One-step Forecasts Research Articles

Ultra-short-term global horizontal irradiance forecasting based on a novel and hybrid GRU-TCN model

The need for energy is increasing globally due to a several factors, including population growth and economic development. Achieving this energy demand in the face of global warming and the depletion of fossil fuels requires the use of renewable energy. Photovoltaic energy is one of the renewable energy sources that is widely used in many nations across the world. Photovoltaic (PV) energy integration into the grid has significant benefits for the environment and economy, but at high penetration levels, its intermittent nature makes system stability difficult to maintain. Accurate ultra-short-term global horizontal irradiance forecasting is necessary in order to guarantee the most optimal use of photovoltaic power production sources. For GHI forecasting, a novel GRU-TCN-based model is proposed in this paper. It is composed of two neural networks: a temporal convolutional network and a gated recurrent unit. After extracting the temporal features from time-series solar irradiance data using GRU, the spatial features are obtained from the correlation matrix of different meteorological variables of the target and its neighbor position using TCN. Univariate and multivariate GRU-TCN models have been used for GHI ultra short-term forecasting. This paper compares the univariate and multivariate GRU-TCN models with TCN, LSTM, and GRU models based on three evaluation metrics in order to investigate how different combinations of variables affect the accuracy of the GRU-TCN models for one-step forecasting. The findings indicate the adoption of a univariate model using historical data of GHI is suitable to obtain reliable forecasting with 23.02 (W/m2) in MAE as opposed to the best multivariate model that achieved 25.67 (W/m2) in MAE. According to the results, the proposed model outperforms the other models assessed and offers a practical alternative for ultra-short-term GHI forecasting.

Enhancing Multi-step Brent Oil Price Forecasting with Ensemble Multi-scenario Bi-GRU Networks

Accurate crude oil price forecasting is crucial for economic stability, investment planning, and strategic decision-making across various industries. Despite numerous research efforts in applying deep learning to time-series forecasting, achieving high accuracy in multi-step predictions for volatile time-series like crude oil prices remains a significant challenge. Moreover, most existing approaches primarily focus on one-step forecasting, and the performance often varies depending on the dataset and specific case study. This paper introduces ensemble-based deep-learning models to capture Brent oil price volatility and enhance the multi-step price prediction. Our methodology employs a two-pronged approach. First, we present an empirical comparison of deep-learning models and architectures, including RNNs, CNNs, and transformers, for forecasting Brent oil prices. We also examine the impact of various external factors on forecasting accuracy. Then, we introduce a novel approach that employs ensemble GRU-based models to enhance prediction accuracy across multiple forecasting scenarios. Extensive experiments were conducted using a dataset of historical Brent prices encompassing the COVID-19 pandemic, which significantly impacted energy markets. The results demonstrate that the proposed model outperforms benchmark and established models, achieving a 9.3% reduction in MSE compared to the closest benchmark model for a 3-day forecasting horizon.

Real-time forecasting of TBM cutterhead torque and thrust force using aware-context recurrent neural networks

Tunnel Boring Machines (TBMs) are instrumental in the construction of modern tunnels, known for their operational reliability and efficiency. The real-time prediction of cutterhead loads is essential for effective project scheduling, cost management, and risk reduction. This study develops machine learning models for predicting future cutterhead torque and thrust force, simultaneously. The dataset is derived from the Yingsong water diversion project and consists of 12,962 steady-state boring cycles. Because geological conditions and setting values are available in advance, we build a geological parameter-based model and a setting value-based model for regression analysis of cutterhead torque and thrust force. In one-step forecasts, the operational parameter-based model closely captures the trend of the measured results, employing the recurrent neural network (RNN), long short-term memory (LSTM) and gated recurrent unit (GRU). We propose an aware-context recurrent neural network (AC-RNN) model that integrates historical operational parameters and the aware context of setting values, leading to a significant improvement in forecasting accuracy. A sensitivity analysis is carried out to quantify the relative importance of input parameters, revealing that the setting value of revolutions per minute is crucial in forecasting. The results of this study offer valuable practical insights into real-time forecasting, thereby informing engineering applications in this field.

