Time Series Prediction Problems Research Articles

A novel intuitionistic fuzzy time series prediction model with cascaded structure for financial time series

Financial time series prediction problems, for decision-makers, are always crucial as they have a wide range of applications in the public and private sectors. This study presents a cascaded intuitionistic fuzzy model for financial time series prediction. The proposed prediction model has the ability to jointly and simultaneously model linear and nonlinear relationships in financial time series. Thus, it can adapt itself to both linear and non-linear surfaces of the data and can produce satisfactory predictions for financial time series. Moreover, the other reason why to be produced better predictions, the proposed model reckons non-membership degrees in addition to membership degrees in the prediction process. With these aspects, the proposed prediction model is different and superior to all models in the literature. This superiority has been proven by the analysis of 48 different financial time series containing TAIEX, DIJ, SSEC, and IEX data sets. The results have been evaluated in terms of RMSE, MAPE, and MdRAE metrics and some other perspectives as well. The proposed prediction model has achieved progress in prediction performance, up to 80% for TAIEX 2000–2004 datasets, 60% for TAIEX 2008–2018 datasets, approximately 50% for DJI and SSEC, and up to 70% for IEX. All the discussed indicators demonstrated the outstanding prediction performance of the proposed cascaded intuitionistic prediction model compared to some other state-of-the-art prediction tools.

Graph Convolutional Network-Guided Mine Gas Concentration Predictor

Coal mining work has always been a high-risk job, although mining technology is now regularly very mature, many accidents still occur every year in various countries around the world, most of which are due to gas explosions, poisoning, asphyxiation and other accidents. Therefore it is important to monitor and predict both underground mine air quality. In this paper, we use the GCN spatio-temporal graph convolution method based on spectral domain for multivariate time series prediction of underground mine air environment. The correlation of these sequences is learned by a self-attentive mechanism, without a priori graph, and the adjacency matrix with an attention mechanism is created dynamically. The temporal and spatial features are learned by graph Fourier transform and inverse Fourier transform in TC module (temporal convolution) and GC module (graph convolution), respectively. Besides, the corresponding experimental predictions are performed on other public datasets. And a new loss function is designed based on the idea of residuals, which greatly improves the prediction accuracy. In addition, the corresponding experimental predictions were performed on other public datasets. The results show that this model has outstanding prediction ability and high prediction accuracy on most time-series prediction data sets. Through experimental verification, this model has high prediction accuracy for dealing with multivariate time series prediction problems, both for long-term and short-term prediction.

Solar power time series forecasting utilising wavelet coefficients

Accurate and reliable prediction of photovoltaic (PV) power output is critical to electricity grid stability and power dispatching capabilities. However, PV power generation is highly volatile and unstable due to different reasons. The wavelet transform (WT) has been utilised in time series applications, such as PV power prediction, to model the stochastic volatility and reduce prediction errors. Yet the existing WT approach has a limitation in terms of time complexity. It requires reconstructing the decomposed components and modelling them separately and thus needs more time for reconstruction, model configuration and training. The aim of this study is to improve the efficiency of applying WT by proposing a new method that uses a single simplified model. Given a time series and its WT coefficients, it trains one model with the coefficients as features and the original time series as labels. This eliminates the need for component reconstruction and training numerous models. This work contributes to the day-ahead aggregated solar PV power time series prediction problem by proposing and comprehensively evaluating a new approach of employing WT. The proposed approach is evaluated using 17 months of aggregated solar PV power data from two real-world datasets. The evaluation includes the use of a variety of prediction models, including Linear Regression, Random Forest, Support Vector Regression, and Convolutional Neural Networks. The results indicate that using a coefficients-based strategy can give predictions that are comparable to those obtained using the components-based approach while requiring fewer models and less computational time.

