Short-term wind speed prediction method integrating deep learning and data denoising

A technology of wind speed forecasting and deep learning, applied in forecasting, data processing applications, electrical digital data processing, etc., can solve problems such as difficult to use, affecting the initial decomposition subsequence value, and reducing accuracy

Active Publication Date: 2020-06-05
CHONGQING UNIV
View PDF4 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the newly obtained data will greatly affect the value of the initial decomposed subsequence, so the forecasting method based on EMD is difficult to use in actual forecasting, which leads to the final forecasting The accuracy of

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
  • Short-term wind speed prediction method integrating deep learning and data denoising
  • Short-term wind speed prediction method integrating deep learning and data denoising
  • Short-term wind speed prediction method integrating deep learning and data denoising

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0063] see Figure 1 to Figure 3 , a short-term wind speed prediction method integrating deep learning and data denoising, mainly includes the following steps:

[0064] 1) Obtain the original wind speed sequence, and preprocess the original wind speed sequence to obtain the wind speed sequence A:{α 1 ,α 2 ,…,α l}. l is the length of the data sequence. The preprocessing includes linear interpolation and normalization. alpha l is the wind speed.

[0065] 2) Use discrete wavelet transform to decompose the wind speed sequence A to obtain the proportional coefficient c containing low-frequency information m,n and wavelet coefficients d containing high-frequency information m,n .

[0066] The main steps of decomposing the wind speed sequence A using discrete wavelet transform are as follows:

[0067] 2.1) Set the wavelet function as a binary wavelet, that is, let a=2 m ,b=n2 m ,m,n∈Z. a is a scale factor, b is a time shift factor, m is a scale, and n is a parameter rela...

Embodiment 2

[0116] An experiment to verify a short-term wind speed prediction method integrating deep learning and data denoising mainly includes the following steps:

[0117] 1) Obtain the original wind speed sequence; perform linear interpolation of missing values ​​and data normalization on the collected original wind speed sequence through data preprocessing, and obtain the sorted wind speed sequence A:{α 1 ,α 2 ,…,α l} (l is the length of the data sequence).

[0118] Since there are many missing and erroneous values ​​in the wind speed raw data, the data is preprocessed using a linear interpolation technique: first, missing and erroneous points are replaced with a flag called "Not a Number (NaN)". Then, the NaN positions are patched using a linear interpolation method. Similarly, convert a raw data series with a resolution of minutes to data with a resolution of hours. Additionally, these time series are normalized to the interval -1 to 1.

[0119] 2) Use discrete wavelet transf...

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 short-term wind speed prediction method integrating deep learning and data denoising. The method comprises the following steps: 1) decomposing a wind speed sequence A by using discrete wavelet transform; 2) distinguishing a noise signal and an effective signal after discrete wavelet transform by using a wavelet soft threshold denoising method; 3) setting a wavelet transform coefficient of the noise signal as zero, and reconstructing a wind speed time sequence to obtain a denoised wind speed sequence B; 4) training a gating cycle unit neural network by using the wind speed sequence B to obtain a wind speed prediction model, and 6) inputting real-time wind speed into the wind speed prediction model to complete future multi-step wind speed prediction. The short-termwind speed high-quality prediction can be realized so as to ensure the economic dispatching and safe operation of a power system.

Description

technical field [0001] The invention relates to the field of wind power prediction, in particular to a short-term wind speed prediction method integrating deep learning and data denoising. Background technique [0002] In the context of today's energy crisis, the development and utilization of renewable energy has become the focus of government and research institutions. As of the end of 2017, the global installed capacity of renewable energy has maintained an average annual growth rate of 8% for three consecutive years, and the installed capacity has reached 2179GW. Among them, wind power installed capacity is an important part of renewable energy, accounting for the second largest share (23.59%). At the same time, the installed capacity of wind power increased by 94GW in 2017, a year-on-year increase of 10%. Compared with other energy sources such as hydraulic power, fossil fuels, and nuclear energy, the magnitude of wind power mainly depends on natural conditions such a...

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): G06Q10/04G06Q50/06G06N3/04G06F16/215G06F16/2458
CPCG06Q10/04G06Q50/06G06F16/215G06F16/2474G06N3/048G06N3/045
Inventor 唐俊杰彭志云陆彬春符礼丹刘梦洁林星宇
Owner CHONGQING UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap