A method for predicting a sewage outlet index based on a random forest and a gradient lifting tree
A gradient boosting tree and random forest technology, applied in prediction, neural learning methods, computer components, etc., can solve the problems of model accuracy to be improved, slow training speed, etc., to improve the training rate and data quality, and reduce processing costs , the effect of improving the prediction accuracy
Active Publication Date: 2019-03-01
SHANGHAI MARITIME UNIVERSITY
View PDF2 Cites 18 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The prediction of sewage effluent indicators based on neural network solves such problems to a certain extent, but there are still shortcomings such as slow training speed and model accuracy that needs to be improved.
And such research does not avoid irrelevant factors in the response process, which negatively affects the training speed and accuracy of the model
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment approach
[0069] As an implementation manner, the calculation of feature importance specifically includes the following steps:
[0070] For each regression tree (regression problem) in the random forest, use the corresponding out-of-bag sample to calculate its out-of-bag sample error, which is recorded as err 1 ;
[0071] Randomly add noise interference to the characteristics of the out-of-bag samples, and calculate its out-of-bag error again, which is recorded as err 2 ;
[0072] Among them, out-of-bag samples refer to sample data that are not used as training samples;
[0073] The formula for calculating the importance of a feature is:
[0074]
[0075] Among them, n is the sample number of out-of-bag samples, f is the sum of the out-of-bag sample error and the out-of-bag error added with noise interference, according to f as the value of the importance of a certain feature;
[0076] According to the calculated feature importance, the features are sorted and the important featu...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a method for predicting a sewage effluent index based on a random forest and a gradient lifting tree, which comprises the following steps of: 1, extracting samples in an original data training set in a put-back manner to form a plurality of sample sets; 2, constructing a random forest according to the samples; calculating feature importance according to the random forest, and carrying out attribute screening; Step 3, constructing a gradient lifting tree model according to samples formed by the screened attributes; 4, according to the real-time monitoring data, putting the real-time monitoring data into the gradient lifting tree model to predict a sewage outlet index of the sewage plant in a future period of time. According to the method, the random forest and the gradient lifting tree model are combined to establish the relation model of the sewage outlet index data, and the sewage outlet index data in a future period of time can be accurately predicted throughdimensionality reduction of the random forest and high-precision training of the gradient lifting tree.
Description
technical field [0001] The invention relates to the technical fields of sewage treatment and machine learning, in particular to a method for predicting sewage effluent indicators based on random forests and gradient boosting trees. Background technique [0002] The urban sewage treatment process is a complex biochemical reaction process, accompanied by physical and chemical reactions, biochemical reactions, phase transition processes, and the transformation and transfer of matter and energy. The process is complex and traditional mathematical modeling is difficult. Many scholars have conducted research on using neural networks to solve such problems. The prediction of sewage effluent indicators based on neural network solves such problems to a certain extent, but there are still shortcomings such as slow training speed and model accuracy that needs to be improved. And such research does not avoid irrelevant factors in the response process, which negatively affects the train...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/18G06K9/62G06N3/08G06Q10/04
CPCG06F17/18G06N3/08G06Q10/04G06F18/2411Y02A20/152
Inventor 张天麟高俊波孙伟赵友标孙峰
Owner SHANGHAI MARITIME UNIVERSITY