Online monitoring data early warning method and system based on massive data trend change

Abstract

The application discloses an online monitoring data early warning method and system based on massive data trend change, comprising massive real-time collection data and storage, the data comprising end screen current, capacitance, dielectric loss, system voltage phase and system frequency; screening bad point data in the stored massive data, the bad point data comprising end screen current, capacitance, dielectric loss data; performing data average value calculation on the screened bad point data; performing massive data trend analysis, and publishing early warning information according to the trend analysis result. The application performs trend analysis on the massive data through an algorithm, obtains early warning information of different levels and publishes the information, so that effective and key data can be accurately extracted from the online monitoring massive data, accurate early warning is realized through the early warning method, and the false alarm rate is greatly reduced.

Description

Technical Field

[0001] This invention relates to the field of electrical equipment condition monitoring technology, and in particular to an online monitoring data early warning method and system based on the trend changes of massive data. Background Technology

[0002] With the digital transformation of the power system, there is a need to monitor the status of more and more electrical equipment. As a result, many electrical devices have been equipped with online monitoring devices. The application of online monitoring devices can reflect the operating status of electrical equipment in real time, detect potential safety hazards early, reduce the failure rate of equipment, effectively ensure the safe and stable operation of equipment, and improve the reliability of power grid operation.

[0003] Current research mainly focuses on the operational stability, data acquisition accuracy, and data transmission timeliness of online monitoring devices. Early warning is only achieved using a simple threshold method. However, with the increasing application of online monitoring devices, the amount of data is growing exponentially. There is a lack of an early warning method for online monitoring data that can handle massive amounts of data. The threshold method used in the past has led to frequent false alarms in online monitoring, with a high false alarm rate, which greatly affects the practicality of online monitoring devices and restricts their development.

[0004] Based on this, an online monitoring data early warning method and system based on the trend changes of massive data is proposed to solve the above problems. Summary of the Invention

[0005] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0006] In view of the problem that the threshold method used in the above and / or existing online data monitoring devices leads to frequent false alarms and a high false alarm rate, which greatly affects the practicality of the online monitoring devices, this invention is proposed.

[0007] To address the aforementioned technical problems, this invention provides the following technical solution: an online monitoring data early warning method and system based on massive data trend changes, comprising: collecting and storing massive amounts of real-time data, the data including end-screen current, capacitance, dielectric loss, system voltage phase, and system frequency; filtering out bad data from the stored massive data, the bad data including end-screen current, capacitance, and dielectric loss data; calculating the average value of the filtered bad data; performing massive data trend analysis; and issuing early warning information based on the trend analysis results.

[0008] In a preferred embodiment of the present invention, the filtering of bad pixel data in the massive stored data includes: statistically analyzing all collected end-screen current data for the day to obtain the maximum and minimum end-screen current values, and then filtering using the following formula:

[0009]

[0010] When A I If the current is greater than or equal to the first preset value, the maximum current value of the last screen is discarded.

[0011] As a preferred embodiment of the present invention, the filtering of bad data in the massive data storage further includes: statistically analyzing all collected capacity data for the day to obtain the maximum and minimum capacity values, and filtering using the following formula:

[0012]

[0013] When A C If the value is greater than or equal to the second preset value, the maximum capacitance value is discarded.

[0014] As a preferred embodiment of the present invention, the filtering of bad pixel data in the massive amount of stored data further includes: statistically analyzing all media loss data collected on the same day to obtain the maximum and minimum media loss values, and filtering using the following formula:

[0015]

[0016] When A T If the value is greater than or equal to the third preset value, the maximum value of the medium loss is removed.

[0017] As a preferred embodiment of the present invention, the step of calculating the average value of the filtered bad pixel data includes:

[0018] The formula for calculating the average current of the last screen is as follows:

[0019]

[0020] Among them, I v I is the average current of the last screen. n n1 represents the number of last screen current data collected.

[0021] The formula for calculating the average capacitance is as follows:

[0022]

[0023] Among them, C v C is the average capacitance. nn is the capacitance, and n2 is the number of capacitance data collected;

[0024] The formula for calculating the average dielectric loss is as follows:

[0025]

[0026] Among them, T v T represents the average value of the dielectric loss. n n represents the dielectric loss, and n3 represents the number of dielectric loss data collected.

