing method and system based on big data

Abstract

The invention discloses a big data-based power equipment fault data processing method and system, relates to the technical field of power equipment, and solves the problem of misjudgment of power data faults caused by data abnormity of other monitoring items due to data missing of some monitoring items. According to the invention, based on the confirmed abnormal signal, whether the monitoring item associated with the corresponding abnormal signal has the data missing condition is identified, if the data missing condition exists, association adjustment of the time characteristic is carried out through a mode of confirming the missing moment, and based on the adjusted data verification result, whether the data missing condition exists is evaluated, so that the accuracy of the data missing condition is improved. According to the method, the signal display is carried out in time if the data missing condition exists, the data abnormity condition exists if the data missing condition does not exist, and the identification processing mode can effectively guarantee the comprehensiveness of the fault data confirmation and processing of the power equipment, and can achieve a better data processing effect.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power equipment, and in particular to a method and system for processing electric power equipment fault data based on big data. Background Art

[0002] Power equipment refers to various equipment used in power generation, transmission, transformation, distribution and consumption; The application with patent publication number CN118035924A discloses a method and system for processing power equipment fault data based on power big data, including: real-time collection of fault data of electric energy meters, preprocessing of the collected fault data, and qualified detection to provide comprehensive and accurate data support for fault judgment of electric energy meters, calculation of fault characterization coefficients of electric energy meters, determination of abnormal degree of fault of electric energy meters, which is conducive to accurate diagnosis of fault problems of electric energy meters, calculation of fault characterization difference of electric energy meters based on fault data of electric energy meters, calculation of influence value of data deviation on fault characterization coefficient of electric energy meters based on obtained fault characterization difference of electric energy meters and t-statistic of two groups of data, analysis of influence degree of data deviation on fault of electric energy meters according to influence value, which is conducive to reducing misjudgment rate of fault degree of electric energy meters, and corresponding optimization and adjustment of electric energy meters to improve their metering accuracy and operation stability.

