A hybrid machine learning optimization algorithm for multivariable pore pressure prediction

Pore pressure is essential data in drilling design,and its accurate prediction is necessary to ensure drilling safety and improve drilling efficiency.Traditional methods for predicting pore pressure are limited when forming particular structures and lithology.In this paper,a machine learning algorithm and effective stress theorem are used to establish the transformation model between rock physical parameters and pore pressure.This study collects data from three wells.Well 1 had 881 data sets for model training,and Wells 2 and 3 had 538 and 464 data sets for model testing.In this paper,support vector machine(SVM),random forest(RF),extreme gradient boosting(XGB),and multilayer perceptron(MLP)are selected as the machine learning algorithms for pore pressure modeling.In addition,this paper uses the grey wolf optimization(GWO)algorithm,particle swarm optimization(PSO)algorithm,sparrow search algorithm(SSA),and bat algorithm(BA)to establish a hybrid machine learning optimization algorithm,and proposes an improved grey wolf optimization(IGWO)algorithm.The IGWO-MLP model obtained the minimum root mean square error(RMSE)by using the 5-fold cross-validation method for the training data.For the pore pressure data in Well 2 and Well 3,the coefficients of determination(R^(2))of SVM,RF,XGB,and MLP are 0.9930 and 0.9446,0.9943 and 0.9472,0.9945 and 0.9488,0.9949 and 0.9574.MLP achieves optimal performance on both training and test data,and the MLP model shows a high degree of generalization.It indicates that the IGWO-MLP is an excellent predictor of pore pressure and can be used to predict pore pressure.

Application of multi-algorithm ensemble methods in high-dimensional and small-sample data of geotechnical engineering:A case study of swelling pressure of expansive soils

Geotechnical engineering data are usually small-sample and high-dimensional,which brings a lot of challenges in predictive modeling.This paper uses a typical high-dimensional and small-sample swell pressure(P_(s))dataset to explore the possibility of using multi-algorithm hybrid ensemble and dimensionality reduction methods to mitigate the uncertainty of soil parameter prediction.Based on six machine learning(ML)algorithms,the base learner pool is constructed,and four ensemble methods,Stacking(SG),Blending(BG),Voting regression(VR),and Feature weight linear stacking(FWL),are used for the multi-algorithm ensemble.Furthermore,the importance of permutation is used for feature dimensionality reduction to mitigate the impact of weakly correlated variables on predictive modeling.The results show that the proposed methods are superior to traditional prediction models and base ML models,where FWL is more suitable for modeling with small-sample datasets,and dimensionality reduction can simplify the data structure and reduce the adverse impact of the small-sample effect,which points the way to feature selection for predictive modeling.Based on the ensemble methods,the feature importance of the five primary factors affecting P_(s) is the maximum dry density(31.145%),clay fraction(15.876%),swell percent(15.289%),plasticity index(14%),and optimum moisture content(13.69%),the influence of input parameters on P_(s) is also investigated,in line with the findings of the existing literature.

Research on Multi-Wave Pore Pressure Prediction Method Based on Three Field Velocity Fusion

The optimization of velocity field is the core issue in reservoir seismic pressure prediction. For a long time, the seismic processing velocity analysis method has been used in the establishment of pressure prediction velocity field, which has a long research period and low resolution and restricts the accuracy of seismic pressure prediction;This paper proposed for the first time the use of machine learning algorithms, based on the feasibility analysis of wellbore logging pressure prediction, to integrate the CVI velocity inversion field, velocity sensitive post stack attribute field, and AVO P-wave and S-wave velocity reflectivity to obtain high-precision seismic P and S wave velocities. On this basis, high-resolution formation pore pressure and other parameters prediction based on multi waves is carried out. The pressure prediction accuracy is improved by more than 50% compared to the P-wave resolution of pore pressure prediction using only root mean square velocity. Practice has proven that the research method has certain reference significance for reservoir pore pressure prediction.

Pressure Regulation for Earth Pressure Balance Control on Shield Tunneling Machine by Using Adaptive Robust Control