[0027] As a preferred embodiment of the present invention, the massive data trend analysis includes: taking several sets of end-screen current data, capacitance data, and dielectric loss data within a first time period for analysis, wherein the current end-screen current data, capacitance data, and dielectric loss data are respectively I n C n T n Then the current data, capacitance data, and dielectric loss data of the previous screen are respectively I n-1 C n-1 T n-1 , if I n-1 -I n C n-1 -C n T n-1 -T n If any value of n is greater than 0 and n is greater than the number of times threshold, a warning will be issued.

[0028] As a preferred embodiment of the present invention, the step of issuing early warning information based on trend analysis results includes: performing m trend analyses at fixed intervals within a second time period, wherein each trend analysis result includes whether there is a certain number of days in which the data shows an increasing trend compared to the previous day, and issuing early warning information based on the trend analysis results.

[0029] As a preferred embodiment of the present invention, the step of issuing early warning information based on trend analysis results further includes: issuing early warning information on the end-screen current, capacitance, and dielectric loss data, including:

[0030] If, during the first trend analysis in the second time period, there are daily data showing an increasing trend compared to the previous day for a period of more than or equal to the preset number of days, a special attention warning will be issued.

[0031] If, during the second trend analysis within the second time period, there are daily data showing an increasing trend compared to the previous day for a period of more than or equal to the preset number of days, then an on-site inspection and attention warning will be issued.

[0032] If, during the m-th trend analysis within the second time period, there are daily data showing an increasing trend compared to the previous day for a period greater than or equal to the preset number of days, a power outage inspection warning will be issued.

[0033] An online monitoring and early warning system based on the trend changes of massive data includes a data acquisition module, a data storage module, a data change calculation module, and an early warning information display and dissemination module.

[0034] The data acquisition module includes a real-time acquisition unit for capacitive equipment end-screen current, a real-time acquisition unit for capacitive equipment capacitance, a real-time acquisition unit for capacitive equipment dielectric loss, a real-time acquisition unit for system voltage phase, and a real-time acquisition unit for system frequency, which respectively acquire the capacitive equipment end-screen current, capacitance, dielectric loss, system voltage phase, and frequency data in real time.

[0035] The data storage module includes a capacitive device end-screen current storage unit, a capacitive device capacitance storage unit, a capacitive device dielectric loss storage unit, a system voltage phase storage unit, and a system frequency storage unit, which respectively store the capacitive device end-screen current, capacitance, dielectric loss, system voltage phase, and frequency data.

[0036] The data change calculation module includes a capacitive device end-screen current change calculation unit, a capacitive device capacitance data change calculation unit, a capacitive device dielectric loss data change calculation unit, a system voltage phase data change calculation unit, and a system frequency data change unit, which respectively calculate the changes in capacitive device end-screen current, capacitance, dielectric loss, system voltage phase, and frequency data.

[0037] As a preferred embodiment of the present invention, the early warning information publishing module includes an icon flashing unit, an audio reminder unit, and an information sending unit to realize the audio-visual display and transmission of early warning information.

[0038] The beneficial effects of this invention are as follows: This invention stores massive amounts of online monitoring data and uses algorithms to remove bad data, thus eliminating interference from external factors. Furthermore, it performs trend analysis on massive amounts of data using algorithms to derive different levels of early warning information and disseminate the information. This enables the accurate extraction of effective and key data from massive amounts of online monitoring data, and provides accurate early warnings, greatly reducing the false alarm rate. This method has the advantages of real-time operation, high efficiency, and accuracy, improving the operational reliability of online monitoring devices. Attached Figure Description

[0039] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0040] Figure 1 This is a flowchart of the online monitoring data early warning method of the present invention.

[0041] Figure 2 This is a flowchart of the online monitoring data early warning triggering process of the present invention.

[0042] Figure 3 This is a structural diagram of the online monitoring data early warning system of the present invention. Detailed Implementation

[0043] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0044] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0045] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0046] Example 1

[0047] Reference Figure 1 This is the first embodiment of the present invention, which provides an online monitoring data early warning method based on the trend changes of massive data, including:

[0048] S1. Collect and store massive amounts of real-time data, including end-screen current, capacitance, dielectric loss, system voltage phase, and system frequency.

[0049] S2. Filter out bad data from the massive amount of stored data. Bad data includes end-screen current, capacitance, and dielectric loss data. Due to external interference, the data may be corrupted at a certain moment and cannot accurately reflect the equipment status. Therefore, it is necessary to filter the data first.

[0050] S3. Calculate the average value of the filtered bad data. There are two methods for calculating the average value: one is to calculate the average value of the data within one hour to obtain the hourly average value, and the other is to calculate the average value of the data within the day to obtain the daily average value.