[0003] During the fault analysis and processing process, the data associated with the power equipment is generally evaluated based on the corresponding data and the corresponding preset interval to determine whether the corresponding data is abnormal data, thereby making a fault judgment. However, this method is rather one-sided, because the corresponding power equipment data is prone to data missing during the actual processing. The missing data of certain monitoring items will cause data abnormalities in other monitoring items, leading to misjudgment of power data faults, which in turn causes major problems in the accuracy of the power equipment fault processing process. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method and system for processing power equipment fault data based on big data, which solves the problem that the missing data of certain monitoring items will cause data anomalies of other monitoring items, leading to misjudgment of power data faults.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for processing power equipment fault data based on big data, comprising the following steps: Step 1: Analyze the periodic data generated by the specified power equipment in the past historical data, determine the data trends associated with different monitoring items from the monitoring data associated with several different monitoring items, and then lock the related monitoring items with association from the data trends confirmed in the same period and calibrate them. The specific sub-steps are: S11. For each group of power equipment, determine the data monitoring items associated with the power equipment, and then extract a set of periodic data from the historical completion data monitored by the power equipment, wherein the periodic duration in the periodic data is T, where T is a preset value, and each moment in the periodic data includes monitoring data of all data monitoring items of the power equipment; S12, based on the periodic data confirmed by the corresponding power equipment, and synchronously sorting the monitoring data belonging to different data monitoring items in the periodic data based on their time sequence, to confirm the data sequence belonging to different data monitoring items of this power equipment; S13, confirm the change data of each group of data sequence at adjacent moments, the change data is the monitoring data of the next moment minus the monitoring data of the previous moment, and mark the change data between different adjacent moments of different data monitoring items as BH. i-k , where i represents different data monitoring items and k represents different adjacent moments; S14. Based on the confirmed data sequences belonging to different data monitoring items, a group of data sequences is randomly selected as the main sequence, and then a group of data sequences is randomly selected from the remaining data sequences as the secondary sequence. The combination of the main sequence and the secondary sequence is calibrated as a combined sequence. The change data of the main sequence corresponding to adjacent moments is taken as the main change data, and the change data of the secondary sequence corresponding to adjacent moments is taken as the secondary change data. The following is adopted: Q k-q =Main variable data÷Secondary variable data to confirm the comparison trend between the same period Q k-q , where k represents different adjacent moments, q represents different combination sequences, and in each different combination sequence, the secondary sequences are different, and the main sequences are the same; Based on the different comparison trends Q confirmed at different adjacent moments in the corresponding combined sequence k-q , compare several groups to trend Q k-q Perform variance processing to confirm the associated verification value H q , select H q min is taken as the pending sequence, and H is identified. q The result of checking min with the preset value Y1: If H q min>Y1, then the data monitoring item associated with the main sequence in the pending sequence is marked as a single monitoring item. q min≤Y1, the data monitoring items associated with the main sequence and the sub-sequence in the pending sequence are marked as associated monitoring items, where Y1 is a preset value; S15, for other data sequences that have not been associated with monitoring items or single monitoring items, the remaining data sequences are processed in sequence in the same manner as step S14 until several data sequences are marked as associated monitoring items or single monitoring items; Step 2: confirm the fault of the monitoring parameters associated with each group of monitoring items in the power equipment, identify whether the monitoring parameters associated with the corresponding monitoring items are abnormal, if there is an abnormality, calibrate the abnormal monitoring items, if there is no abnormality, do not calibrate, the specific sub-steps are: The monitoring parameters of each group of monitoring items in the power equipment are calibrated as CS i , and the monitoring parameter CS associated with the corresponding monitoring item i The endpoint values ​​of the preset intervals associated with each group of monitoring items are all preset values. i ∈ preset interval, it represents the monitoring parameter CS of this monitoring item at the current moment i No abnormality, if CS i ∉Preset interval, it represents the monitoring parameter CS associated with the corresponding monitoring item at the current moment i There is an abnormality, and the monitoring item corresponding to the current moment is marked as an abnormal monitoring item; Step 3: Based on the abnormal monitoring items calibrated by the corresponding power equipment, it is preferred to identify whether such abnormal monitoring items belong to single monitoring items or associated monitoring items. Based on the identification results, the analysis signal or fault signal is confirmed, and the confirmed signal is displayed. The specific sub-steps are: Based on the abnormal monitoring item marked at the current moment, identify whether this abnormal monitoring item belongs to the single monitoring item shown. If it does, directly generate a fault signal for direct display. If it does not, it means that this abnormal monitoring item has other associated monitoring items, that is, this abnormal monitoring item belongs to the associated monitoring item, so an analysis signal is generated, and the generated analysis signal is subsequently processed; Step 4: Based on the abnormal monitoring items calibrated by the corresponding power equipment and the associated analysis signals, initially lock other monitoring items associated with the abnormal monitoring item from the associated monitoring items of the abnormal monitoring item, and then lock a group of verification cycles, and perform verification analysis on the monitoring data associated with the abnormal monitoring item and other monitoring items within the verification cycle to assess whether the abnormal monitoring item is a data missing abnormality or a fault abnormality. The specific sub-steps are: S41, based on the abnormal monitoring item and the analysis signal, determining other monitoring items from the associated monitoring items to which the abnormal monitoring item belongs, and marking the other monitoring items determined by the abnormal monitoring item as associated sub-items; S42, defining a set of verification periods, wherein the verification period is a preset period, and the abnormal time associated with the abnormal monitoring item is located at the middle time of the verification period, sequentially confirming the monitoring data generated by the abnormal monitoring item within the verification period, and sorting the confirmed monitoring data according to the time sequence, and confirming the abnormal monitoring item data sequence; Then, the monitoring data associated with the corresponding associated sub-items within the verification period are sorted according to the time sequence to confirm the data sequence of the associated sub-items; S43, according to the abnormal moment confirmed in the abnormal monitoring item data sequence, mark a group of moments before the abnormal moment as the previous moment, confirm the calibration moment associated with the previous moment from the associated sub-item data sequence, the calibration moment and the previous moment belong to the same moment, confirm a group of missing moments after the calibration moment confirmed in the associated sub-item data sequence, make the missing moment and the abnormal moment belong to the same moment, and adjust the time features associated with other different monitoring data in the associated sub-item data sequence; S44, perform data verification on the adjusted associated sub-item data sequence and the abnormal monitoring item data sequence: confirm the change data between adjacent moments in the two groups of data sequences, mark the change data at adjacent moments in the associated sub-item data sequence as the first data, and then synchronously lock the change data at adjacent moments in the abnormal monitoring item data sequence and mark it as the second data. The determination method of the change data is consistent with the determination method of the corresponding change data in step S13, wherein the change data of the two groups of adjacent moments before and after the abnormal moment are not confirmed, and similarly, the change data of the two groups of adjacent moments before and after the missing data are not synchronously confirmed; S45, identifying whether the data sequence associated with the abnormal monitoring item in step S14 belongs to the primary sequence or the secondary sequence in the associated monitoring item: If it belongs to the main sequence, the first data ÷ the second data = data trend is used to confirm the data trend of the associated sub-item data sequence and the abnormal monitoring item data sequence between adjacent moments; If it belongs to a subsequence, the following formula is used: second data ÷ first data = data trend to confirm the data trend of the associated sub-item data sequence and the abnormal monitoring item data sequence between adjacent moments; S46. Perform variance processing on the confirmed data trends of several groups, lock the calibration variance, and then confirm the associated verified value determined by this associated monitoring item from the processed historical data. If the calibration variance is ≤ the associated verified value, a data missing signal is generated for display, indicating that there is data missing in the associated monitoring item of this abnormal monitoring item, resulting in data abnormality in this abnormal monitoring item; if the calibration variance is greater than the associated verified value, an abnormal monitoring fault signal is generated for display.

