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.

Data-driven modeling on anisotropic mechanical behavior of brain tissue with internal pressure

Brain tissue is one of the softest parts of the human body,composed of white matter and grey matter.The mechanical behavior of the brain tissue plays an essential role in regulating brain morphology and brain function.Besides,traumatic brain injury(TBI)and various brain diseases are also greatly influenced by the brain's mechanical properties.Whether white matter or grey matter,brain tissue contains multiscale structures composed of neurons,glial cells,fibers,blood vessels,etc.,each with different mechanical properties.As such,brain tissue exhibits complex mechanical behavior,usually with strong nonlinearity,heterogeneity,and directional dependence.Building a constitutive law for multiscale brain tissue using traditional function-based approaches can be very challenging.Instead,this paper proposes a data-driven approach to establish the desired mechanical model of brain tissue.We focus on blood vessels with internal pressure embedded in a white or grey matter matrix material to demonstrate our approach.The matrix is described by an isotropic or anisotropic nonlinear elastic model.A representative unit cell(RUC)with blood vessels is built,which is used to generate the stress-strain data under different internal blood pressure and various proportional displacement loading paths.The generated stress-strain data is then used to train a mechanical law using artificial neural networks to predict the macroscopic mechanical response of brain tissue under different internal pressures.Finally,the trained material model is implemented into finite element software to predict the mechanical behavior of a whole brain under intracranial pressure and distributed body forces.Compared with a direct numerical simulation that employs a reference material model,our proposed approach greatly reduces the computational cost and improves modeling efficiency.The predictions made by our trained model demonstrate sufficient accuracy.Specifically,we find that the level of internal blood pressure can greatly influence stress distribution and determine the possible related damage behaviors.

Excess pore pressure behavior and evolution in deep coalbed methane reservoirs

Deep coalbed methane(DCBM),an unconventional gas reservoir,has undergone significant advancements in recent years,sparking a growing interest in assessing pore pressure dynamics within these reservoirs.While some production data analysis techniques have been adapted from conventional oil and gas wells,there remains a gap in the understanding of pore pressure generation and evolution,particularly in wells subjected to large-scale hydraulic fracturing.To address this gap,a novel technique called excess pore pressure analysis(EPPA)has been introduced to the coal seam gas industry for the first time to our knowledge,which employs dual-phase flow principles based on consolidation theory.This technique focuses on the generation and dissipation for excess pore-water pressure(EPWP)and excess pore-gas pressure(EPGP)in stimulated deep coal reservoirs.Equations have been developed respectively and numerical solutions have been provided using the finite element method(FEM).Application of this model to a representative field example reveals that excess pore pressure arises from rapid loading,with overburden weight transferred under undrained condition due to intense hydraulic fracturing,which significantly redistributes the weight-bearing role from the solid coal structure to the injected fluid and liberated gas within artificial pores over a brief timespan.Furthermore,field application indicates that the dissipation of EPWP and EPGP can be actually considered as the process of well production,where methane and water are extracted from deep coalbed methane wells,leading to consolidation for the artificial reservoirs.Moreover,history matching results demonstrate that the excess-pressure model established in this study provides a better explanation for the declining trends observed in both gas and water production curves,compared to conventional practices in coalbed methane reservoir engineering and petroleum engineering.This research not only enhances the understanding of DCBM reservoir behavior but also offers insights applicable to production analysis in other unconventional resources reliant on hydraulic fracturing.

