Gated Recurrent Unit Neural Network Research Articles

Short-term wind power prediction model based on improved ACA-GRU neural network

To improve the accuracy of wind power forecasting, the improved ACA (Ant Colony Algorithm) is used to optimize the GRU (Gated Recurrent Unit) model. First, the original power generation data is normalized; Second, the GRU neural network model is established, and the ant colony algorithm is used to optimize it; Finally, the optimized GRU model and the non-optimized model are used to predict the short-term wind power output, and the prediction results are compared to verify that the improved ACA-GRU prediction model has higher prediction accuracy for short-term wind power output.

A hybrid CNN-GRU based probabilistic model for load forecasting from individual household to commercial building

In load forecasting, build load is relatively difficult to predict due to its variability and uncertainty. This study intends to develop a hybrid deep learning models and quantile regression loss function for building load prediction. Then, through the numerical study for the individual and commercial building, it is provided to demonstrate the practicality of proposed method which showed the accuracy with lower loss than existing models. In academic contribution, this study proves that Convolutional Neural Network (CNN) can extract useful information from high uncertainty power load, and Gated Recurrent Unit (GRU) has the benefit in the times-series forecasting. Furthermore, the proposed hybrid framework outperforms the conventional LSTM and GRU neural network. In practical contribution, the issues of reducing energy waste and demand side management can be improved through more accurate load forecasting. Also, RE100 company can realize the risk of the power purchasing planning and customized decision-making through prediction intervals.

Short-Term Electricity Price Forecasting Using EEMD and GRU-NN

A vital necessity for the reliable operation of the power system as well as for the financial well-being of the consumers is an accurate forecast of the price of electricity. The price of electricity is highly volatile, nonlinear, and subject to seasonal fluctuations. The price series is also affected by electrical market uncertainty and demand, which is strongly reliant on weather and electricity usage time. This study presents a short-term electricity price forecasting method that follows the similar days approach comprising Ensemble Empirical Mode Decomposition technique. The suggested method utilizes a feature selection methodology that combines the Random Forest Regressor and the Gradient Boosting Regressor to determine which features are most important for the machine learning model. To forecast the electricity price, a Gated Recurrent Unit Neural Network (GRU-NN) is employed as the machine learning model. The GRU-NN has the capability of accurately capturing complicated temporal relationships in electricity price time series, which enables it to make correct predictions. To evaluate the validity of the proposed method, data from the PJM electricity market have been used. The simulation results demonstrate that the suggested method is superior to the existing technique, with significantly improved values achieved for both the mean absolute percentage error (MAPE) and the root mean square error (RMSE).

Simulated annealing algorithm optimized GRU neural network for urban rainfall-inundation prediction

Abstract In the context of global climate change and the continuous development of urban areas, rainfall-inundation modeling is a common approach that provides critical support for the protection and early warning of urban waterlogging protection. The present study conducts a data-driven model for hourly urban rainfall-inundation depth prediction, which is based on a gated recurrent unit (GRU) neural network and uses the simulated annealing (SA) algorithm for the hyperparameter optimization of GRU, namely the SA-GRU model. To verify the performance of the proposed model, backpropagation, long short-term memory (LSTM), and bidirectional LSTM (BiLSTM) neural networks are set as benchmarks. Results show that the SA-GRU has high accuracy in the case of short-term inundation prediction, with the Nash–Sutcliffe efficiency from 0.999 to 0.596 for the 1-h-ahead to 8-h-ahead predictions. And further research reveals that the SA-GRU integrates the significant optimization of SA, with an average 20% reduction of the root mean square error within the first eight prediction periods, and the efficient training speed of GRU, with 23.7% faster than LSTM and 44.2% faster than BiLSTM. In conclusion, the SA-GRU excels in urban inundation prediction, demonstrating its value in flood management and decision-making.