[0006] Preferably, a power equipment fault data processing system based on big data comprises: The associated monitoring item calibration end performs data analysis on the periodic data generated by the specified power equipment in the past historical data, determines the data trends associated with different monitoring items from the monitoring data associated with several different monitoring items, and then locks the associated related monitoring items from the data trends confirmed in the same period and calibrates them, and locks the associated monitoring items or single monitoring items based on the calibration results; The abnormal monitoring item calibration end performs fault confirmation on the monitoring parameters associated with each group of monitoring items in the power equipment, and identifies whether the monitoring parameters associated with the corresponding monitoring items are abnormal. If there is an abnormality, the abnormal monitoring item is calibrated, and if there is no abnormality, no calibration is performed; The signal preliminary confirmation end, based on the abnormal monitoring items calibrated by the corresponding power equipment, preferentially identifies whether such abnormal monitoring items belong to single monitoring items or associated monitoring items, confirms the analysis signal or fault signal based on the identification result, and displays the confirmed signal; The abnormality determination center, based on the abnormal monitoring items calibrated by the corresponding power equipment and the associated analysis signals, initially locks other monitoring items associated with the abnormal monitoring item from the associated monitoring items of the abnormal monitoring item, and then locks a group of verification cycles to perform verification and analysis on the monitoring data associated with the abnormal monitoring item and other monitoring items within the verification cycle, and assesses whether the abnormal monitoring item is a data missing abnormality or a fault abnormality.

[0007] The present invention provides a method and system for processing power equipment fault data based on big data. Compared with the prior art, it has the following beneficial effects: The present invention monitors the monitoring data associated with different monitoring items in each power device, and identifies the associated monitoring items based on the relevant characteristics between the data. By confirming the associated monitoring items and the unassociated monitoring items, it can effectively confirm whether the monitoring data of the monitoring items are abnormal, thereby achieving a better data accuracy confirmation effect and ensuring the specific accuracy of subsequent data confirmation; In the subsequent data verification and evaluation process, based on the confirmed abnormal signal, identify whether there is any data missing in the monitoring item associated with the corresponding abnormal signal. If there is data missing, the time feature association adjustment is performed by confirming the missing time. Based on the adjusted data verification result, assess whether there is data missing. If there is data missing, display the signal in time. If there is no data missing, it means that there is data abnormality. This identification and processing method can effectively ensure the comprehensiveness of the confirmation and processing of the power equipment fault data, and can achieve better data processing effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 It is a schematic diagram of the process of the present invention; Figure 2 It is a schematic diagram of the principle framework of the present invention. DETAILED DESCRIPTION

