Analytical Model of Micropyramidal Capacitive Pressure Sensors and Machine‐Learning‐Assisted Design

Automated lying-posture tracking is important in preventing bed-related disorders, such as pressure injuries, sleep apnea, and lower-back pain. Prior research studied in-bed lying posture tracking using sensors of different modalities (e.g., accelerometer and pressure sensors). However, there remain significant gaps in research regarding how to design efficient in-bed lying posture tracking systems. These gaps can be articulated through several research questions, as follows. First, can we design a single-sensor, pervasive, and inexpensive system that can accurately detect lying postures? Second, what computational models are most effective in the accurate detection of lying postures? Finally, what physical configuration of the sensor system is most effective for lying posture tracking? To answer these important research questions, in this article we propose a comprehensive approach for designing a sensor system that uses a single accelerometer along with machine learning algorithms for in-bed lying posture classification. We design two categories of machine learning algorithms based on deep learning and traditional classification with handcrafted features to detect lying postures. We also investigate what wearing sites are the most effective in the accurate detection of lying postures. We extensively evaluate the performance of the proposed algorithms on nine different body locations and four human lying postures using two datasets. Our results show that a system with a single accelerometer can be used with either deep learning or traditional classifiers to accurately detect lying postures. The best models in our approach achieve an F1 score that ranges from 95.2% to 97.8% with a coefficient of variation from 0.03 to 0.05. The results also identify the thighs and chest as the most salient body sites for lying posture tracking. Our findings in this article suggest that, because accelerometers are ubiquitous and inexpensive sensors, they can be a viable source of information for pervasive monitoring of in-bed postures.

Download Full-text

A data-driven analytical model for wind turbine wakes using machine learning method

Energy Conversion and Management ◽

10.1016/j.enconman.2021.115130 ◽

2022 ◽

Vol 252 ◽

pp. 115130

Author(s):

Guo Nai-Zhi ◽

Zhang Ming-Ming ◽

Li Bo

Keyword(s):

Machine Learning ◽

Wind Turbine ◽

Analytical Model ◽

Data Driven ◽

Machine Learning Method ◽

Learning Method

Download Full-text

Smart Inventory Management System and Predictive Analysis

International Journal for Research in Applied Science and Engineering Technology ◽

10.22214/ijraset.2021.38493 ◽

2021 ◽

Vol 9 (10) ◽

pp. 729-732

Author(s):

Harsh Srivastava

Keyword(s):

Machine Learning ◽

Analytical Model ◽

Inventory Management ◽

Business Intelligence ◽

Management System ◽

Data Science ◽

Predictive Analysis ◽

Inventory Management System ◽

Business Intelligence Systems

Abstract: The idea of a smarter inventory management and business intelligence systems is a challenge in its implementation for various organization and businesses especially small and medium sized retailers. The concept of the paper is based on the business aspects while it merges various technological features to make a Predictive Analytical Model. Very few organizations and companies implement this which amount to loss of business, redundancies and errors that can be largely solved by system that uses data science, machine learning and visualization. This paper aims to present various aspects that can be infused with technology to aid Business Techniques and enable automation.

Download Full-text

Unobtrusive and Continuous Monitoring of Alcohol-impaired Gait Using Smart Shoes

Methods of Information in Medicine ◽

10.3414/me15-02-0008 ◽

2017 ◽

Vol 56 (01) ◽

pp. 74-82 ◽

Cited By ~ 6

Author(s):

Sunghoon I. Lee ◽

Hyo Suk Nam ◽

Jordan H. Garst ◽

Alex Huang ◽

Andrew Campion ◽

...

Keyword(s):

Machine Learning ◽

Monitoring System ◽

Alcohol Intoxication ◽

Pressure Sensors ◽

Well Being ◽

Value Analysis ◽

Single Feature ◽

Ubiquitous Environment ◽

Gait Features ◽

Gait Monitoring

SummaryBackground: Alcohol ingestion influences sensory-motor function and the overall well-being of individuals. Detecting alcoholinduced impairments in gait in daily life necessitates a continuous and unobtrusive gait monitoring system.Objectives: This paper introduces the development and use of a non-intrusive monitoring system to detect changes in gait induced by alcohol intoxication.Methods: The proposed system employed a pair of sensorized smart shoes that are equipped with pressure sensors on the insole. Gait features were extracted and adjusted based on individual’s gait profile. The adjusted gait features were used to train a machine learning classifier to discriminate alcohol-impaired gait from normal walking. In experiment of pilot study, twenty participants completed walking trials on a 12 meter walkway to measure their sober walking and alcohol-impaired walking using smart shoes.Results: The proposed system can detect alcohol-impaired gait with an accuracy of 86.2% when pressure value analysis and person-dependent model for the classifier are applied, while statistical analysis revealed that no single feature was discriminative for the detection of gait impairment.Conclusions: Alcohol-induced gait disturbances can be detected with smart shoe technology for an automated monitoring in ubiquitous environment. We demonstrated that personal monitoring and machine learning-based prediction could be customized to detect individual variation rather than applying uniform boundary parameters of gait.