Effects of data smoothing and recurrent neural network (RNN) algorithms for real-time forecasting of tunnel boring machine (TBM) performance

Tunnel boring machines (TBMs) have been widely utilised in tunnel construction due to their high efficiency and reliability. Accurately predicting TBM performance can improve project time management, cost control, and risk management. This study aims to use deep learning to develop real-time models for predicting the penetration rate (PR). The models are built using data from the Changsha metro project, and their performances are evaluated using unseen data from the Zhengzhou Metro project. In one-step forecast, the predicted penetration rate follows the trend of the measured penetration rate in both training and testing. The autoregressive integrated moving average (ARIMA) model is compared with the recurrent neural network (RNN) model. The results show that univariate models, which only consider historical penetration rate itself, perform better than multivariate models that take into account multiple geological and operational parameters (GEO and OP). Next, an RNN variant combining time series of penetration rate with the last-step geological and operational parameters is developed, and it performs better than other models. A sensitivity analysis shows that the penetration rate is the most important parameter, while other parameters have a smaller impact on time series forecasting. It is also found that smoothed data are easier to predict with high accuracy. Nevertheless, over-simplified data can lose real characteristics in time series. In conclusion, the RNN variant can accurately predict the next-step penetration rate, and data smoothing is crucial in time series forecasting. This study provides practical guidance for TBM performance forecasting in practical engineering.

Short-Term Electrical Load Forecasting in Power Systems Using Deep Learning Techniques

The use of big data in deep neural networks has recently surpassed traditional machine learning techniques in many application areas. The main reasons for the use of deep neural networks are the increase in computational power made possible by graphics processing units and tensor processing units, and the new algorithms created by recurrent neural networks and CNNs. In addition to traditional machine learning methods, deep neural networks have applications in anticipating electricity load. Using a real dataset for one-step forecasting, this article compares three deep learning algorithms for short-term power load forecasting: LSTM, GRUs, and CNN. The statistics come from the Turkish city of Zonguldak and include hourly electricity usage loads and temperatures over a period of three years, commencing in 2019 and ending in 2021. The mean absolute percentage error is used to compare the performances of the techniques. Forecasts are made for twelve representative months from each season. The main reason for the significant deviations in the forecasts for January, May, September, and December is the presence of religious and national holidays in these months. This was solved by adding the information obtained from religious and national holidays to the modeling. This is not to say that CNNs are not good at capturing long-term dependencies and modeling sequential data. In all experiments, LSTM, GRUs and encoder-decoder LSTM outperformed simple CNN designs. In the future, these new architectural methods can be applied to long- or short-term electric charge predictions and their results can be compared to LSTM, GRUs and their variations.

Original article Forecasting trends in the development of time series by estimating the Hodges-Lehman median

Introduction. The lack of stationarity in the time series’ development, small samples, the outliers presence, jumps do not allow to find estimates of model parameters that have good properties of statistical estimates and, as a result, to find reliable forecasts, both in the development trend of the process and in its numerical expression. The means to solve these problems is the use of ordinal or robust statistics. Scientific monographs and special chapters in books on mathematical statistics contain a deep and extensive theory on the study of the properties of ordinal statistics, which are the justification for their application in forecasting methods. The aim of the work is to develop and verify method for obtaining one-step forecasts of trends in the development of time series based on stable statistical estimates.Materials and methods. The article presents the results of the development of a method for obtaining one-step forecasts of trends in the development of time series based on the construction of confidence intervals of a selective stable HodgesLehman estimate based on Walsh averages. In particular, the Hodges-Lehman median is used to solve the problem of small samples obtained during the procedure of shifting the time series window. The proposed method is considered in detail in the article: the basic definitions, the theoretical justification of the method, calculation formulas, a detailed description of the algorithm, formulas for calculating the quality metric of forecasts are given.The results of the study. The method was implemented in computational experiment using the example of forecasting the Urals crude oil’s spot price. The article presents the computational experiment’s results. The parameters of the proposed method can be configured to obtain reliable one-step forecasts.Discussion and conclusions. The method proposed in the article has shown its effectiveness on experimental data and can be used as an independent method for constructing one-step forecasts of trends in the development of time series. Further development of the method involves the improvement of computational procedures, verification of the method in case of jumps in the dynamics of the time series.