[0009] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0010] First embodiment: Please refer to Figure 1 The present application provides a method for processing power equipment fault data based on big data, comprising the following steps: Step 1: Perform data analysis on the periodic data generated by the specified power equipment in the past historical data, determine the data trends associated with different monitoring items from the monitoring data associated with several different monitoring items, and then lock the related monitoring items with correlation from the data trends confirmed in the same period and calibrate them. Specifically, in the corresponding power equipment, there are related monitoring items with strong correlation, such as voltage and current. The correlation is very strong. When the voltage changes, the related current will also change. Similarly, in the actual data fault verification and identification process, for the related monitoring items with correlation, one group of the two groups of related related monitoring items has a fault and the other group does not have a fault. In this case, there may be a data missing problem in the actual processing process. Under normal circumstances, the values ​​of the two groups of equipment with strong enough correlation are all in a state of correlation change, so that the relevant signals are displayed; The specific sub-steps for determining the existence of associated monitoring items in the power equipment are: S11. For each group of power equipment, determine the data monitoring items associated with the power equipment (that is, the specific numerical items monitored by the designated system, and the insignificant parameters in the corresponding power equipment are not recorded in the data monitoring items), and then extract a set of periodic data from the historical completion data monitored by the power equipment, wherein the period duration in the periodic data is T, where T is a preset value, and its specific value is determined by the operator based on experience, and each moment in the periodic data includes the monitoring data of all the data monitoring items of the power equipment (that is, each data monitoring item has corresponding monitoring data at the corresponding moment); S12, based on the periodic data confirmed by the corresponding power equipment, and synchronously sorting the monitoring data belonging to different data monitoring items in the periodic data based on their time sequence, to confirm the data sequence belonging to different data monitoring items of this power equipment; S13, confirm the change data of each group of data sequence at adjacent moments, the change data is the monitoring data of the next moment minus the monitoring data of the previous moment, and mark the change data between different adjacent moments of different data monitoring items as BH. i-k , where i represents different data monitoring items and k represents different adjacent moments; S14. Based on the confirmed data sequences belonging to different data monitoring items, randomly select a group of data sequences as the main sequence, and then randomly select a group of data sequences from the remaining data sequences as the secondary sequence. The combination of the main sequence and the secondary sequence is calibrated as a combined sequence (the combined sequence includes two groups of data sequences). The change data of the main sequence corresponding to adjacent moments is taken as the main variable data, and the change data of the secondary sequence corresponding to adjacent moments is taken as the secondary variable data. Use: Q k-q =Main variable data÷Secondary variable data to confirm the comparison trend between the same period Q k-q , where k represents different adjacent moments (that is, the same period in two data sequences), q represents different combination sequences, and in each different combination sequence, the secondary sequences are different, and the primary sequences are the same; Based on the different comparison trends Q confirmed at different adjacent moments in the corresponding combined sequence k-q , compare several groups to trend Q k-q Perform variance processing to confirm the associated verification value H q , select H q min is taken as the pending sequence, and H is identified. q The result of checking min with the preset value Y1: If H q min>Y1, then the data monitoring item associated with the main sequence in the pending sequence is marked as a single monitoring item. q min≤Y1, the data monitoring items associated with the main sequence and the secondary sequence in the pending sequence are marked as associated monitoring items, where Y1 is a preset value, and its specific value is determined by the operator based on experience; S15, for other data sequences that have not been associated with monitoring items or single monitoring items, the remaining data sequences are processed in sequence in the same manner as step S14 until several data sequences are marked as associated monitoring items or single monitoring items; Specifically, the so-called associated monitoring items are two sets of data monitoring items that are highly correlated. For example, data monitoring items such as voltage and current will definitely be classified into corresponding associated monitoring items. The so-called single monitoring item is that the data associated with this item has no correlation with the data of any other item. In this case, such non-correlated data monitoring items are single monitoring items. Step 2: Perform fault confirmation on the monitoring parameters associated with each group of monitoring items in the power equipment, and identify whether the monitoring parameters associated with the corresponding monitoring items are abnormal. If there is an abnormality, calibrate the abnormal monitoring items. If there is no abnormality, no calibration is performed. The specific sub-steps for identifying whether the corresponding monitoring items are abnormal are: S21. Calibrate the monitoring parameters of each group of monitoring items in the power equipment in real time as CS i , and the monitoring parameter CS associated with the corresponding monitoring item i The endpoint values ​​of the preset intervals associated with each group of monitoring items are preset values, which are prepared in advance by relevant operators based on experience. i ∈ preset interval, it represents the monitoring parameter CS of this monitoring item at the current moment i No abnormality, if CS i ∉Preset interval, it represents the monitoring parameter CS associated with the corresponding monitoring item at the current moment i There is an abnormality, and the monitoring item corresponding to the current moment is marked as an abnormal monitoring item, wherein the preset interval associated with each different monitoring item is different, and the specific value range of the corresponding preset interval is determined by the relevant operator according to the standard operation of the corresponding equipment, to lock the standard interval associated with the corresponding monitoring item for comprehensive confirmation; Step 3: Based on the abnormal monitoring items calibrated by the corresponding power equipment, it is preferred to identify whether such abnormal monitoring items belong to single monitoring items or associated monitoring items. Based on the identification results, the analysis signal or fault signal is confirmed, and the confirmed signal is displayed. The specific sub-steps for identification are: S31, based on the abnormal monitoring item calibrated at the current moment, identify whether the abnormal monitoring item belongs to the single monitoring item shown, if it does, directly generate a fault signal for direct display, if it does not, it means that the abnormal monitoring item has other associated monitoring items, that is, the abnormal monitoring item belongs to the associated monitoring item, so an analysis signal is generated, and the generated analysis signal is subsequently processed; Step 4: Based on the abnormal monitoring items calibrated by the corresponding power equipment and the associated analysis signals, initially lock other monitoring items associated with the abnormal monitoring item from the associated monitoring items of the abnormal monitoring item, and then lock a group of verification cycles, and perform verification analysis on the monitoring data associated with the abnormal monitoring item and other monitoring items within the verification cycle, and assess whether the abnormal monitoring item is a data missing abnormality or a fault abnormality, wherein the specific sub-steps of the assessment are: S41, based on the abnormal monitoring item and the analysis signal, determining other monitoring items from the associated monitoring items to which the abnormal monitoring item belongs, and marking the other monitoring items determined by the abnormal monitoring item as associated sub-items; S42, defining a set of verification cycles, wherein the verification cycle is a preset cycle, and the specific duration of the preset cycle is determined by relevant operators based on experience, and the abnormal time associated with the abnormal monitoring item (that is, the corresponding time when the monitoring item is determined to be an abnormal monitoring item) is located in the middle of the verification cycle, and the monitoring data generated by the abnormal monitoring item in the verification cycle are confirmed in sequence, and the confirmed monitoring data are sorted according to the time sequence relationship to confirm the abnormal monitoring item data sequence; Then, the monitoring data associated with the corresponding associated sub-items within the verification period are sorted according to the time sequence to confirm the data sequence of the associated sub-items; S43, according to the abnormal moment confirmed in the abnormal monitoring item data sequence, mark the group of moments before the abnormal moment as the previous moment, confirm the calibration moment associated with the previous moment from the associated sub-item data sequence, the calibration moment and the previous moment belong to the same moment, confirm a group of missing moments after the calibration moment confirmed in the associated sub-item data sequence, so that the missing moment and the abnormal moment belong to the same moment, and adjust the time features associated with other subsequent different monitoring data in the associated sub-item data sequence (when a group of missing moments is inserted, then the other subsequent associated moments need to be added by 1, so that the associated related moments are all extended one moment backward to become another group of moments); S44, perform data verification on the adjusted associated sub-item data sequence and the abnormal monitoring item data sequence: confirm the change data between adjacent moments in the two groups of data sequences, mark the change data at adjacent moments in the associated sub-item data sequence as the first data, and then synchronously lock the change data at adjacent moments in the abnormal monitoring item data sequence and mark it as the second data. The determination method of the change data is consistent with the determination method of the corresponding change data in step S13, wherein the change data of the two groups of adjacent moments before and after the abnormal moment are not confirmed, and similarly, the change data of the two groups of adjacent moments before and after the missing data are not synchronously confirmed; S45, identifying whether the data sequence associated with the abnormal monitoring item in step S14 belongs to the primary sequence or the secondary sequence in the associated monitoring item: If it belongs to the main sequence, the first data ÷ the second data = data trend is used to confirm the data trend of the associated sub-item data sequence and the abnormal monitoring item data sequence between adjacent moments; If it belongs to a subsequence, the following formula is used: second data ÷ first data = data trend to confirm the data trend of the associated sub-item data sequence and the abnormal monitoring item data sequence between adjacent moments; S46, performing variance processing on the confirmed data trends, locking the calibration variance, and then confirming the associated verified value determined by the associated monitoring item from the processed historical data (that is, the data in step S14). If the calibration variance is ≤ the associated verified value, a data missing signal is generated for display, indicating that the associated monitoring item of the abnormal monitoring item has data missing, resulting in data abnormality of the abnormal monitoring item. If the calibration variance is greater than the associated verified value, an abnormal monitoring fault signal is generated for display for external personnel to view, indicating that such processing process has a monitoring fault and needs to be processed in a timely manner. Specifically, with respect to whether there are abnormalities in the corresponding data monitoring items, priority is given to whether there is missing data. If there is missing data, the correlation adjustment of the time characteristics is performed by confirming the missing time. Based on the adjusted data verification results, it is assessed whether there is missing data. If there is missing data, the signal is displayed in time. If there is no missing data, it means that there is data abnormality, which causes the signal abnormality of the corresponding monitoring item.

