Water quality parameter time series prediction method based on relevance vector machine regression

Abstract

The invention provides a water quality parameter time series prediction method based on relevance vector machine regression. The water quality parameter time series prediction method comprises the following steps of 1 acquiring water quality parameter historical data from an automatic water quality monitoring station and performing data pre-processing; 2 using front 2 / 3 data in the pro-processed water quality parameter historical data as a training sample set and using rear 1 / 3 data as a testing sample set; 3 using the training sample set to train an RVM, using the testing sample set to test the trained RVM so as to obtain a water quality parameter time series prediction model based on the RVM regression; 4 using the water quality parameter time series prediction model based on the RVM regression to predict new water quality parameters. The water quality parameter time series prediction method can perform time series prediction, is large in prediction range, high in accuracy and good in prediction stability, and can provide probabilistic output, give a predicted confidence interval while performing prediction, reduce the prediction time and timely observe water quality parameter change.

Description

technical field [0001] The invention relates to the field of water quality monitoring, in particular to a water quality parameter time series prediction method based on correlation vector machine regression. Background technique [0002] The time series of water quality parameters is an ordered monitoring data sequence, which reflects the distribution of a certain water quality parameter in time, such as the monitoring data of the water quality parameter PH value of a watershed section from the first week to the 50th week of a certain year. The water quality parameter time series prediction method is to use the obtained historical time series collection, analyze the inherent statistical characteristics and development laws of the historical data in the collection, establish a water quality parameter time series prediction model, and use the model to obtain forecast data to indicate the future. Data trends. Time series prediction of water quality parameters is the basic work...

AI Technical Summary

Problems solved by technology

The Chinese patent application number 20131013190.7 provides a sewage water quality monitoring method and device. The prediction model used in this method is a soft sensor method based on correlation vector machine. Compared with the model established by neural network and support vector machine modeling method , has better applicability and higher prediction accuracy, but this method has the following defects: 1: Since the relevant parameters are used for analysis to obtain the content of effluent total nitrogen or effluent total phosphorus, the uncertainty of relevant data will be It greatly affects the final output data results. Although the data output has been greatly improved compared with the models established by neural network and support vector machine modeling methods, the instability of the final output data results still exists; Two: Simply analyzing the content of total nitrogen or total phosphorus in the effluent at that time cannot realize time series prediction, the prediction range is small, and the accuracy is low

Images