Cutting state estimation and time series prediction using deep learning for Cutter Suction Dredger

With the increasing environmental protection standard, the Cutter Suction Dredger (CSD) construction is required to guarantee precise dredging in some special areas. Specifically, dredging the side slope of channeling, which requires the operator to accurately control the CSD. However, the underwater cutting environment of CSD construction is complicated and the soil kinds are sometimes spatially different, which makes it extremely difficult for the operator to accurately control the CSD. Therefore, we propose a state-of-the-art time series prediction model based on CNN-LSTM Encode-Decode, which can predict the cutting status of the cutter in advance and also ahead estimate the soil kinds to be cut. First, we establish the relationship between the cutting torque and the cutter motor power (CMP) and rotation speed (CMRS). By ahead forecasting the CMP and CMRS, the cutting torque at the next moment can be obtained. Then, the experienced operator can estimate the kind of underwater cutting soil by torque. In addition, we also compare the ahead one-step forecasting and ahead multi-step forecasting of this method with other models and conduct an experiment to verify the performance. The comparison results demonstrate that our approach significantly outperforms, which can help the operator's ahead perception of the underwater cutter status and ensure precise dredging.

Deep Reinforcement Learning-Based Operation of Transmission Battery Storage with Dynamic Thermal Line Rating

It is well known that dynamic thermal line rating has the potential to use power transmission infrastructure more effectively by allowing higher currents when lines are cooler; however, it is not commonly implemented. Some of the barriers to implementation can be mitigated using modern battery energy storage systems. This paper proposes a combination of dynamic thermal line rating and battery use through the application of deep reinforcement learning. In particular, several algorithms based on deep deterministic policy gradient and soft actor critic are examined, in both single- and multi-agent settings. The selected algorithms are used to control battery energy storage systems in a 6-bus test grid. The effects of load and transmissible power forecasting on the convergence of those algorithms are also examined. The soft actor critic algorithm performs best, followed by deep deterministic policy gradient, and their multi-agent versions in the same order. One-step forecasting of the load and ampacity does not provide any significant benefit for predicting battery action.

Success and challenges in predicting TBM penetration rate using recurrent neural networks

Tunnel Boring Machines (TBMs) have been increasingly used in tunnelling projects. Forecasting future TBM performance would be desirable for project time management and cost control. We aim to use recurrent neural networks to predict the near future TBM penetration rate from historical data. Our datasets are composed of Changsha and Zhengzhou metro lines, with totally different geological conditions. In the experiments, the one-step forecast of TBM penetration rate by the traditional recurrent neural network (RNN) or long short-term memory (LSTM) is relatively accurate, irrespective of the different geological conditions used in training and evaluation. Predicting the next Nth step penetration rate proves to be more challenging and depends on the time to the future or the distance ahead of the TBM cutterhead. There are generally time lags between measured and predicted results. The recursive RNN is then developed to address the lag problems, but to no avail. Alternative methods for predicting future penetration rates are studied, including the penetration rate at the Nth step in the future and the average penetration rate of the next N steps, with the latter being trained by long-input or short-input methods. The average N-step forecast using short inputs provides the best results, and its performance over other alternatives becomes more distinct as the number N increases. We also discuss the possibility of the forecast problem as a quasi-random walk, which means that forecasting penetration rate cannot easily be achieved using low-frequency data with RNNs, and that the accuracy depends on the correlation between the last and predicted steps in the data.

