Multi-correlation vector machine water quality prediction method based on quantum genetic algorithm optimization

Abstract

The invention provides a multi-correlation vector machine water quality prediction method and system based on quantum genetic algorithm optimization. The method comprises: establishing a correlation vector machine water quality prediction model according to a monitored historical data time sequence of a certain water quality index; establishing different association vector machine sub-models, wherein each established sub-model performs description of different angles on the object from different characteristics of each piece of data; learning and training the BP network by using the predictionresults obtained by the sub-models to obtain a nonlinear combination function model, and further completing effective fusion of the sub-models; embedding a quantum genetic optimization algorithm to optimize and select initial parameters of the BP network, so that the convergence rate of the network is increased to achieve global optimization, and finally water quality prediction of the multi-correlation vector machine based on quantum genetic algorithm optimization is achieved. The selection of the kernel function is not limited by any condition, and the credibility of the prediction result can be given when the prediction value is given.

Description

technical field [0001] The present invention relates to the field of water quality prediction, and more specifically, relates to a multi-correlation vector machine water quality prediction method and system based on quantum genetic algorithm optimization. Background technique [0002] Water quality prediction, as an important means of water environmental pollution control, can predict the change trend of water quality, so as to effectively control the deterioration of water quality; at the same time, in the field of aquaculture, the prediction of various water quality parameters is an important guarantee for the modernization and automation of aquaculture , to effectively predict water quality, and provide effective guidance and reference for aquaculture enterprises and individuals to increase production and save investment. In order to grasp the current situation of water quality and its development trend, effectively control and improve the deterioration of water quality, ...

AI Technical Summary

Problems solved by technology

[0004] In the existing technology, a single RVM model is not easy to adapt to the corresponding characteristics of different water quality indicators. When predicting multiple water quality indicators, it is inevitable that the prediction accuracy will decrease, and the generalization ability is relatively weak.

Images