首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到2条相似文献,搜索用时 0 毫秒
1.
Data-driven techniques are used extensively for hydrologic time-series prediction. We created various data-driven models (DDMs) based on machine learning: long short-term memory (LSTM), support vector regression (SVR), extreme learning machines, and an artificial neural network with backpropagation, to define the optimal approach to predicting streamflow time series in the Carson River (California, USA) and Montmorency (Canada) catchments. The moderate resolution imaging spectroradiometer (MODIS) snow-coverage dataset was applied to improve the streamflow estimate. In addition to the DDMs, the conceptual snowmelt runoff model was applied to simulate and forecast daily streamflow. The four main predictor variables, namely snow-coverage (S-C), precipitation (P), maximum temperature (Tmax), and minimum temperature (Tmin), and their corresponding values for each river basin, were obtained from National Climatic Data Center and National Snow and Ice Data Center to develop the model. The most relevant predictor variable was chosen using the support vector machine-recursive feature elimination feature selection approach. The results show that incorporating the MODIS snow-coverage dataset improves the models' prediction accuracies in the snowmelt-dominated basin. SVR and LSTM exhibited the best performances (root mean square error = 8.63 and 9.80) using monthly and daily snowmelt time series, respectively. In summary, machine learning is a reliable method to forecast runoff as it can be employed in global climate forecasts that require high-volume data processing.  相似文献   

2.
ABSTRACT

Wind speed forecasting plays an important role in power grid dispatching management. This article proposes a short-term wind speed forecasting method based on random forest model combining ensemble empirical modal decomposition and improved harmony search algorithm. First, the initial wind speed data set is decomposed into several ensemble empirical mode functions by EEMD, then feature extraction of each sub-modal IMF is performed using fast Fourier transform to solve the cycle of each sub-modal IMF. Next, combining the high-performance parameter optimization ability of the improved harmony search algorithm, two optimal parameters of random forest model, number of decision trees, and number of split features are determined. Finally, the random forest model is used to forecast the processing results of each submodal IMF. The proposed model is applied to the simulation analysis of historical wind data of Chaoyang District, Liaoning Province from April 27, 2015 to May 22, 2015. To illustrate the suitability and superiority of the EEMD-RF-IHS model, three types of models are used for comparison: single models including ANN, SVM, RF; EMD combination models including EMD-ANN, EMD-SVM, EMD-RF; EEMD combination models including EEMD-ANN, EEMD-SVM, EEMD-RF. The analysis results of evaluation indicators show that the proposed model can effectively forecast short-term wind data with high stability and precision, providing a reference for forecasting application in other industry fields.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号