Time series data restoration method for smart campus based on coefficient of variation constraint
A technology of time series and coefficient of variation, applied in the field of data repair, can solve problems such as data distortion, and achieve the effect of improving accuracy, reducing repair difference, and improving authenticity and correctness
Pending Publication Date: 2022-01-14
HANGZHOU DIANZI UNIV
View PDF0 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0006] The purpose of the present invention is to overcome the problem that the traditional time series data repair method greatly modifies the original value, resulting in data distortion, and provides a smart campus time series data repair method based on the constraint of variation coefficient
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 Construction
[0033] The specific embodiments of the present invention will be further described in detail below in conjunction with the accompanying drawings.
[0034] A time series data restoration method for smart campuses based on variation coefficient constraints, the specific steps are described as follows figure 1 shown, where:
[0035] Step 1: Obtain the time series data to be repaired; window the time series data to obtain the data sequence under N windows, the size of each window is w, and the step size is 1;
[0036]Step 2: Select the clean data in the time series data to obtain the preprocessed time series data; calculate the coefficient of variation under each window of the preprocessed time series according to formula (1), and calculate the average coefficient of variation under each window The value serves as the coefficient of variation threshold v for the entire preprocessed time series.
[0037]
[0038] where V s (X i ) represents the data sequence X under the i-th...
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 discloses a time series data restoration method for a smart campus based on coefficient of variation constraint. According to the method, abnormal data points can be detected and candidate restoration values can be generated by using coefficient of variation constraints, and changes among all data points within a period of time are comprehensively considered. According to the method, the local optimal solution under the window can be generated by utilizing the minimum restoration principle, the restoration difference value of the abnormal data points is reduced as much as possible, and the authenticity and correctness of the restored data are improved. According to the method, the weighted average value of the candidate restoration sets obtained under all windows containing abnormal data points can be used as the final restoration value, the weight conforms to the principle of 'high near and low far', namely, the candidate restoration value obtained by the front window has better confidence than the candidate restoration value obtained by the rear window, and compared with other algorithms directly using the average value, the correctness of the restored data is improved. According to the invention, the condition of delayed arrival of data points can be processed by using a disordered processing algorithm, and the condition of data timestamp disorder is supported.
Description
technical field [0001] The invention belongs to the technical field of data restoration, and in particular relates to a time series data restoration method for smart campuses based on variation coefficient constraints. Background technique [0002] Time series is an important data type, and time series data refers to data collected at different times for describing phenomena that change over time. This type of data reflects the state or degree of a certain thing, phenomenon, etc. over time. Compared with traditional static data, due to the wide application of tools based on sensor networks, the growth rate of time series data can reach gigabytes per minute, resulting in a very large amount of data; in addition, time series data comes from real systems, and its influencing factors There are so many, so that a certain attribute changes randomly on an individual basis, and as a whole, it presents a slow and long-term trend of continuous rise, fall, and stay of the same nature ...
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 Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/215
CPCG06F16/215
Inventor 钱瑞祥张纪林陈军相袁俊峰金峻帆刘涛刘峰
Owner HANGZHOU DIANZI UNIV