Unlock instant, AI-driven research and patent intelligence for your innovation.

Time sequence forecast method based on quantum gravity algorithm

A technology of time series and prediction methods, applied in computing, special data processing applications, instruments, etc., can solve the problems of insufficient model training, affecting model prediction performance, local minima, etc.

Inactive Publication Date: 2016-02-10
UNIV OF ELECTRONIC SCI & TECH OF CHINA
View PDF2 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, it is not an easy task to establish a Volterra series model for a nonlinear system. The estimation of its high-order kernel function is one of the biggest difficulties it faces, which largely restricts the effectiveness of the Volterra functional series model. application
However, the existing Volterra kernel function identification mainly adopts the traditional least squares method, which uses gradient information to search, which is easy to fall into a local minimum and cannot obtain satisfactory results. Especially in the case of severe interference, the stability of identification is not ideal
As a result, the modeling is inaccurate and affects the prediction performance of the model, especially when the training set contains too little information related to the predicted trajectory or contains a lot of irrelevant information, it is easy to affect the training effect, resulting in insufficient model training. or overfitting, leading to poor multi-step prediction performance

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
  • Time sequence forecast method based on quantum gravity algorithm
  • Time sequence forecast method based on quantum gravity algorithm
  • Time sequence forecast method based on quantum gravity algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0055] figure 1 It is a flow chart of a specific embodiment of the time series prediction method based on the quantum gravity algorithm of the present invention.

[0056] In this example, if figure 1 As shown, the time series prediction method based on the quantum gravity algorithm of the present invention can be divided into three stages: data preprocessing stage, prediction model training stage and prediction stage. Each stage is described in detail below:

[0057] (1) Data preprocessing stage

[0058] S101: Suppose the preselected data set DS={x(1), x(2),...,x(D)}, where D is the number of data;

[0059] First, determine the optimal embedding dimension d and delay time τ of the data set DS according to the method of differential entropy rate;

[0060] Secondly, use the function windowize in Matlab to reconstruct the data set DS in frames, and map each frame data to the d-dimensional feature space, and use Volterra series to p(p≥1) order truncation, so as to get N 0 (N ...

example

[0098] In order to illustrate the technical effects of the present invention, the present invention will conduct experimental analysis from simulation and actual chaotic time series to measure the validity of the prediction model. Among them, the simulation sequence is the famous Mackey-Glass (MG) data, and the actual sequence is several sets of data sets provided by the TimeSeriesPredictionGroup group: Laser, DailyMinimumTemperatures (DMT), Electricity_Demand (ED), CATS_Benchmark (CATS_B), SunspotNumber (SN), PolandElectricity ( PE), Dslp. At the same time, the present invention also uses a set of degradation measurement data (RDD) generated by the random degradation model to verify the method proposed by the present invention.

[0099] For these 9 different simulated and actual chaotic time series, they are first normalized, and then 2 / 3 of the data are used as training data, and 1 / 3 of the data are used as test data. In the parameter setting of the prediction model, the me...

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 time sequence forecast method based on a quantum gravity algorithm. A time sequence forecast model is constructed by adopting a Volterra series expanded formula. The method comprises the steps that firstly, an input signal vector subjected to Volterra series P-order truncation is constructed on the basis of quoting a phase space framing reconstruction technology; secondly, acceleration is introduced to serve as a variable parameter into the quantum gravity algorithm put forward by Mohadeseh Soleimanpour et al., and a Volterra kernel function of the forecast model is effectively trained by means of the algorithm; finally, a forecast value of forecast point time is obtained through a linear combination of the input signal vector of the forecast point time and the Volterra kernel function. It is verified through experiments that the accuracy of time sequence forecast can be effectively improved by introducing the quantum gravity algorithm into identification of the Volterra kernel function of a non-linear system.

Description

technical field [0001] The invention belongs to the fields of signal processing and signal prediction, and more specifically, relates to a time series prediction method based on a quantum gravity algorithm. Background technique [0002] Nonlinear time series forecasting methods have been developed since the mid-1980s, and have been studied more deeply and applied more widely in recent years. Among them, the Volterra series model analysis method has become one of the widely used prediction models because the output is a linear function of its filter kernel, and the existing linear tools can be used to analyze its filtering performance. Volterra series is a kind of universal function, and most nonlinear dynamic systems can be approximated to any degree of accuracy by Volterra series. In other words, a large class of nonlinear dynamic systems has its inherent Volterra kernel function. The Volterra kernel function can fully characterize the properties of nonlinear systems. Th...

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): G06F19/00
Inventor 刘震曾现萍程玉华田书林龙伊雯
Owner UNIV OF ELECTRONIC SCI & TECH OF CHINA