Performance evaluation of sequence-to-sequence-Attention model for short-term multi-step ahead building energy predictions

Traditional building energy prediction(BEP) methods usually solve time-series prediction problems using either recursive strategy or direct strategy, which may ignore time-dependence between continuous building energy data in building energy systems. To overcome this issue, a sequence-to-sequence(Seq2seq) model combined with attention mechanism(Seq2seq-Att) is developed to realize multi-step ahead BEP. Compared with the original Seq2seq, both parameter-tuning and attention mechanism in the Seq2seq-Att model have great impacts on BEP performance improvement. To obtain quantitative analyses of performance improvement of these two aspects, this study conducted a comprehensive performance evaluation of four Seq2seq models (i.e., before and after parameter-tuning, adding attention and without attention). In this study, the length of sliding window is 24-h and prediction time steps ranges from 1-h to 12-h ahead. From the open-source Building Data Genome Project 2, 36 buildings are selected. Results indicate that adding attention to Seq2seq together with parameter-tuning, the multi-step ahead prediction performance can be increased by 8%(parameter-tuning around 6% while adding attention about 2%) on average. For prediction time step less than 3-h, parameter-tuning is a convenient way to improve the Seq2seq-based multi-step ahead BEP model. But for cases of prediction time step over 3-h, combining attention to the Seq2seq after parameter-tuning is recommended.

Customer active power consumption prediction for the next day based on historical profile

Energy consumption prediction application is one of the most important fields that is artificially controlled with Artificial Intelligence technologies to maintain accuracy for electricity market costs reduction. This work presents a way to build and apply a model to each costumer in residential buildings. This model is built by using Long Short Term Memory (LSTM) networks to address a demonstration of time-series prediction problem and Deep Learning to take into consideration the historical consumption of customers and hourly load profiles in order to predict future consumption. Using this model, the most probable sequence of a certain industrial customer’s consumption levels for a coming day is predicted. In the case of residential customers, determining the particular period of the prediction in terms of either a year or a month would be helpful and more accurate due to changes in consumption according to the changes in temperature and weather conditions in general. Both of them are used together in this research work to make a wide or narrow prediction window. A test data set for a set of customers is used. Consumption readings for any customer in the test data set applying LSTM model are varying between minimum and maximum values of active power consumption. These values are always alternating during the day according to customer consumption behavior. This consumption variation leads to leveling all readings to be determined in a finite set and deterministic values. These levels could be then used in building the prediction model. Levels of consumption’s are modeling states in the transition matrix. Twenty five readings are recorded per day on each hour and cover leap years extra ones. Emission matrix is built using twenty five values numbered from one to twenty five and represent the observations. Calculating probabilities of being in each level (node) is also covered. Logistic Regression Algorithm is used to determine the most probable nodes for the next 25 hours in case of residential or industrial customers. Index Terms—Smart Grids, Load Forecasting, Consumption Prediction, Long Short Term Memory (LSTM), Logistic Regression Algorithm, Load Profile, Electrical Consumption.

Reservoir production prediction with optimized artificial neural network and time series approaches

Numerical simulation of oil reservoirs is one of the most commonly used methods for reservoir production prediction, but its accuracy is based on accurate geological modeling and high-quality history matching. Therefore, numerical simulation is time-consuming and costly and requires extensive information. Traditional back-propagation neural networks and their improved algorithms are widely used for production prediction, but they are not suitable for time-series prediction problems. Based on variations in oil production, this study proposes a reservoir production prediction model based on a combined convolutional neural network (CNN) and a long short-term memory (LSTM) neural network model optimized by the particle swarm optimization (PSO) algorithm. First, the model extracts important temporal data features through the upper CNN, which is next imported to the lower LSTM network to further extract correlation features in the time dimension; then, it iteratively optimizes the key hyperparameters in the CNN-LSTM model through the PSO algorithm; finally, it uses the trained model for reservoir prediction. Compared with the training results of the LSTM neural network and CNN model, the PSO–CNN-LSTM model has higher prediction accuracy in time-series production prediction. Our proposed hybrid model is a data-driven method and is based on routinely available production data. Quick and accurate production prediction can lead to better informed operational decisions and optimization of recovery and economics.