Using Time-Series Generative Adversarial Networks to Synthesize Sensing Data for Pest Incidence Forecasting on Sustainable Agriculture

A sufficient amount of data is crucial for high-performance and accurate trend prediction. However, it is difficult and time-consuming to collect agricultural data over long periods of time; the consequence of such difficulty is datasets that are characterized by missing data. In this study we use a time-series generative adversarial network (TimeGAN) to synthesize multivariate agricultural sensing data and train RNN (Recurrent Neural Network), LSTM (Long Short-Term Memory), and GRU (Gated Recurrent Unit) neural network prediction models on the original and generated data to predict future pest populations. After our experiment, the data generated using TimeGAN and the original data have the smallest EC value in the GRU model, which is 9.86. The results show that the generative model effectively synthesizes multivariate agricultural sensing data and can be used to make up for the lack of actual data. The pest prediction model trained on synthetic data using time-series data generation yields results that are similar to that of the model trained on actual data. Accurate prediction of pest populations would represent a breakthrough in allowing for accurate and timely pest control.

Differential Settlement of Track Foundations Identification Based on GRU Neural Network

The timely identification of differential settlement of track foundations is of great significance for the safety of train operation and the maintenance of track structures. However, traditional monitoring techniques cannot meet the requirements of efficient, real-time, and automatic monitoring of track foundation settlement. In order to solve these problems, a real-time identification method based on a gated recurrent unit (GRU) neural network is proposed for the differential settlement of track foundations monitoring. According to parameter sensitivity analysis, the vertical acceleration of the vehicle is selected as the known data fed into the GRU network for differential settlement identification. Then the GRU network is employed to establish the nonlinear relationship between the vertical acceleration of the vehicle and the differential settlement of the track foundation. The results indicate that the longitudinal continuous differential settlement distribution curve of track foundations could be accurately identified with GRU neural network through the real-time vibration response of the vehicle–track. The current method may provide a new means for the real-time and efficient identification of the differential settlement of track foundations.

Forecasting atmospheric turbulence conditions from prior environmental parameters using artificial neural networks.

Atmospheric turbulence (C n2) modeling has been proposed by physics-based models, but they are unable to capture the many cases. Recently, machine learning surrogate models have been used to learn the relationship between local meteorological conditions and turbulence strength. These models predict C n2 at time t from weather at time t. This work expands modeling capabilities by proposing a technique to forecast 3h of future turbulence conditions at 30min intervals from prior environmental parameters using artificial neural networks. First, local weather and turbulence measurements are formatted to pairs of the input sequence and output forecast. Next, a grid search is used to find the best combination of model architecture, input variables, and training parameters. The architectures investigated are the multilayer perceptron and three variants of the recurrent neural network (RNN): the simple RNN, the long short-term memory RNN (LSTM-RNN), and the gated recurrent unit RNN (GRU-RNN). A GRU-RNN architecture that uses 12h of prior inputs is found to have the best performance. Finally, this model is applied to the test dataset and analyzed. It is shown that the model has generally learned the relationship between prior environmental and future turbulence conditions.

