A data prediction method based on improved pso‑bp neural network

A BP neural network, PSO-BP technology, applied in the direction of biological neural network model, etc., can solve the problem of inability to take into account reliability and hardware cost, and achieve the effect of easy implementation, scientific and reasonable results, and improved accuracy

Active Publication Date: 2017-12-05
ZHEJIANG UNIV OF TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the shortcomings of existing data "bad point" processing methods that cannot take into account both reliability and hardware cost, the present invention proposes a data prediction method based on an improved PSO-BP neural network, which can effectively process data "bad point", Balancing reliability and hardware cost

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
  • A data prediction method based on improved pso‑bp neural network
  • A data prediction method based on improved pso‑bp neural network
  • A data prediction method based on improved pso‑bp neural network

Examples

Experimental program
Comparison scheme
Effect test

example

[0080] Example: Taking the temperature data prediction of the upper guide bearing of a water turbine as an example to illustrate, the data prediction method based on the improved PSO-BP neural network includes the following process:

[0081] 1), BP neural network model construction

[0082] In this patent, taking the prediction of the temperature data of the upper guide bearing of the water turbine as an example, according to the characteristics of the data, it is determined to use 4 consecutive data collected by the sensor to predict the next data to be collected, so the number of input layer nodes of the neural network model is determined to be 4 , and the number of output layer nodes is 1. According to the design concept of this patent, one layer is used to determine the number of hidden layers of the BP neural network. Use formula (1) to determine the selection of hidden layer nodes.

[0083]

[0084] In the formula: n is the number of nodes in the input layer, q is t...

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

A data prediction method based on the improved PSO-BP neural network, which predicts the next data to be collected by several consecutive data collected by the sensor, compares the predicted data with the collected data and judges whether the collected data is "bad" or not. Points", first determine the number of input nodes, output nodes and hidden layer nodes of the BP neural network according to the characteristics of the data collected by the sensor; then use the improved particle swarm optimization algorithm to optimize the connection weights and thresholds of the BP neural network, And get the final BP neural network prediction model; then use MATLAB7.1 and VC6 to generate a DLL file for the prediction model; finally use the programming software to call the DLL file to compare the predicted data with the collected data, and judge whether the collected data is data" bad point". The invention can effectively deal with data "bad spots", taking into account both reliability and hardware cost.

Description

technical field [0001] The invention relates to the field of data prediction, in particular to a data prediction method based on an improved PSO-BP neural network. Background technique [0002] In the field of automation control, especially in petrochemical, power station and other control fields, real-time monitoring of operating equipment is very important. Different types of data need to be collected during real-time monitoring, including temperature, water level, pressure, flow, etc. In the actual operation of equipment monitoring, the sensor will have data "bad spots", and these data "bad spots" will directly affect the judgment of the equipment monitoring system. There are two main cases of data "bad spots": 1. Due to the instability of the data source, the data suddenly jumps. 2. Since the sensor has not been repaired or replaced for a long time during use, it will gradually fail, resulting in more and more "bad pixels" of collected data. In the above two cases, 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
Patent Type & Authority Patents(China)
IPC IPC(8): G06N3/02
Inventor 赵燕伟应伟军任设东陈相云寿开荣冷龙龙
Owner ZHEJIANG 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