A temperature forecasting model was created firstly based on the Kalman filter method,and then used to predict the highest and lowest temperature in Nanchang station from October 27 to November 1,2017.Finally,accordin...A temperature forecasting model was created firstly based on the Kalman filter method,and then used to predict the highest and lowest temperature in Nanchang station from October 27 to November 1,2017.Finally,according to the empirical forecasting method,guidance forecasts were established for the northern,central,and southern parts of Nanchang City.After inspection,it was found that the temperature prediction model established based on the Kalman filter method in Nanchang station had good prediction performance,and especially in the 24-hour forecast,it had advantages over the European Center.The accuracy of low temperature forecast was better than that of high temperature forecast.展开更多
矿山环境由于其复杂性往往会增加车辆事故发生的可能性。为提高矿山车辆预警系统的准确性,研究通过V2X(Vehicle to Everything)技术实现车辆信息共享,结合具备卡尔曼滤波器的神经网络与双向循环神经网络(KalmanNet-BRNN)模型对车辆状态...矿山环境由于其复杂性往往会增加车辆事故发生的可能性。为提高矿山车辆预警系统的准确性,研究通过V2X(Vehicle to Everything)技术实现车辆信息共享,结合具备卡尔曼滤波器的神经网络与双向循环神经网络(KalmanNet-BRNN)模型对车辆状态进行估计。研究结果中,在4种数据集上,研究方法的均方误差平均降低了7.60、1.70、11.01和3.19,动态时间规整平均降低了2.05 m、3.50 m、2.73 m和5.38 m。结果表明,融合V2X技术和KalmanNet-BRNN网络架构可用于矿山车辆的预警,同时研究方法在多个数据集上均方误差最低,动态时间规整值最小,表明了研究方法能够提升矿山车辆预警系统的准确性,同时预测的车辆状态信息更为接近车辆真实状态;研究不仅提高了矿山作业的安全性,同时可为矿山企业的运营效率的提升提供数据支撑。展开更多
文摘A temperature forecasting model was created firstly based on the Kalman filter method,and then used to predict the highest and lowest temperature in Nanchang station from October 27 to November 1,2017.Finally,according to the empirical forecasting method,guidance forecasts were established for the northern,central,and southern parts of Nanchang City.After inspection,it was found that the temperature prediction model established based on the Kalman filter method in Nanchang station had good prediction performance,and especially in the 24-hour forecast,it had advantages over the European Center.The accuracy of low temperature forecast was better than that of high temperature forecast.
