Data visualization monitoring platform

Abstract

The invention relates to the technical field of data monitoring, and discloses a data visualization monitoring platform, the monitoring platform performs data processing by a monitoring system, and the monitoring system is composed of a data reading unit, an abnormal data calling unit, a data projection screen, a data backtracking unit, a data abnormity analysis unit and a suggestion unit. And the data projection screen is in network connection with other units of the monitoring system by using a local area network. The data projection screen is connected with other units of the monitoring system through a local area network, the data condition of a monitoring area is displayed in real time, after an abnormal data calling unit detects an abnormality, marked abnormal data is transmitted to the data projection screen for comparative analysis, and an analysis result is transmitted to the suggestion unit through a reporting function; and subsequent decisions and improvement measures are supported, and the beneficial effects of data visual projection, abnormal data monitoring and reporting are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of data monitoring, and in particular to a data visualization monitoring platform. Background Art

[0002] Big data monitoring refers to the acquisition, collection, and analysis of data through big data technology, and can accurately analyze information and effectively predict the dynamic trend of information development. Big data monitoring mainly revolves around massive data across the entire network, and most of them require the help of monitoring systems to assist in data analysis. Since data monitoring mainly revolves around massive data across the entire network, real-time data monitoring can be performed when monitoring data, but visual real-time monitoring cannot be performed. In order to solve the problem of not being able to perform visual monitoring, a data centralized monitoring platform with a publication number of CN117785060 A includes a login module, a centralized monitoring module, a data processing module, and a centralized monitoring and data storage module. The login module authenticates the user who controls the data centralized monitoring platform. The authentication is the basis for ensuring the safe operation of the platform. The centralized monitoring module is used to monitor and collect multiple data acquisition devices for monitoring data of various equipment in the substation, monitor and classify the platform in real time, and obtain the transaction content after data collection. The data processing module is used to automatically analyze and process the execution results according to the transaction content after data collection. The invention provides a data centralized monitoring platform, which realizes visual real-time monitoring, can quickly diagnose data faults, enables the platform to classify and store data, facilitates the classification and search of data transaction content after the platform monitors, enables the platform to automatically clean up and release space, and has intelligent circulation of memory space, which is highly intelligent.

[0003] Although the existing technology has solved the problem of visually projecting data, the visualization system itself does not have the function of discovering data anomalies. Anomalies in the visualized data projected onto the display screen still need to be monitored and discovered by users. Watching the display screen for a long time is very labor-intensive, and people may miss things when observing with the naked eye. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the deficiencies in the prior art, the present invention provides a data visualization monitoring platform, which has the advantages of data visualization projection and simultaneous monitoring of abnormal data and reporting, and directly projecting the abnormal data onto the display screen after discovery, so that users will not miss the abnormal data, thereby solving the problems of the above-mentioned technologies.

[0006] (II) Technical solution

[0007] To achieve the above object, the present invention provides the following technical solutions: a data visualization monitoring platform, wherein the monitoring platform processes data by a monitoring system, wherein the monitoring system is composed of a data reading unit, an abnormal data retrieval unit, a data projection screen, a data backtracking unit, a data abnormality analysis unit and a suggestion unit, and the data projection screen is connected to other units of the aforementioned monitoring system through a local area network;

[0008] The data reading unit is used to read the monitoring area connected to the monitoring platform. The data reading unit also records the data information in the monitoring area and sends the recorded data to the abnormal data retrieval unit at regular intervals;

[0009] The abnormal data retrieval unit is used to collect the rated growth rate, average value, rated decline rate, interval value and the positive and negative proportional data of the statistical information over time, and perform abnormal detection. The detected abnormal data is marked and retrieved to the data projection screen for display;

[0010] The data backtracking unit is used to read the abnormal data archive and backtrack the record information corresponding to the abnormal data, and the record information is: video, picture, audio and text information corresponding to the abnormal data in the time period when the abnormal data occurred;

[0011] The data anomaly analysis unit reads the data backtracking unit information and performs comparative analysis on the recorded information, where the comparison source is the recorded information under the normal state of the occurrence source, and analyzes the anomaly points;

[0012] The suggestion unit is used to read the abnormal points analyzed by the data abnormality analysis unit and send them to the data projection screen for display;

[0013] The abnormal detection threshold ycyz is ±5% of the rated growth rate, average value, rated decline rate, interval value and the positive and negative proportional data growing over time, that is, -5%>ycyz<5%.