Download Full-text

Intelligent Production Monitoring with Continuous Deep Learning Models

10.2118/206525-ms ◽

2021 ◽

Author(s):

Anton Gryzlov ◽

Sergey Safonov ◽

Muhammad Arsalan

Keyword(s):

Machine Learning ◽

Deep Learning ◽

Multiphase Flow ◽

Pressure Sensors ◽

Production Optimization ◽

Learning Models ◽

Learning Methods ◽

Machine Learning Methods ◽

Production Monitoring ◽

Oil Gas

Abstract Monitoring of production rates is essential for reservoir management, history matching, and production optimization. Traditionally, such information is provided by multiphase flow meters or test separators. The growth of the availability of data, combined with the rapid development of computational resources, enabled the inception of digital techniques, which estimate oil, gas, and water rates indirectly. This paper discusses the application of continuous deep learning models, capable of reproducing multiphase flow dynamics for production monitoring purposes. This technique combines time evolution properties of a dynamical system and the ability of neural networks to quantitively describe poorly understood multiphase phenomena and can be considered as a hybrid solution between data-driven and mechanistic approaches. The continuous latent ordinary differential equation (Latent ODE) approach is compared to other known machine learning methods, such as linear regression, ensemble-based model, and recurrent neural network. In this work, the application of Latent ordinary differential equations for the problem of multiphase flow rate estimation is introduced. The considered example refers to a scenario, where the topside oil, gas, and water flow rates are estimated using the data from several downhole pressure sensors. The predictive capabilities of different types of machine learning and deep learning instruments are explored using simulated production data from a multiphase flow simulator. The results demonstrate the satisfactory performance of the continuous deep learning models in comparison to other machine learning methods in terms of accuracy, where the normalized root mean squared error (RMSE) and mean absolute error (MAE) of prediction below 5% were achieved. While LODE demonstrates the significant time required to train the model, it outperforms other methods for irregularly sampled time-series, which makes it especially attractive to forecast values of multiphase rates.

Download Full-text

Identifying Perverse Incentives in Buyer Profiling on Online Trading Platforms

Information Systems Research ◽

10.1287/isre.2021.1077 ◽

2021 ◽

Author(s):

Karthik Kannan ◽

Rajib L. Saha ◽

Warut Khern-am-nuai

Keyword(s):

Artificial Intelligence ◽

Machine Learning ◽

Analytical Model ◽

Perverse Incentive ◽

Intelligence Models ◽

Online Trading ◽

Artificial Intelligence Models ◽

Perverse Incentives ◽

Consumer Profiling ◽

Underlying Mechanisms

With advance machine learning and artificial intelligence models, the capability of online trading platforms to profile consumers to identify and understand their needs has substantially increased. In this study, we use an analytical model to study whether these platforms have an incentive to profile their customers as accurately as possible. We find that “payments-for-transactions” platforms (i.e., platforms that charge for transactions that occur on the platform) indeed have such incentives to accurately profile the customers. However, surprisingly, “payments-for-discoveries” platform (i.e., platforms that charge customers for discoveries) have a perverse incentive to deviate from accurate consumer profiling. Our study provides insights into underlying mechanisms that drive this perverse incentive and discuss circumstances that lead to such a perverse incentive.

Download Full-text

Optimization of the Reservoir Management System and the ESP Operation Control Process by Means of Machine Learning on the Oilfields of Salym Petroleum Development N.V.

10.2118/206518-ms ◽

2021 ◽

Author(s):

Anastasia Dmitrievna Musorina ◽

Grigory Sergeyevich Ishimbayev

Keyword(s):

