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

Multi-dimensional expansion prediction method and device for non-stationary time series data

A technology for time series and forecasting methods, applied in the field of multi-dimensional expansion forecasting of non-stationary time series data, can solve problems such as processing, incompetence in time series forecasting tasks, lack of modeling, etc., to achieve the effect of strong expression ability

Pending Publication Date: 2020-09-11
CHENGDU QIANJIA TECH CO LTD
View PDF19 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] In the existing regression analysis methods related to time series data, the processing and prediction of stationary time series data are mostly based on fixed time intervals, and it is impossible to deal with the situation of irregular time intervals in the data.
At the same time, the modeling dimension of the classic time series model mainly focuses on the absolute scale of time and the dimension of sampling value, and lacks the modeling of other information dimensions such as irregular time interval differences. The level and depth of the model are insufficient in the feature extraction process, and the model expression ability With certain limitations, it is not competent for time series forecasting tasks under complex conditions

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
  • Multi-dimensional expansion prediction method and device for non-stationary time series data
  • Multi-dimensional expansion prediction method and device for non-stationary time series data
  • Multi-dimensional expansion prediction method and device for non-stationary time series data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0034] The present invention is realized through the following technical solutions, as figure 1 As shown, a multi-dimensional extended forecasting method for non-stationary time series data, including the following steps:

[0035] Step S1: Collect the original meter reading data, process the original meter reading data, and expand the data dimension to construct the meter reading data set.

[0036] The original meter reading data collected by the meter is composed of two parts: the remote transmission data of the meter and the manual meter reading data. The original meter reading data has differences in time steps and belongs to a non-stationary time series. The dimension of the original meter reading data is expanded, and the dimension of the expanded original meter reading data includes meter number, meter reading time, and meter reading reading.

[0037] Perform granular processing on the meter reading time of the expanded original meter reading data, align the remote data...

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 relates to a multi-dimensional expansion prediction method and device for non-stationary time series data, and the method comprises the following steps: collecting original meter readingdata, processing the original meter reading data, and expanding data dimensions to construct a meter reading data set; constructing RNN sequence expression models with different sequence dependent lengths, carrying out model training by adopting total data, and determining the optimal dependent length according to model performance expression; establishing an RNN sequence model based on the optimal sequence dependence length, and performing model training based on the meter reading data set to obtain optimized RNN sequence model parameters; predicting the latest remote transmission data of each meter by using the optimized RNN sequence model parameters; and comparing the predicted remote transmission data value with the actual remote transmission data value, obtaining a correction value according to a strategy, and recording the correction value into a database. According to the method, the non-stationary time sequence can be modeled, and meanwhile, the information dimension of the difference time step length can be fused, so that the time sequence model has stronger expression capability.

Description

technical field [0001] The invention relates to the technical field of time series data prediction, in particular to a method and device for multi-dimensional expansion prediction of non-stationary time series data. Background technique [0002] In the existing regression analysis methods related to time series data, the processing and prediction of stationary time series data are mostly based on fixed time intervals, and it is impossible to deal with the situation of irregular time intervals in the data. At the same time, the modeling dimension of the classic time series model is mainly concentrated in the absolute time scale and sampling value dimension, and lacks the modeling of other information dimensions such as irregular time interval differences. With certain limitations, it is not competent for time series forecasting tasks under complex conditions. Contents of the invention [0003] The purpose of the present invention is to improve the deficiencies existing in ...

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): G06F30/27G06Q10/04G06N3/04G06N3/08
CPCG06F30/27G06Q10/04G06N3/084G06N3/045
Inventor 常关羽赵勇张彬胡芸华朱炼牛富增
Owner CHENGDU QIANJIA TECH CO LTD
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