Considering the recent developments in deep learning, it has become increasingly important to verify what methods are valid for the prediction of multivariate time-series data. In this study, we propose a novel method...Considering the recent developments in deep learning, it has become increasingly important to verify what methods are valid for the prediction of multivariate time-series data. In this study, we propose a novel method of time-series prediction employing multiple deep learners combined with a Bayesian network where training data is divided into clusters using K-means clustering. We decided how many clusters are the best for K-means with the Bayesian information criteria. Depending on each cluster, the multiple deep learners are trained. We used three types of deep learners: deep neural network (DNN), recurrent neural network (RNN), and long short-term memory (LSTM). A naive Bayes classifier is used to determine which deep learner is in charge of predicting a particular time-series. Our proposed method will be applied to a set of financial time-series data, the Nikkei Average Stock price, to assess the accuracy of the predictions made. Compared with the conventional method of employing a single deep learner to acquire all the data, it is demonstrated by our proposed method that F-value and accuracy are improved.展开更多
为了提高短期风电功率预测的准确性,提出一种基于贝叶斯优化和特征融合的xLSTM(extended Long Short-Term Memory)-Transformer模型。该模型综合应用长短期记忆(LSTM)网络的时序处理能力和Transformer的自注意力机制的动态特征融合能力...为了提高短期风电功率预测的准确性,提出一种基于贝叶斯优化和特征融合的xLSTM(extended Long Short-Term Memory)-Transformer模型。该模型综合应用长短期记忆(LSTM)网络的时序处理能力和Transformer的自注意力机制的动态特征融合能力。借助贝叶斯优化方法,模型可在较少的迭代次数条件下优化超参数,显著降低模型对计算资源的依赖。实验结果表明,内蒙古某风电场数据集上,与单一的LSTM模型、Transformer模型、门控循环单元(GRU)模型以及未采用贝叶斯优化和特征融合的xLSTM-Transformer模型相比,当步长(LookBack)为4和8时,所提模型的决定系数R2较基准模型平均提升1.2%~11.3%;平均绝对误差(MAE)平均降低12.8%~38.4%;均方根误差(RMSE)平均降低8.6%~35.8%。结果表明,所提模型在短历史输入条件下具有更高的预测精度与稳定性。展开更多
姿态控制系统是卫星系统中重要的组成部分,由于其高昂的造价,发生故障会引发恶劣的影响。随着航天科技的发展,卫星姿态控制系统也逐渐复杂,其可能发生故障的概率也随之增大。针对传统神经网络故障诊断结果缺少置信度、鲁棒性较差以及易...姿态控制系统是卫星系统中重要的组成部分,由于其高昂的造价,发生故障会引发恶劣的影响。随着航天科技的发展,卫星姿态控制系统也逐渐复杂,其可能发生故障的概率也随之增大。针对传统神经网络故障诊断结果缺少置信度、鲁棒性较差以及易发生过拟合的缺点,在对贝叶斯统计和深度学习理论研究的基础上,提出了一种基于贝叶斯线性层与贝叶斯卷积层的Bayesian Le Net结合的网络模型。通过对卫星姿态控制系统飞轮部件的故障数据分析和处理,进而采用该模型对故障仿真,并与贝叶斯全连接神经网络与传统Le Net进行对比,实验结果表明:在飞轮可能发生的三种故障前提下,上述网络模型准确率较高,过拟合现象较轻。验证了上述网络模型的有效性。展开更多
文摘Considering the recent developments in deep learning, it has become increasingly important to verify what methods are valid for the prediction of multivariate time-series data. In this study, we propose a novel method of time-series prediction employing multiple deep learners combined with a Bayesian network where training data is divided into clusters using K-means clustering. We decided how many clusters are the best for K-means with the Bayesian information criteria. Depending on each cluster, the multiple deep learners are trained. We used three types of deep learners: deep neural network (DNN), recurrent neural network (RNN), and long short-term memory (LSTM). A naive Bayes classifier is used to determine which deep learner is in charge of predicting a particular time-series. Our proposed method will be applied to a set of financial time-series data, the Nikkei Average Stock price, to assess the accuracy of the predictions made. Compared with the conventional method of employing a single deep learner to acquire all the data, it is demonstrated by our proposed method that F-value and accuracy are improved.
文摘为了提高短期风电功率预测的准确性,提出一种基于贝叶斯优化和特征融合的xLSTM(extended Long Short-Term Memory)-Transformer模型。该模型综合应用长短期记忆(LSTM)网络的时序处理能力和Transformer的自注意力机制的动态特征融合能力。借助贝叶斯优化方法,模型可在较少的迭代次数条件下优化超参数,显著降低模型对计算资源的依赖。实验结果表明,内蒙古某风电场数据集上,与单一的LSTM模型、Transformer模型、门控循环单元(GRU)模型以及未采用贝叶斯优化和特征融合的xLSTM-Transformer模型相比,当步长(LookBack)为4和8时,所提模型的决定系数R2较基准模型平均提升1.2%~11.3%;平均绝对误差(MAE)平均降低12.8%~38.4%;均方根误差(RMSE)平均降低8.6%~35.8%。结果表明,所提模型在短历史输入条件下具有更高的预测精度与稳定性。
文摘姿态控制系统是卫星系统中重要的组成部分,由于其高昂的造价,发生故障会引发恶劣的影响。随着航天科技的发展,卫星姿态控制系统也逐渐复杂,其可能发生故障的概率也随之增大。针对传统神经网络故障诊断结果缺少置信度、鲁棒性较差以及易发生过拟合的缺点,在对贝叶斯统计和深度学习理论研究的基础上,提出了一种基于贝叶斯线性层与贝叶斯卷积层的Bayesian Le Net结合的网络模型。通过对卫星姿态控制系统飞轮部件的故障数据分析和处理,进而采用该模型对故障仿真,并与贝叶斯全连接神经网络与传统Le Net进行对比,实验结果表明:在飞轮可能发生的三种故障前提下,上述网络模型准确率较高,过拟合现象较轻。验证了上述网络模型的有效性。