[0051] S4. Conduct massive data trend analysis and issue early warning information based on the trend analysis results.

[0052] This embodiment stores real-time collected data, filters the data, calculates the average value of the filtered data, performs massive data trend analysis on an hourly and daily basis, and finally issues early warning information. This enables the accurate extraction of effective and key data from massive online monitoring data and provides accurate early warnings, greatly reducing the false alarm rate. This invention has the advantages of real-time, high efficiency, and accuracy, and improves the operational reliability of online monitoring devices.

[0053] Example 2

[0054] Reference Figure 1 , 2 This is the second embodiment of the present invention, which is based on the previous embodiment.

[0055] S2. Filtering out bad data from the massive amount of stored data includes:

[0056] S11. Statistically analyze all the current data collected on the last screen that day to obtain the maximum and minimum values ​​of the last screen current, and then filter them using the following formula:

[0057]

[0058] When A I When the current is greater than or equal to the first preset value, the maximum current of the last screen is removed. The first preset value is 0.4.

[0059] S12. Statistically analyze all the capacitance data collected that day to obtain the maximum and minimum capacitance values, and then filter them using the following formula:

[0060]

[0061] When A C If the value is greater than or equal to the second preset value, the maximum capacitance value is removed. The second preset value is 0.4.

[0062] S13. Statistically analyze all the media loss data collected that day to obtain the maximum and minimum media loss values, and then filter them using the following formula:

[0063]

[0064] When AT When the value is greater than or equal to the third preset value, the maximum value of the medium loss is removed. The third preset value is 1.

[0065] S3. Calculate the average value of the filtered bad data:

[0066] S31, The formula for calculating the average current of the last screen is as follows:

[0067]

[0068] Among them, I v I is the average current of the last screen. n n1 represents the number of last screen current data collected.

[0069] S32, The formula for calculating the average capacitance is as follows:

[0070]

[0071] Among them, C v C is the average capacitance. n n is the capacitance, and n2 is the number of capacitance data collected;

[0072] S33. The formula for calculating the average value of dielectric loss is as follows:

[0073]

[0074] Among them, T v T represents the average value of the dielectric loss. n n represents the dielectric loss, and n3 represents the number of dielectric loss data collected.

[0075] S41. Conducting massive data trend analysis includes: taking several sets of end-screen current data, capacitance data, and dielectric loss data within the first time period for analysis. The current end-screen current data, capacitance data, and dielectric loss data are respectively I... n C n T n Then the current data, capacitance data, and dielectric loss data of the previous screen are respectively I n-1 C n-1 T n-1 , if I n-1 -I n C n-1 -C n T n-1 -T n If any value of n is greater than 0 and n is greater than the number of times threshold, a warning will be issued.

[0076] S42. Issuing early warning information based on trend analysis results includes: conducting m trend analyses at fixed intervals within the second time period, with each trend analysis result including whether there is an increasing trend in data for a certain number of days compared to the previous day, and issuing early warning information based on the trend analysis results.

[0077] The early warning information dissemination for data on end-screen current, capacitance, and dielectric loss includes:

[0078] If, during the first trend analysis in the second time period, there are daily data showing an increasing trend compared to the previous day for a period of more than or equal to the preset number of days, a special attention warning will be issued.

[0079] If, during the second trend analysis within the second time period, there are daily data showing an increasing trend compared to the previous day for a period of more than or equal to the preset number of days, then an on-site inspection and attention warning will be issued.

[0080] If, during the m-th trend analysis within the second time period, there are daily data showing an increasing trend compared to the previous day for a period greater than or equal to the preset number of days, a power outage inspection warning will be issued.

[0081] In this embodiment, massive data trend analysis is performed on the end-screen current, capacitance, and dielectric loss after average value calculation. Massive data trend analysis is performed on the end-screen current data, capacitance data, and dielectric loss data over 24 hours, 3 days, 5 days, and 7 days.

[0082] Trend analysis was performed on the 24 sets of end-screen current data each day: the current value is I. n The previous value was I. n-1 If I n-1 -I n Check if the number of >0 exceeds 18 times; perform trend analysis on the current data of the last screen within 3 days, 5 days and 7 days: analyze whether there are 2 days within 3 days with data that increase compared to the previous day, 3 days within 5 days with data that increase compared to the previous day, and 5 days within 7 days with data that increase compared to the previous day.