[0011] Second embodiment: Combination Figure 2 , a power equipment fault data processing system based on big data, comprising: The associated monitoring item calibration end performs data analysis on the periodic data generated by the specified power equipment in the past historical data, determines the data trends associated with different monitoring items from the monitoring data associated with several different monitoring items, and then locks the associated related monitoring items from the data trends confirmed in the same period and calibrates them, and locks the associated monitoring items or single monitoring items based on the calibration results; The abnormal monitoring item calibration end performs fault confirmation on the monitoring parameters associated with each group of monitoring items in the power equipment, and identifies whether the monitoring parameters associated with the corresponding monitoring items are abnormal. If there is an abnormality, the abnormal monitoring item is calibrated, and if there is no abnormality, no calibration is performed; The signal preliminary confirmation end, based on the abnormal monitoring items calibrated by the corresponding power equipment, preferentially identifies whether such abnormal monitoring items belong to single monitoring items or associated monitoring items, confirms the analysis signal or fault signal based on the identification result, and displays the confirmed signal; The abnormality determination center, based on the abnormal monitoring items calibrated by the corresponding power equipment and the associated analysis signals, initially locks other monitoring items associated with the abnormal monitoring item from the associated monitoring items of the abnormal monitoring item, and then locks a group of verification cycles to perform verification and analysis on the monitoring data associated with the abnormal monitoring item and other monitoring items within the verification cycle, and assesses whether the abnormal monitoring item is a data missing abnormality or a fault abnormality.

