Short-term prediction method for marine environment monitoring factors

Abstract

The invention discloses a short-term prediction method for marine environment monitoring factors. According to the method, an autoregressive model is adopted for a short-term prediction of marine environment input factors, and then a deep neural network method is adopted for a short-term prediction of output factors. When the characteristics of the autoregressive model and the characteristics of a deep neural network model are efficiently used, short-term input predicted values are obtained by means of the autoregressive model, then predicted values of the output factors are obtained by means of the deep neural network model, and thus the limitations that prediction of a traditional neural network model to time series data is insufficient in extracted substantive characteristics and limited in gradient descent are overcome. The method achieves the prediction of the marine environment monitoring factors, and has an important significance in a short-term forecast of the marine ecological environment, a short-term pre-warning of marine disasters and the like.

Description

technical field [0001] The invention relates to a short-term prediction method for marine environment monitoring elements, in particular to a short-term prediction method for marine environment monitoring elements. Background technique [0002] Accurate and real-time prediction of marine environmental monitoring elements can effectively prevent various marine disasters such as red tides and oil spills, and at the same time help to have a more accurate understanding of the time trend of the entire sea area, which is of great significance in practical application. [0003] The autoregressive model (AR) is often used to simulate the prediction of time series, and its effect on the future prediction of nonlinear elements is not obvious, but it has a good accuracy in the future prediction of a single marine environmental element. Machine learning allows the machine to simulate the human brain through training, and at the same time input a large amount of historical data from the ...

AI Technical Summary

Problems solved by technology

Compared with the classical statistical regression model, this artificial intelligence-based neural network method shows superiority in terms of multiple factors and full of randomness, but the problems of local minimum in this type of neural network method have not been thoroughly solved. At the same time, how to use multiple network layers to extract useful feature values ​​more efficiently has puzzled researchers for a long time.