Method for predicting dissolved oxygen in sewage biochemical treatment

Abstract

The invention provides a method for predicting dissolved oxygen in sewage biochemical treatment. The method comprises the steps: S1, collecting and preprocessing water quality data in sewage biochemical treatment, obtaining a data set, and performing variational mode decomposition on the factors influencing the water quality to obtain decomposition components; S2, according to the data set in the step S1, judging the relationship between the decomposition components and the dissolved oxygen, carrying out clustering analysis, and dividing the decomposition components into a tight group and a loose group; and S3, according to the data set in the step S1, carrying out multi-task learning training on the tight group, carrying out multi-view learning training on the loose group, establishing a prediction model, and carrying out dissolved oxygen prediction and auxiliary task prediction. According to the prediction method, dissolved oxygen prediction and auxiliary task prediction of related factors are combined, and the accuracy of future dissolved oxygen prediction is remarkably improved.

Description

technical field [0001] The invention relates to the technical field of sewage biochemical treatment, in particular to a method for predicting dissolved oxygen in sewage biochemical treatment. Background technique [0002] In sewage treatment, the stable control of dissolved oxygen in the biochemical pool is the key to biological denitrification. At present, the control of dissolved oxygen in sewage treatment plants is based on the data of the online dissolved oxygen meter, manual adjustment, and the dissolved oxygen is often adjusted according to the dissolved oxygen meter. There is a hysteresis, which is not conducive to ensuring the stable compliance of the effluent. [0003] In order to improve the prediction effect of dissolved oxygen in biochemical pools, some studies have used related prediction methods to predict dissolved oxygen in biochemical pools in advance. [0004] CN113033861A discloses a water quality prediction method and system based on a time series model,...

AI Technical Summary

Problems solved by technology

[0004] CN113033861A discloses a water quality prediction method and system based on a time series model, which optimizes the time segment of the water quality time series data in combination with a genetic algorithm, and then sends the optimized time segment data into the model training as a single piece of data; combined with the attention mechanism In the encoding and decoding stage of the model, a multi-layer attention mechanism is implemented for the water quality data, and external information is introduced to realize joint modeling and realize the final water quality data prediction. However, this prediction method requires a large-scale water quality feature sequence, and in practice many Water quality characteristic data is missing or cannot find enough characteristic data related to dissolved oxygen prediction

[0005] CN111898673A discloses a method for predicting dissolved oxygen content based on EMD and LSTM, including obtaining water quality data and performing data cleaning, using the KNN algorithm to complete the missing data in the water quality data, and using the EMD algorithm to calculate the original dissolved oxygen content in the water quality data. Decompose the time series of monitoring data to obtain multiple components including residuals and finite eigenmode functions, train and verify multiple sub-LSTM networks, and use multiple sub-LSTM networks that have passed the verification to obtain multiple components corresponding to For the predicted value of dissolved oxygen in the next unit time, the predicted values ​​corresponding to all components are accumulated to obtain the predicted result of dissolved oxygen in the next unit time. The success of signal processing techniques and the shortcomings of work on multi-task, multi-view learning models

[0006] CN109147875A discloses a method for predicting the concentration of dissolved oxygen in sewage based on the support vector regression algorithm of fuzzy clustering, which predicts the content of dissolved oxygen in sewage. Aiming at the problem of difficult real-time measurement of dissolved oxygen in the process of sewage treatment, this method first uses fuzzy Clustering divides the entire sample into multiple sub-samples, and then builds a support vector regression model on each sub-sample, and then integrates it to predict the dissolved oxygen content in sewage online. However, in this prediction method, fuzzy clustering needs to know the number of classifications in advance. , requires a lot of prior knowledge; secondly, the amount of fuzzy clustering calculations, when the amount of data is large, fuzzy clustering will not be able to achieve the purpose of clustering; and SVM needs a lot of prior characteristic factors and their data to achieve better results.

[0007] CN106802563A discloses a sewage process optimization control method based on fruit fly optimization and LSSVM. By collecting sewage process data, a system model including dissolved oxygen and nitrate nitrogen in the sewage process is established to accurately describe the real-time state of the system, and the fruit fly algorithm is used for rolling optimization. , embodies the control objectives and various constraints in the optimization performance index, and updates the model online based on real-time data, but in this prediction method, the sparsity of LSSVM in iterative optimization is not good, time-consuming and laborious; and the effect of fruit fly optimization is largely It lies in the selection of parameters, and the selection of parameters requires rich prior experience and knowledge, which has a great influence on the accuracy of the final result

[0008] Therefore, it is necessary to develop a new prediction method for dissolved oxygen in sewage biochemical treatment to solve the problem of difficult data acquisition in existing prediction methods

Images