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

Incremental online characteristic extraction and analysis method and system

A feature extraction and analysis method technology, applied in the field of data analysis, can solve problems such as poor performance, large time overhead, and unclear feature structure, and achieve the effects of good scalability, poor scalability, and high efficiency

Inactive Publication Date: 2016-08-10
ZHEJIANG UNIV
View PDF4 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the feature extraction process of time series data, the data changes with time. The traditional method is to uniformly read all the data from the database and extract features every time the data is analyzed. This method takes a lot of time and has poor performance. The extracted feature structure is not clear, and the scalability of the system is poor

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
  • Incremental online characteristic extraction and analysis method and system
  • Incremental online characteristic extraction and analysis method and system
  • Incremental online characteristic extraction and analysis method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0028] The system architecture realized by the incremental online feature extraction and analysis method of the present invention is as follows: figure 1 shown, including the following specific steps:

[0029] (1) The incremental online feature extraction method of the present invention first sets up a database system module, the database system module mainly stores the relevant original time series data of industrial control, and the database system module provides data connection query with other modules simultaneously; The module includes multiple database tables, which mainly store time-series data, and the data storage structure is complex.

[0030] (2) The data preprocessing module performs data preprocessing on the original time series data set in step (1). The original time series data is incomplete and noisy, so the data preprocessing module can process these rough data and finally get a complete Correct timing data.

[0031] The data preprocessing module includes t...

Embodiment 2

[0040] (1) In this embodiment, the time-series data of the production of a certain drug is selected as the original data, and a corresponding database system is established.

[0041] (2) Preprocess the original data. The preprocessing mainly eliminates error values, null values ​​and values ​​with low acquisition frequency. The preprocessed values ​​are visualized as figure 2 and image 3 As shown, it corresponds to the data signal of the temperature of the upper part of the heating and the steam pressure of the heating in the process of drug production and refining.

[0042] (3) Feature extraction For time series data, the benchmark features are: average value, variance and time length, and the extracted feature data is stored in the feature data table.

[0043] (4) The trigger threshold of the incremental trigger is set to 1000, that is, when the amount of data corresponding to the database system module increases to 1000, a feature extraction operation is triggered to inc...

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 an incremental online characteristic extraction and analysis method and system. The method comprises the steps of storing industrial control time sequence data to a database; preprocessing original data to obtain clean data; extracting characteristic data by a characteristic extraction module and storing the characteristic data to a characteristic data table; monitoring an original data amount of a database system in real time by an incremental trigger; and when a triggering threshold is exceeded, triggering the characteristic extraction module to incrementally read original data and extract corresponding characteristics, and storing the original data and the corresponding characteristics into the characteristic data table. The method and system have the advantages that an incremental characteristic extraction and analysis framework is proposed; an incremental triggering supervision program is added; the monitoring of the database system and the incremental extraction of the characteristics are realized; the real-time online extraction of the characteristics and the online analysis of the data are finally realized; the efficiency is high; and the extendibility is high.

Description

technical field [0001] The invention belongs to the technical field of data analysis, and in particular relates to an incremental online feature extraction and analysis method and system. Background technique [0002] With the development of the Internet, more and more data are accumulated, and we are submerged in the data. Big data analysis and data mining have brought hope to people. Data mining is to identify effective, novel and potentially useful information from the data. , the process of eventually comprehensible patterns. A key step in data mining is feature extraction. Feature extraction is based on the original rough data for proper reduction and transformation, and extracts a feature set to represent the original rough data. The quality of the features directly affects the effect of the data mining model. [0003] In an industrial control process, there are many industrial control parameters. Industrial controllers and their associated I / O devices are central to...

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): G06F17/30
CPCG06F16/284G06F16/2465
Inventor 姜晓红包友军付钊李金昌
Owner ZHEJIANG UNIV
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