Most current studies about shield tunneling machine focus on the construction safety and tunnel structure stability during the excavation. Behaviors of the machine itself are also studied, like some tracking control of the machine. Yet, few works concern about the hydraulic components, especially the pressure and flow rate regulation components. This research focuses on pressure control strategies by using proportional pressure relief valve, which is widely applied on typical shield tunneling machines. Modeling of a commercial pressure relief valve is done. The modeling centers on the main valve, because the dynamic performance is determined by the main valve. To validate such modeling, a frequency-experiment result of the pressure relief valve, whose bandwidth is about 3 Hz, is presented as comparison. The modeling and the frequency experimental result show that it is reasonable to regard the pressure relief valve as a second-order system with two low corner frequencies. PID control, dead band compensation control and adaptive robust control（ARC） are proposed and simulation results are presented. For the ARC, implements by using first order approximation and second order approximation are presented. The simulation results show that the second order approximation implement with ARC can track 4 Hz sine signal very well, and the two ARC simulation errors are within 0.2 MPa. Finally, experiment results of dead band compensation control and adaptive robust control are given. The results show that dead band compensation had about 30° phase lag and about 20% off of the amplitude attenuation. ARC is tracking with little phase lag and almost no amplitude attenuation. In this research, ARC has been tested on a pressure relief valve. It is able to improve the valve＇s dynamic performances greatly, and it is capable of the pressure control of shield machine excavation.

Machine learning and data-driven prediction of pore pressure from geophysical logs:A case study for the Mangahewa gas field,New Zealand

Pore pressure is an essential parameter for establishing reservoir conditions,geological interpretation and drilling programs.Pore pressure prediction depends on information from various geophysical logs,seismic,and direct down-hole pressure measurements.However,a level of uncertainty accompanies the prediction of pore pressure because insufficient information is usually recorded in many wells.Applying machine learning(ML)algorithms can decrease the level of uncertainty of pore pressure prediction uncertainty in cases where available information is limited.In this research,several ML techniques are applied to predict pore pressure through the over-pressured Eocene reservoir section penetrated by four wells in the Mangahewa gas field,New Zealand.Their predictions substantially outperform,in terms of prediction performance,those generated using a multiple linear regression(MLR)model.The geophysical logs used as input variables are sonic,temperature and density logs,and some direct pore pressure measurements were available at the reservoir level to calibrate the predictions.A total of 25,935 data records involving six well-log input variables were evaluated across the four wells.All ML methods achieved credible levels of pore pressure prediction performance.The most accurate models for predicting pore pressure in individual wells on a supervised basis are decision tree(DT),adaboost(ADA),random forest(RF)and transparent open box(TOB).The DT achieved root mean square error(RMSE)ranging from 0.25 psi to 14.71 psi for the four wells.The trained models were less accurate when deployed on a semi-supervised basis to predict pore pressure in the other wellbores.For two wells(Mangahewa-03 and Mangahewa-06),semi-supervised prediction achieved acceptable prediction performance of RMSE of 130—140 psi;while for the other wells,semi-supervised prediction performance was reduced to RMSE>300 psi.The results suggest that these models can be used to predict pore pressure in nearby locations,i.e.similar geology at corresponding depths within a field,but they become less reliable as the step-out distance increases and geological conditions change significantly.In comparison to other approaches to predict pore pressures,this study has identified that application of several ML algorithms involving a large number of data records can lead to more accurate prediction results.

Classifying Machine Learning Features Extracted from Vibration Signal with Logistic Model Tree to Monitor Automobile Tyre Pressure

Tyre pressure monitoring system(TPMS)is compulsory in most countries like the United States and European Union.The existing systems depend on pressure sensors strapped on the tyre or on wheel speed sensor data.A difference in wheel speed would trigger an alarm based on the algorithm implemented.In this paper,machine learning approach is proposed as a new method to monitor tyre pressure by extracting the vertical vibrations from a wheel hub of a moving vehicle using an accelerometer.The obtained signals will be used to compute through statistical features and histogram features for the feature extraction process.The LMT(Logistic Model Tree)was used as the classifier and attained a classification accuracy of 92.5%with 10-fold cross validation for statistical features and 90.5% with 10-fold cross validation for histogram features.The proposed model can be used for monitoring the automobile tyre pressure successfully.

A machine learning based solver for pressure Poisson equations

When using the projection method(or fractional step method)to solve the incompressible Navier-Stokes equations,the projection step involves solving a large-scale pressure Poisson equation(PPE),which is computationally expensive and time-consuming.In this study,a machine learning based method is proposed to solve the large-scale PPE.An machine learning(ML)-block is used to completely or partially(if not sufficiently accurate)replace the traditional PPE iterative solver thus accelerating the solution of the incompressible Navier-Stokes equations.The ML-block is designed as a multi-scale graph neural network(GNN)framework,in which the original high-resolution graph corresponds to the discrete grids of the solution domain,graphs with the same resolution are connected by graph convolution operation,and graphs with different resolutions are connected by down/up prolongation operation.The well trained MLblock will act as a general-purpose PPE solver for a certain kind of flow problems.The proposed method is verified via solving two-dimensional Kolmogorov flows(Re=1000 and Re=5000)with different source terms.On the premise of achieving a specified high precision(ML-block partially replaces the traditional iterative solver),the ML-block provides a better initial iteration value for the traditional iterative solver,which greatly reduces the number of iterations of the traditional iterative solver and speeds up the solution of the PPE.Numerical experiments show that the ML-block has great advantages in accelerating the solving of the Navier-Stokes equations while ensuring high accuracy.

