A sensor data sequence analysis reporting method

By analyzing and filtering the sensor data sequences, the terminal can accurately report abnormal data, solving the problem of abnormal data during long-term operation of IoT smart terminals and reducing the data analysis pressure on the platform.

CN116347376BActive Publication Date: 2025-12-19AEROSPACE SCI & IND INERTIA TECH CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202111580314.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-22
Publication Date
2025-12-19
Estimated Expiration
2041-12-22

AI Technical Summary

Technical Problem

During long-term operation, IoT smart terminals may report abnormal data, which increases the pressure on platform data analysis. This is especially true for sensors with triggering functions, where existing technologies are unable to effectively handle abnormal data.

Method used

By analyzing the sensor data sequence through the terminal, the changes in the initial and updated sequences are used to identify data anomalies. Filtering and multiple data acquisitions are employed to confirm data changes, thereby enabling accurate reporting of abnormal data.

Benefits of technology

It reduced the pressure on the platform for abnormal data analysis, improved the accuracy and reliability of data reporting, and reduced the platform's data processing burden.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116347376B_ABST
    Figure CN116347376B_ABST
Patent Text Reader

Abstract

The application provides a sensor data sequence analysis reporting method. After a terminal is installed and powered on, initial sequence 1 and sequence 2 are formed by using sensor data collected within a certain time after power-on. When the terminal reaches a reporting period, sensor data collection is performed. Multiple sensor data samples are collected each time data is collected, and the collected samples are filtered to obtain a current period collection value. Sequence 1 is updated using the current period collection value, and sequence 2 is updated using sensor data collected between two period collections. Sequence 1 and sequence 2 are used as auxiliary judgment basis for device triggering. The application uses a method of analyzing and reporting by using sensor data sequence characteristics, thereby reducing the pressure of platform-side abnormal data analysis.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things, and in particular to a method for analyzing and reporting sensor data sequences. BACKGROUND

[0002] An Internet of Things intelligent terminal (referred to as a terminal) usually has functions such as sensor data acquisition, data processing, and remote data transmission. For a sensor without a triggering function, the terminal sends a data acquisition command at a certain period, and the sensor responds. For a sensor with a triggering function, when the sensor does not detect a data change and sends a triggering signal, the terminal acquires sensor data and reports the data at a reporting period. When the sensor detects a change in its own data that reaches a threshold, the sensor sends a triggering signal (command or triggering IO pin change) to the terminal, and the terminal acquires sensor data immediately after receiving the signal and reports the data immediately after determining that the acquisition result exceeds the set threshold.

[0003] After data acquisition is completed, the terminal performs preliminary processing on the data and transmits the data remotely to an Internet of Things platform. However, due to influences such as vibration, human interference, electromagnetic interference, and sensor faults, the terminal may report abnormal data during long-term operation, which affects data analysis and early warning judgment on the platform. Usually, the processing of such abnormal data is performed on the platform, and when the number of devices is large, this situation causes great pressure on data analysis on the platform. Therefore, for a terminal that acquires a sensor with a triggering function, it is necessary to design a method for analyzing and reporting sensor data sequences based on the characteristics of the sequences to reduce the pressure on abnormal data analysis on the platform. SUMMARY

[0004] The present application provides a sensor data sequence analysis and reporting method, which enables the terminal to determine abnormal data based on sequence analysis of acquired data and reduces the pressure on abnormal data analysis on the platform.

[0005] A sensor data sequence analysis and reporting method technical solution is as follows:

[0006] After the terminal is installed and powered on, sensor data acquired at a certain time interval is used as initial sequence 1 and initial sequence 2. When the terminal reaches a reporting period, sensor data is acquired, a plurality of sensor data samples are acquired each time data is acquired, and the acquired samples are filtered to obtain a current period acquisition value. Sequence 1 is updated using each period acquisition value, and sequence 2 is updated using sensor data acquired between two period acquisitions. Sequence 1 and sequence 2 are used as auxiliary judgment criteria for device triggering.

[0007] Further, the method specifically includes the following steps:

[0008] (1) Terminal power-on initial value confirmation, using a certain time interval to collect sensor data as the initial sequence 1 and initial sequence 2;

[0009] (2) The terminal does not reach the reporting period according to the collection period to collect sensor data to form sequence 2;

[0010] (3) The terminal reaches the collection period to collect sensor data;

[0011] (4) The terminal performs data change detection to determine whether the sensor data change is normal;

[0012] (5) The data change is normal, report the collected data and update sequence 1, end this cycle collection process;

[0013] (6) The data change is abnormal and the trigger collection flag is triggered to report, report the collected data and update sequence 1, end this cycle collection process;

[0014] (7) The data change is abnormal and the trigger collection flag is triggered to collect, delay and re-collect data, report the data and update sequence 1, end this cycle collection process;