[0083] Trend analysis was performed on 24 sets of daily capacitance data: the current value is C. n The previous value is C n-1 If C n-1 -C n Check if the number of >0 exceeds 18 times; perform trend analysis on the capacity data for 3 days, 5 days and 7 days: analyze whether there are 2 days in 3 days where the data is higher than the previous day, 3 days in 5 days where the data is higher than the previous day, and 5 days in 7 days where the data is higher than the previous day.

[0084] Trend analysis was performed on 24 sets of media loss data each day: the current value is T. n The previous value is T n-1 If T n-1 -T n Check if the number of >0 exceeds 18 times; perform trend analysis on the media loss data within 3 days, 5 days and 7 days: analyze whether there are 2 days within 3 days with data that increase compared to the previous day, 3 days within 5 days with data that increase compared to the previous day, and 5 days within 7 days with data that increase compared to the previous day.

[0085] Early warning information is issued based on the current data of the last screen:

[0086] If I n-1 -I n If the number of occurrences of >0 exceeds 18, a warning will be issued.

[0087] If data shows an increase on two out of three days compared to the previous day, a special warning will be issued.

[0088] If data shows an increase for 3 out of 5 days compared to the previous day, an on-site inspection and monitoring alert will be issued.

[0089] If data shows an increase for 5 out of 7 days compared to the previous day, a power outage inspection warning will be issued.

[0090] Early warning information will be issued based on the power capacity data.

[0091] If C n-1 -C n If the number of occurrences of >0 exceeds 18, a warning will be issued.

[0092] If data shows an increase on two out of three days compared to the previous day, a special warning will be issued.

[0093] If data shows an increase for 3 out of 5 days compared to the previous day, an on-site inspection and monitoring alert will be issued.

[0094] If data shows an increase for 5 out of 7 days compared to the previous day, a power outage inspection warning will be issued.

[0095] Early warning information will be issued based on media loss data.

[0096] If T n-1 -T n If the number of occurrences of >0 exceeds 18, a warning will be issued.

[0097] If data shows an increase on two out of three days compared to the previous day, a special warning will be issued.

[0098] If data shows an increase for 3 out of 5 days compared to the previous day, an on-site inspection and monitoring alert will be issued.

[0099] If data shows an increase for 5 out of 7 days compared to the previous day, a power outage inspection warning will be issued.

[0100] Example 3

[0101] Reference Figure 3 This is the third embodiment of the present invention, which is an online monitoring data early warning system based on the trend changes of massive data: including a data acquisition module 100, a data storage module 200, a data change calculation module 300, and an early warning information display and release module 400.

[0102] The data acquisition module 100 includes a real-time acquisition unit 101 for capacitive equipment end-screen current, a real-time acquisition unit 102 for capacitive equipment capacitance, a real-time acquisition unit 103 for capacitive equipment dielectric loss, a real-time acquisition unit 104 for system voltage phase, and a real-time acquisition unit 105 for system frequency, which respectively acquire the capacitive equipment end-screen current, capacitance, dielectric loss, system voltage phase, and frequency data in real time.

[0103] The data storage module 200 includes a capacitive device end-screen current storage unit 201, a capacitive device capacitance storage unit 202, a capacitive device dielectric loss storage unit 203, a system voltage phase storage unit 204, and a system frequency storage unit 205, which respectively store the capacitive device end-screen current, capacitance, dielectric loss, system voltage phase, and frequency data.

[0104] The data change calculation module 300 includes a capacitive device end screen current change calculation unit 301, a capacitive device capacitance data change calculation unit 302, a capacitive device dielectric loss data change calculation unit 303, a system voltage phase data change calculation unit 304, and a system frequency data change unit 305, which respectively calculate the changes in capacitive device end screen current, capacitance, dielectric loss, system voltage phase, and frequency data.

[0105] The early warning information release module 400 includes an icon flashing unit 401, an audio reminder unit 402, and an information sending unit 403, which realizes the audio-visual display and transmission of early warning information.