[0014] Preferably, the data reading unit steps are:

[0015] S1.1. Determine the specific area that the monitoring platform needs to monitor, with custom cubic meters as the unit. Each unit contains multiple monitoring devices, and the equipment ratio is 10m 3 The number of monitoring devices jksl is more than two and less than five;

[0016] S1.2. Determine the data format and frequency output by each monitoring point or device, including sensor temperature, sensor humidity, image or video, sound, and operating status;

[0017] S1.3. Choose commercial monitoring software: SolarWinds, Splunk or PRTG Network Monitor, which is used to read the data of monitoring points or devices at regular intervals, so that the data reading unit can be connected to various data sources in the monitoring area. The data reading unit records the data information in the monitoring area, obtains the data and stores it in the local cache or database, including timestamp, data source and specific data value information;

[0018] S1.4. Design a scheduled task to send the recorded data to the abnormal data retrieval unit regularly. Use polling or event-based methods to determine the sending frequency and data volume according to monitoring requirements. Before sending the data, the data reading unit performs preliminary abnormality detection to check whether the data exceeds the preset threshold or meets the expected data range;

[0019] S1.5. Mark the abnormal data or send it to the abnormal data retrieval unit in priority for subsequent abnormal analysis and display.

[0020] Preferably, the timing rule of the data reading unit is:

[0021] When jksl>2, it is the basic timing unit, which is performed once every 1 hour. If the number of jksl is +1, the timing unit is -20% based on the basic timing unit. The principle expression of the data reading unit sending data regularly is:

[0022] T interval =T base ×0.8 (jksl-2)

[0023] in:

[0024] T base It is the basic timing unit, that is, once per hour, expressed as T base = 1 hour;

[0025] T interval The time interval for sending data regularly.

[0026] Preferably, the abnormal data retrieving unit specifically comprises the following steps:

[0027] S2.1. The data reading unit is responsible for regularly reading the data of the monitoring equipment, including sensor data temperature, humidity, image or video data, sound data and the operating status of the equipment, according to the predetermined time interval T interval Sent to data storage or local cache, including timestamp, data source and specific value information;

[0028] S2.2, the anomaly detection threshold ycyz is set to ±5% of the rated growth rate, average value, rated decline rate, interval value and the positive and negative proportional data of growth over time. These thresholds will serve as reference standards for anomaly detection;

[0029] S2.3. The abnormal data retrieval unit receives the data sent from the data reading unit, performs statistics on the received data, calculates the growth rate, average value, decline rate, interval value and other indicators of the data, and calculates its deviation from the preset threshold value ycyz, and applies a pre-set anomaly detection algorithm or rule, such as checking whether the data exceeds the threshold range of ±5%. If the data is identified as abnormal, the data is marked and subsequently processed.

[0030] Preferably, the rated growth rate expression in the abnormal data retrieval unit is:

[0031]

[0032] in,

[0033] edzz is the rated growth rate;

[0034] X(t) is the data value at time t, and X(t-1) is the data value at time t-1;

[0035] Rated growth rate anomaly detection:

[0036] |edzz|>ycyz marks X(t) as abnormal data.

[0037] Preferably, the average numerical expression in the abnormal data retrieval unit is:

[0038]

[0039] in,

[0040] Pjsz is the average value

[0041] X i is the i-th data point, n is the total number of data points;

[0042] Average value anomaly detection:

[0043] |pjsz-pjsz|>ycyz, then X(t) is marked as abnormal data.

[0044] Preferably, the expression of the rated drop rate in the abnormal data retrieval unit is:

[0045]

[0046] in:

[0047] edxj is the rated drop, X(t) is the data value at time t, and X(t-1) is the data value at time t-1;

[0048] The expression for rated drop rate anomaly detection is:

[0049] |edxj|>ycyz, then X(t) is marked as abnormal data.

[0050] Preferably, the interval numerical expression in the abnormal data retrieval unit is:

[0051] Range Value = max(X range )-min(X range )

[0052] The expression for interval numerical anomaly detection is:

[0053] If Range Value>ycyz, X(t) is marked as abnormal data.