Data-Driven State Prediction and Analysis of SOFC System Based on Deep Learning Method

A solid oxide fuel cell (SOFC) system is a kind of green chemical-energy–electric-energy conversion equipment with broad application prospects. In order to ensure the long-term stable operation of the SOFC power-generation system, prediction and evaluation of the system’s operating state are required. The mechanism of the SOFC system has not been fully revealed, and data-driven single-step prediction is of little value for practical applications. The state-prediction problem can be regarded as a time series prediction problem. Therefore, an innovative deep learning model for SOFC system state prediction is proposed in this study. The model uses a two-layer LSTM network structure that supports multiple sequence feature inputs and flexible multi-step prediction outputs, which allows multi-step prediction of system states using SOFC system experimental data. Comparing the proposed model with the traditional ARIMA model and LSTM recursive prediction model, it is shown that the multi-step LSTM prediction model performs better than the ARIMA and LSTM recursive prediction models in terms of two evaluation criteria: root mean square error and mean absolute error. Thus, the proposed multi-step LSTM prediction model can effectively and accurately predict and evaluate the SOFC system’s state.

COME-UP: Computation Offloading in Mobile Edge Computing with LSTM Based User Direction Prediction

In mobile edge computing (MEC), mobile devices limited to computation and memory resources offload compute-intensive tasks to nearby edge servers. User movement causes frequent handovers in 5G urban networks. The resultant delays in task execution due to unknown user position and base station lead to increased energy consumption and resource wastage. The current MEC offloading solutions separate computation offloading from user mobility. For task offloading, techniques that predict the user’s future location do not consider user direction. We propose a framework termed COME-UP Computation Offloading in mobile edge computing with Long-short term memory (LSTM) based user direction prediction. The nature of the mobility data is nonlinear and leads to a time series prediction problem. The LSTM considers the previous mobility features, such as location, velocity, and direction, as input to a feed-forward mechanism to train the learning model and predict the next location. The proposed architecture also uses a fitness function to calculate priority weights for selecting an optimum edge server for task offloading based on latency, energy, and server load. The simulation results show that the latency and energy consumption of COME-UP are lower than the baseline techniques, while the edge server utilization is enhanced.

An end-to-end air pollutant concentrations prediction algorithm based on attention mechanism for targeted season: A case study in North China

Accurate air pollutant concentrations prediction allows effective environment management to reduce the impact of pollution. The encoder-decoder model based on long short-term memory (LSTM) demonstrated great potential in air pollutant concentrations prediction. However, the influence of the hidden vector on the output of the decoder at each moment may be different during the long time-series prediction problem. In this paper, an attention mechanism is introduced in the decoder part to further improve the final prediction of the pollutant concentrations by filtering out some noise. In the experimental stage, we exploited the data collected from five representative cities in North China (Beijing, Tianjin, Shijiazhuang, Taiyuan, and Baotou) from 2014 to 2019 as the experimental dataset, and added the auxiliary data unique to North China. Then we divided the datasets into four seasonal datasets (spring, summer, autumn, and winter) to obtain targeted seasonal prediction models for the different seasons. The experimental results show that the model can predict the future trend of the pollutant concentration in a certain season relatively accurately.

Research on Multi-step Mixed Predictiom Model of Coal Gasifier Furnace Temperature Based on Machine Learning

In order to ensure the safe and stable operation of the coal gasifier, it is particularly important to study a new and reliable method for predicting the temperature of the gasifier. In this paper, the coal gasifier furnace temperature prediction is transformed into a time series prediction problem, and according to the non-linear characteristics of the gasifier furnace temperature data set, the LSTM (Long Short-Term Memory) model in the machine learning algorithm is selected for the gasifier furnace.Then use the PSO (Particle Swarm optimization) algorithm to optimize the hyperparameters of the model. The results show that the LSTM model is feasible but the accuracy is slightly lacking. The prediction accuracy based on the POS-LSTM model is significantly improved. It is of great significance to accurately predict the temperature of the gasifier.