[0106] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An online monitoring data early warning method based on the trend changes of massive data, characterized in that: include, Massive amounts of real-time data are collected and stored, including end-screen current, capacitance, dielectric loss, system voltage phase, and system frequency. Filter out bad data from the massive amount of stored data, including end screen current, capacitance, and dielectric loss data; Calculate the average value of the filtered bad data; Conduct trend analysis on massive amounts of data and issue early warning information based on the trend analysis results; The bad data in the filtered and stored massive amount of data includes: All the current data collected on the last screen that day were statistically analyzed to obtain the maximum and minimum values ​​of the last screen current. The following formula was used for filtering: when When the current value of the last screen is greater than or equal to the first preset value, the maximum current value of the last screen is discarded. The aforementioned massive data trend analysis includes: Within the first time period, several sets of end-screen current data, capacitance data, and dielectric loss data are analyzed. The current end-screen current data, capacitance data, and dielectric loss data are as follows: , , Then the current data, capacitance data, and dielectric loss data of the previous last screen are respectively , , ,like , , If any value of n is greater than 0 and n is greater than the number of times threshold, a warning will be issued. The issuance of early warning information based on trend analysis results includes: conducting m trend analyses at fixed intervals within the second time period; the trend analysis includes whether each trend analysis result shows an increasing trend in data for a certain number of days compared to the previous day; issuing early warning information on the end-screen current, capacitance, and dielectric loss data based on the trend analysis results; and further includes: If, during the first trend analysis in the second time period, there are daily data showing an increasing trend compared to the previous day for a period of more than or equal to the preset number of days, a special attention warning will be issued. If, during the second trend analysis within the second time period, there are daily data showing an increasing trend compared to the previous day for a period of more than or equal to the preset number of days, then an on-site inspection and attention warning will be issued. If, during the m-th trend analysis within the second time period, there are daily data showing an increasing trend compared to the previous day for a period greater than or equal to the preset number of days, a power outage inspection warning will be issued.

2. The online monitoring data early warning method based on massive data trend changes as described in claim 1, characterized in that: The bad data in the filtered and stored massive amount of data also includes: All collected capacitance data for the day were statistically analyzed to determine the maximum and minimum capacitance values, which were then filtered using the following formula: when If the value is greater than or equal to the second preset value, the maximum capacitance value is discarded.

3. The online monitoring data early warning method based on massive data trend changes as described in claim 2, characterized in that: The bad data in the filtered and stored massive amount of data also includes: All collected media loss data for the day were statistically analyzed to determine the maximum and minimum media loss values, which were then filtered using the following formula: when If the value is greater than or equal to the third preset value, the maximum value of the medium loss is removed.

4. The online monitoring data early warning method based on massive data trend changes as described in claim 3, characterized in that: The step of calculating the average value of the filtered bad data includes: The formula for calculating the average current of the last screen is as follows: in, This represents the average current of the last screen. This is the current of the last screen. The number of current data collected at the end screen; The formula for calculating the average capacitance is as follows: in, This is the average capacitance. For capacitance, The number of capacitance data collected; The formula for calculating the average dielectric loss is as follows: in, This represents the average value of the dielectric loss. For dielectric loss, This refers to the number of media loss data collected.

5. An online monitoring data early warning system based on the trend changes of massive data, employing the online monitoring data early warning method based on the trend changes of massive data as described in any one of claims 1 to 4, characterized in that: The online monitoring data early warning system includes a data acquisition module (100), a data storage module (200), a data change calculation module (300), and an early warning information display and release module (400). The data acquisition module (100) includes a real-time acquisition unit for capacitive equipment end screen current (101), a real-time acquisition unit for capacitive equipment capacitance (102), a real-time acquisition unit for capacitive equipment dielectric loss (103), a real-time acquisition unit for system voltage phase (104), and a real-time acquisition unit for system frequency (105), which respectively acquire the capacitive equipment end screen current, capacitance, dielectric loss, system voltage phase, and frequency data in real time; The data storage module (200) includes a capacitive device end-screen current storage unit (201), a capacitive device capacitance storage unit (202), a capacitive device dielectric loss storage unit (203), a system voltage phase storage unit (204), and a system frequency storage unit (205), which respectively store the capacitive device end-screen current, capacitance, dielectric loss, system voltage phase, and frequency data; The data change calculation module (300) includes a capacitive device end screen current change calculation unit (301), a capacitive device capacitance data change calculation unit (302), a capacitive device dielectric loss data change calculation unit (303), a system voltage phase data change calculation unit (304), and a system frequency data change unit (305), which respectively calculate the changes in capacitive device end screen current, capacitance, dielectric loss, system voltage phase, and frequency data.

6. The online monitoring data early warning system based on massive data trend changes as described in claim 5, characterized in that: The warning information display and publishing module (400) includes an icon flashing unit (401), an audio reminder unit (402), and an information sending unit (403) to realize the audio-visual display and sending of warning information.

Images