Nonlinear Sensor Research Articles

A Fast-Response Breathing Monitoring System for Human Respiration Disease Detection

This paper presents a sensing system for the real-time monitoring of human respiration. The system is equipped with a fast response thermoresistive micro calorimetric flow (TMCF) sensor and a dedicated data processing algorithm. The TMCF sensor is designed with a proposed nonlinear sensor model and fabricated in a CMOS compatible process, which obtains a high sensitivity of 114 mV/SLM and a fast response time of less than 6 ms. By using this high-performance micro flow sensor and its proprietary data processing algorithm, critical human respiration information including respiratory rate (RR) and minute ventilation (MV) can be easily obtained. The proposed sensing system achieves a very small mean absolute error (MAE) of less than 2.7 mHz for RR, and the extracted MV is also in good agreement with the commonly reported value of 4 - 6 L/min. In addition, benefiting from the very short time constant of the developed TMCF sensor, the proposed sensing system can successfully distinguish different respiratory diseases, such as apnea, hypopnea, polypnea, etc. Therefore, this proposed human respiration monitoring system will be a promising sensing technology for respiration diagnosis in medical applications.

Quantized Fault-Tolerant Control for Descriptor Systems with Intermittent Actuator Faults, Randomly Occurring Sensor Non-Linearity, and Missing Data

This paper examines the fault-tolerant control problem for discrete-time descriptor systems that are susceptible to intermittent actuator failures, nonlinear sensor data, and probability-based missing data. The discrete-time non-homogeneous Markov chain was adopted to describe the stochastic behavior of actuator faults. Moreover, Bernoulli-distributed stochastic variables with known conditional probabilities were employed to describe the practical features of random sensor non-linearity and missing data. In this study, the output signals were quantized and a dynamic output feedback controller was synthesized such that the closed-loop system was stochastically admissible and satisfied the strictly (Q,S,R)-γ-dissipative performance index. The theoretical developments are illustrated through numerical simulations of an infinite machine bus.

Linear and Non-Linear Soft Sensors for Predicting the Research Octane Number (RON) through Integrated Synchronization, Resolution Selection and Modelling.

The Research Octane Number (RON) is a key quality parameter for gasoline, obtained offline through complex, time-consuming, and expensive standard methods. Measurements are usually only available a few times per week and after long delays, making process control very challenging. Therefore, alternative methods have been proposed to predict RON from readily available data. In this work, we report the development of inferential models for predicting RON from process data collected in a real catalytic reforming process. Data resolution and synchronization were explicitly considered during the modelling stage, where 20 predictive linear and non-linear machine learning models were assessed and compared using a robust Monte Carlo double cross-validation approach. The workflow also handles outliers, missing data, multirate and multiresolution observations, and processes dynamics, among other features. Low RMSE were obtained under testing conditions (close to 0.5), with the best methods belonging to the class of penalized regression methods and partial least squares. The developed models allow for improved management of the operational conditions necessary to achieve the target RON, including a more effective use of the heating utilities, which improves process efficiency while reducing costs and emissions.

Electromagnetic wave-based extreme deep learning with nonlinear time-Floquet entanglement

Wave-based analog signal processing holds the promise of extremely fast, on-the-fly, power-efficient data processing, occurring as a wave propagates through an artificially engineered medium. Yet, due to the fundamentally weak non-linearities of traditional electromagnetic materials, such analog processors have been so far largely confined to simple linear projections such as image edge detection or matrix multiplications. Complex neuromorphic computing tasks, which inherently require strong non-linearities, have so far remained out-of-reach of wave-based solutions, with a few attempts that implemented non-linearities on the digital front, or used weak and inflexible non-linear sensors, restraining the learning performance. Here, we tackle this issue by demonstrating the relevance of time-Floquet physics to induce a strong non-linear entanglement between signal inputs at different frequencies, enabling a power-efficient and versatile wave platform for analog extreme deep learning involving a single, uniformly modulated dielectric layer and a scattering medium. We prove the efficiency of the method for extreme learning machines and reservoir computing to solve a range of challenging learning tasks, from forecasting chaotic time series to the simultaneous classification of distinct datasets. Our results open the way for optical wave-based machine learning with high energy efficiency, speed and scalability.

Adaptive event‐triggered H∞ state estimation of semi‐Markovian jump neural networks with randomly occurred sensor nonlinearity

AbstractThis article mainly discusses the problem for adaptive event‐triggered H state estimation of semi‐Markovian jump neural networks (s‐MJNNs) subject to random sensor nonlinearity. To reduce the communication load, adaptive event‐triggered scheme (AETS) is introduced to decide whether to transmit sampled data or not. Also, considering the possible sensor nonlinearity, a new estimation error model is established under the framework of AETS. An appropriate Lyapunov–Krasovskii functional (LKF) containing the proposed adaptive event trigger condition is constructed, and sufficient conditions are obtained to guarantee the asymptotic stability of the estimation error system. Then, through a set of feasible linear matrix inequalities (LMIs), the co‐design method of estimator and AETS is proposed. Finally, the feasibility of this paper is proved by three numerical examples.