Design of a Regional Economic Forecasting Model Using Optimal Nonlinear Support Vector Machines.

Forecasting regional economic activity is a progressively significant element of regional economic research. Regional economic prediction can directly assist local, national, and subnational policymakers. Regional economic activity forecast can be employed for defining macroeconomic forces, such as prediction of stock market and cyclicality of national labor market movement. The recent advances of machine learning (ML) models can be employed to solve the time series prediction problem. Since the parameters involved in the ML model considerably influence the performance, the parameter tuning process also becomes essential. With this motivation, this study develops a quasioppositional cuckoo search algorithm (QOCSA) with a nonlinear support vector machine (SVM)-based prediction model, called QOCSO-NLSVM for regional economic prediction. The goal of the QOCSO-NLSVM technique is to identify the present regional economic status. The QOCSO-NLSVM technique has different stages such as clustering, preprocessing, prediction, and optimization. Besides, the QOCSO-NLSVM technique employs the density-based clustering algorithm (DBSCAN) to determine identical states depending upon the per capita NSDP growth trends and socio-economic-demographic features in a state. Moreover, the NLSVM model is employed for the time series prediction process and the parameters involved in it are optimally tuned by the use of the QOCSO algorithm. To showcase the effective performance of the QOCSO-NLSVM technique, a wide range of simulations take place using regional economic data. To determine the current economic situation in a region, the QOCSO-NLSVM technique is used. The simulation results reported the better performance of the QOCSO-NLSVM technique over recent approaches. The QOCSO-NLSVM technique generated effective results with a minimal mean square error of 70.548 or greater. Astonishingly good results were obtained using the QOCSO-NLSVM approach, which had the lowest root mean square error (RMSE) of 8.399.

Passive Sonar Target Tracking Based on Deep Learning

At present, the tracking accuracy of underwater passive target tracking is often limited due to models that are overly simple, with low complexity, poor universality, and an inability to learn. In this paper, a cubature Kalman filter (CKF) algorithm based on a gated recurrent unit (GRU) network is proposed. The filter innovation, prediction error, and filter gain obtained from the CKF are used as the input to the GRU network, and the filter error value is used as the output to train the network. End-to-end online learning is carried out using the designed fully connected network, and the current state of the target is predicted. In this paper, a deep neural network based on the GRU architecture is used to convert the tracking prediction problem into a time series prediction problem in the field of artificial intelligence, and its strong fitting ability is used to resolve the uncertainty of the target motion. Simulation results show that an unmanned underwater vehicle (UUV) state estimation method based on the GRU filter proposed in this paper offers better accuracy and stability than the traditional state estimation method.

ESS 최적운영 전략이 계통의 신뢰도에 미치는 영향에 관한 연구

Based on the empirical data published by the Korea Electric Power Exchange and the Korea Meteorological Administration, we present a methodology that predicts the system marginal prices. For this purpose, this study uses the long-short term memory (LSTM) technique, which is commonly used in time-series prediction problems. As a result, we present a strategy for operating energy storage systems (ESSs) from the perspective of a prosumer rather than a system operator or conventional utilities. This study also examines the effects of the proposed prosumer actions on the system. The proposed operating strategy is established based on dynamic programming. In particular, this study takes the proposed operation strategy of ESSs the point of view of system reliability.