A Study of Forest Phenology Prediction Based on GRU Models

Investigating forest phenology prediction is a key parameter for assessing the relationship between climate and environmental changes. Traditional machine learning models are not good at capturing long-term dependencies due to the problem of vanishing gradients. In contrast, the Gated Recurrent Unit (GRU) can effectively address the problem of vanishing gradients and allow the neural network to capture longer-range dependencies. In this study, an optical camera was used as experimental equipment to obtain forest images. The absolute greenness index (GEI) data of the region of interest (ROI) in the images were calculated to fit the seasonal variation curve of forest phenology. The GRU neural network model was introduced to train and analyze the GEI data, and the performance of the GRU neural network was evaluated using the Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE) methods. Finally, the model was used to predict the trend of GEI data for the next 60 days. The results showed that: (1) In terms of training and predicting forest phenology, the GRU model was validated using histograms and autocorrelation graphs, which indicated that the distribution of predicted data was consistent with the trend of actual data, the GRU model data was feasible, and the model was stable. (2) The MSE values of the GRU model at rain-fed-CK (preset point 1), sufficient drip irrigation-DIFI (preset point 3), and sufficient furrow irrigation-BIFI (preset point 5) were 9.055 × 10−5, 12.91 × 10−5, and 8.241 × 10−5, respectively. The RMSE values were 9.516 × 10−3, 11.36 × 10−3, and 7.313 × 10−3, respectively. The MAE values were 7.174 × 10−3, 8.241 × 10−3, and 5.351 × 10−3, respectively. These results indicate that the overall performance of the GRU model was good. (3) The predicted GEI data for the next 60 days showed a trend consistent with actual changes in GEI data, as demonstrated by the GRU model. The GRU model has become the preferred method for phenological prediction due to its simple internal structure and relatively short training time. Results show that the GRU model can achieve forest phenological change prediction and can reveal in-depth insights into future forest growth and climate change, providing a theoretical basis for the application of forest phenological prediction.

From Stochastic to Rough Volatility: A New Deep Learning Perspective on Hedging

The Black–Scholes model assumes that volatility is constant, and the Heston model assumes that volatility is stochastic, while the rough Bergomi (rBergomi) model, which allows rough volatility, can perform better with high-frequency data. However, classical calibration and hedging techniques are difficult to apply under the rBergomi model due to the high cost caused by its non-Markovianity. This paper proposes a gated recurrent unit neural network (GRU-NN) architecture for hedging with different-regularity volatility. One advantage is that the gating network signals embedded in our architecture can control how the present input and previous memory update the current activation. These gates are updated adaptively in the learning process and thus outperform conventional deep learning techniques in a non-Markovian environment. Our numerical results also prove that the rBergomi model outperforms the other two models in hedging.

The optimized gate recurrent unit based on improved evolutionary algorithm to predict stock market returns

In order to accelerate the learning ability of neural network structure parameters and improve the prediction accuracy of deep learning algorithms, an evolutionary algorithm, based on a prior Gaussian mutation (PGM) operator, is proposed to optimize the structure parameters of a gated recurrent unit (GRU) neural network. In this algorithm, the sensitivity learning process of GRU model parameters into the Gaussian mutation operator, used the variance of the GRU model parameter training results as the Gaussian mutation variance to generate the optimal individual candidate set. Then, the optimal GRU neural network structure is constructed using the evolutionary algorithm of the prior Gaussian mutation operator. Moreover, the PGM-EA-GRU algorithm is applied to the prediction of stock market returns. Experiments show that the prediction model effectively overcomes the GRU neural network, quickly falling into a local optimum and slowly converging. Compared to the RF, SVR, RNN, LSTM, GRU, and EA-GRU benchmark models, the model significantly improves the searchability and prediction accuracy of the optimal network structure parameters. It also validates the effectiveness and the progressive nature of the PGM-EA-GRU model proposed in this paper with stock market return prediction.

Improved prediction of chlorophyll-a concentrations in reservoirs by GRU neural network based on particle swarm algorithm optimized variational modal decomposition

The accurate and reliable prediction of chlorophyll-a (Chl-a) concentration is of great significance in reservoir environment management and pollution control. To improve the accuracy of Chl-a index prediction, a novel hybrid water quality prediction method was proposed for gated recurrent unit (GRU) neural network based on particle swarm algorithm optimized variational modal decomposition (PV-GRU). The results showed that the variational mode decomposition (VMD) optimized by particle swarm optimization (PSO) in this study effectively reduced the non-smooth of water quality data. In addition, the GRU neural network reduced the risk of overfitting the deep-learning model with small sample data. Overall, the PV-GRU prediction model exhibited significant superiority in predicting non-smooth and non-linear Chl-a sequences with a relatively small sample size. The prediction errors of PV-GRU model were all less than those of other comparative models, and the fitting determination coefficient R2 was 94.21%. These results indicated that the proposed PV-GRU model can effectively predict the content of Chl-a in reservoirs, which provides an alternative new method for water quality prediction to prevent and control eutrophication in reservoirs.

