Anomaly detection method for continuous space-time refueling data
An anomaly detection and time-series data technology, applied in digital data information retrieval, neural learning methods, electrical digital data processing, etc., can solve the problems of spatio-temporal data processing in the same frame, difficult to define, difficult to label, etc.
Active Publication Date: 2019-09-13
XINJIANG TECHN INST OF PHYSICS & CHEM CHINESE ACAD OF SCI
View PDF9 Cites 10 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Solve the problems that are extremely difficult to define and label in real application scenarios, and it is difficult for existing methods to process spatio-temporal data through the same framework
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0028] An abnormality detection method for continuous spatio-temporal refueling data according to the present invention, the method involves real-time data collection for multiple gas stations, based on the combination of statistics and machine learning, through the preset unsupervised time series data anomaly detection module, semi-supervised time series data anomaly detection module and multi-view based spatio-temporal depth anomaly detection module, the three anomaly detection modules mine and detect potential abnormal objects, and finally discriminate abnormal objects through weighted methods. The specific operation Follow these steps:
[0029] a. Anomaly detection module based on unsupervised time series data: automatically encode and extract features through AutoEncoder, then train through deep learning sequence model, and finally perform anomaly detection through residual criterion;
[0030] The automatic encoding machine is used to extract the features of the original ...
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
Login to View More Abstract
The invention relates to an anomaly detection method for continuous space-time refueling data. The method relates to real-time data acquisition for a plurality of gas stations. In combination of statistics and machine learning, potential abnormal objects are mined and detected through three abnormal detection modules including a preset time sequence data abnormal detection module based on unsupervised, a semi-supervised time sequence data abnormal detection module based on semi-supervised and a space-time depth abnormal detection module based on multiple views, and finally the abnormal objectsare judged in a weighting mode. The problems that abnormity is difficult to define and mark in a real application scene and space-time data is difficult to process through the same framework in an existing method are solved. The detection method provided by the invention can improve the detection accuracy of the spatio-temporal data exception in the refueling field, thereby meeting the spatio-temporal data analysis and processing requirements in the refueling field.
Description
technical field [0001] The invention discloses an anomaly detection device for continuous spatio-temporal refueling data. Specifically, it uses deep learning, data analysis and data visualization technologies to carry out automatic anomaly detection for spatio-temporal data of gas stations, and relates to information extraction and data preprocessing in the field of information technology , deep learning, data analysis and anomaly detection. Background technique [0002] Anomaly detection refers to the problem of finding patterns that do not meet expectations from data. These incompatible patterns have different names in different application fields, such as: anomalies, outliers, inconsistent observations, exceptions, deviations Among them, abnormality and outlier are the two most widely used terms, and sometimes they can be used interchangeably. Spatio-temporal anomaly detection belongs to the subfield of anomaly detection, and its purpose is to mine various patterns that ...
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
Login to View More IPC IPC(8): G06F16/2458G06K9/62G06N3/04G06N3/08
CPCG06F16/2458G06F16/2465G06N3/08G06N3/049G06N3/045G06F18/213G06F18/23G06F18/23213
Inventor 马博蒋同海周喜杨雅婷王磊马玉鹏赵凡王轶
Owner XINJIANG TECHN INST OF PHYSICS & CHEM CHINESE ACAD OF SCI