Search

The predictions from time series data can help us sense development trends and make scientific decisions in advance. The commonly used forecasting methods with backpropagation consume a lot of computational resources. The deep echo state network (DeepESN) is an advanced prediction method with a deep neural network structure and training algorithm without backpropagation. In this paper, a Bayesian optimization algorithm (BOA) is proposed to optimize DeepESN to address the problem of increasing parameter scale. Firstly, the DeepESN was studied and constructed as the basic prediction model for the time series data. Secondly, the BOA was reconstructed, based on the DeepESN, for optimal parameter searching. The algorithm is proposed within the framework of the DeepESN. Thirdly, an experiment was conducted to verify the DeepESN with a BOA within three datasets: simulation data generated from computer programs, a real humidity dataset collected from Beijing, and a power load dataset obtained from America. Compared with the models of BP (backpropagation), LSTM (long short-term memory), GRU (gated recurrent unit), and ESN (echo state network), DeepESN obtained optimal results, which were 0.0719, 18.6707, and 764.5281 using RMSE evaluation. While getting better accuracy, the BOA optimization time was only 323.4 s, 563.2 s, and 9854 s for the three datasets. It is more efficient than grid search and grey wolf optimizer. [ABSTRACT FROM AUTHOR]

Accurate PM 2.5 forecasting provides a possibility for establishing an early warning system to notify the public and take precautionary measures to prevent negative effects on ambient air quality and public health. Considering strong seasonal variation in meteorological conditions, in this paper, a seasonal stacked autoencoder model combining seasonal analysis and deep feature learning is proposed for forecasting the hourly PM 2.5 concentration, named DL-SSAE model. The original data are firstly decomposed into four seasonal subseries according to the Chinese calendar, and then the Kendall correlation coefficient method is employed to search inherent relationships between PM 2.5 concentrations and meteorological parameters within 1-h ahead for each seasonal time series. The inherent relationships of each seasonal subseries are finally extracted, learned, and modeled by different deep neural networks (stacked autoencoders for regression), and the hourly PM 2.5 forecasts are yielded. The addressed model is tested by the dataset collected from three environmental monitoring stations in Beijing, China. The results demonstrate that the proposed model outperforms all other considered models with/without seasonality consideration in this paper. [ABSTRACT FROM AUTHOR]

Short-term traffic prediction (STTP) is one of the most critical capabilities in Intelligent Transportation Systems (ITS), which can be used to support driving decisions, alleviate traffic congestion and improve transportation efficiency. However, STTP of large-scale road networks remains challenging due to the difficulties of effectively modeling the diverse traffic patterns by high-dimensional time series. Therefore, this paper proposes a framework that involves a deep clustering method for STTP in large-scale road networks. The deep clustering method is employed to supervise the representation learning in a visualized way from the large unlabeled dataset. More specifically, to fully exploit the traffic periodicity, the raw series is first divided into a number of sub-series for triplet generation. The convolutional neural networks (CNNs) with triplet loss are utilized to extract the features of shape by transforming the series into visual images. The shape-based representations are then used to cluster road segments into groups. Thereafter, a model sharing strategy is further proposed to build recurrent NNs-based predictions through group-based models (GBMs). GBM is built for a type of traffic patterns, instead of one road segment exclusively or all road segments uniformly. Our framework can not only significantly reduce the number of prediction models, but also improve their generalization by virtue of being trained on more diverse examples. Furthermore, the proposed framework over a selected road network in Beijing is evaluated. Experiment results show that the deep clustering method can effectively cluster the road segments and GBM can achieve comparable prediction accuracy against the IBM with less number of prediction models. [ABSTRACT FROM AUTHOR]

Ultra-short-term load demand forecasting is significant to the rapid response and real-time dispatching of the power demand side. Considering too many random factors that affect the load, this paper combines convolution, long short-term memory (LSTM), and gated recurrent unit (GRU) algorithms to propose an ultra-short-term load forecasting model based on deep learning. Firstly, more than 100,000 pieces of historical load and meteorological data from Beijing in the three years from 2016 to 2018 were collected, and the meteorological data were divided into 18 types considering the actual meteorological characteristics of Beijing. Secondly, after the standardized processing of the time-series samples, the convolution filter was used to extract the features of the high-order samples to reduce the number of training parameters. On this basis, the LSTM layer and GRU layer were used for modeling based on time series. A dropout layer was introduced after each layer to reduce the risk of overfitting. Finally, load prediction results were output as a dense layer. In the model training process, the mean square error (MSE) was used as the objective optimization function to train the deep learning model and find the optimal super parameter. In addition, based on the average training time, training error, and prediction error, this paper verifies the effectiveness and practicability of the load prediction model proposed under the deep learning structure in this paper by comparing it with four other models including GRU, LSTM, Conv-GRU, and Conv-LSTM. [ABSTRACT FROM AUTHOR]

With the boom of big data, the Internet contains more and more personal behavior information, but it is difficult to extract effectively. A model involving multivariate processing capability must be constructed to deal with these time series with complex characteristics. In this paper, a novel hybrid model embedding Baidu Search Index is therefore proposed to implement multi-step ahead subway passenger flow forecasting. Firstly, we collect data from informative Baidu Search Index, reduce dimensionality, and screen out the powerful predictors by statistical analysis. Secondly, we extract matching common modes at similar time scales between the subway passenger flow and screened Baidu Search Index via multivariate mode decomposition being optimized by multi-objective algorithm. Furthermore, to eliminate pseudo statistical causality, we select the optimal combination of modal components between subway passenger flow and its corresponding Baidu Search Index at each time scale by an innovative multi–modal analysis strategy. Thirdly, we reconstruct the forecasting values of each selected optimal combination as the final results. The empirical results of Beijing, Shanghai and Guangzhou show that the proposed model can significantly outperform six benchmark models in both the level and directional accuracy. So introducing Baidu Search Index creates a sound opportunity to enhance the subway passenger flow forecasting ability. [ABSTRACT FROM AUTHOR]