[0015] (8) The data change is abnormal and there is no sensor trigger signal, analyze the change trend of sequence 1 and sequence 2 data;

[0016] (9) When the change trend is a slow change trend data, re-collect data, report the data and update sequence 1, end this cycle collection process;

[0017] (10) When the change trend is a jump data, re-collect data N times to form sequence 3 and re-judge the data change trend;

[0018] (11) When the sequence 3 change trend is still a jump data, report the data and report the data abnormal fault, update sequence 1 and end this cycle collection process;

[0019] (12) When the sequence 3 change trend is still a slow change trend data, report the data, update sequence 1 and end this cycle collection process;

[0020] (13) After the cycle collection process is completed, clear the trigger collection flag;

[0021] (14) When the terminal receives a sensor trigger signal, trigger collection, set the trigger collection flag to trigger collection, after data change detection, if the change value does not exceed the threshold, the trigger process is ended;

[0022] (15) After data change detection, if the change value exceeds the set threshold, immediately report the data, set the trigger collection flag to trigger report, and update sequence 1;

[0023] (16) Trigger report after a certain time when no new trigger is detected to determine the current time to the next time interval of periodic collection, re-collection of data N times and update sequence 2.

[0024] Further, the normal change criterion is that the change amount compared with the initial value is not more than the threshold value, and the change amount compared with the last collection value is not more than the threshold value.

[0025] The present application is aimed at the intelligent terminal with the triggered function sensor, the method of analyzing and reporting by using the sensor data sequence characteristics, thereby reducing the pressure of platform end abnormal data analysis. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 The flow chart of sensor data sequence analysis and reporting is given.

[0027] Figure 2 The flow chart of sensor initial value determination is given. DETAILED DESCRIPTION

[0028] The present application will be further described in detail below in combination with the drawings.

[0029] The present application uses the triggered function of the sensor as an auxiliary determination, analyzes and processes the data sequence within a certain time to remove the influence of abnormal data. The specific technical scheme adopted is as follows.

[0030] The terminal is installed and powered on, and the terminal collects sensor data multiple times and automatically confirms the initial value when powered on. The terminal collects sensor data when reaching the reporting period, collects multiple samples each time of data collection, and obtains the current sensor data collection value after filtering the collected samples.

[0031] The N times of periodic collection data before each periodic collection are updated as sequence 1, and the N times of periodic collection data between every two periodic collections are updated as sequence 2. The data change detection is performed when collecting data in the reporting period to determine whether the data change is normal. The reference of the data change detection is the initial value and the last periodic collection value. The normal change criterion is that the change amount compared with the initial value is not more than the threshold value, and the change amount compared with the last collection value is not more than the threshold value. The data change is normal, and then it is directly reported.

[0032] When the data changes abnormally, the terminal judges whether a trigger signal is received between the current cycle collection and the last cycle collection. If a trigger signal is received between the two collections, it is judged whether the trigger signal is reported at the same time. If the terminal reports the trigger, the data collected this time is directly reported. If the terminal does not report the trigger after the trigger, the terminal re-collects data after a certain time delay and reports the data. If no trigger signal is received between the two collections and the data changes abnormally, it is judged whether the data is trend data or jump data according to the data change trend of sequence 1 and sequence 2. If the data is slowly changing trend data, the data is re-collected and reported. If the data is jump data, the data is re-collected N times to form sequence 3 to judge the data change trend, and then the data is collected once more to report the data, and the sequence 3 is still judged as jump data when the data is reported.

[0033] When the terminal receives a sensor trigger signal, the terminal collects the data, detects the data change, and reports the data immediately when the change value exceeds the set threshold. At this time, if there is no new trigger within a certain time, the time interval between the current time and the next cycle collection is judged, the data is re-collected N times, and sequence 2 is updated.

[0034] As shown in Figure 1 , 2 , the flow of sensor data sequence analysis and reporting with trigger function is as follows.

[0035] (1) After the terminal is powered on, the initial value is confirmed, and the sensor data collected at a certain time interval is used as the initial sequence 1 and the initial sequence 2;

[0036] (2) When the terminal does not reach the reporting cycle, the sensor data is collected according to the collection cycle to form sequence 2;

[0037] (3) When the terminal reaches the collection cycle, the sensor data is collected;

[0038] (4) The terminal judges whether the sensor data change is normal;

[0039] (5) If the data changes normally, the collected data is reported and sequence 1 is updated, and the current cycle collection process is ended;

[0040] (6) If the data changes abnormally and the trigger collection flag is trigger reporting, the collected data is reported and sequence 1 is updated, and the current cycle collection process is ended;