Univariate and Multivariate LSTM Models for One Step and Multistep PV Power Forecasting

The energy demand is increasing due to population growth and economic development. To satisfy this energy demand, the use of renewable energy is essential to face global warming and the depletion of fossil fuels. Photovoltaic energy is one of the renewable energy sources, widely used by several countries over the world. The integration of PV energy into the grid brings significant benefits to the economy and environment, however, high penetration of this energy also brings some challenges to the stability of the electrical grid, due to the intermittency of solar energy. To overcome this issue, the use of a forecasting system is one of the solutions to guarantee an effective integration of PV plants in the electrical grid. In this paper, a PV power ultra short term forecasting has been done by using univariate and multivariate LSTM models. Different combinations of input variables of the models and different timesteps forecasting were tested and compared. The main aim of this work is to study the influence of the different combinations of variables on the accuracy of the LSTM models for one-step forecasting and multistep forecasting and comparing the univariate and multivariate LSTM models with MLP and CNN models . The results show that for one step forecasting, the use of a univariate model based on historical data of PV output power is sufficient to get accurate forecasting with 28.98W in MAE compared to multivariate models that can reach 35.39W. Meanwhile, for multistep forecasting, it is mandatory to use a multivariate model that has historical data of meteorological variables and PV output power in the input of LSTM model. Moreover, The LSTM model shows great accuracy compared to MLP and CNN especially in multistep PV power forecasting.

A hybrid deep learning model approach for performance index prediction of mechanical equipment

To forecast the health status of mechanical equipment in industrial production, fault diagnosis systems need a fast and accurate algorithm to forecast the important performance indexes of mechanical equipment. According to the characteristics of time series, a composite variable wavelet transform, deep autoencoder and long short-term memory (CWD-LSTM) hybrid neural network forecast algorithm is proposed to carry out one-step forecast experiments on air compressor datasets. As one of the important indexes reflecting the performance of the air compressor, loading time is usually a parameter that the fault diagnosis system needs to forecast and analyze. The experimental results show that compared with the original neural network and other similar algorithms, the CWD-LSTM algorithm has obvious advantages in forecasting the loading time under a variety of detection indexes. More importantly, CWD-LSTM does not require a high update frequency of the neural network, and manufacturers do not need a frequent training model to ensure the reliability of forecast.

Multi-Step-Ahead Electricity Price Forecasting Based on Temporal Graph Convolutional Network

Traditional electricity price forecasting tends to adopt time-domain forecasting methods based on time series, which fail to make full use of the regional information of the electricity market, and ignore the extra-territorial factors affecting electricity price within the region under cross-regional transmission conditions. In order to improve the accuracy of electricity price forecasting, this paper proposes a novel spatio-temporal prediction model, which is combined with the graph convolutional network (GCN) and the temporal convolutional network (TCN). First, the model automatically extracts the relationships between price areas through the graph construction module. Then, the mix-jump GCN is used to capture the spatial dependence, and the dilated splicing TCN is used to capture the temporal dependence and forecast electricity price for all price areas. The results show that the model outperforms other models in both one-step forecasting and multi-step forecasting, indicating that the model has superior performance in electricity price forecasting.

Review of load forecasting based on artificial intelligence methodologies, models, and challenges

Accurate load forecasting can efficiently reduce the day-ahead dispatch stress of power system or microgrid. The overview of load forecasting based on artificial intelligence models are comprehensively summarized in this paper. As the steps of load forecasting based on artificial intelligence model mainly include data processing, setting up forecasting strategy and model forecasting, the paper firstly reviewed the data processing studies. According to the kinds of data obtained, the data can be classified into two categories: multivariate time series and single variate time series. Secondly the forecasting methodologies including one-step forecasting and rolling forecasting are summarized and compared. In addition, according to the form of the prediction results, point prediction, interval prediction and probability prediction are summarized. Thirdly, the paper reviews the artificial intelligence models used in load forecasting. In light of the application scenarios, it can be classified into single model and combination model. Finally, we also discussed the future trend for the research, such as fuzzy reasoning, intelligent optimization in forecasting, novel machine learning and transfer learning technologies, etc.

