Chaotic time sequence prediction method based on particle swarm optimization and auto-encoder

A chaotic time series, particle swarm optimization technology, applied in forecasting, neural learning methods, biological neural network models, etc., can solve problems such as air pollution

Pending Publication Date: 2020-06-02
DALIAN UNIV OF TECH
View PDF0 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

A large amount of pollutants, which are increasingly released into the atmosphere, which leads to serious air pollution problems

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
  • Chaotic time sequence prediction method based on particle swarm optimization and auto-encoder
  • Chaotic time sequence prediction method based on particle swarm optimization and auto-encoder
  • Chaotic time sequence prediction method based on particle swarm optimization and auto-encoder

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment

[0102] Taking the time series of Air Quality Index (AQI) in Beijing as the research object, the data comes from the UCI machine learning database.

[0103] The data interval is from 0:00 on January 1, 2010 to 23:59 on December 31, 2014, a total of 43800 groups of samples, each group of samples includes 5-dimensional variables, which are PM collected every hour 2.5 Concentration, dew point temperature, air temperature, air pressure and cumulative wind speed.

[0104] The first 75% of the data (32850 samples, from 0:00 on January 1, 2010 to 17:59 on September 1, 2013) were used as training samples for this method, and the output weights of the feed-forward neural network were optimized with the MPSO algorithm.

[0105] Such as image 3 As shown in , the trained MPSO-SAE is simulated (at this time, the feedforward neural network has obtained the optimal output weight), and the fitting situation between the simulated value and the sample is compared to test the generalization abili...

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 chaotic time sequence prediction method based on particle swarm optimization and an auto-encoder, and belongs to the field of chaotic time sequence modeling analysis of complex systems. Aiming at the defects of a traditional simple model during chaotic time sequence prediction, the prediction method improves the mode of directly predicting the acquired data by the traditional model by combining the characteristics that the stack self-encoding network can perform unsupervised feature extraction for multiple times and the intelligent optimization algorithm does not require strict mathematical conditions. The method comprises the following steps: firstly, mapping original data to a high-dimensional space by applying chaos and phase space reconstruction theories, revealing evolution information contained in the hybrid power system, then carrying out feature extraction by utilizing a stack self-encoding network, and finally, carrying out prediction; optimizing theoutput weight of the prediction model through the particle swarm optimization algorithm, so that the model has better generalization performance, and finally, the prediction precision of the chaotic time sequence model is effectively improved.

Description

technical field [0001] The invention belongs to the field of chaotic time series modeling and analysis of complex systems, in particular to a chaotic time series prediction method based on particle swarm optimization and an autoencoder. Background technique [0002] Time series is some orderly observed data x(t) obtained by sampling at regular intervals t by researchers who analyze dynamic systems. Time series forecasting is to establish an appropriate model based on past information and quantitatively predict the trend in a certain period of time in the future. Chaos refers to the long-term unpredictable, random-like motion of deterministic dynamical systems due to their sensitivity to initial values. Chaotic time series is a kind of time series with chaotic characteristics. Chaotic time series is sensitive to initial conditions. The trajectory of motion is limited to a limited area, and the trajectory does not repeat. It is predictable in the short term and unpredictable ...

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): G06Q10/04G06N3/08G06N3/04
CPCG06N3/0418G06N3/086G06Q10/04
Inventor 任伟杰李昕韩敏
Owner DALIAN UNIV OF TECH
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