[0041] (7) If the data changes abnormally and the trigger collection flag is trigger collection, the data is re-collected after a delay of 30 seconds, the data is reported and sequence 1 is updated, and the current cycle collection process is ended;

[0042] (8) Data change anomaly and no sensor trigger signal, data change trend analysis is performed on sequence 1 and sequence 2;

[0043] (9) When the change trend is slowly changing trend data, data collection is performed again, data reporting is performed, sequence 1 is updated, and the current cycle collection process is ended;

[0044] (10) When the change trend is jump data, data is collected N times to form sequence 3, and data change trend judgment is performed again;

[0045] (11) When the change trend of sequence 3 is still jump data, data reporting is performed, data abnormality fault is reported, sequence 1 is updated, and the current cycle collection process is ended;

[0046] (12) When the change trend of sequence 3 is still slowly changing trend data, data reporting is performed, sequence 1 is updated, and the current cycle collection process is ended;

[0047] (13) After the cycle collection process is ended, the trigger collection flag is cleared;

[0048] (14) When the terminal receives a sensor trigger signal, trigger collection is performed, the trigger collection flag is set to trigger collection, after data change detection, it is determined that the change value does not exceed the threshold, and the trigger process is ended;

[0049] (15) After data change detection, it is determined that the change value exceeds the set threshold, the data is immediately reported, the trigger collection flag is set to trigger reporting, and sequence 1 is updated;

[0050] (16) After trigger reporting, no new trigger is detected within a certain time, the time interval from the current time to the next cycle collection is determined, data is collected N times, and sequence 2 is updated.

[0051] The above specific embodiments are only used to explain and illustrate the technical solutions of the present application, but cannot constitute a limitation on the protection scope of the claims. It should be clear to those skilled in the art that any simple deformation or replacement of the technical solutions of the present application to obtain new technical solutions falls within the protection scope of the present application.

Claims

1. A method for analyzing and reporting sensor data sequences, characterized in that, After the terminal is installed and powered on, sensor data collected at certain time intervals are used as initial sequence 1 and initial sequence 2. When the terminal reaches the reporting cycle, sensor data is collected. Multiple sensor data samples are collected each time, and the collected samples are filtered to obtain the collected value for the current cycle. Sequence 1 is updated using the data collected in each cycle, and Sequence 2 is updated using the sensor data collected between two cycles; Sequence 1 and Sequence 2 are used as auxiliary judgment criteria for device triggering. The method specifically includes the following steps: (1) After the terminal is powered on, the initial value is confirmed, and the sensor data collected at a certain time interval is used as the initial sequence 1 and the initial sequence 2. (2) When the terminal has not reached the reporting cycle, sensor data is collected according to the collection cycle to form sequence 2; (3) The terminal collects sensor data when the acquisition cycle is reached; (4) The terminal performs data change detection to determine whether the sensor data changes are normal; (5) If the data changes normally, report the collected data and update sequence 1, and end the current periodic collection process; (6) If the data changes abnormally and the trigger collection flag is set to trigger reporting, report the collected data and update sequence 1, and end the current periodic collection process; (7) If the data changes abnormally and the trigger collection flag is set to trigger collection, the data collection will be restarted after a delay, the data will be reported and sequence 1 will be updated, and the current collection cycle will end. (8) If the data changes abnormally and there is no sensor trigger signal, perform trend analysis on the data of sequence 1 and sequence 2; (9) When the data shows a slow trend, re-collect the data, report the data and update sequence 1, and end the current collection process; (10) When the trend of change is a jump, the data is collected N times to form sequence 3 and the data trend is judged again; (11) When the trend of sequence 3 is still a jump data, data is reported and data abnormality fault is reported at the same time, sequence 1 is updated and the current cycle collection process is ended; (12) When the trend of sequence 3 is still a slow trend, data is reported, sequence 1 is updated and the current periodic data collection process ends; (13) Clear the trigger acquisition flag after the periodic acquisition process is completed; (14) When the terminal receives the sensor trigger signal, it performs trigger acquisition, sets the trigger acquisition flag to trigger acquisition, and the trigger process ends when the change value does not exceed the threshold after data change detection. (15) After the data change detection results that the change value exceeds the set threshold, the data is immediately reported, and the trigger collection flag is set to trigger reporting, and sequence 1 is updated at the same time; (16) If no new trigger is detected within a certain period of time after the trigger is reported, determine the time interval from the current time to the next periodic collection, collect data N times again and update sequence 2.

2. The sensor data sequence analysis and reporting method according to claim 1, characterized in that, The criteria for determining normal data changes are that the change compared to the initial data collection value does not exceed the threshold, and the change compared to the previous data collection value does not exceed the threshold.

Citation Information

Patent Citations

  • Service monitoring method and device, electronic equipment and computer readable storage medium

    CN113568802A