Multivariate intuitionistic fuzzy inference system for stock market prediction: The cases of Istanbul and Taiwan

Many of decision-making and policy planning processes involve a time-series prediction problem and so this area has extensive literature including a great variety of time-series prediction tools and inferences systems. An important part of these is based on fuzzy sets. However, it is known that fuzzy sets may fail to satisfy or characterize the uncertainty of the data in a comprehensive manner because they cannot depict the neutrality degree of time-series. Another important and decisive deficiency of current inference systems is to based on the univariate structure. However, the time series dealt with in a prediction problem generally interact with other time series. Considering these issues, creating an inference system based on intuitionistic fuzzy sets and multivariate relationships for a time series prediction problem is a requirement even an obligation. With these regards, this study presents a multivariate intuitionistic fuzzy time-series definition and its prediction models and introduces a multivariate intuitionistic fuzzy inference system (M-IFIS). The basic novelty of the article can be expressed as the definition of a multivariate intuitionistic fuzzy time series, as well as the creation of a relevant analysis mechanism, first-time in the literature. Sigma-pi neural network is used as an inference tool in M-IFIS and membership and non-membership values and lagged crisp observations of multivariable time-series are used as inputs of it. In order to reveal the performance of the proposed system, Istanbul Stock Exchange (IEX) and Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX) are analysed and the results are evaluated as comprehensive and comparative. All findings reveal the superiority M-IFIS in predictive accuracy.

A New Neural Networks Approach Used to Improve Wind Speed Time Series Forecasting

Generally, wind turbines convert the energy of wind into electricity. In this order, it is essential to predict accurately this source’s availability and intensity at the same location and height where wind electric generators will be installed, and therefore obtain reliable time-series data. The problem of meteorological time series prediction can be formulated as a system identification problem. To improve the prediction of these meteorological time series, we describe then use an application of a new neural networks approach in this paper. This novel, robust, and reliable forecasting method is based on the application of a new learning algorithm that allows a renewal of learning data, with time. For our algorithm a neural network is developed to estimate just one value y (t+1), then it is taken up with a new learning set enriched by data freshly measured. The obtained results showed a good agreement between measured and predicted series, and the mean relative error over the whole data set, which are not exceeding 5 %.

Proportional integral derivative booster for neural networks-based time-series prediction: Case of water demand prediction

Multi-step time-series prediction is an essential supportive step for decision-makers in several industrial areas. Artificial intelligence techniques, which use a neural network component in various forms, have recently frequently been used to accomplish this step. However, the complexity of the neural network structure still stands up as a critical problem against prediction accuracy. In this paper, a method inspired by the proportional–integral–derivative (PID) control approach is investigated to enhance the performance of neural network models used for multi-step ahead prediction of periodic time-series information while maintaining a negligible impact on the complexity of the system. The PID-based method is applied to the predicted value at each time step to bring that value closer to the real value. The water demand forecasting problem is considered as a case study, where two deep neural network models from the literature are used to prove the effectiveness of the proposed boosting method. Furthermore, to prove the applicability of this PID-based booster to other types of periodic time-series prediction problems, it is applied to enhance the accuracy of a neural network model used for multi-step forecasting of hourly energy consumption. The comparison between the results of the original prediction models and the results after using the proposed technique demonstrates the superiority of the proposed method in terms of prediction accuracy and system complexity.

Application of Deep Learning Architectures for Satellite Image Time Series Prediction: A Review

Satellite image time series (SITS) is a sequence of satellite images that record a given area at several consecutive times. The aim of such sequences is to use not only spatial information but also the temporal dimension of the data, which is used for multiple real-world applications, such as classification, segmentation, anomaly detection, and prediction. Several traditional machine learning algorithms have been developed and successfully applied to time series for predictions. However, these methods have limitations in some situations, thus deep learning (DL) techniques have been introduced to achieve the best performance. Reviews of machine learning and DL methods for time series prediction problems have been conducted in previous studies. However, to the best of our knowledge, none of these surveys have addressed the specific case of works using DL techniques and satellite images as datasets for predictions. Therefore, this paper concentrates on the DL applications for SITS prediction, giving an overview of the main elements used to design and evaluate the predictive models, namely the architectures, data, optimization functions, and evaluation metrics. The reviewed DL-based models are divided into three categories, namely recurrent neural network-based models, hybrid models, and feed-forward-based models (convolutional neural networks and multi-layer perceptron). The main characteristics of satellite images and the major existing applications in the field of SITS prediction are also presented in this article. These applications include weather forecasting, precipitation nowcasting, spatio-temporal analysis, and missing data reconstruction. Finally, current limitations and proposed workable solutions related to the use of DL for SITS prediction are also highlighted.