A data and physical model dual-driven based trajectory estimator for long-term navigation

Long-term navigation ability based on consumer-level wearable inertial sensors plays an essential role towards various emerging fields, for instance, smart healthcare, emergency rescue, soldier positioning et al. The performance of existing long-term navigation algorithm is limited by the cumulative error of inertial sensors, disturbed local magnetic field, and complex motion modes of the pedestrian. This paper develops a robust data and physical model dual-driven based trajectory estimation(DPDD-TE) framework, which can be applied for long-term navigation tasks. A Bi-directional Long Short-Term Memory(Bi-LSTM) based quasi-static magnetic field(QSMF) detection algorithm is developed for extracting useful magnetic observation for heading calibration, and another Bi-LSTM is adopted for walking speed estimation by considering hybrid human motion information under a specific time period. In addition, a data and physical model dual-driven based multi-source fusion model is proposed to integrate basic INS mechanization and multi-level constraint and observations for maintaining accuracy under long-term navigation tasks, and enhanced by the magnetic and trajectory features assisted loop detection algorithm. Real-world experiments indicate that the proposed DPDD-TE outperforms than existing algorithms, and final estimated heading and positioning accuracy indexes reaches 5° and less than 2 m under the time period of 30 min, respectively.

Long-term decay of the water pressure in the osmotic tensiometer

Matric suction is an important state variable required for the assessment of unsaturated soil properties.Tensiometers are commonly used for direct matric suction measurement but have a limited measuring range up to 90 kPa due to the cavitation problem.Osmotic tensiometer(OT)can improve the measuring range of tensiometers by increasing the osmotic pressure of water to avoid the cavitation.However,the long-term water pressure decay that appeared in OTs caused a gradual decrease in their measuring range.In this study,crosslinked poly(acrylamide-co-acrylic acid)potassium salt(PAM-co-PAAK)was used for the preparation of OTs(five in total)to explore the mechanism of water pressure decay of OTs.The maximum water pressure in the OT versus the volume fraction of polymer filled in the OT was described based on the Flory-Huggins polymer theories and validated using WP4C dewpoint hygrometer.The long-term pressure decay of OT-1,OT-2,and OT-3 was observed for 130 d and constant pressures were found for OT-1 and OT-2,indicating that the pressure decay of OT was mainly caused by the stress relaxation of the polymer hydrogels,and standard linear solid(SLS)rheological model was appropriate to fit the decay data.For OT-1,OT-2 and OT-3,the theoretical osmotic pressure that was calculated based on the mass of retrieved polymer from OTs after 130-d pressure observation was higher than the actual osmotic pressure as observed,indicating that polymer leakage cannot explain the pressure decay of the OT.The ultravioletevisible(UVevisible)spectrophotometry examined the change in polymer concentrations in the water containers of OT-4 and OT-5 and demonstrated that there was no increase in polymer leakage during the period of pressure decay of OT-4 and OT-5.As a result,the pressure decay of OT was not caused by polymer leakage.The results of this research suggested that the viscoelastic properties of polymers should be taken into consideration in further OT development.

Validation of the Relationship between Precipitable Water and Surface Vapor Pressure by Means of Reanalysis Data

By means of ERA-40, JRA-25, NCEP/NCAR and NCEP/DOE reanalysis data, empirical relations between precipitable water and surface vapor pressure in spatial and temporal scale were calculated. The reliabilities of precipitable water from reanalysis data were validated based on comparing different W-e empirical relations of various reanalysis data, in order to provide basis and reference for reasonable application. The results showed that W-e empirical relation of ERA-40 was closest to that of sounding data in China, and precipitable water from ERA-40 was the most credible. The worldwide comparison among W-e empirical relations of four reanalysis data showed that there was little difference in annual mean W-e empirical relations in the middle latitudes and great differences in low and high latitudes. Seasonal mean W-e empirical relations in the middle latitudes of the northern Hemisphere had little difference in spring, autumn and winter, but great difference in summer. Therefore, the reliabilities of precipitable water from reanalysis data in spring, autumn and winter in the middle latitudes of the northern hemisphere were higher than other areas and seasons. W-e empirical relations of NCEP/NCAR and NCEP/DOE had good stability in different years, while there was poor stability in ERA-40 and JRA-25.