Machine Learning ◽

Input Parameter ◽

Pressure Sensors ◽

Development Program ◽

Gas Production ◽

Measuring Instruments ◽

Target Variable ◽

Pump Design ◽

Field Development ◽

Operation Control

Abstract Under the present conditions of oil and gas production, which are characterized by mature production fields and the focus shifted towards digitalization of production processes and use of machine learning (ML) models, the issues related to the improvement of accuracy and consistency of the well operation control data are becoming increasingly important. As a result, SPD has successfully implemented the project of using annular pressure sensors in combination with machine learning models to control the well annular pressure as part of the field development program compliance. Under the field development program, echosounder and telemetry system readings are typically used to control the annular pressure and the dynamic flowing level. Echosounders, however, are not designed as measuring instruments, the accuracy of their readings being low and making it impossible to reliably evaluate the well's dynamic flowing level and annular pressure, as well as to achieve the well's maximum potential, and the telemetry systems used to measure the pump intake pressure may go wrong. This manuscript describes the approach to the producer well annular pressure assessment based on the machine learning model data. The machine learning (ML) model is a function of the target variable (bottom-hole pressure), which is predicted on the basis of the actual data: static parameters (well schematic, pump design) and dynamic parameters (annular and line pressures, flowrate). The input parameter interpretation results in the most probable value of the target variable based on the historic data.

Download Full-text

Machine learning of electro-hydraulic motor dynamics

SN Applied Sciences ◽

10.1007/s42452-019-1889-y ◽

2019 ◽

Vol 2 (1) ◽

Cited By ~ 1

Author(s):

Luca Baronti ◽

Biao Zhang ◽

Marco Castellani ◽

Duc Truong Pham

Keyword(s):

Neural Network ◽

Machine Learning ◽

Analytical Model ◽

Random Search ◽

Bees Algorithm ◽

Multi Layer Perceptron ◽

Hydraulic Motor ◽

Neural Network Models ◽

Load Simulator ◽

Intelligent Optimisation

AbstractIn this paper we propose an innovative machine learning approach to the hydraulic motor load balancing problem involving intelligent optimisation and neural networks. Two different nonlinear artificial neural network approaches are investigated, and their accuracy is compared to that of a linearised analytical model. The first neural network approach uses a multi-layer perceptron to reproduce the load simulator dynamics. The multi-layer perceptron is trained using the Rprop algorithm. The second approach uses a hybrid scheme featuring an analytical model to represent the main system behaviour, and a multi-layer perceptron to reproduce unmodelled nonlinear terms. Four techniques are tested for the optimisation of the parameters of the analytical model: random search, an evolutionary algorithm, particle swarm optimisation, and the Bees Algorithm. Experimental tests on 4500 real data samples from an electro-hydraulic load simulator rig reveal that the accuracy of the hybrid and the neural network models is comparable, and significantly superior to the accuracy of the analytical model. The results of the optimisation procedures suggest also that the inferior performance of the analytical model is likely due to the non-negligible magnitude of the unmodelled nonlinearities, rather than suboptimal setting of the parameters. Despite its limitations, the analytical linear model performs comparably to the state-of-the-art in the literature, whilst the neural and hybrid approaches compare favourably.

Download Full-text

An Analytical Model for Porous Polymer-Ceramic Capacitive Pressure Sensors

IEEE Sensors Journal ◽

10.1109/jsen.2014.2332357 ◽

2014 ◽

Vol 14 (12) ◽

pp. 4411-4422 ◽

Cited By ~ 3

Author(s):

Timothy L. Weadon ◽

Thomas H. Evans ◽

Edward M. Sabolsky

Keyword(s):

Analytical Model ◽

Pressure Sensors ◽

Porous Polymer

Download Full-text

Fault detection and accommodation of a thermal-type flow sensor in an aircraft bleed air system

10.32920/ryerson.14660643 ◽

2021 ◽

Author(s):

Christopher Day

Keyword(s):

Fault Detection ◽

Analytical Model ◽

Mass Flow ◽

Flow Rate ◽

Pressure Sensors ◽

Flow Sensor ◽

Rule Base ◽

Sensor Fault ◽

Air System ◽

Mass Flow Sensor

A fault in the primary mass flow sensor of an aircraft engine bleed air system can cause significant deterioration of overall system performance. This project uses an analytical model of the bleed air system to create a fault detection and accommodation scheme for the mass flow sensor. The analytical model uses information from the upstream and downstream pressure sensors to predict the output of the mass flow sensor. Faults are detected by comparing the output from the sensor with the predicted output from the analytical model. A fuzzy logic rule base is used to determine the degree of the flow sensor fault. The degree of the sensor fault is used to determine the inaccuracy of the faulty sensor output. A corrected estimation of the flow rate is then created using a weighted algorithm consisting of the predicted flow rate from the analytical model and the flow rate from the faulty sensor. The analytical model is also used to detect and accommodate transient responses from the flow sensor including signal overshoot, oscillations and time constant errors. A MATLAB computer simulation is conducted to evaluate the performance of the bleed air system degrades slightly in the event of a fault of the flow sensor. While the sensor fault will degrade the performance of the bleed air system, the degradation is not significant, and the bleed air system is able to maintain acceptable performance in the presence of faults.

Download Full-text