[0054] Preferably, the inverse proportional data expression of the abnormal data retrieval unit that increases with time is:

[0055]

[0056] Where: zfbs is the inverse proportional data, X(t) is the data value at time t, and X(t-1) is the data value at time t-1;

[0057] The expression for anomaly detection of inversely proportional data growing over time is:

[0058] |zfbs-1|>ycyz, then X(t) is marked as abnormal data.

[0059] Preferably, the abnormality analysis unit specifically comprises the following steps:

[0060] S3.1. Read data: obtain the data information to be analyzed from the data traceability unit;

[0061] S3.2. Obtain normal record information: Obtain normal record information from the source of occurrence as a benchmark for comparative analysis;

[0062] S3.3, Comparative analysis: Compare and analyze the read data information with the normal recorded information to find abnormal points in the data;

[0063] S3.4. Anomaly analysis: Analyze the anomalies found to determine their causes and impacts.

[0064] Compared with the prior art, the present invention provides a data visualization monitoring platform, which has the following beneficial effects:

[0065] 1. The present invention is connected to other units of the monitoring system through a local area network via a data projection screen to display the data situation of the monitoring area in real time. After the abnormal data retrieval unit detects the abnormality, it transmits the marked abnormal data to the data projection screen, so that the operator can immediately notice the abnormal situation, thereby realizing real-time data visualization monitoring. The abnormal data retrieval unit detects the abnormality according to the preset abnormality detection algorithm, and the deviation of the rated growth rate, average value, etc. exceeds the set threshold value ±5%. Once the abnormality is found, the data is immediately marked and fed back to the data projection screen, so that the operator can quickly identify and understand the abnormal data. The data abnormality analysis unit uses the historical abnormal data provided by the data backtracking unit and the normal record information of the occurrence source to conduct comparative analysis and transmit the analysis results to the suggestion unit through the reporting function, so as to support subsequent decision-making and improvement measures, thereby achieving the beneficial effect of data visualization projection while monitoring abnormal data and reporting.

[0066] 2. The present invention is equipped with a real-time anomaly detection algorithm through the abnormal data retrieval unit in the data visualization monitoring platform, such as deviation detection of rated growth rate, average value, rated decline rate, etc., and detection of the rate of change of the inverse proportional data growing over time. Once the abnormal data is detected, the system immediately marks it and transmits it to the data projection screen. The abnormal data retrieval unit displays the marked abnormal data on the data projection screen in priority according to the preset priority rules. This mechanism ensures that the abnormal data can be significantly highlighted on the projection screen and will not be covered or submerged by other normal data, achieving the beneficial effect that the abnormal data will be directly projected on the display screen after it is discovered, and the user will not miss the abnormal data. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a schematic diagram of the structure of the present invention; DETAILED DESCRIPTION

[0068] 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.

[0069] See also Figure 1 , a data visualization monitoring platform, the monitoring platform is processed by a monitoring system, the monitoring system is composed of a data reading unit, an abnormal data retrieval unit, a data projection screen, a data backtracking unit, a data abnormality analysis unit and a suggestion unit, the data projection screen uses a local area network to connect to other units of the aforementioned monitoring system;

[0070] The data reading unit is used to read the monitoring area connected to the monitoring platform. The data reading unit also records the data information in the monitoring area and sends the recorded data to the abnormal data retrieval unit at regular intervals;

[0071] The abnormal data retrieval unit is used to collect the rated growth rate, average value, rated decline rate, interval value and the positive and negative proportional data of the statistical information over time, and perform abnormal detection. The detected abnormal data is marked and retrieved to the data projection screen for display;

[0072] The data backtracking unit is used to read the abnormal data archive and backtrack the record information corresponding to the abnormal data, and the record information is: video, picture, audio and text information corresponding to the abnormal data in the time period when the abnormal data occurred;

[0073] The data anomaly analysis unit reads the data backtracking unit information and performs comparative analysis on the recorded information, where the comparison source is the recorded information under the normal state of the occurrence source, and analyzes the anomaly points;

[0074] The suggestion unit is used to read the abnormal points analyzed by the data abnormality analysis unit and send them to the data projection screen for display;

[0075] The abnormal detection threshold ycyz is ±5% of the rated growth rate, average value, rated decline rate, interval value and the positive and negative proportional data growing over time, that is, -5%>ycyz<5%.