[0012] The third embodiment: In the specific implementation process, this embodiment includes the entire implementation process of the above two groups of embodiments.

[0013] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0014] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for processing power equipment fault data based on big data, characterized in that: The following steps are involved: Step 1: Analyze the periodic data generated by the specified power equipment in the past historical data, determine the data trends associated with different monitoring items from the monitoring data associated with several different monitoring items, and then lock the related monitoring items with association from the data trends confirmed in the same period and calibrate them; Step 2: Perform fault confirmation on the monitoring parameters associated with each group of monitoring items in the power equipment to identify whether the monitoring parameters associated with the corresponding monitoring items are abnormal. If there is an abnormality, calibrate the abnormal monitoring items. If there is no abnormality, no calibration is performed. Step 3: Based on the abnormal monitoring items calibrated by the corresponding power equipment, it is preferred to identify whether such abnormal monitoring items belong to single monitoring items or associated monitoring items, and based on the identification results, confirm the analysis signal or fault signal, and display the confirmed signal; Step 4. Based on the abnormal monitoring items calibrated by the corresponding power equipment and the associated analysis signals, initially lock other monitoring items associated with the abnormal monitoring item from the associated monitoring items of the abnormal monitoring item, and then lock a group of verification cycles, and perform verification and analysis on the monitoring data associated with the abnormal monitoring item and other monitoring items within the verification cycle to assess whether the abnormal monitoring item is a data missing abnormality or a fault abnormality.

