Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Sampling GPR method of continuous anomaly detection in collecting data flow of environment sensor

An environmental sensor and anomaly detection technology, applied in electrical digital data processing, special data processing applications, instruments, etc., can solve the problems of large amount of data calculation, inability to real-time anomaly detection, etc., to achieve the effect of increasing the efficiency of algorithm execution

Active Publication Date: 2013-10-02
哈尔滨工业大学高新技术开发总公司
View PDF0 Cites 52 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem that the abnormality detection cannot be performed in real time due to the large amount of data calculation in the traditional environmental sensor data flow anomaly detection, and provides a sampling GPR method for continuous anomaly detection in the collected data flow of the environmental sensor

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Sampling GPR method of continuous anomaly detection in collecting data flow of environment sensor
  • Sampling GPR method of continuous anomaly detection in collecting data flow of environment sensor
  • Sampling GPR method of continuous anomaly detection in collecting data flow of environment sensor

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0029] Specific implementation mode one: the following combination figure 1 Describe this embodiment, the sampling GPR method of continuous anomaly detection in the collection data flow of environment sensor described in this embodiment, it comprises the following steps:

[0030] Step 1: Set the size of the sliding window of the environmental sensor sensing data to N, and set the sampling ratio to B:1, and sample the first N*B data streams in the sliding window as offline data, and obtain N*B data as the initial forecast window data, and form the forecast window D according to the initial forecast window data T ;

[0031] Step 2: Use the next moment data element index adjacent to the current moment in the environmental sensor sensing data stream as the prediction window D T The input value of the prediction window D T Output the predicted mean value of the data elements in the environmental sensor sensing data stream at the next moment, and obtain the variance corresponding...

specific Embodiment approach 2

[0036] Specific implementation mode 2: This implementation mode further explains the implementation mode 1, and the prediction window D in this implementation mode T ={x i-Q ,x i-Q+1 ,...,x i}, where i represents the current moment, Q is the prediction window D T The size of , and Q=N*B, x is the prediction window data at the time corresponding to its subscript;

[0037] Data element x at the next moment i+1 The index of as the prediction window D T The input value of , get the data element x i+1 The predicted mean of and the variance q corresponding to the predicted mean;

[0038] The 95% confidence interval for determining the data elements at the next moment is

specific Embodiment approach 3

[0039] Specific implementation mode three: the following combination Figure 5 Describe this implementation mode. This implementation mode will further explain the second embodiment mode. In the fifth step of this embodiment mode, use the UBCS algorithm to determine whether the data element at the next moment is added to the prediction window D T The specific method is:

[0040] Step 51: According to the sliding window size N of the environmental sensor sensing data, set its sampling size to k, then the size of each basic window is N / k, and the ratio of N / k is the ratio of rounding down, then the first The data element index in one basic window is [1,2,3,...,N / k], and the data element index in the second basic window is [N / k+1,N / k+2, ...,2*N / k],..., the data element index in the I-th basic window is [(I-1)*N / k+1,...,I*N / k];

[0041] Step 52: From the prediction window D T Randomly select the next data element index in as the representative index;

[0042] Step 53: When the...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a sampling GPR method of continuous anomaly detection in a collecting data flow of an environment sensor, and belongs to the technical field of data monitoring of environment sensors. The sampling GPR method of the continuous anomaly detection in the collecting data flow of the environment sensor is used for solving the problem that anomaly detection can not be conducted in real time, wherein the problem is caused by the fact that data calculation amount is large in data flow anomaly detection of a traditional environment sensor. The sampling GPR method of the continuous anomaly detection in the collecting data flow of the environment sensor is based on a prediction-model method, a prediction model is built through historical data, the mean value and the confidence interval of current data are obtained, a current data value is compared with the confidence interval, and the current data value is regarded as exceptional data if the current data value exceeds the confidence interval. According to the sampling GPR method of the continuous anomaly detection in the collecting data flow of the environment sensor, less historical data are needed, algorithm operation efficiency is improved, and input training data are not required to be provided with category tags. The sampling GPR method of the continuous anomaly detection in the collecting data flow of the environment sensor can detect an exceptional situation in a self-adaptive mode according to real-time arrival data, and is applied to continuous exceptional data detection in collecting data flow of the environment sensor.

Description

technical field [0001] The invention relates to a sampling GPR method for continuous anomaly detection in the collected data stream of an environmental sensor, and belongs to the technical field of data monitoring of the environmental sensor. Background technique [0002] The wide application of environmental sensors puts forward higher requirements for real-time analysis of data. Environmental sensors are distributed in the environment they monitor, and the collected data continuously generates databases in the form of time series through telemetry technology. This continuously generated data form for the data flow model. At present, the real-time application of environmental sensor data has received widespread attention, but because environmental sensors are generally used in relatively harsh environments, their data is transmitted through communication networks, and the data transmission process will be affected by the environment and easily corroded. Undetected errors w...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F19/00
Inventor 刘大同彭宇庞景月罗清华彭喜元
Owner 哈尔滨工业大学高新技术开发总公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com