Search

Accurate photovoltaic (PV) power forecast is crucial for carbon neutrality. Current researches on PV power forecast mainly focus on using temporal information from single PV station, and the spatial information in multiple PV power stations are often neglected. To address this problem, this paper introduces Moran index to verify the spatial autocorrelation of PV power for the first time, uses Granger causality test and transfer entropy to reveal the spatial information gain in PV power forecast for the first time, and proposes a novel spatial–temporal probabilistic PV forecast method using deep Graph Gated Recurrent Unit (GraphGRU) network-based spatial–temporal multi-task learning and Kernel Density Estimation (KDE). Deep GraphGRU combines the advantages of Graph Convolutional Network (GCN) in spatial feature extraction and the advantages of Gated Recurrent Unit (GRU) network in temporal feature extraction, and thus has strong ability to extract spatial–temporal information in historical data of multiple different PV power stations. Through GraphGRU, temporal dependency information extracted from historical data of multiple PV stations can promote each other to improve the forecast accuracy of each PV stations. KDE is used for estimating the joint probabilistic density function and giving the spatial–temporal probabilistic confidence interval of PV power. Experiments were performed in the five-year actual PV power data from 11 provinces of Belgium and the three-year solar irradiation data from 12 places in China to verify the superiorities of the proposed method. Comparison with conventional spatial–temporal and temporal forecast methods show that the proposed GraphGRU-based spatial–temporal forecast method can extract the spatial–temporal information from multiple PV power stations well and significantly outperform conventional temporal forecast methods. [ABSTRACT FROM AUTHOR]

Accurate forecasting of the agricultural power load is important for rural electricity networks' safe and stable operation. But it is more uncertain and difficult to forecast than industrial, commercial, and residential loads. Multivariate time series prediction methods have problems such as poor refinement of weather services and difficulty in quantifying the influencing factors. Univariate time series prediction methods should be emphasized, and the characteristic information in the historical load data should be fully exploited. Therefore, this paper systematically investigates the decomposition strategy and attention-based recurrent neural network for univariate time-series multi-step prediction. The decomposition strategy is employed to generate derived variable sequences from the target sequence as inputs for the model, while the dual-stage attention mechanism is used for dynamically selecting features. A one-year real-world dataset of agricultural power loads in southern China is used for validation. The effectiveness of two different decomposition algorithms on dual-stage attention-based recurrent neural network (DARNN) is discussed, then the proposed ensemble empirical mode decomposition and dual-stage attention-based recurrent neural network (EEMD-DARNN) is compared with four univariate time series prediction models and three ablation models. The proposed model is optimal in four evaluation indicators, with R 2 , R M S E , M A P E , and M A E of 0.970, 12.298, 0.040, and 7.335 respectively in the shortest forecast horizon (15 min), and 0.941, 17.305, 0.065 and 11.347 respectively in the longest forecast horizon (1 h). It is a significant improvement over the existing models. In conclusion, EEMD-DARNN can achieve more accurate ultra-short-term agricultural power load forecasting. [ABSTRACT FROM AUTHOR]

The air pollutant diffusion trend prediction plays an important role in the environment protection. The approaches in the existing studies rarely consider spatial and temporal information sequentially or cannot use spatiotemporal structure information effectively. In this paper, we propose a new deep learning-based air pollutant diffusion trend prediction model, called GATBL-Learning. It is an end-to-end neural network model consisting of Graph Attention Network (GAT) and Encoder-Decoder model. Since each city in the region will set as a node, GAT can extract spatial information from them by aggregation operation. Then Encoder-Decoder model, consisting of Bi-LSTM as Encoder and LSTM as Decoder, mines time series features and outputs the data containing spatiotemporal features. In addition, we also observe that the changes of air pollutants show seasonality discipline and training the seasonal model can improve the model adaptability. The results demonstrate the proposed prediction model is effective and superior. [ABSTRACT FROM AUTHOR]

Integrating return prediction of traditional time series models in portfolio formation can improve the performance of original portfolio optimization model. Since machine learning and deep learning models have shown overwhelming superiority than time series models, this paper combines return prediction in portfolio formation with two machine learning models, i.e., random forest (RF) and support vector regression (SVR), and three deep learning models, i.e., LSTM neural network, deep multilayer perceptron (DMLP) and convolutional neural network. To be specific, this paper first applies these prediction models for stock preselection before portfolio formation. Then, this paper incorporates their predictive results in advancing mean–variance (MV) and omega portfolio optimization models. In order to present the superiority of these models, portfolio models with autoregressive integrated moving average's return prediction are used as benchmarks. Evaluation is based on historical data of 9 years from 2007 to 2015 of component stocks of China securities 100 index. Experimental results show that MV and omega models with RF return prediction, i.e., RF+MVF and RF+OF, outperform the other models. Further, RF+MVF is superior to RF+OF. Due to the high turnover of these two models, this paper discusses their performance after deducting the transaction fee cased by turnover. Experiments present that RF+MVF still performs the best among MVF models and omega model with SVR prediction (SVR+OF) performs the best among OF models. Moreover, RF+MVF performs better than SVR+OF and high turnover erodes nearly half of their total returns especially for RF+OF and RF+MVF. Therefore, this paper recommends investors to build MVF with RF return prediction for daily trading investment. • Compares the performance of machine learning and deep learning in stock preselection. • Combining return prediction of machine learning and deep learning in portfolio formation. • Emphasis on advancing portfolio optimization with return prediction. • Advanced mean–variance model with random forest forecasts performs the best. [ABSTRACT FROM AUTHOR]