2. A method for processing power equipment fault data based on big data according to claim 1, characterized in that: In step 1, the specific sub-steps of determining that there are related monitoring items associated with the power equipment are: S11. For each group of power equipment, determine the data monitoring items associated with the power equipment, and then extract a set of periodic data from the historical completion data monitored by the power equipment, wherein the periodic duration in the periodic data is T, where T is a preset value, and each moment in the periodic data includes monitoring data of all data monitoring items of the power equipment; S12, based on the periodic data confirmed by the corresponding power equipment, and synchronously sorting the monitoring data belonging to different data monitoring items in the periodic data based on their time sequence, to confirm the data sequence belonging to different data monitoring items of this power equipment; S13, confirm the change data of each group of data sequence at adjacent moments, the change data is the monitoring data of the next moment minus the monitoring data of the previous moment, and mark the change data between different adjacent moments of different data monitoring items as BH. i-k , where i represents different data monitoring items and k represents different adjacent moments; S14. Based on the confirmed data sequences belonging to different data monitoring items, a group of data sequences is randomly selected as the main sequence, and then a group of data sequences is randomly selected from the remaining data sequences as the secondary sequence. The combination of the main sequence and the secondary sequence is calibrated as a combined sequence. The change data of the main sequence corresponding to adjacent moments is taken as the main change data, and the change data of the secondary sequence corresponding to adjacent moments is taken as the secondary change data. The following is adopted: Q k-q =Main variable data÷Secondary variable data to confirm the comparison trend between the same period Q k-q , where k represents different adjacent moments, q represents different combination sequences, and in each different combination sequence, the secondary sequences are different, and the main sequences are the same; Based on the different comparison trends Q confirmed at different adjacent moments in the corresponding combined sequence k-q , compare several groups to trend Q k-q Perform variance processing to confirm the associated verification value H q , select H q min is taken as the pending sequence, and H is identified. q The result of checking min with the preset value Y1: If H q min>Y1, then the data monitoring item associated with the main sequence in the pending sequence is marked as a single monitoring item. q min≤Y1, the data monitoring items associated with the main sequence and the sub-sequence in the pending sequence are marked as associated monitoring items, where Y1 is a preset value; S15. For other data sequences that have not been associated with monitoring items or single monitoring items, the same method as step S14 is adopted to process the remaining data sequences in sequence until several data sequences are marked as associated monitoring items or single monitoring items.

3. The method for processing power equipment fault data based on big data according to claim 1, characterized in that: In step 2, the specific sub-steps for identifying whether the corresponding monitoring item is abnormal are: The monitoring parameters of each group of monitoring items in the power equipment are calibrated as CS i , and the monitoring parameter CS associated with the corresponding monitoring item i The endpoint values ​​of the preset intervals associated with each group of monitoring items are all preset values. i ∈ preset interval, it represents the monitoring parameter CS of this monitoring item at the current moment i No abnormality, if CS i ∉Preset interval, it represents the monitoring parameter CS associated with the corresponding monitoring item at the current moment i There is an anomaly, and the monitoring item corresponding to the current moment is marked as an abnormal monitoring item.

4. The method for processing power equipment fault data based on big data according to claim 1, characterized in that: In step 3, the specific sub-steps of confirming the analysis signal or the fault signal are: Based on the abnormal monitoring item calibrated at the current moment, identify whether this abnormal monitoring item belongs to the single monitoring item represented. If it does, directly generate a fault signal for display. If it does not, it means that this abnormal monitoring item has other associated monitoring items, that is, this abnormal monitoring item belongs to the associated monitoring item, so an analysis signal is generated and the generated analysis signal is subsequently processed.

