An adaptive physics-informed deep learning method for pore pressure prediction using seismic data

Accurate prediction of formation pore pressure is essential to predict fluid flow and manage hydrocarbon production in petroleum engineering.Recent deep learning technique has been receiving more interest due to the great potential to deal with pore pressure prediction.However,most of the traditional deep learning models are less efficient to address generalization problems.To fill this technical gap,in this work,we developed a new adaptive physics-informed deep learning model with high generalization capability to predict pore pressure values directly from seismic data.Specifically,the new model,named CGP-NN,consists of a novel parametric features extraction approach(1DCPP),a stacked multilayer gated recurrent model(multilayer GRU),and an adaptive physics-informed loss function.Through machine training,the developed model can automatically select the optimal physical model to constrain the results for each pore pressure prediction.The CGP-NN model has the best generalization when the physicsrelated metricλ=0.5.A hybrid approach combining Eaton and Bowers methods is also proposed to build machine-learnable labels for solving the problem of few labels.To validate the developed model and methodology,a case study on a complex reservoir in Tarim Basin was further performed to demonstrate the high accuracy on the pore pressure prediction of new wells along with the strong generalization ability.The adaptive physics-informed deep learning approach presented here has potential application in the prediction of pore pressures coupled with multiple genesis mechanisms using seismic data.

Prediction on Failure Pressure of Pipeline Containing Corrosion Defects Based on ISSA-BPNNModel

Oil and gas pipelines are affected by many factors,such as pipe wall thinning and pipeline rupture.Accurate prediction of failure pressure of oil and gas pipelines can provide technical support for pipeline safety management.Aiming at the shortcomings of the BP Neural Network(BPNN)model,such as low learning efficiency,sensitivity to initial weights,and easy falling into a local optimal state,an Improved Sparrow Search Algorithm(ISSA)is adopted to optimize the initial weights and thresholds of BPNN,and an ISSA-BPNN failure pressure prediction model for corroded pipelines is established.Taking 61 sets of pipelines blasting test data as an example,the prediction model was built and predicted by MATLAB software,and compared with the BPNN model,GA-BPNN model,and SSA-BPNN model.The results show that the MAPE of the ISSA-BPNN model is 3.4177%,and the R2 is 0.9880,both of which are superior to its comparison model.Using the ISSA-BPNN model has high prediction accuracy and stability,and can provide support for pipeline inspection and maintenance.

Pressure Prediction for High-Temperature and High-Pressure Formation and Its Application to Drilling in the Northern South China Sea

There are plentiful potential hydrocarbon resources in the Yinggehai and Qiongdongnan basins in the northern South China Sea. However, the special petrol-geological condition with high formation temperature and pressure greatly blocked hydrocarbon exploration. The conventional means of drills, including methods in the prediction and monitoring of underground strata pressure, can no longer meet the requirements in this area. The China National Offshore Oil Corporation has allocated one well with a designed depth of 3200 m and pressure coefficient of 2.3 in the Yinggehai Basin (called test well in the paper) in order to find gas reservoirs in middle-deep section in the Miocene Huangliu and Meishan formations at the depth below 3000 m. Therefore, combined with the '863' national high-tech project, the authors analyzed the distribution of overpressure in the Yinggehai and Qiongdongnan basins, and set up a series of key technologies and methods to predict and monitor formation pressure, and then apply the results to pressure prediction of the test well. Because of the exact pressure prediction before and during drilling, associated procedure design of casing and their allocation in test well has been ensured to be more rational. This well is successfully drilled to the depth of 3485 m (nearly 300 m deeper than the designed depth) under the formation pressure about 2.3 SG (EMW), which indicate that a new step in the technology of drilling in higher temperature and pressure has been reached in the China National Offshore Oil Corporation.