[0076] Preferably, the data reading unit steps are:

[0077] S1.1. Determine the specific area that the monitoring platform needs to monitor, with custom cubic meters as the unit. Each unit contains multiple monitoring devices, and the equipment ratio is 10m 3 The number of monitoring devices jksl is more than two and less than five;

[0078] S1.2. Determine the data format and frequency output by each monitoring point or device, including sensor temperature, sensor humidity, image or video, sound, and operating status;

[0079] S1.3. Choose commercial monitoring software: SolarWinds, Splunk or PRTG Network Monitor, which is used to read the data of monitoring points or devices at regular intervals, so that the data reading unit can be connected to various data sources in the monitoring area. The data reading unit records the data information in the monitoring area, obtains the data and stores it in the local cache or database, including timestamp, data source and specific data value information;

[0080] S1.4. Design a scheduled task to send the recorded data to the abnormal data retrieval unit regularly. Use polling or event-based methods to determine the sending frequency and data volume according to monitoring requirements. Before sending the data, the data reading unit performs preliminary abnormality detection to check whether the data exceeds the preset threshold or meets the expected data range;

[0081] S1.5. Mark the abnormal data or send it to the abnormal data retrieval unit in priority for subsequent abnormal analysis and display.

[0082] Customize monitoring areas in cubic meters, accurately manage and monitor equipment, quickly locate anomalies, support output of multiple data types, including sensor data, images, and videos, etc., to enhance the comprehensiveness and applicability of the monitoring system. Select well-known monitoring software: SolarWinds, Splunk or PRTG Network Monitor to ensure the stability of data reading and the richness of functions. Store data in local cache or database and record key information to facilitate fast access and subsequent analysis. Design scheduled tasks and event-based data collection methods, cooperate with preliminary anomaly detection to ensure the timeliness and accuracy of monitoring data.

[0083] Preferably, the timing rule of the data reading unit is:

[0084] When jksl>2, it is the basic timing unit, which is performed once every 1 hour. If the number of jksl is +1, the timing unit is -20% based on the basic timing unit. The principle expression of the data reading unit sending data regularly is:

[0085] T interval =T base ×0.8 (jksl-2)

[0086] in:

[0087] T base It is the basic timing unit, that is, once per hour, expressed as T base = 1 hour;

[0088] T interval The time interval for sending data regularly.

[0089] According to the number of monitoring points or devices jksl, the time interval for sending data is dynamically adjusted. When the number of monitoring devices exceeds two, jksl>2, the frequency of sending data will be adjusted according to the formula T interval =T base ×0.8 (jksl-2)This flexibility allows the system to automatically adjust the frequency of data transmission according to actual monitoring needs, effectively balancing the relationship between data timeliness and system load. By dynamically adjusting the time interval for scheduled transmission, the system can manage resources more effectively and reduce unnecessary data transmission and processing costs. This optimization can reduce the energy consumption of system operation, extend the service life of equipment and network resources, and thus improve the overall cost-effectiveness of the system.

[0090] Preferably, the abnormal data retrieving unit specifically comprises the following steps:

[0091] S2.1. The data reading unit is responsible for regularly reading the data of the monitoring equipment, including sensor data temperature, humidity, image or video data, sound data and the operating status of the equipment, according to the predetermined time interval T interval Sent to data storage or local cache, including timestamp, data source and specific value information;

[0092] S2.2, the anomaly detection threshold ycyz is set to ±5% of the rated growth rate, average value, rated decline rate, interval value and the positive and negative proportional data of growth over time. These thresholds will serve as reference standards for anomaly detection;

[0093] S2.3. The abnormal data retrieval unit receives the data sent from the data reading unit, performs statistics on the received data, calculates the growth rate, average value, decline rate, interval value and other indicators of the data, and calculates its deviation from the preset threshold value ycyz, and applies a pre-set anomaly detection algorithm or rule, such as checking whether the data exceeds the threshold range of ±5%. If the data is identified as abnormal, the data is marked and subsequently processed.