5. The method for processing power equipment fault data based on big data according to claim 1, characterized in that: In step 4, the specific sub-steps for evaluating whether the abnormal monitoring item is a data missing abnormality or a fault abnormality are: S41, based on the abnormal monitoring item and the analysis signal, determining other monitoring items from the associated monitoring items to which the abnormal monitoring item belongs, and marking the other monitoring items determined by the abnormal monitoring item as associated sub-items; S42, defining a set of verification periods, wherein the verification period is a preset period, and the abnormal time associated with the abnormal monitoring item is located at the middle time of the verification period, sequentially confirming the monitoring data generated by the abnormal monitoring item within the verification period, and sorting the confirmed monitoring data according to the time sequence, and confirming the abnormal monitoring item data sequence; Then, the monitoring data associated with the corresponding associated sub-items within the verification period are sorted according to the time sequence to confirm the data sequence of the associated sub-items; S43, according to the abnormal moment confirmed in the abnormal monitoring item data sequence, mark a group of moments before the abnormal moment as the previous moment, confirm the calibration moment associated with the previous moment from the associated sub-item data sequence, the calibration moment and the previous moment belong to the same moment, confirm a group of missing moments after the calibration moment confirmed in the associated sub-item data sequence, make the missing moment and the abnormal moment belong to the same moment, and adjust the time features associated with other different monitoring data in the associated sub-item data sequence; S44, perform data verification on the adjusted associated sub-item data sequence and the abnormal monitoring item data sequence: confirm the change data between adjacent moments in the two groups of data sequences, mark the change data at adjacent moments in the associated sub-item data sequence as the first data, and then synchronously lock the change data at adjacent moments in the abnormal monitoring item data sequence and mark it as the second data. The determination method of the change data is consistent with the determination method of the corresponding change data in step S13, wherein the change data of the two groups of adjacent moments before and after the abnormal moment are not confirmed, and similarly, the change data of the two groups of adjacent moments before and after the missing data are not synchronously confirmed; S45, identifying whether the data sequence associated with the abnormal monitoring item in step S14 belongs to the primary sequence or the secondary sequence in the associated monitoring item: If it belongs to the main sequence, the first data ÷ the second data = data trend is used to confirm the data trend of the associated sub-item data sequence and the abnormal monitoring item data sequence between adjacent moments; If it belongs to a subsequence, the following formula is used: second data ÷ first data = data trend to confirm the data trend of the associated sub-item data sequence and the abnormal monitoring item data sequence between adjacent moments; S46. Perform variance processing on the confirmed data trends, lock the calibration variance, and then confirm the associated verified value determined by this associated monitoring item from the processed historical data. If the calibration variance is ≤ the associated verified value, a data missing signal is generated for display, indicating that there is data missing in the associated monitoring item of this abnormal monitoring item, resulting in data anomalies in this abnormal monitoring item.

6. A method for processing power equipment fault data based on big data according to claim 5, characterized in that: In the step S46, if the calibration variance>the associated verified value, an abnormal monitoring fault signal is generated for display.

7. A power equipment fault data processing system based on big data, the processing system operates according to the power equipment fault data processing method based on big data according to claims 1-6, characterized in that: include: The associated monitoring item calibration end performs data analysis on the periodic data generated by the specified power equipment in the past historical data, determines the data trends associated with different monitoring items from the monitoring data associated with several different monitoring items, and then locks the associated related monitoring items from the data trends confirmed in the same period and calibrates them, and locks the associated monitoring items or single monitoring items based on the calibration results; The abnormal monitoring item calibration end performs fault confirmation on the monitoring parameters associated with each group of monitoring items in the power equipment, and identifies whether the monitoring parameters associated with the corresponding monitoring items are abnormal. If there is an abnormality, the abnormal monitoring item is calibrated, and if there is no abnormality, no calibration is performed; The signal preliminary confirmation end, based on the abnormal monitoring items calibrated by the corresponding power equipment, preferentially identifies whether such abnormal monitoring items belong to single monitoring items or associated monitoring items, confirms the analysis signal or fault signal based on the identification result, and displays the confirmed signal; The abnormality determination center, based on the abnormal monitoring items calibrated by the corresponding power equipment and the associated analysis signals, initially locks other monitoring items associated with the abnormal monitoring item from the associated monitoring items of the abnormal monitoring item, and then locks a group of verification cycles to perform verification and analysis on the monitoring data associated with the abnormal monitoring item and other monitoring items within the verification cycle, and assesses whether the abnormal monitoring item is a data missing abnormality or a fault abnormality.

Images