Water quality prediction method, device and system and computer readable storage medium

Abstract

The invention discloses a water quality prediction method, device and system and a computer readable storage medium, and the method comprises the steps: monitoring the water quality data of to-be-predicted water quality, carrying out the data processing of the water quality data, and obtaining the to-be-analyzed water quality data corresponding to the to-be-predicted water quality; carrying out component analysis on the to-be-analyzed water quality data to obtain corresponding dimension reduction data, carrying out hypothesis testing on the dimension reduction data, and determining a testing result of the to-be-predicted water quality; and performing time series prediction on the to-be-analyzed water quality data based on the inspection result to determine a prediction result of the to-be-predicted water quality. Non-linear water quality data are processed through a machine learning method of data processing, component analysis, hypothesis testing and time sequence prediction, correlation between different water quality data is mined, the water quality data do not need to be processed manually, and therefore the difficulty of processing the non-linear water quality data manually is avoided, and the accuracy of sewage quality prediction is improved.

Description

technical field [0001] The present application relates to the technical fields of machine learning and sewage treatment, and in particular to a water quality prediction method, device, system and computer-readable storage medium. Background technique [0002] At present, sewage treatment is mainly divided into industrial sewage treatment and civil water treatment, among which the treatment of sewage water quality is related to the production and health of residents. In the process of treating sewage water quality, it is an important link to detect the water quality and detect its abnormality. [0003] The prediction of sewage water quality belongs to the abnormal detection part of the sewage treatment process. The traditional detection method basically relies on the experience of the operator, and manually designs or extracts the water quality data characteristics of the sewage water quality, so as to realize the treatment of the sewage water quality. However, the use of hi...

AI Technical Summary

Problems solved by technology

However, the use of historical data by manual methods is very low, resulting in very low efficiency in the process of sewage treatment

Furthermore, the amount of operational data of sewage treatment plants is very large, and different data are non-linearly correlated, and a single data is time-autocorrelated. Due to long-term dynamic changes, manual processing often cannot use the data, making it difficult to predict. Changes in sewage water quality make it impossible to provide early warning and timely treatment of upcoming water quality problems

Images