Learning Car-Following Behaviors for a Connected Automated Vehicle System: An Improved Sequence-to-Sequence Deep Learning Model

Data-driven car-following modeling is of great significance to traffic behavior analysis and the development of connected automated vehicle (CAV) technology. The existing researches focus on reproducing the car-following process by capturing the behavior of the host vehicle using the information of its nearest preceding vehicle. While the other preceding vehicles may affect the host vehicle as well. To fill the gap above, this paper presents an improved sequence-to-sequence deep learning-based (ISDL) car-following model for a CAV system. Firstly, the kinematics information considering the multiple preceding vehicles are organized as the input characteristics. Secondly, an improved sequence-to-sequence deep learning framework is proposed by integrating an encoder with the bidirectional gated recurrent unit (GRU) neural network and a decoder using an attention-based GRU neural network in an end-to-end fashion. Finally, the car-following data with multiple preceding vehicles captured from the NGSIM dataset are employed to train and calibrate the proposed model. Experimental results indicate that the deep learning-based models’ performance in learning heterogeneous driving behavior can be enhanced by adding information about multiple preceding vehicles. In addition, the proposed ISDL model outperforms the benchmark car-following models in terms of the accuracy of the simulated speeds and simulated positions. Through tests on platoon simulation, the ISDL model is capable of reshaping the traffic oscillation phenomenon as well.

Personalized ontology and deep training tree‐based optimal gated recurrent unit‐recurrent neural network for prediction of students' behavior

SummaryDigital libraries are now widely used by students all around the world due to the hassle‐free services it offers in the form of digitized study materials, ebooks, multimedia content and so forth. However, to increase user satisfaction and allegiance, a personalized digital library needs to be deployed to offer customized services to the users. To accomplish this objective, this article presents a novel scheme to design personalized digital libraries via the usage of the Protégé editor and machine learning techniques. The Protégé 4.3 tool is used to generate the digital library ontologies and identify the interrelated concepts. Domain knowledge acquisition, ontology organization, ontology elaboration, ensuring the consistency of the information, and ontology validation. The gated recurrent unit‐recurrent neural network (GRU‐RNN) with a deep training tree (DTT) is used to predict the different user behavior styles such as cognitive behavior, learning speed behavior, sedentary behavior, and aggressive behavior. The DTT is mainly incorporated to overcome the vanishing gradient issue associated with the GRU‐RNN and used in the place of gradient‐based optimization. A black widow optimization approach is employed to update the weights of the GRU‐RNN network in order to increase its accuracy. The efficiency of the proposed methodology is analyzed by using different performance metrics such as F‐score, accuracy, loss, precision, and recall scores. The results show that the personalized digital library created can be efficient in terms of usability.

Research on soft compensation of the potential drift signal of a pH electrode based on a gated recurrent neural network

As a convenient chemical sensor, the pH electrode is widely used in the measurement of the pH value of water bodies. However, due to structural aging and environmental influences, the pH electrode is prone to drift, which directly results in the inability to obtain accurate measurement results. Based on the above problems, this paper proposes a cascade structure soft compensation model with the gated recurrent unit (GRU) as the main body. The model uses the complete ensemble empirical mode decomposition with adaptive noise with permutation entropy (CEEMDAN-PE) method to obtain the main characteristics of the pH electrode potential drift signal to reduce the interference of noise in the actual measurement environment, and uses its output as the input of the GRU neural network to obtain the prediction results and compensate for the drift signal. This model is called the CEEMDA-PE & GRU (CPG) model. In this paper, the CPG model is compared with the commonly used time series prediction model, and the results show that the prediction effect of this model is better than other models. The root mean squared error, mean absolute error, and mean absolute percentage error of the prediction model are reduced by 60.97%, 65.53%, and 66.55%, respectively. Finally, this paper proposes the concept of the degree of compensation to evaluate the compensation effect. The average degree of compensation of the soft compensation method is above 83%. The results show that the soft compensation method can improve the measurement accuracy of the pH electrode and has good robustness.