Tyre Pressure Supervision of Two Wheeler Using Machine Learning

The regulation of tyre pressure is treated as a significant aspect of‘tyre maintenance’in the domain of autotronics.The manual supervision of a tyre pressure is typically an ignored task by most of the users.The existing instru-mental scheme incorporates stand-alone monitoring with pressure and/or temperature sensors and requires reg-ular manual conduct.Hence these schemes turn to be incompatible for on-board supervision and automated prediction of tyre condition.In this perspective,the Machine Learning(ML)approach acts appropriate as it exhi-bits comparison of specific performance in the past with present,intended for predicting the same in near future.The current investigation experimentally assesses the suitability of ML scheme for vibration based on-board supervision of tyre pressure of two wheeled vehicle.In order to examine the vibration response of a wheel hub,the in-house design&development of DAQ(Data Acquisition System)is described.Micro Electro-Mechanical Scheme(MEMS)built accelerometer is incorporated with open source hardware and software to collect and store the data.This framework is easy to develop,monitor and can be retrofitted in two wheeled vehicle.For various pressure conditions,the change in response of wheel hub vibration with respect to time is collected.The statistical parameters describing these vibration signals are determined and the decision tree is applied to select distinguishing parameters between extracted parameters.The classification of different conditions of tyre pressure is carried out using ML classifiers.

Analysis of Static Pressure in Area between Back Plate and Cylinder of a Carding Machine with CFD

To analyze static pressure between back plate and cylinder in an A186 carding machine,a fluid model is established. The model takes into account static pressure of airflow near back plate with the numerical simulation method of Computational Fluid Dynamics (CFD) in FLUENT software. The result of the simulation in the model shows that static pressure in this area quickly increases to its maximum then rapidly decreases to a lower fixed value from inlet to outlet along a zone between back plate and cylinder. Both rotating speeds of the cylinder and the taker-in affect static pressure from the inlet to the outlet,of which the cylinder rotating speed has more influence than that of taker-in. Numerical simulations reveal that static pressure on surface of back plate are in good agreement with the former result of experimental analysis.

The mechanism of external pressure suppressing dendrites growth in Li metal batteries

Li metal is considered an ideal anode material for application in the next-generation secondary batteries.However,the commercial application of Li metal batteries has not yet been achieved due to the safety concern caused by Li dendrites growth.Despite the fact that many recent experimental studies found that external pressure suppresses the Li dendrites growth,the mechanism of the external pressure effect on Li dendrites remains poorly understood on the atomic scale.Herein,the large-scale molecular dynamics simulations of Li dendrites growth under different external pressure were performed with a machine learning potential,which has the quantum-mechanical accuracy.The simulation results reveal that the external pressure promotes the process of Li self-healing.With the increase of external pressure,the hole defects and Li dendrites would gradually fuse and disappear.This work provides a new perspective for understanding the mechanism for the impact of external pressure on Li dendrites.

Predicting triaxial compressive strength of high-temperature treated rock using machine learning techniques

The accurate prediction of the strength of rocks after high-temperature treatment is important for the safety maintenance of rock in deep underground engineering.Five machine learning(ML)techniques were adopted in this study,i.e.back propagation neural network(BPNN),AdaBoost-based classification and regression tree(AdaBoost-CART),support vector machine(SVM),K-nearest neighbor(KNN),and radial basis function neural network(RBFNN).A total of 351 data points with seven input parameters(i.e.diameter and height of specimen,density,temperature,confining pressure,crack damage stress and elastic modulus)and one output parameter(triaxial compressive strength)were utilized.The root mean square error(RMSE),mean absolute error(MAE)and correlation coefficient(R)were used to evaluate the prediction performance of the five ML models.The results demonstrated that the BPNN shows a better prediction performance than the other models with RMSE,MAE and R values on the testing dataset of 15.4 MPa,11.03 MPa and 0.9921,respectively.The results indicated that the ML techniques are effective for accurately predicting the triaxial compressive strength of rocks after different high-temperature treatments.