[0094] Preferably, the rated growth rate expression in the abnormal data retrieval unit is:

[0095]

[0096] in,

[0097] edzz is the rated growth rate;

[0098] X(t) is the data value at time t, and X(t-1) is the data value at time t-1;

[0099] Rated growth rate anomaly detection:

[0100] |edzz>ycyz marks X(t) as abnormal data.

[0101] Preferably, the average numerical expression in the abnormal data retrieval unit is:

[0102]

[0103] in,

[0104] Pjsz is the average value

[0105] X i is the i-th data point, n is the total number of data points;

[0106] Average value anomaly detection:

[0107] |pjsz-pjsz|>ycyz, then X(t) is marked as abnormal data.

[0108] Preferably, the expression of the rated drop rate in the abnormal data retrieval unit is:

[0109]

[0110] in:

[0111] edxj is the rated drop, X(t) is the data value at time t, and X(t-1) is the data value at time t-1;

[0112] The expression for rated drop rate anomaly detection is:

[0113] |edxj|>ycyz, then X(t) is marked as abnormal data.

[0114] Preferably, the interval numerical expression in the abnormal data retrieval unit is:

[0115] Range Value = max(X range )-min(X range )

[0116] The expression for interval numerical anomaly detection is:

[0117] If Range Value>ycyz, X(t) is marked as abnormal data.

[0118] Preferably, the inverse proportional data expression of the abnormal data retrieval unit that increases with time is:

[0119]

[0120] Where: zfbs is the inverse proportional data, X(t) is the data value at time t, and X(t-1) is the data value at time t-1;

[0121] The expression for anomaly detection of inversely proportional data growing over time is:

[0122] |zfbs-1|>ycyz, then X(t) is marked as abnormal data.

[0123] Preferably, the abnormality analysis unit specifically comprises the following steps:

[0124] S3.1. Read data: obtain the data information to be analyzed from the data traceability unit;

[0125] S3.2. Obtain normal record information: Obtain normal record information from the source of occurrence as a benchmark for comparative analysis;

[0126] S3.3, Comparative analysis: Compare and analyze the read data information with the normal recorded information to find abnormal points in the data;

[0127] S3.4. Anomaly analysis: Analyze the anomalies found to determine their causes and impacts.

[0128] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A data visualization monitoring platform, characterized in that: The monitoring platform is processed by the monitoring system, which is composed of a data reading unit, an abnormal data retrieval unit, a data projection screen, a data backtracking unit, a data abnormality analysis unit and a suggestion unit. The data projection screen is connected to the other units of the aforementioned monitoring system through a local area network. The data reading unit is used to read the monitoring area connected to the monitoring platform. The data reading unit also records the data information in the monitoring area and sends the recorded data to the abnormal data retrieval unit at regular intervals; The abnormal data retrieval unit is used to collect the rated growth rate, average value, rated decline rate, interval value and the positive and negative proportional data of the statistical information over time, and perform abnormal detection. The detected abnormal data is marked and retrieved to the data projection screen for display; The data backtracking unit is used to read the abnormal data archive and backtrack the record information corresponding to the abnormal data, and the record information is: video, picture, audio and text information corresponding to the abnormal data in the time period when the abnormal data occurred; The data anomaly analysis unit reads the data backtracking unit information and performs comparative analysis on the recorded information, where the comparison source is the recorded information under the normal state of the occurrence source, and analyzes the anomaly points; The suggestion unit is used to read the abnormal points analyzed by the data abnormality analysis unit and send them to the data projection screen for display; The abnormality detection threshold ycyz is ±5% of the rated growth rate, average value, rated decline rate, interval value and the positive and negative proportional data growing with time, that is, -5%>ycyz<5%.

2. A data visualization monitoring platform according to claim 1, characterized in that: The steps of the data reading unit are as follows: S1.

1. Determine the specific area that the monitoring platform needs to monitor, with custom cubic meters as the unit. Each unit contains multiple monitoring devices, and the equipment ratio is 10m 3 The number of monitoring devices jksl is more than two and less than five; S1.

2. Determine the data format and frequency output by each monitoring point or device, including sensor temperature, sensor humidity, image or video, sound, and operating status; S1.

3. Choose commercial monitoring software: SolarWinds, Splunk or PRTG Network Monitor, which is used to read the data of monitoring points or devices at regular intervals, so that the data reading unit can be connected to various data sources in the monitoring area. The data reading unit records the data information in the monitoring area, obtains the data and stores it in the local cache or database, including timestamp, data source and specific data value information; S1.