Transfer Learning with Recurrent Neural Networks for Long-Term Production Forecasting in Unconventional Reservoirs

Summary Robust production forecasting allows for optimal resource recovery through efficient field management strategies. In hydraulically fractured unconventional reservoirs, the physics of fluid flow and transport processes is not well understood and the presence of and transitions between multiple flow regimes further complicate forecasting. An important goal for field operators is to obtain a fast and reliable forecast with minimal historical production data. The abundance of wells drilled in fractured tight formations and continuous data acquisition effort motivate the use of data-driven forecast methods. However, traditional data-driven forecast methods require sufficient training data from an extended period of production for any target well, which may have limited practical use when the effective production life of wells is relatively short. In this paper, a deep recurrent neural network (RNN) model is developed for long-term production forecasting in unconventional reservoirs. As input data, the model takes completion parameters, formation and fluid properties, operating controls, and early (i.e., 3–6 months) production response data. The model is trained on a collection of historical production data across multiple flow regimes, control settings, and the corresponding well properties from multiple shale plays. The proposed RNN model can predict oil, water, and gas production as multivariate time series under varying operating controls. Once the forecast model is trained, it can be used to obtain a one-step forecast by feeding the model with input well properties, operating controls, and a short initial production. The long-term forecast is obtained by either recursively feeding the model with forecast results from the preceding timesteps or by training the model for multistep ahead predictions. Unlike other applications of RNN that require a long history of production data for training, our model employs transfer learning by combining early production data from the target well with the long-term dynamics captured from historical production data in other wells. We illustrate our approach using synthetic data sets and a case study from Bakken Play in North Dakota.

Multistep Forecasting of Soil Moisture Using Spatiotemporal Deep Encoder–Decoder Networks

Abstract Accurate spatiotemporal predictions of surface soil moisture (SM) are important for many critical applications. Machine learning models provide a powerful method for building an accurate and reliable predictive model of SM. However, the models used in recent studies have some limitations, including lack of spatial autocorrelation (SAC), vague representation of important features, and primarily focused on the one-step forecast. Thus, we proposed an attention based convolutional long-short term memory model (AttConvLSTM) for multistep forecasting. The model includes three layers; spatial compression, axial attention, and encoder-decoder prediction, which are used for compressing spatial information, feature extraction, and multistep prediction, respectively. The model was trained using surface SM from Soil Moisture Active Passive L4 product at 18km spatial resolution over the United States. The results show that AttConvLSTM predicts 24 hours ahead SM with mean R2 and RMSE is equal to 0.82 and 0.02, respectively. Compared with LSTM, AttConvLSTM improves the model performance over 73.6% of regions, with an improvement of 8.4% and 17.4% in R2 and RMSE, respectively. The performance of the model is mainly influenced by temporal autocorrelation (TAC). Moreover, we also highlight the importance of SAC on model performance, especially over regions with high SAC and low TAC. Moreover, our model is also competent for SM predictions from several hours to several days, which could be a useful tool for predicting all meteorological variables and forecasting extremes.

A class of max-INAR(1) processes with explanatory variables

The sample paths generated by the integer-valued autoregressive (INAR) processes usually fluctuate up and down in both directions. In some real examples, we will encounter that the path increases in one direction, then decreases, and then goes back and forth. If such count data sets are concerned, the conventional INAR processes cannot be able to capture this up-and-down movement tendency. To solve this problem, the max-AR process was extended to the discrete case, called max-INAR(1) process. To make this process more flexible, we introduce a class of max-INAR processes with explanatory variables, where the autoregressive parameter in the thinning operator depends on explanatory variables. The conditional maximum likelihood method is used to estimate parameters. Further, we consider the one-step forecast for the proposed models. The performance of the estimators in finite sample is numerically assessed. The proposed models are also applied to three real examples.

Improvement of methods of hydrological forecasting using geoinformation technologies

This article discusses the possibility of improving hydrological forecasting methods based on a neural network. The hydrological series, its importance and forecasting features are considered. For hydrological forecasting using the MapInfoProfessional geoinformation system, an electronic map has been developed containing information about the rivers of Russia, as well as gauging stations on the Ob River. The electronic map is the basis for creating a module for short-term hydrological forecasting based on an artificial neural network. The features of a neural network, methods of its training and implementation are considered. The developed artificial neural network is a layer of neurons with a linear activation function and a delay line at the input. To predict the levels of hydrological series, real water levels at gauging stations of the Ob River in the Novosibirsk region will be used. The developed module and its capabilities have been tested. The study was carried out on the basis of models of hydrological series, as well as on the basis of levels of real hydrological series. Based on the study, dependence of the root-mean-square error on the number of previous values of series was revealed. The study also shows that it is possible to use a neural network for the current one-step forecasting of levels of hydrological series in conditions of insufficient information about the runoff region and its characteristics.