• A grey seasonal trend support vector machine for energy forecasting is proposed. • The model can efficiently identify the seasonal fluctuations in energy data. • The model is powerful to predict China's energy production. • The predictive performance of the new model is much superior to the benchmarks. This paper proposed a hybrid ensemble forecasting technique that incorporates the merits of the accumulation generation operation (AGO), least-square support vector regression (LSSVR), dummy variable, and time trend item to forecast the seasonal time series characterized by nonlinearity and uncertainty. This proposed framework, a grey seasonal trend least squares support vector machine (GSTLSSVR), can make it easy to add arbitrary seasonality to any feature, which improves model realism significantly. Furthermore, the grid search method and the last block validation are developed to identify the optimal hyperparameter, making it possible for the developed method to capture seasonal and nonlinear patterns of the original data. For illustration and verification, the quarterly petroleum coke production and monthly crude oil production in China are employed to construct diverse competing models, including the seasonal fluctuation grey model (SGM(1,1)), seasonal autoregressive integrated moving average (SARIMA), Holt-Winters (HW), backpropagation neural network (BPNN), a grey seasonal least squares support vector machine (GSLSSVR), long short term memory (LSTM) and LSSVR. The forecasting values of quarterly petroleum coke production are 668.74, 682.68, 679.99, 704.35 for 2019 and 717.59, 734.99, 727.16, 741.53 for 2020, with the lowest MAPEP value (3.7566%). Moreover, the monthly predicted crude oil production for 2019 are 3113.9, 1645, 1562.2, 1625.4, 1599.3, 1619.1, 1627, 1576.9, 1629.6, 1583.6, and 1615.6, while those counterparts for 2020 are 3093.2, 1625, 1547.5, 1621.5, 1610.6, 1644.9, 1662.9, 1615.4, 1667.4, 1615, 1667.4, 1615, and 1640.2, with the lowest MAPEP value (1.0034%). Generally, the empirical results show that the proposed model obtains superior overall performance to other benchmarks in both case studies. Therefore, this proposed model exhibits promising prospects for identifying seasonal patterns in energy production sequences. [ABSTRACT FROM AUTHOR]

• Enhanced linear and nonlinear framework for interval-valued time series is proposed. • Multi-step-ahead forecasting for decomposition-ensemble method is performed. • Multi-factor interval division can mine the internal information of the interval. • Bivariate decomposition algorithm is introduced for multi-scale modeling. • The proposed model outperforms other state-of-the-art methods. To forecast possible air pollution risks, a large number of models have been built to predict the daily average or hourly value of PM 2.5 concentrations. In fact, due to the complexity of PM 2.5 formation and the diversity of its influencing factors, PM 2.5 concentrations may fluctuate greatly during a certain period, single-valued methods cannot present this uncertainty and variability. Therefore, this paper develops a hybrid modeling framework that combines the elastic net (EN) and multivariate relevance vector machine (MVRVM) for interval-valued PM 2.5 time series (IPMTS) forecasting. Following the philosophy of the "linear and nonlinear" modeling framework, the EN and MVRVM are applied to capture the linear and nonlinear patterns hidden in the IPMTS. Different from existing studies that directly model linear and nonlinear patterns of time series, we introduce the multi-factor interval division (MFID) strategy and bivariate empirical mode decomposition (BEMD) algorithm into linear and nonlinear pattern modeling, respectively. This enhanced linear and nonlinear framework (termed MFID-EN-BEMD-MVRVM) can consider the possible linear relationship between PM 2.5 and its related factors within the intervals and model the nonlinear pattern of IPMTS at multiple scales. The IPMTS extracted from four municipalities in China are used to validate the efficiency of the proposed model. Experimental results show that the MFID-EN-BEMD-MVRVM model achieves better forecasting performance than some benchmarks and state-of-the-art models in terms of comprehensive evaluation criteria. In addition, from the application analysis, the proposed interval-valued PM 2.5 modeling architecture has the possibility of real-time forecasting. [ABSTRACT FROM AUTHOR]

Abstract: How to effectively predict financial distress is an important problem in corporate financial management. Though much attention has been paid to financial distress prediction methods based on single classifier, its limitation of uncertainty and benefit of multiple classifier combination for financial distress prediction has also been neglected. This paper puts forward a financial distress prediction method based on weighted majority voting combination of multiple classifiers. The framework of multiple classifier combination system, model of weighted majority voting combination, basic classifiers’ voting weight model and basic classifiers’ selection principles are discussed in detail. Empirical experiment with Chinese listed companies’ real world data indicates that this method can greatly improve the average prediction accuracy and stability, and it is more suitable for financial distress prediction than single classifiers. [Copyright &y& Elsevier]

The aim of this paper is the generation of a time-series based statistical data-driven procedure in order to track an outbreak. At first are used univariate time series models in order to predict the evolution of the reported cases. Moreover, are considered combinations of the models in order to provide more accurate and robust results. Additionally, statistical probability distributions are considered in order to generate future scenarios. Final step is the build and use of an epidemiological model (tSIR) and the calculation of an epidemiological ratio (R 0) for estimating the termination of the outbreak. The time series models include Exponential Smoothing and ARIMA approaches from the classical models, also Feed-Forward Artificial Neural Networks and Multivariate Adaptive Regression Splines from the machine learning toolbox. Combinations include simple mean, Newbolt-Granger and Bates-Granger approaches. Finally, the tSIR model and the R 0 ratio are used for estimating the spread and the reversion of the pandemic. The suggested procedure is used to track the COVID-19 epidemic in Greece. This epidemic has appeared in China in December 2019 and has been widespread since then to all over the world. Greece is the center of this empirical study as is considered an early successful paradigm of resistance against the virus. [ABSTRACT FROM AUTHOR]