4. Design a scheduled task to send the recorded data to the abnormal data retrieval unit regularly. Use polling or event-based methods to determine the sending frequency and data volume according to monitoring requirements. Before sending the data, the data reading unit performs preliminary abnormality detection to check whether the data exceeds the preset threshold or meets the expected data range; S1.

5. Mark the abnormal data or send it to the abnormal data retrieval unit in priority for subsequent abnormal analysis and display.

3. A data visualization monitoring platform according to claim 2, characterized in that: The data reading unit timing rule is: When jksl>2, it is the basic timing unit, which is performed once every 1 hour. If the number of jksl is +1, the timing unit is -20% based on the basic timing unit. The principle expression of the data reading unit sending data regularly is: T interval =T base ×0.8 (jksl-2) in: T base It is the basic timing unit, that is, once per hour, expressed as T base = 1 hour; T interval The time interval for sending data regularly.

4. A data visualization monitoring platform according to claim 1, characterized in that: The specific steps of the abnormal data retrieval unit are as follows: S2.

1. The data reading unit is responsible for regularly reading the data of the monitoring equipment, including sensor data temperature, humidity, image or video data, sound data and the operating status of the equipment, according to the predetermined time interval T interval Sent to data storage or local cache, including timestamp, data source and specific value information; S2.2, the anomaly detection threshold ycyz is set to ±5% of the rated growth rate, average value, rated decline rate, interval value and the positive and negative proportional data of growth over time. These thresholds will serve as reference standards for anomaly detection; S2.

3. The abnormal data retrieval unit receives the data sent from the data reading unit, performs statistics on the received data, calculates the growth rate, average value, decline rate, interval value and other indicators of the data, and calculates its deviation from the preset threshold value ycyz, and applies a pre-set anomaly detection algorithm or rule, such as checking whether the data exceeds the threshold range of ±5%. If the data is identified as abnormal, the data is marked and subsequently processed.

5. A data visualization monitoring platform according to claim 4, characterized in that: The rated growth rate expression in the abnormal data retrieval unit is: in, edzz is the rated growth rate; X(t) is the data value at time t, and X(t-1) is the data value at time t-1; Rated growth rate anomaly detection: |edzz|>ycyz marks X(t) as abnormal data.

6. A data visualization monitoring platform according to claim 4, characterized in that: The average numerical expression in the abnormal data retrieval unit is: in, Pjsz is the average value X i is the i-th data point, n is the total number of data points; Average value anomaly detection: |pjsz-pjsz|>ycyz, then X(t) is marked as abnormal data.

7. A data visualization monitoring platform according to claim 4, characterized in that: The rated drop rate expression in the abnormal data retrieval unit is: in: edxj is the rated drop, X(t) is the data value at time t, and X(t-1) is the data value at time t-1; The expression for rated drop rate anomaly detection is: |edxj|>ycyz, then X(t) is marked as abnormal data.

8. A data visualization monitoring platform according to claim 4, characterized in that: The interval numerical expression in the abnormal data retrieval unit is: Range Value=max(X range )-min(X range ) The expression for interval numerical anomaly detection is: If Range Value>ycyz, X(t) is marked as abnormal data.

9. A data visualization monitoring platform according to claim 4, characterized in that: The expression of the inverse proportional data that increases with time in the abnormal data retrieval unit is: Where: zfbs is the inverse proportional data, X(t) is the data value at time t, and X(t-1) is the data value at time t-1; The expression for anomaly detection of inversely proportional data growing over time is: |zfbs-1|>ycyz, then X(t) is marked as abnormal data.

10. A data visualization monitoring platform according to claim 1, characterized in that: The specific steps of the abnormal analysis unit are: S3.

1. Read data: obtain the data information to be analyzed from the data traceability unit; S3.

2. Obtain normal record information: Obtain normal record information from the source of occurrence as a benchmark for comparative analysis; S3.3, Comparative analysis: Compare and analyze the read data information with the normal recorded information to find abnormal points in the data; S3.

4. Anomaly analysis: Analyze the anomalies found to determine their causes and impacts.

Images