Evaluation and prediction of earth pressure balance shield performance in complex rock strata:A case study in Dalian,China

This research explores the potential for the evaluation and prediction of earth pressure balance shield performance based on a gray system model.The research focuses on a shield tunnel excavated for Metro Line 2 in Dalian,China.Due to the large error between the initial geological exploration data and real strata,the project construction is extremely difficult.In view of the current situation regarding the project,a quantitative method for evaluating the tunneling efficiency was proposed using cutterhead rotation(R),advance speed(S),total thrust(F)and torque(T).A total of 80 datasets with three input parameters and one output variable(F or T)were collected from this project,and a prediction framework based gray system model was established.Based on the prediction model,five prediction schemes were set up.Through error analysis,the optimal prediction scheme was obtained from the five schemes.The parametric investigation performed indicates that the relationships between F and the three input variables in the gray system model harmonize with the theoretical explanation.The case shows that the shield tunneling performance and efficiency are improved by the tunneling parameter prediction model based on the gray system model.

Modeling viscosity of methane,nitrogen,and hydrocarbon gas mixtures at ultra-high pressures and temperatures using group method of data handling and gene expression programming techniques

Accurate gas viscosity determination is an important issue in the oil and gas industries.Experimental approaches for gas viscosity measurement are timeconsuming,expensive and hardly possible at high pressures and high temperatures(HPHT).In this study,a number of correlations were developed to estimate gas viscosity by the use of group method of data handling(GMDH)type neural network and gene expression programming(GEP)techniques using a large data set containing more than 3000 experimental data points for methane,nitrogen,and hydrocarbon gas mixtures.It is worth mentioning that unlike many of viscosity correlations,the proposed ones in this study could compute gas viscosity at pressures ranging between 34 and 172 MPa and temperatures between 310 and 1300 K.Also,a comparison was performed between the results of these established models and the results of ten wellknown models reported in the literature.Average absolute relative errors of GMDH models were obtained 4.23%,0.64%,and 0.61%for hydrocarbon gas mixtures,methane,and nitrogen,respectively.In addition,graphical analyses indicate that the GMDH can predict gas viscosity with higher accuracy than GEP at HPHT conditions.Also,using leverage technique,valid,suspected and outlier data points were determined.Finally,trends of gas viscosity models at different conditions were evaluated.

Psychological stress and long-term blood pressure variability of military young males:The cardiorespiratory fitness and hospitalization events in armed forces study

BACKGROUND Acute stress might increase short-term heart rate variability and blood pressure variability(BPV);however,chronic stress would not alter short-term BPV in animal models.AIM To examine the association of psychological stress with long-term BPV in young male humans.METHODS We prospectively examined the association of chronic psychological stress with long-term BPV in 1112 healthy military males,averaged 32.2 years from the cardiorespiratory fitness and hospitalization events in armed forces study in Taiwan.Psychological stress was quantitatively evaluated with the Brief Symptom Rating Scale(BSRS-5),from the least symptom of 0 to the most severe of 20,and the five components of anxiety,insomnia,depression,interpersonal sensitivity,and hostility(the severity score in each component from 0 to 4).Longterm BPV was assessed by standard deviation(SD)for systolic and diastolic blood pressure(SBP and DBP),and average real variability(ARV),defined as the average absolute difference between successive measurements of SBP or DBP,across four visits in the study period from 2012 to 2018(2012-14,2014-15,2015-16,and 2016-18).RESULTS The results of multivariable linear regressions showed that there were no correlations of the BSRS-5 score with SDSBP,SDDBP,ARVSBP,and ARVDBP after adjusting for all the covariates[β(SE):-0.022(0.024),-0.023(0.026),-0.001(0.018),and 0.001(0.020),respectively;P>0.05 for all].In addition,there were also no correlations between each component of the BSRS score and the long-term BPV indexes.CONCLUSION Our findings suggest that chronic psychological stress might not be associated with long-term BPV in military young male humans.