Prediction of Dynamic Wellbore Pressure in Gasified Fluid Drilling

The basis of designing gasified drilling is to understand the behavior of gas/liquid two-phase flow in the wellbore. The equations of mass and momentum conservation and equation of fluid flow in porous media were used to establish a dynamic model to predict wellbore pressure according to the study results of Ansari and Beggs-Brill on gas-liquid two-phase flow. The dynamic model was solved by the finite difference approach combined with the mechanistic steady state model. The mechanistic dynamic model was numerically implemented into a FORTRAN 90 computer program and could simulate the coupled flow of fluid in wellbore and reservoir. The dynamic model revealed the effects of wellhead back pressure and injection rate of gas/liquid on bottomhole pressure. The model was validated against full-scale experimental data, and its 5.0% of average relative error could satisfy the accuracy requirements in engineering design.

Impacts of Initial Perturbations on 24-h Sea-Level Pressure Predictions Near Typhoon Matsa

This study investigates the impact of uncertainty in initial conditions on 24-h sea-level pressure predictions near 0509 Typhoon Matsa by using conditional nonlinear optimal perturbation (CNOP).The CNOP is calculated by using a newly proposed fast algorithm.The model used is the Global/Regional Assimilation and PrEdiction System (GRAPES).The sensitivity of the 24-h predictions is studied in terms of horizontal and vertical ranges and also in terms of different initial state variables.To study the sensitivity of 24-h sea-level pressure predictions to different initial state variables,four functions are given as metrics to find the sensitive initial locations.The results show that the main prediction errors come from initial uncertainty on the levels below 200 hPa and at the region south of about 37.6°N,with more sensitivity to initial winds than to other initial state variables.

Prediction of Shale Gas Pressure based on Multi-channel Seismic Inversion in Fuling

The accurate prediction of formation pressure is important in oil/gas exploration and development.However,the achievement of this goal remains challenging,due to insufficient logging data and the low predictive data accuracy from seismic data.In this work,a case study was carried out in the Baima area of Wulong,in order to develop a workflow for accurately predicting shale gas formation pressure.The multi-channel stack method was first used,as well as the inversion of single-channel seismic data,to construct velocity and density models of the formation.Combined with the existing welllogging data,the velocity and density models of the whole well section were established.The shale gas formation pressure was then estimated using the Eaton method.The results show that the multi-channel seismic stacking method has a higher accuracy than the inversion of the formation velocity obtained by the single-channel seismic method.The discrepancies between our predicted formation pressure and the actual formation pressure measurement are within an acceptable range,indicating that our workflow is effective.

Study on Dynamic Prediction of Two-Phase Pipe Flow in Inclined Wellbore with Middle and High Yield

Gas-liquid two-phase flow is ubiquitous in the process of oil and gas exploitation,gathering and transportation.Flow pattern,liquid holdup and pressure drop are important parameters in the process of gas-liquid two-phase flow,which are closely related to the smooth passage of the two-phase fluid in the pipe section.Although Mukherjee,Barnea and others have studied the conventional viscous gas-liquid two-phase flow for a long time at home and abroad,the overall experimental scope is not comprehensive enough and the early experimental conditions are limited.Therefore,there is still a lack of systematic experimental research and wellbore pressure for gas-liquid two-phase flow under the conditions of middle and high yield and high gas-liquid ratio in conventional viscosity,and the prediction accuracy is low.In view of this,this study carried out targeted systematic research,and from the flow pattern,liquid holdup and pressure drop aspects,established the relevant model,obtained a set of inclined wellbore gas-liquid two-phase pipe flow dynamic prediction method.At the same time,firstly,the model is tested by experimental data,and then the model is compared and verified by a number of field measured wells,which proves that the model is reliable and the prediction accuracy of wellbore pressure is high.

Non-Contact Physiological Measurement System for Wearing Masks During the Epidemic