Machine Learning for Detecting Blood Transfusion Needs Using Biosignals

Adequate oxygen in red blood cells carrying through the body to the heart and brain is important to maintain life.For those patients requiring blood,blood transfusion is a common procedure in which donated blood or blood components are given through an intravenous line.However,detecting the need for blood transfusion is time-consuming and sometimes not easily diagnosed,such as internal bleeding.This study considered physiological signals such as electrocardiogram(ECG),photoplethysmogram(PPG),blood pressure,oxygen saturation(SpO2),and respiration,and proposed the machine learning model to detect the need for blood transfusion accurately.For the model,this study extracted 14 features from the physiological signals and used an ensemble approach combining extreme gradient boosting and random forest.The model was evaluated by a stratified five-fold crossvalidation:the detection accuracy and area under the receiver operating characteristics were 92.7%and 0.977,respectively.

Soft ground tunnel lithology classification using clustering-guided light gradient boosting machine

During tunnel boring machine(TBM)excavation,lithology identification is an important issue to understand tunnelling performance and avoid time-consuming excavation.However,site investigation generally lacks ground samples and the information is subjective,heterogeneous,and imbalanced due to mixed ground conditions.In this study,an unsupervised(K-means)and synthetic minority oversampling technique(SMOTE)-guided light-gradient boosting machine(LightGBM)classifier is proposed to identify the soft ground tunnel classification and determine the imbalanced issue of tunnelling data.During the tunnel excavation,an earth pressure balance(EPB)TBM recorded 18 different operational parameters along with the three main tunnel lithologies.The proposed model is applied using Python low-code PyCaret library.Next,four decision tree-based classifiers were obtained in a short time period with automatic hyperparameter tuning to determine the best model for clustering-guided SMOTE application.In addition,the Shapley additive explanation(SHAP)was implemented to avoid the model black box problem.The proposed model was evaluated using different metrics such as accuracy,F1 score,precision,recall,and receiver operating characteristics(ROC)curve to obtain a reasonable outcome for the minority class.It shows that the proposed model can provide significant tunnel lithology identification based on the operational parameters of EPB-TBM.The proposed method can be applied to heterogeneous tunnel formations with several TBM operational parameters to describe the tunnel lithologies for efficient tunnelling.

Pore pressure prediction in offshore Niger delta using data-driven approach: Implications on drilling and reservoir quality

Despite exploration and production success in Niger Delta,several failed wells have been encountered due to overpressures.Hence,it is very essential to understand the spatial distribution of pore pressure and the generating mechanisms in order to mitigate the pitfalls that might arise during drilling.This research provides estimates of pore pressure along three offshore wells using the Eaton's transit time method,multi-layer perceptron artificial neural network(MLP-ANN)and random forest regression(RFR)algorithms.Our results show that there are three pressure magnitude regimes:normal pressure zone(hydrostatic pressure),transition pressure zone(slightly above hydrostatic pressure),and over pressured zone(significantly above hydrostatic pressure).The top of the geopressured zone(2873 mbRT or 9425.853 ft)averagely marks the onset of overpressurization with the excess pore pressure above hydrostatic pressure(P∗)varying averagely along the three wells between 1.06−24.75 MPa.The results from the three methods are self-consistent with strong correlation between the Eaton's method and the two machine learning models.The models have high accuracy of about>97%,low mean absolute percentage error(MAPE<3%)and coefficient of determination(R2>0.98).Our results have also shown that the principal generating mechanisms responsible for high pore pressure in the offshore Niger Delta are disequilibrium compaction,unloading(fluid expansion)and shale diagenesis.

Process Window Expansion of Laser Chemical Machining by Using High Pressure

Laser Chemical Machining (LCM) is a non-conventional removal process, based on a precise thermal activation of heterogeneous chemical reactions between an electrolyte and a metallic surface. Due to local overheating during the process, boiling bubbles occur, which can impair the removal quality. In order to reduce the amount of bubbles, the laser chemical process is performed at different process pressures. Removal experiments were performed on Titanium Grade 1 using the electrolyte phosphoric acid at various process pressures, machining speeds and laser powers in order to determine the limit of the process window by evaluating the characteristics of the removal cavities. As a result, the process window for non-disturbed laser chemical machining is widened at higher process pressures. The process pressures have no influence on the geometric shape of the removal. The expansion of the process window is attributed to the fact that at higher process pressures the saturation temperature of the electrolyte rises, so that bubble boiling starts at a higher surface temperature on the workpiece induced by the laser power. The removal rate could be increased by a factor of 2.48 by increasing the process pressures from ambient pressure to 6 bar, thus taking an important step towards the economic efficiency of the laser chemical machining.