Prediction of dissolved gas concentration in transformer oil considering data loss scenarios in power system

The trend prediction of dissolved gas concentration in transformer oil can provide basis for transformer fault diagnosis, which is of vital significance to the safe operation of power system. However, due to the inevitable malfunction of monitoring equipment, it is difficult to collect all needed data in actual operation scenarios. Therefore, a method for predicting dissolved gas concentration in transformer oil for data loss scenarios is proposed based on Bayesian probabilistic matrix factorization (BPMF) and gated recurrent unit (GRU) neural network. Firstly, aiming at the problem of data loss in actual monitoring of dissolved gas in oil, BPMF is used to fill in the missing data. Then, a GRU neural network model is established to predict the trend of dissolved gas concentration in oil. Finally, the hyperparameters of the prediction model are selected and optimized by Bayesian theory. The examples show that this method can effectively fill in the missing part of the measured data. Compared with traditional prediction methods, the proposed method has higher prediction accuracy.

Land Subsidence Evolution and Simulation in the Western Coastal Area of Bohai Bay, China

Groundwater overexploitation and loading of buildings have been the main factors triggering land subsidence along the west coast of Bohai Bay, China, since the 2000s. Uneven subsidence has been causing damage to buildings and civil facilities, loss of elevation, increasing the risk of flood and seawater intrusion, and threatening the safety of people’s lives and property. This paper analyzed the spatial and temporal features of land subsidence along the coastal area from 2003 to 2010 and from 2015 to 2020, respectively. The relations between the initiating factors and land subsidence were explored. Then, the simulation model of land subsidence was constructed through a deep learning method. During the process, multiple data were collected, including land satellite (Landsat), environmental satellite advanced synthetic aperture radar (ENVISAT ASAR) and Sentinel-1 images, leveling data, lithological data, and groundwater level data. The area occupied by buildings and vertical displacement were extracted by using supervised classification, small baseline subset (SBAS), and persistent scatterer interferometry (PSI) technologies. The gated recurrent unit (GRU) neural network was adopted to simulate the evolution of land subsidence. Results showed that the maximum annual vertical displacement rate decreased from −94 mm/yr during 2003–2010 to −87 mm/yr during 2015–2020. The correlation efficiency between the groundwater level of the third confined aquifer group and land subsidence was larger than the area occupied by buildings and the compressible layer thickness with subsidence. The constructed GRU neural network model can simulate subsidence from September 2019 to December 2019, with the overall RMSE and MAE being 3.16 mm and 2.19 mm, respectively. This work can facilitate an understanding of the evolution and prevention of land subsidence along the west coast of Bohai Bay, which will provide information for policy decisions and flood-fighting plans of the worldwide coastal cities.

Performance enhancement of CAP-VLC system utilizing GRU neural network based equalizer

In this paper, a gated recurrent unit (GRU) neural network based equalizer is proposed for the first time to compensate for linear and nonlinear distortions in carrier-less amplitude phase (CAP) band-limited visible light communication (VLC) systems. The equalization scheme is mainly based on the GRU algorithm, which captures the dependencies between sequences within the memory size through a gating mechanism. Moreover, we experimentally compare the performance of traditional finite impulse response (FIR) based equalizer using real-valued GRU neural network (RV-GRUNN) with and without memory as well as separate complex-valued GRU neural network (SCV-GRUNN) with and without memory. Compared with the memoryless networks, the networks with memory improve the convergence speed and fitting accuracy at the expense of complexity. Compared with real-valued networks, the separate complex-valued networks have lower complexity and better generalization ability. By deploying SCV-GRUNN (M=3), a 1.8-m 560 Mb/s CAP128 VLC system is successfully demonstrated with Q factor improvement of 3.81 dB, 2.92 dB, 1.42 dB and 1.05 dB over the second-order Volterra Series, RV-GRUNN(M=1), SCV-GRUNN (M=1) and RV-GRUNN (M=3), respectively.

