Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

On-line time series data prediction method, system and storage medium based on fuzzy inference

A time-series data and fuzzy reasoning technology, applied in biological neural network models, neural architectures, etc., can solve problems such as the inability to effectively solve time-varying system identification problems

Pending Publication Date: 2018-12-28
SHANDONG NORMAL UNIV
View PDF0 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the Interval Type II Fuzzy Neural Network (IT2FNN) with a fixed structure cannot effectively solve the problem of time-varying system identification.

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
  • On-line time series data prediction method, system and storage medium based on fuzzy inference
  • On-line time series data prediction method, system and storage medium based on fuzzy inference
  • On-line time series data prediction method, system and storage medium based on fuzzy inference

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0298] Example 1: Identification of nonlinear systems

[0299] This example uses eIT2FNN-LSTM to identify a nonlinear dynamical system, which is J.B. Theocharis, A high-order recurrent neuro-fuzzy system with internal dynamics: application to the adaptive noise cancellation, Fuzzy Sets and Systems 157(4)(2006)471– 500. Questions under study. A dynamical system with input delays is guided by a difference equation:

[0300] the y p (k+1)=f(y p (k),y p (k-1),y p (k-2),u(k),u(k-1)), (55)

[0301] in,

[0302]

[0303] Only the current state y p (k) and control input u(k) as the input of eIT2FNN-LSTM. The training of eIT2FNN-LSTM consists of 10 epochs with 900 time steps in each epoch. The number of training periods is the same as that in the previous research Abiyev R H, Kaynak O. Type 2 Fuzzy Neural Structure for Identification and Control of Time-Varying Plants[J]. IEEE Transactions on Industrial Electronics,2010,57(12):4147-4159. The training period used is the sam...

example 2

[0308] Example 2: Abalone Age Prediction

[0309] The purpose of the abalone age prediction problem is to predict the age of abalone based on the physical characteristics of the age of the abalone. The dataset is collected from the UCI Machine Learning Repository. It includes 4177 samples; 3342 samples are used for training and the remaining 835 samples are used for testing. Using length, diameter, height, total weight, shell weight, visceral weight, and shell weight as input features, the number of rings is predicted. The performance of eIT2FNN-LSTM is compared with McIT2FIS-US, RIT2NFS-WB and SEIT2FNN. The results are given in Table 2. Table 2 shows the number of rules used by all algorithms, training and testing RMSE. It can be seen from Table 2 that the generalization ability of eIT2FNN-LSTM is better than other algorithms.

[0310] Table 2 Performance comparison on the abalone age prediction problem

[0311]

[0312]

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 an on-line time series data prediction method based on fuzzy inference, a system and a storage medium. Training process: extracting the time series data of the training samples, inputting the time series data of the training samples into the LSTM fuzzy neural network model based on interval type II, outputting each sample, first carrying out the structure learning, and thencarrying out the parameter learning on the basis of the structure learning; the trained LSTM fuzzy neural network model based on interval type 2 being obtained. Forecasting process: extracting the time series data of the sample to be forecasted, inputting the time series data of the sample to be forecasted into the trained LSTM fuzzy neural network model based on interval type II, and outputtingthe forecasting result.

Description

technical field [0001] The invention relates to an online time series data prediction method, system and storage medium based on fuzzy reasoning. Background technique [0002] For a dynamic system, the output is a nonlinear function of past outputs, past inputs, or both. Problems in dynamic systems arise in many fields, such as control and pattern recognition. Fuzzy neural network (FNN), which has the advantages of fuzzy logic and neural network, is often used to deal with this series of problems, but in order to identify dynamic systems or identify time series, it is usually necessary to use a recurrent model that represents some kind of memory. The general internal feed-forward network structure is limited to dealing with static problems and cannot effectively deal with temporal problems, while the recurrent model can be considered as a closed-loop system in which the recurrent path introduces the dynamics of the model, so it is different from the traditional fuzzy neural...

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
IPC IPC(8): G06N3/04
CPCG06N3/043G06N3/045
Inventor 骆超王海月
Owner SHANDONG NORMAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com