Time sequence noise reduction method and device based on improved singular spectrum analysis
A time series, singular spectrum technology, applied in the direction of instruments, complex mathematical operations, design optimization/simulation, etc., can solve problems such as difficulty in distinguishing noise components, and achieve the effect of strengthening recognition ability, improving stability and noise reduction ability
Pending Publication Date: 2021-12-24
GUANGZHOU UNIVERSITY
View PDF3 Cites 4 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0006] In order to overcome the deficiencies in the above-mentioned prior art, one object of the present invention is to provide a time series noise reduction method and device based on improved singular spectrum analysis, by utilizing the noise data identification ability of detrend analysis (DFA), identify and Eliminate the noise components in the SSA sub-sequence, overcoming the defect that the traditional SSA algorithm is difficult to distinguish the noise components
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
[0096] The following is an example to illustrate the time series noise reduction method based on the improved singular spectrum analysis technology of the present invention. In this embodiment, its specific implementation process is as follows:
[0097] S101, collect a city gas load sequence as basic test data, such as image 3 shown.
[0098] S102, setting the test range of the window length m to [10, 100], and setting the selection threshold S of the SSA algorithm to a value range of [0.5, 0.95]. Then, SSA algorithm and SSA-DFA algorithm are used to denoise the test data respectively.
[0099] S103, calculate the sample entropy and signal-to-noise ratio of each noise reduction result of the SSA and SSA-DFA algorithms, and obtain the sensitivity analysis results as shown in Table 1, wherein the sample entropy is used to describe the complexity of the time series, the larger the value, The greater the complexity; the signal-to-noise ratio is used to describe the proportion ...
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 sequence noise reduction method and device based on improved singular spectrum analysis. The method comprises the following steps: S1, constructing a trajectory matrix for an original time sequence according to a set window length; S2, constructing a symmetric matrix by using the constructed trajectory matrix, performing singular value decomposition on the symmetric matrix to obtain a feature vector and a feature value, and obtaining a right singular vector of the symmetric matrix and a subclass matrix of the trajectory matrix; S3, adopting a diagonal averaging method for each subclass matrix to obtain an SSA decomposition subsequence; S4, analyzing whether each sub-sequence belongs to a white noise component, a pink noise component or a Brown noise component or not by adopting a de-trending analysis method; and S5, eliminating noise components in the original data and reconstructing residual main components to reduce noise of the original sequence.
Description
technical field [0001] The invention relates to a time series noise reduction method and device, in particular to a time series noise reduction method and device based on improved singular spectrum analysis. Background technique [0002] Noise data in time series is the main factor affecting the accuracy of time series forecasting. In order to remove the noise in the original time series and improve the prediction accuracy of the model, it is often necessary to use noise reduction technology to preprocess the original time series, and the noise reduction algorithm using singular spectrum analysis is widely used. For example, the Chinese patent application with publication number CN108647824A uses singular spectrum analysis (abbreviation: SSA) to filter the noise component of the electricity price sequence, and reconstructs the trend component and oscillation component to obtain the training data input of the extreme learning machine, thereby improving the extreme learning ma...
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): G06F30/20G06F17/18G06F119/02
CPCG06F30/20G06F17/18G06F2119/02
Inventor韦南殷丽华李超孙哲罗熙
OwnerGUANGZHOU UNIVERSITY