For accurately forecasting the liquid steel temperature in ladle furnace （LF）, a novel temperature predic tion model based on optimally pruned Bagging combined with modified extreme learning machine （ELM） is pro p...For accurately forecasting the liquid steel temperature in ladle furnace （LF）, a novel temperature predic tion model based on optimally pruned Bagging combined with modified extreme learning machine （ELM） is pro posed. By analyzing the mechanism of LF thermal system, a thermal model with partial linear structure is obtained. Subsequently, modified ELM, named as partial linear extreme learning machine （PLELM）, is developed to estimate the unknown coefficients and undefined function of the thermal model. Finally, a pruning Bagging method is pro- posed to establish the aggregated prediction model for the sake of overcoming the limitation of individual predictor and further improving the prediction performance. In the pruning procedure, AdaBoost is adopted to modify the ag- gregation order of the original Bagging ensembles, and a novel early stopping rule is designed to terminate the aggre- gation earlier. As a result, an optimal pruned Bagging ensemble is achieved, which is able to retain Bagging＇s ro- bustness against highly influential points, reduce the storage needs as well as speed up the computing time. The pro- posed prediction model is examined by practical data, and comparisons with other methods demonstrate that the new ensemble predictor can improve prediction accuracy, and is usually consisted compactly.展开更多
y consumption efficiency and to increase the crop yield.With the increase of agri-cultural data generated by the Internet of Things(IoT),more feasible models are necessary to get full usage of such information.In this...y consumption efficiency and to increase the crop yield.With the increase of agri-cultural data generated by the Internet of Things(IoT),more feasible models are necessary to get full usage of such information.In this research,a Gradient Boost Decision Tree(GBDT)model based on the newly-developed Light Gradient Boosting Machine algorithm(LightGBM or LGBM)was proposed to model the internal temperature of a greenhouse.Fea-tures including climate variables,control variables and additional temporal information collected within five years were used to construct a suitable dataset to train and validate the LGBM model.An adaptive cross-validation method was developed as a novelty to improve the LGBM model performance and self-adaptive ability.For comparison of the pre-dictive accuracy,a Back-Propagation(BP)Neural Network model and a Recurrent Neural Network(RNN)model were built under the same process.Another two GBDT algorithms,Extreme Gradient Boosting(Xgboost)and Stochastic Gradient Boosting(SGB),were also introduced to compare the predictive accuracy with LGBM model.Results suggest that the LGBM has best fitting ability for the temperature curves with RMSE value at 0.645℃,as well as the fastest training speed among all algorithms with 60 times faster than the other two neural network algorithms.The LGBM has strongly potential application pro-spect on both greenhouse environment prediction and real-time predictive control.展开更多
基金Sponsored by Fundamental Research Funds for Central Universities of China(110604011,110304006)National Natural Science Foundation of China(61074098)
文摘For accurately forecasting the liquid steel temperature in ladle furnace （LF）, a novel temperature predic tion model based on optimally pruned Bagging combined with modified extreme learning machine （ELM） is pro posed. By analyzing the mechanism of LF thermal system, a thermal model with partial linear structure is obtained. Subsequently, modified ELM, named as partial linear extreme learning machine （PLELM）, is developed to estimate the unknown coefficients and undefined function of the thermal model. Finally, a pruning Bagging method is pro- posed to establish the aggregated prediction model for the sake of overcoming the limitation of individual predictor and further improving the prediction performance. In the pruning procedure, AdaBoost is adopted to modify the ag- gregation order of the original Bagging ensembles, and a novel early stopping rule is designed to terminate the aggre- gation earlier. As a result, an optimal pruned Bagging ensemble is achieved, which is able to retain Bagging＇s ro- bustness against highly influential points, reduce the storage needs as well as speed up the computing time. The pro- posed prediction model is examined by practical data, and comparisons with other methods demonstrate that the new ensemble predictor can improve prediction accuracy, and is usually consisted compactly.
基金This work was supported in part by Shanghai Agriculture Applied Technology Development Program,China(Grant No.G 2020-02-08-00-07-F01480)Shanghai Municipal Science and Technology Commission Innovation Action Plan(Grant No.17391900900)National Natural Science Foundation of China(Grant No.61573258).
文摘y consumption efficiency and to increase the crop yield.With the increase of agri-cultural data generated by the Internet of Things(IoT),more feasible models are necessary to get full usage of such information.In this research,a Gradient Boost Decision Tree(GBDT)model based on the newly-developed Light Gradient Boosting Machine algorithm(LightGBM or LGBM)was proposed to model the internal temperature of a greenhouse.Fea-tures including climate variables,control variables and additional temporal information collected within five years were used to construct a suitable dataset to train and validate the LGBM model.An adaptive cross-validation method was developed as a novelty to improve the LGBM model performance and self-adaptive ability.For comparison of the pre-dictive accuracy,a Back-Propagation(BP)Neural Network model and a Recurrent Neural Network(RNN)model were built under the same process.Another two GBDT algorithms,Extreme Gradient Boosting(Xgboost)and Stochastic Gradient Boosting(SGB),were also introduced to compare the predictive accuracy with LGBM model.Results suggest that the LGBM has best fitting ability for the temperature curves with RMSE value at 0.645℃,as well as the fastest training speed among all algorithms with 60 times faster than the other two neural network algorithms.The LGBM has strongly potential application pro-spect on both greenhouse environment prediction and real-time predictive control.