Small perturbations are enough: Adversarial attacks on time series prediction

Time-series data are widespread in real-world industrial scenarios. To recover and infer missing information in real-world applications, the problem of time-series prediction has been widely studied as a classical research topic in data mining. Deep learning architectures have been viewed as next-generation time-series prediction models. However, recent studies have shown that deep learning models are vulnerable to adversarial attacks. In this study, we prospectively examine the problem of time-series prediction adversarial attacks and propose an attack strategy for generating an adversarial time series by adding malicious perturbations to the original time series to deteriorate the performance of time-series prediction models. Specifically, a perturbation-based adversarial example generation algorithm is proposed using the gradient information of the prediction model. In practice, unlike the imperceptibility to humans in the field of image processing, time-series data are more sensitive to abnormal perturbations and there are more stringent requirements regarding the amount of perturbations. To address this challenge, we craft an adversarial time series based on the importance measurement to slightly perturb the original data. Based on comprehensive experiments conducted on real-world time-series datasets, we verify that the proposed adversarial attack methods not only effectively fool the target time-series prediction model LSTNet, they also attack state-of-the-art CNN-, RNN-, and MHANET-based models. Meanwhile, the results show that the proposed methods achieve a good transferability. That is, the adversarial examples generated for a specific prediction model can significantly affect the performance of the other methods. Moreover, through a comparison with existing adversarial attack approaches, we can see that much smaller perturbations are sufficient for the proposed importance-measurement based adversarial attack method. The methods described in this paper are significant in understanding the impact of adversarial attacks on a time-series prediction and promoting the robustness of such prediction technologies.

CNN-BiLSTM water level prediction method with attention mechanism

Hydrological time series data is stochastic and complex, and the importance of its historical features is different. A single model is difficult to overcome its own limitations when dealing with hydrological time series prediction problems, and the prediction accuracy of a single model can be further improved. According to the characteristics of hydrological time series data, a CNN-BiLSTM water level prediction method with attention mechanism is proposed. In this paper, CNN extracts the spatial characteristics of water level data and BiLSTM learns the time period characteristics by combining the past and future sequence information, attention mechanism is introduced to focus the salient features in the sequence. Taking the hourly water level data of Pinghe basin in China as experimental basis, experimental result shows that this method is more accuracy than Support Vector Machine (SVM), Temporal Convolutional Neural network (TCN), and Bidirectional Long Short-Term Memory network (BiLSTM) model.

Energy Load Forecasting Using a Dual-Stage Attention-Based Recurrent Neural Network.

Providing a stable, low-price, and safe supply of energy to end-users is a challenging task. The energy service providers are affected by several events such as weather, volatility, and special events. As such, the prediction of these events and having a time window for taking preventive measures are crucial for service providers. Electrical load forecasting can be modeled as a time series prediction problem. One solution is to capture spatial correlations, spatial-temporal relations, and time-dependency of such temporal networks in the time series. Previously, different machine learning methods have been used for time series prediction tasks; however, there is still a need for new research to improve the performance of short-term load forecasting models. In this article, we propose a novel deep learning model to predict electric load consumption using Dual-Stage Attention-Based Recurrent Neural Networks in which the attention mechanism is used in both encoder and decoder stages. The encoder attention layer identifies important features from the input vector, whereas the decoder attention layer is used to overcome the limitations of using a fixed context vector and provides a much longer memory capacity. The proposed model improves the performance for short-term load forecasting (STLF) in terms of the Mean Absolute Error (MAE) and Root Mean Squared Errors (RMSE) scores. To evaluate the predictive performance of the proposed model, the UCI household electric power consumption (HEPC) dataset has been used during the experiments. Experimental results demonstrate that the proposed approach outperforms the previously adopted techniques.