PCA-FSA-MLR Model and Its Application in Runoff Forecast
To improve the accuracy of runoff forecast,and establish a runoff forecast model combining principal component analysis (PCA),future search algorithm (FSA),and multiple linear regression (MLR),this paper reduces the dimensionality of the sample data by PCA,selects 8 standard test functions and simul...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Pearl River
2021-01-01
|
| Series: | Renmin Zhujiang |
| Subjects: | |
| Online Access: | http://www.renminzhujiang.cn/thesisDetails#10.3969/j.issn.1001-9235.2021.06.014 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841535640489951232 |
|---|---|
| author | GUO Cunwen CUI Dongwen |
| author_facet | GUO Cunwen CUI Dongwen |
| author_sort | GUO Cunwen |
| collection | DOAJ |
| description | To improve the accuracy of runoff forecast,and establish a runoff forecast model combining principal component analysis (PCA),future search algorithm (FSA),and multiple linear regression (MLR),this paper reduces the dimensionality of the sample data by PCA,selects 8 standard test functions and simulates and verifies FSA under different dimensional conditions,optimizes MLR constant terms and partial regression coefficients by FSA,proposes a PCA-FSA-MLR runoff forecast model,constructs PCA-LS-MLR,PCA-FSA-SVM,and PCA-SVM models with dimensionality reduction processing by PCA and FSA-MLR,LS-MLR,FSA-SVM,and SVM without dimensionality reduction processing as a comparison model,and verifies each model through forecasting the annual runoff and monthly runoff in December of Longtan station in Yunnan Province.The results show that:①FSA has better optimization accuracy and global extremum search ability under different dimensional conditions;②The average absolute relative error of the annual runoff and monthly runoff in December of Longtan station through PCA-FSA-MLR model are 1.63% and 3.91% respectively,and its forecast accuracy is better than the other 7 models,with higher forecast accuracy and stronger generalization ability;③For the same model,the forecast accuracy after dimensionality reduction processing by PCA is better than that without dimensionality reduction processing,so the data dimensionality reduction by PCA is helpful to improve the forecast accuracy of models. |
| format | Article |
| id | doaj-art-9afc5ce426ed4aeca9115311b33ec6c8 |
| institution | Kabale University |
| issn | 1001-9235 |
| language | zho |
| publishDate | 2021-01-01 |
| publisher | Editorial Office of Pearl River |
| record_format | Article |
| series | Renmin Zhujiang |
| spelling | doaj-art-9afc5ce426ed4aeca9115311b33ec6c82025-01-15T02:30:37ZzhoEditorial Office of Pearl RiverRenmin Zhujiang1001-92352021-01-014247650206PCA-FSA-MLR Model and Its Application in Runoff ForecastGUO CunwenCUI DongwenTo improve the accuracy of runoff forecast,and establish a runoff forecast model combining principal component analysis (PCA),future search algorithm (FSA),and multiple linear regression (MLR),this paper reduces the dimensionality of the sample data by PCA,selects 8 standard test functions and simulates and verifies FSA under different dimensional conditions,optimizes MLR constant terms and partial regression coefficients by FSA,proposes a PCA-FSA-MLR runoff forecast model,constructs PCA-LS-MLR,PCA-FSA-SVM,and PCA-SVM models with dimensionality reduction processing by PCA and FSA-MLR,LS-MLR,FSA-SVM,and SVM without dimensionality reduction processing as a comparison model,and verifies each model through forecasting the annual runoff and monthly runoff in December of Longtan station in Yunnan Province.The results show that:①FSA has better optimization accuracy and global extremum search ability under different dimensional conditions;②The average absolute relative error of the annual runoff and monthly runoff in December of Longtan station through PCA-FSA-MLR model are 1.63% and 3.91% respectively,and its forecast accuracy is better than the other 7 models,with higher forecast accuracy and stronger generalization ability;③For the same model,the forecast accuracy after dimensionality reduction processing by PCA is better than that without dimensionality reduction processing,so the data dimensionality reduction by PCA is helpful to improve the forecast accuracy of models.http://www.renminzhujiang.cn/thesisDetails#10.3969/j.issn.1001-9235.2021.06.014runoff forecastprincipal component analysis (PCA)future search algorithm (FSA)multiple linear regression (MLR)data dimensionality reductionsimulation and verificationparameter optimization |
| spellingShingle | GUO Cunwen CUI Dongwen PCA-FSA-MLR Model and Its Application in Runoff Forecast Renmin Zhujiang runoff forecast principal component analysis (PCA) future search algorithm (FSA) multiple linear regression (MLR) data dimensionality reduction simulation and verification parameter optimization |
| title | PCA-FSA-MLR Model and Its Application in Runoff Forecast |
| title_full | PCA-FSA-MLR Model and Its Application in Runoff Forecast |
| title_fullStr | PCA-FSA-MLR Model and Its Application in Runoff Forecast |
| title_full_unstemmed | PCA-FSA-MLR Model and Its Application in Runoff Forecast |
| title_short | PCA-FSA-MLR Model and Its Application in Runoff Forecast |
| title_sort | pca fsa mlr model and its application in runoff forecast |
| topic | runoff forecast principal component analysis (PCA) future search algorithm (FSA) multiple linear regression (MLR) data dimensionality reduction simulation and verification parameter optimization |
| url | http://www.renminzhujiang.cn/thesisDetails#10.3969/j.issn.1001-9235.2021.06.014 |
| work_keys_str_mv | AT guocunwen pcafsamlrmodelanditsapplicationinrunoffforecast AT cuidongwen pcafsamlrmodelanditsapplicationinrunoffforecast |