Nonlinear Autoregressive Neural Networks to Predict Hydraulic Fracturing Fluid Leakage into Shallow Groundwater

Taherdangkoo, Reza; Tatomir, Alexandru; Taherdangkoo, Mohammad; Qiu, Pengxiang; Sauter, Martin

doi:10.3390/w12030841

Cited by 40 publications

(16 citation statements)

References 57 publications

(69 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ANNs have been used for data-driven modelling in different fields. In hydrology [31], ANNs have been successfully applied, for example to predict rainfall-runoff [32,33], groundwater levels [34] and groundwater contamination [35] . Moreover, ANNs have been used in energy system applications [36], for example to predict the performance of [37], reliability of [38] and design [39] renewable energy systems .…”

Section: Physics-inspired Artificial Neural Networkmentioning

confidence: 99%

Improving Thermochemical Energy Storage Dynamics Forecast with Physics-Inspired Neural Network Architecture

et al. 2020

View full text Add to dashboard Cite

Thermochemical Energy Storage (TCES), specifically the calcium oxide (CaO)/calcium hydroxide (Ca(OH)2) system is a promising energy storage technology with relatively high energy density and low cost. However, the existing models available to predict the system’s internal states are computationally expensive. An accurate and real-time capable model is therefore still required to improve its operational control. In this work, we implement a Physics-Informed Neural Network (PINN) to predict the dynamics of the TCES internal state. Our proposed framework addresses three physical aspects to build the PINN: (1) we choose a Nonlinear Autoregressive Network with Exogeneous Inputs (NARX) with deeper recurrence to address the nonlinear latency; (2) we train the network in closed-loop to capture the long-term dynamics; and (3) we incorporate physical regularisation during its training, calculated based on discretized mole and energy balance equations. To train the network, we perform numerical simulations on an ensemble of system parameters to obtain synthetic data. Even though the suggested approach provides results with the error of 3.96×10−4 which is in the same range as the result without physical regularisation, it is superior compared to conventional Artificial Neural Network (ANN) strategies because it ensures physical plausibility of the predictions, even in a highly dynamic and nonlinear problem. Consequently, the suggested PINN can be further developed for more complicated analysis of the TCES system.

show abstract

Section: Physics-inspired Artificial Neural Networkmentioning

confidence: 99%

Improving Thermochemical Energy Storage Dynamics Forecast with Physics-Inspired Neural Network Architecture

et al. 2020

View full text Add to dashboard Cite

show abstract

“…e effect of saturation on the coupled thermohydromechanical process in porous media is of great significance to understand the properties of the multifield coupling response, even including the influence of chemical factors [19][20][21]. However, all the constitutive models described above are derived based on the general elastic-plastic theory.…”

Section: Introductionmentioning

confidence: 99%

Thermo-Hydromechanical Coupling Responses Driven by a Central Heat Source in Unsaturated Soils

Wang¹,

Liu²,

Chen

2022

Advances in Materials Science and Engineering

View full text Add to dashboard Cite

Based on the thermodynamic and thermoelastic theory, the coupled governing equations of deformation, heat transfer, and moisture migration in unsaturated soils were given. The coupled calculation process was realized by the finite element method. The hydrothermal coupling characteristics of two cases of unsaturated kaolin clay were studied by using a self-developed test device, and the test results were compared with the numerical results. The results showed that when the initial saturation of the soil is high, the volumetric water content of the measured point increases, and then it is in a stable state during the heating process. When the initial saturation is low, the volumetric water content near the heat source increases firstly and then decreases during the heating process. The rise and fall of temperature make the volumetric water content of the soil irreversible. The volumetric water content of each measured point is lower than the initial state. The closer it is to the internal measured heat source, the more obvious this phenomenon is; the lower the initial saturation is, the more obvious it is.

show abstract

“…It mends AR with nonlinear trait and black box solving of parameters and confines the input features of ANN to reduce overfit and difficulties on training. Because of the nonlinear trait of ARNN, it is also called nonlinear autoregressive neural network (NARNN) [28][29][30][31].…”

Section: Introductionmentioning

confidence: 99%

An Improved Recursive ARIMA Method with Recurrent Process for Remaining Useful Life Estimation of Bearings

Luo

Wang

Yang

2022

Shock and Vibration

View full text Add to dashboard Cite

A typical way to predict the remaining useful life (RUL) of bearings is to predict certain health indicators (HIs) according to the historical HI series and forecast the end of life (EOL). The autoregressive neural network (ARNN) is an early idea to combine the artificial neural network (ANN) and the autoregressive (AR) model for forecasting, but the model is limited to linear terms. To overcome the limitation, this paper proposes an improved autoregressive integrated moving average with the recurrent process (ARIMA-R) method. The proposed method adds moving average (MA) components to the framework of ARNN, adding the long-range dependence and nonlinear factors. To deal with the recursive characteristics of the MA term, a process of MA component estimating is constructed based on the expectation-maximum method. In the concrete realization of the method, the rotation tree (RTF) is introduced in place of ANN to improve the prediction performance. The experiment on FEMTO datasets reveals that the proposed ARIMA-R method outperforms the ARNN method in terms of predictive performance evaluation indicators.

show abstract

Nonlinear Autoregressive Neural Networks to Predict Hydraulic Fracturing Fluid Leakage into Shallow Groundwater

Cited by 40 publications

References 57 publications

Improving Thermochemical Energy Storage Dynamics Forecast with Physics-Inspired Neural Network Architecture

Improving Thermochemical Energy Storage Dynamics Forecast with Physics-Inspired Neural Network Architecture

Thermo-Hydromechanical Coupling Responses Driven by a Central Heat Source in Unsaturated Soils

An Improved Recursive ARIMA Method with Recurrent Process for Remaining Useful Life Estimation of Bearings

Contact Info

Product

Resources

About