GRU Neural Network Improved Bioimpedance Based Stroke Volume Estimation during Ergometry Stress Test.

Cardiovascular diseases (CVDs) are one of the leading members of non-communicable diseases. An early diagnosis is essential for effective treatment, to reduce hospitalization time and health care costs. Nowadays, an exercise stress test on an ergometer is used to identify CVDs. To improve the accuracy of diagnostics, the hemodynamic status and parameters of a person can be investigated. For hemodynamic management, thoracic electrical bioimpedance has recently been used. This technique offers beat-to-beat stroke volume calculation but suffers from an artifact-sensitive signal that makes such measurements difficult during movement. We propose a new method based on a gated recurrent unit (GRU) neural network and the ECG signal to improve the measurement of bioimpedance signals, reduce artifacts and calculate hemodynamic parameters. We conducted a study with 23 subjects. The new approach is compared to ensemble averaging, scaled Fourier linear combiner, adaptive filter, and simple neural networks. The GRU neural network performs better with single artifact events than shallow neural networks (mean error −0.0244, mean square error 0.0181 for normalized stroke volume). The GRU network is superior to other algorithms using time-correlated data for the exercise stress test.

State of charge estimation for lithium-ion batteries using gated recurrent unit recurrent neural network and adaptive Kalman filter

The state of charge (SOC) is one of the most important monitoring states for the battery management system. It is still a challenge to estimate the battery SOC accurately and stably. The conventional model-based filtering methods may cause inaccurate SOC estimation in application due to the high dependence on accurate battery model. And the emerging methods based on machine learning often have the problem of estimated SOC fluctuation when the current fluctuates greatly. To solve these problems, this paper proposes a robust and efficient combined SOC estimation method, GRU-AKF, which combines the gated recurrent unit recurrent neural network (GRU-RNN) and the adaptive Kalman filter (AKF). Firstly, the GRU-RNN is used to establish a mapping model between the battery measured variables and SOC in the full temperature range, and to achieve the SOC pre-estimation. Then, an AKF is employed to filter the output SOC of the GRU-RNN for reducing the fluctuation in pre-estimated SOC. Finally, the accurate and stable estimated SOC is obtained. In the experiments, the LiFePO4 battery datasets at various temperatures are used to validate the SOC estimation performance and generalization ability. Specifically, the root mean square error is less than 1.3% and 5.8%, and the maximum error is less than 2.2% and 7.7% for the unknown data at positive and negative temperatures, respectively. By comparing with other methods of the same type, the proposed method is demonstrated to be superior in SOC estimation performance and computation efficiency, especially it has excellent performance in initial SOC convergence ability.

Real-time analysis and prediction of shield cutterhead torque using optimized gated recurrent unit neural network

An accurate prediction of earth pressure balance (EPB) shield moving performance is important to ensure the safety tunnel excavation. A hybrid model is developed based on the particle swarm optimization (PSO) and gated recurrent unit (GRU) neural network. PSO is utilized to assign the optimal hyperparameters of GRU neural network. There are mainly four steps: data collection and processing, hybrid model establishment, model performance evaluation and correlation analysis. The developed model provides an alternative to tackle with time-series data of tunnel project. Apart from that, a novel framework about model application is performed to provide guidelines in practice. A tunnel project is utilized to evaluate the performance of proposed hybrid model. Results indicate that geological and construction variables are significant to the model performance. Correlation analysis shows that construction variables (main thrust and foam liquid volume) display the highest correlation with the cutterhead torque (CHT). This work provides a feasible and applicable alternative way to estimate the performance of shield tunneling.