Sea water temperature and salinity information time sequence prediction method based on shipborne CTD measurement data

A technology for measuring data and seawater, which is applied in the field of data processing to achieve the effect of speeding up calculation efficiency and improving accuracy

Pending Publication Date: 2021-02-02
TIANJIN UNIV
View PDF4 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention proposes a time series prediction method of seawater temperature and salinity information based on ship-borne CTD measurement data to solve the problem of time-series prediction of irregular environmental data collected by ship-borne CTD

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
  • Sea water temperature and salinity information time sequence prediction method based on shipborne CTD measurement data
  • Sea water temperature and salinity information time sequence prediction method based on shipborne CTD measurement data
  • Sea water temperature and salinity information time sequence prediction method based on shipborne CTD measurement data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] Due to the incompleteness and non-uniformity of historical time series data collected by shipborne CTD, traditional supervised learning algorithms cannot effectively select ideal data sets for model training, and supervised learning algorithms such as neural networks rely too much on The accuracy of historical data leads to the inability to stably analyze information trends in the work of information forecasting. In the work of information forecasting, the unsupervised learning method can rely on less historical experience to realize the prediction of the trend, but there are problems such as low prediction accuracy and difficult definition, and the prediction of time series belongs to the regression problem, and the unsupervised learning method does not fully applicable. Therefore, the present invention proposes a time series prediction method combining supervised learning and unsupervised learning, which is based on fuzzy segmentation and neural network fusion algorit...

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 sea water temperature and salinity information time sequence prediction method based on shipborne CTD measurement data. The method comprises the following steps: collectingsea water temperature and conductivity data, and calculating salinity measurement data based on the measured conductivity information so as to obtain original temperature and salinity data; preprocessing original temperature and salinity data collected by a shipborne CTD, wherein the preprocessing process comprises interpolation, filtering and screening so as to obtain time sequence information conforming to a calculation standard; dividing the preprocessed data into time information and attribute information including temperature and salinity, and performing fuzzy rule division on the time information; acquiring a new membership function after threshold segmentation, performing sequence alignment on attribute information and time information after fuzzy segmentation to establish a new time sequence data set, and constructing the new membership data set by using the part of sequence information as a new fuzzy sequence number; designing a long-term and short-term memory neural network prediction model, setting a loss function, and training and establishing the prediction model by using the training set.

Description

technical field [0001] The invention belongs to the field of data processing, and relates to a method for processing data collected by a CTD (conductivity, temperature, and depth) instrument for marine data detection. Background technique [0002] With people's exploration of the ocean, the new generation of marine environment detection equipment has the advantages of high observation accuracy and wide measurement range, but it also increases the measurement volume of ocean data. Faced with the exponential growth of ocean big data processing problem, this study proposes a time series forecasting method based on fuzzy segmentation. At present, marine survey ships measure the temperature and salinity data of the mission sea area by deploying and launching CTD on board. However, the survey ship needs to repeatedly measure a sea area by means of wiring cruise. The measurement efficiency is low, and many wrong data will be collected, which will bring difficulties to the later da...

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): G06F17/10G06K9/62G06N3/04
CPCG06F17/10G06N3/049G06N3/045G06F18/23213G06F18/25
Inventor 杨嘉琛温家宝
Owner TIANJIN UNIV
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