Physiological signals indicate a person’s physical and mental state at any given time.Accordingly,many studies extract physiological signals from the human body with non-contact methods,and most of them require facial feature points.However,under COVID-19,wearing a mask has become a must in many places,so how non-contact physiological information measurements can still be performed correctly even when a mask covers the facial information has become a focus of research.In this study,RGB and thermal infrared cameras were used to execute non-contact physiological information measurement systems for heart rate,blood pressure,respiratory rate,and forehead temperature for peoplewearing masks due to the pandemic.Using the green(G)minus red(R)signal in the RGB image,the region of interest(ROI)is established in the forehead and nose bridge regions.The photoplethysmography(PPG)waveforms of the two regions are obtained after the acquired PPG signal is subjected to the optical flow method,baseline drift calibration,normalization,and bandpass filtering.The relevant parameters in Deep Neural Networks(DNN)for the regression model can correctly predict the heartbeat and blood pressure.In addition,the temperature change in the ROI of the mask after thermal image processing and filtering can be used to correctly determine the number of breaths.Meanwhile,the thermal image can be used to read the temperature average of the ROI of the forehead,and the forehead temperature can be obtained smoothly.The experimental results show that the above-mentioned physiological signals of a subject can be obtained in 6-s images with the error for both heart rate and blood pressure within 2%∼3%and the error of forehead temperature within±0.5°C.

Predict Blood Pressure by Photoplethysmogram with the Fluid-Structure Interaction Modeling

Blood pressure(BP)has been identified as one of the main factors in cardiovascular disease and other related diseases.Then how to accurately and conveniently measure BP is important to monitor BP and to prevent hypertension.This paper proposes an efficient BP measurement model by integrating a fluid-structure interaction model with the photoplethysmogram(PPG)signal and developing a data-driven computational approach to fit two optimization parameters in the proposedmodel for each individual.The developed BPmodel has been validated on a public BP dataset and has shown that the average prediction errors among the root mean square error(RMSE),the mean absolute error(MAE),the systolic blood pressure(SBP)error,and the diastolic blood pressure(DBP)error are all below 5mmHg for normal BP,stage I,and stage II hypertension groups,and,prediction accuracies of the SBP and the DBP are around 96%among those three groups.

Transparent open-box learning network and artificial neural network predictions of bubble-point pressure compared

The transparent open box(TOB)learning network algorithm offers an alternative approach to the lack of transparency provided by most machine-learning algorithms.It provides the exact calculations and relationships among the underlying input variables of the datasets to which it is applied.It also has the capability to achieve credible and auditable levels of prediction accuracy to complex,non-linear datasets,typical of those encountered in the oil and gas sector,highlighting the potential for underfitting and overfitting.The algorithm is applied here to predict bubble-point pressure from a published PVT dataset of 166 data records involving four easy-tomeasure variables(reservoir temperature,gas-oil ratio,oil gravity,gas density relative to air)with uneven,and in parts,sparse data coverage.The TOB network demonstrates high-prediction accuracy for this complex system,although it predictions applied to the full dataset are outperformed by an artificial neural network(ANN).However,the performance of the TOB algorithm reveals the risk of overfitting in the sparse areas of the dataset and achieves a prediction performance that matches the ANN algorithm where the underlying data population is adequate.The high levels of transparency and its inhibitions to overfitting enable the TOB learning network to provide complementary information about the underlying dataset to that provided by traditional machine learning algorithms.This makes them suitable for application in parallel with neural-network algorithms,to overcome their black-box tendencies,and for benchmarking the prediction performance of other machine learning algorithms.

An exploration in remote blood pressure management:Application of daily routine pattern based on mobile data in health management

Blood pressure(BP)is an important indicator of an individuaPs health status and is closely related to daily behaviors.Thus,a continuous daily measurement of BP is critical for hypertension control.To assist continuous measurement,BP prediction based on non-physiological data(ubiquitous mobile phone data)was studied in the research.An algorithm was proposed that predicts BP based on patients'daily routine,which includes activities such as sleep,work,and commuting.The aim of the research is to provide insight into the application of mobile data in telemonitoring and the continuous unobtrusive daily measurement of BP.A half-year data set from October 2017 of 320 individuals,including telecom data and BP measurement data,was analyzed.Two hierarchical Bayesian topic models were used to extract individuals,location-driven daily routine patterns(topics)and calculate probabilities among these topics from their day-level mobile trajectories.Based on the topic probability distribution and patients'contextual data,their BP were predicted using different models.The prediction model comparison shows that the long short-term memory(LSTM)method exceeds others when the data has a high dependency.Otherwise,the Random Forest regression model outperforms the LSTM method.Also,the experimental results validate the effectiveness of the topics in BP prediction.