AB-Net: A Novel Deep Learning Assisted Framework for Renewable Energy Generation Forecasting

Renewable energy (RE) power plants are deployed globally because the renewable energy sources (RESs) are sustainable, clean, and environmentally friendly. However, the demand for power increases on a daily basis due to population growth, technology, marketing, and the number of installed industries. This challenge has raised a critical issue of how to intelligently match the power generation with the consumption for efficient energy management. To handle this issue, we propose a novel architecture called ‘AB-Net’: a one-step forecast of RE generation for short-term horizons by incorporating an autoencoder (AE) with bidirectional long short-term memory (BiLSTM). Firstly, the data acquisition step is applied, where the data are acquired from various RESs such as wind and solar. The second step performs deep preprocessing of the acquired data via several de-noising and cleansing filters to clean the data and normalize them prior to actual processing. Thirdly, an AE is employed to extract the discriminative features from the cleaned data sequence through its encoder part. BiLSTM is used to learn these features to provide a final forecast of power generation. The proposed AB-Net was evaluated using two publicly available benchmark datasets where the proposed method obtains state-of-the-art results in terms of the error metrics.

Extreme events in globally coupled chaotic maps

Understanding and predicting uncertain things are the central themes of scientific evolution. Human beings revolve around these fears of uncertainties concerning various aspects like a global pandemic, health, finances, to name but a few. Dealing with this unavoidable part of life is far tougher due to the chaotic nature of these unpredictable activities. In the present article, we consider a global network of identical chaotic maps, which splits into two different clusters, despite the interaction between all nodes are uniform. The stability analysis of the spatially homogeneous chaotic solutions provides a critical coupling strength, before which we anticipate such partial synchronization. The distance between these two chaotic synchronized populations often deviates more than eight times of standard deviation from its long-term average. The probability density function of these highly deviated values fits well with the generalized extreme value distribution. Meanwhile, the distribution of recurrence time intervals between extreme events resembles the Weibull distribution. The existing literature helps us to characterize such events as extreme events using the significant height. These extremely high fluctuations are less frequent in terms of their occurrence. We determine numerically a range of coupling strength for these extremely large but recurrent events. On-off intermittency is the responsible mechanism underlying the formation of such extreme events. Besides understanding the generation of such extreme events and their statistical signature, we furnish forecasting these events using the powerful deep learning algorithms of an artificial recurrent neural network. This long short-term memory (LSTM) can offer handy one-step forecasting of these chaotic intermittent bursts. We also ensure the robustness of this forecasting model with two hundred hidden cells in each LSTM layer.

Epidemiological Prognosis of Pertussis Incidence in Bulgaria.

Epidemiological forecasting facilitates scientifically sound solutions to upcoming theoretical and practical issues, in the development of public health management, in particular of infectious diseases. To critically analyze the most recent scientific advances in the biosocial nature and methodology of epidemiological forecasting to present a real-life example of pertussis, a disease with shifting epidemiology. For the prediction of pertussis morbidity the autoregressive integrated moving average (ARIMA) the model was established by utilizing the method of time series analysis to construct a model of overall morbidity using Time series modeller in SPSS v.25. To model pertussis morbidity we obtained official data from the Ministry of Health and the National Center for Infectious and Parasitic Diseases, since the beginning of disease registration from 1903 until 2018. We also analyzed the shifting epidemiology of pertussis. The proper identification procedures we applied indicated ARIMA (3,0,0) model to best fit our original time series of the annual whooping cough morbidity for the 1921-2018 period. The model predicts better morbidity in a one-step forecast. The incidence rate is expected to be stable at about 1.35 per 100,000 in the next three years, which is close to the 2016 level and lower than those in 2017-2018. The ARIMA (3,0,0) model in our study is an adequate tool for presenting the pertussis morbidity trend and is suitable to forecast near-future disease dynamics, with acceptable error tolerance.