Synthetic data as an investigative tool in hypertension and renal diseases research

There is a growing body of clinical research on the utility of synthetic data derivatives,an emerging research tool in medicine.In nephrology,clinicians can use machine learning and artificial intelligence as powerful aids in their clinical decision-making while also preserving patient privacy.This is especially important given the epidemiology of chronic kidney disease,renal oncology,and hypertension worldwide.However,there remains a need to create a framework for guidance regarding how to better utilize synthetic data as a practical application in this research.

Influences of Institutional Pressures on Corporate Social Performance： Empirical Analysis on the Panel Data of Chinese Power Generation Enterprises

Institutional theory has proved the influence of institutional pressures on organization practices and structures. Meanwhile, with the soaring use of corporate social performance （CSP）, more researchers are focusing on exploring the relationship between institution pressures and CSP which is still not completely understood yet. Against this background, the paper aims to fill the gap through generally hypothesizing that different types of institutional pressures individually and collectively affect CSP via the mediating effect of corporate environmental strategy. First, based on the previous and extensive literature review, the theoretical framework and research hypotheses are constructed. Next, canonical correlation analysis about the panel data of 51 Chinese large-scale power generation enterprises from 2004 to 2009 is made to test the relevant hypotheses. Finally, based on the data analysis results, the study draws some conclusions and policy implications for promoting the CSP of Chinese enterprises, including enhancing the steering function of government policies and industry regulations and emphasizing the intermediary role of media.

A formation pressure prediction method based on tectonic overpressure

Traditional formation pressure prediction methods all are based on the formation undercompaction mechanism and the prediction results are obviously low when predicting abnormally high pressure caused by compressional structure overpressure.To eliminate this problem,we propose a new formation pressure prediction method considering compressional structure overpressure as the dominant factor causing abnormally high pressure.First,we establish a model for predicting maximum principal stress,this virtual maximum principal stress is calculated by a double stress field analysis.Then we predict the formation pressure by fitting the maximum principal stress with formation pressure. The real maximum principal stress can be determined by caculating the sum of the virtual maximum principal stresses.Practical application to real data from the A1 and A2 wells in the A gas field shows that this new method has higher accuracy than the traditional equivalent depth method.

Development of Overpressure in the Tertiary Damintun Depression, Liaohe Basin, Northern China

The Damintun depression is one of the four depressions in the Liaohe basin in northern China, and is a rift basin developed in the Paleogene. This paper discusses in detail the characteristics of pressure and fluid potential of the Damintun depression based on a synthesis of the data from boreholes, well tests and seismic surveys. Data from sonic logs, well tests and seismic velocity measurements are used to study the pressure characteristics of the areas. From the sonic log data, shales can be characterized as normally pressured, slightly overpressured or highly overpressured; from the well test data, the pressure-depth gradient in oil-producing intervals implies hydrostatic pressure in general. Most seismic profiles in the Damintun depression are of sufficiently high quality for seismic velocities to be measured. The fluid pressures, excess pressures and pressure coefficients in 47 representative seismic profiles are predicted using formula calculation methods, and further transformed to fluid potenti

Processing and analysis of transient pressure measurements from permanent down-hole gauges

With permanent down-hole gauges （PDGs） widely installed in oilfields around the world in recent years, a continuous stream of transient pressure data in real time is now available, which motivates a new round of research interests in further developing pressure transient processing and analysis techniques. Transient pressure measurements from PDG are characterized by long term and high volume data. These data are recorded under unconstrained circumstances, so effects due to noise, rate fluctuation and interference from other wells cannot be avoided. These effects make the measured pressure trends decline or rise and then obscure or distort the actual flow behavior, which makes subsequent analysis difficult. In this paper, the problems encountered in analysis of PDG transient pressure are investigated. A newly developed workflow for processing and analyzing PDG transient pressure data is proposed. Numerical well testing synthetic studies are performed to demonstrate these procedures. The results prove that this new technique works well and the potential for practical application looks very promising.