Geometric Error Identification of Gantry-Type CNC Machine Tool Based on Multi-Station Synchronization Laser Tracers

Laser tracers are a three-dimensional coordinate measurement system that are widely used in industrial measurement.We propose a geometric error identification method based on multi-station synchronization laser tracers to enable the rapid and high-precision measurement of geometric errors for gantry-type computer numerical control(CNC)machine tools.This method also improves on the existing measurement efficiency issues in the single-base station measurement method and multi-base station time-sharing measurement method.We consider a three-axis gantry-type CNC machine tool,and the geometric error mathematical model is derived and established based on the combination of screw theory and a topological analysis of the machine kinematic chain.The four-station laser tracers position and measurement points are realized based on the multi-point positioning principle.A self-calibration algorithm is proposed for the coordinate calibration process of a laser tracer using the Levenberg-Marquardt nonlinear least squares method,and the geometric error is solved using Taylor’s first-order linearization iteration.The experimental results show that the geometric error calculated based on this modeling method is comparable to the results from the Etalon laser tracer.For a volume of 800 mm×1000 mm×350 mm,the maximum differences of the linear,angular,and spatial position errors were 2.0μm,2.7μrad,and 12.0μm,respectively,which verifies the accuracy of the proposed algorithm.This research proposes a modeling method for the precise measurement of errors in machine tools,and the applied nature of this study also makes it relevant both to researchers and those in the industrial sector.

A machine learning approach for the prediction of pore pressure using well log data of Hikurangi Tuaheni Zone of IODP Expedition 372,New Zealand

Pore pressure(PP)information plays an important role in analysing the geomechanical properties of the reservoir and hydrocarbon field development.PP prediction is an essential requirement to ensure safe drilling operations and it is a fundamental input for well design,and mud weight estimation for wellbore stability.However,the pore pressure trend prediction in complex geological provinces is challenging particularly at oceanic slope setting,where sedimentation rate is relatively high and PP can be driven by various complex geo-processes.To overcome these difficulties,an advanced machine learning(ML)tool is implemented in combination with empirical methods.The empirical method for PP prediction is comprised of data pre-processing and model establishment stage.Eaton's method and Porosity method have been used for PP calculation of the well U1517A located at Tuaheni Landslide Complex of Hikurangi Subduction zone of IODP expedition 372.Gamma-ray,sonic travel time,bulk density and sonic derived porosity are extracted from well log data for the theoretical framework construction.The normal compaction trend(NCT)curve analysis is used to check the optimum fitting of the low permeable zone data.The statistical analysis is done using the histogram analysis and Pearson correlation coefficient matrix with PP data series to identify potential input combinations for ML-based predictive model development.The dataset is prepared and divided into two parts:Training and Testing.The PP data and well log of borehole U1517A is pre-processed to scale in between[-1,+1]to fit into the input range of the non-linear activation/transfer function of the decision tree regression model.The Decision Tree Regression(DTR)algorithm is built and compared to the model performance to predict the PP and identify the overpressure zone in Hikurangi Tuaheni Zone of IODP Expedition 372.

Experimental Study of Integration and Polyblends Performance for Improved Mixing-Extruding Machine

One improved mixing-extruding machine was introduced as the second-generation product of the mixing-molding integrated technology. In the extruding system,the conventional single screw extruder was substituted by a special conical twin-screw extruder,resulting in stronger feeding ability,more stable extrusion pressure,and better quality of products. The integrated mathematical model of mixing-extruding process was also established by theoretical derivation and optimization according to the experimental results.Then its accuracy was verified by the influences of the pressure of floating weight and the cooling water temperature of extruder on the mixing-extruding integrated process. The results showed that the changes of both parameters could give rise to the fluctuation of the temperature and apparent viscosity of polyblends, thus further influencing the screw rotation speed.

Discrete element simulation of mechanical characteristic of conditioned sands in earth pressure balance shield tunneling

The discrete element method (DEM) was used to simulate the flow characteristic and strength characteristic of the conditioned sands in the earth pressure balance (EPB) tunneling. In the laboratory the conditioned sands were reproduced and the slump test and the direct shear test of the conditioned sands were implemented. A DEM equivalent model that can simulate the macro mechanical characteristic of the conditioned sands was proposed,and the corresponding numerical models of the slump test and the shear test were established. By selecting proper DEM model parameters,the errors of the slump values between the simulation results and the test results are in the range of 10.3%-14.3%,and the error of the curves between the shear displacement and the shear stress calculated with the DEM simulation is 4.68%-16.5% compared with that of the laboratory direct shear test. This illustrates that the proposed DEM equivalent model can approximately simulate the mechanical characteristics of the conditioned sands,which provides the basis for further simulation of the interaction between the conditioned soil and the chamber pressure system of the EPB machine.