Design of Low-Power Data Logger of Deep Sea for Long-Term Field Observation

This paper describes the implementation of a data logger for the real-time in-situ monitoring of hydrothermal systems. A compact mechanical structure ensures the security and reliability of data logger when used under deep sea. The data logger is a battery powered instrument, which can connect chemical sensors （pH electrode, H2S electrode, H2 electrode） and temperature sensors. In order to achieve major energy savings, dynamic power management is implemented in hardware design and software design. The working current of the data logger in idle mode and active mode is 15 μA and 1.44 mA respectively, which greatly extends the working time of battery. The data logger has been successftdly tested in the first Sino-American Cooperative Deep Submergence Project from August 13 to September 3, 2005.

Approach to minish scattering of results for split Hopkinson pressure bar test

Split Hopkinson pressure bar（SHPB） apparatus, usually used for testing behavior of material in median and high strain-rate, is now widely used in the study of rock dynamic constitutive relation, damage evolvement mechanism and energy consumption. However, the possible reasons of sampling disturbance, machining error and so on often lead to the scattering of test results, and bring ultimate difficulty for forming general test conclusion. Based on the stochastic finite element method, the uncertain parameters of specimen density ps, specimen radius Rs, specimen elastic modulus Es and specimen length Ls in the data processing of SHPB test were considered, and the correlation between the parameters and the test results was analyzed. The results show that the specimen radius Rs has direct correlation with the test result, improving the accuracy in preparing and measuring of specimen is an effective way to improve the accuracy of test and minish the scattering of results for SHPB test.

Influence of random uncertainties of anisotropic fibrous model parameters on arterial pressure estimation

This paper deals with a stochastic approach based on the principle of the maximum entropy to investigate the effect of the parameter random uncertainties on the arterial pressure. Motivated by a hyperelastic, anisotropic, and incompressible constitutive law with fiber families, the uncertain parameters describing the mechanical behavior are considered. Based on the available information, the probability density functions are attributed to every random variable to describe the dispersion of the model parameters. Numerous realizations are carried out, and the corresponding arterial pressure results are compared with the human non-invasive clinical data recorded over a mean cardiac cycle. Furthermore, the Monte Carlo simulations are performed, the convergence of the probabilistic model is proven. The different realizations are useful to define a reliable confidence region, in which the probability to have a realization is equM to 95%. It is shown through the obtained results that the error in the estimation of the arterial pressure can reach 35% when the estimation of the model parameters is subjected to an uncertainty ratio of 5%. Finally, a sensitivity analysis is performed to identify the constitutive law relevant parameters for better understanding and characterization of the arterial wall mechanical behaviors.

Self-calibration and algorithm of sample set of piezoresistive pressure sensors

Aiming at piezoresistive pressure sensors, this paper studies simulation of standard pressure by using benchmark current source and self-calibration of the sampling data characteristics. A data fusion algorithm for sample set is presented which transforms a surface problem into a curve fitting and interpolation problem. The simulation result shows that benchmark current source simulating pressure is successful and data fusion algorithm is effective. The maximum measurement error is only 0.098 kPa and maximum relative error is 0.92% at 0-45 kPa and -10-45~C.

Expert system for evaluating the safety of pressure vessels

With more application of welding technology in important structures more attention was paid to the evaluation of the safety of welded structures, the life prediction and decision to repair the welded structures. Based on material fracture mechanism and Chinese standard of safety evaluations of pressure vessels, an expert system was developed to evaluate the safety of welded pressure vessels. The system can analyze the weld defects in a pressure vessel, convert different kinds of defects into equivalent cracks and obtain their equivalent sizes. Furthermore, the system can calculate the stress and strain in the positions of weld defects and make decision on whether the defects are tolerable or not according to the code. When it is tolerable, the system will calculate the safety margin. The fatigue life can be predicted if the defects undergo fatigue load too. Moreover, data bases are built for storing mechanical properties of material and evaluated results.