Nonlinear Dynamic Soft Sensor Development with a Supervised Hybrid CNN-LSTM Network for Industrial Processes.

A soft sensor is a key component when a real-time measurement is unavailable for industrial processes. Recently, soft sensor models based on deep-learning techniques have been successfully applied to complex industrial processes with nonlinear and dynamic characteristics. However, the conventional deep-learning-based methods cannot guarantee that the quality-relevant features are included in the hidden states when the modeling samples are limited. To address this issue, a supervised hybrid network based on a dynamic convolutional neural network (CNN) and a long short-term memory (LSTM) network is designed by constructing multilayer dynamic CNN-LSTM with improved structures. In each time instant, data augmentation is implemented by dynamic expansion of the original samples. Moreover, multiple supervised hidden units are trained by adding quality variables as part of the layer input to acquire a better quality-related feature learning performance. The effectiveness of the proposed soft senor development is validated through two industrial applications, including a penicillin fermentation process and a debutanizer column.

Closing the Control Loop with Time-Variant Embedded Soft Sensors and Recurrent Neural Networks.

Embedded soft sensors can significantly impact the design and control of soft-bodied robots. Although there have been considerable advances in technology behind these novel sensing materials, their application in real-world tasks, especially in closed-loop control tasks, has been severely limited. This is mainly because of the challenge involved with modeling a nonlinear time-variant sensor embedded in a complex soft-bodied system. This article presents a learning-based approach for closed-loop force control with embedded soft sensors and recurrent neural networks (RNNs). We present learning protocols for training a class of RNNs called long short-term memory (LSTM) that allows us to develop accurate and robust state estimation models of these complex dynamical systems within a short period of time. Using this model, we develop a simple feedback force controller for a soft anthropomorphic finger even with significant drift and hysteresis in our feedback signal. Simulation and experimental studies are conducted to analyze the capabilities and generalizability of the control architecture. Experimentally, we are able to develop a closed-loop controller with a control frequency of 25 Hz and an average accuracy of 0.17 N. Our results indicate that current soft sensing technologies can already be used in real-world applications with the aid of machine learning techniques and an appropriate training methodology.

Phase Demodulation of Rotor Torsional Vibration Measurement under Time-Varying Speed

When torsional vibration is measured with the zebra-tape method, the modulation signals can be contaminated by electrical noise, environmental noise, step noise, sensor nonlinearities, and so on. Although the amplitudes of these types of noise are very weak relative to the amplitude of the ideal pulse test signal, they may introduce large-scale random broad-spectrum noise to the signal phases. Such noise is difficult to remove using frequency domain, time-frequency-domain, or threshold sampling methods and seriously affects the accuracy of torsional vibration measurements. This paper presents a phase demodulation algorithm based on downsampling and local resampling (DSLR) to improve the accuracy of torsional vibration measurements. To verify the proposed DSLR algorithm, torsional vibration simulations and experiments are conducted under a time-varying rotation speed. The results show that the angular displacement signal obtained from the torsional vibration test using zebra tapes is the superposition of the dynamic and static angular displacements, with the latter generated because of the varying rotation speed with respect to time. The DSLR algorithm effectively reduces the phase noise and demodulates the pulse signal phases. The work in this paper provides a method for obtaining reference torsional vibration measurements of rotors based on the zebra-tape method under time-varying speeds.

Nonlinear Optical and Ion Sensor Properties of Novel Molecules Conjugated by Click Chemistry

The molecular structure, luminescence behavior, and electronic energy level of an organic optoelectronic materials are important parameters for its synthesis. The electro-optical properties can be changed by modifying the structure of the molecule to make the electronic energy level adjustable. In this article, a series of organic conjugated micro-molecules are successfully synthesized by linking small compound units. This metal-free [2 + 2] click chemistry process generates donor–acceptor chromophore substances with high yield, high solubility, and adjustable energy levels, which can be widely used for sensors and nonlinear optics in different fields. A-TCNE, A-TCNQ, and A-F4-TCNQ molecules are characterized comprehensively via UV-Vis-NIR spectra, 1H NMR spectra, infrared spectroscopy, and mass spectrometry. The unique nonlinear optical phenomena and powerful intra-molecular charge–transfer interactions of these new materials give them fascinating potential for application as optoelectronic materials.

On the Question of the Intrinsic Nonlinearity of the Induction Sensor

On the Question of the Intrinsic Nonlinearity of the Induction Sensor

Design of active fault-tolerant control system for Air-fuel ratio control of internal combustion engines using fuzzy logic controller.

Fault-Tolerant Control Systems (FTCS) are used in critical and safety applications to improve performance and stability despite failure modes. As a result, costly production losses related to unusual and unplanned shutdowns can be prevented by incorporating these systems in the critical process plant machines. The Internal Combustion (IC) engines are highly used process plant machines and faults in their sensors will cause their shutdown instigating the need to install FTCS in them. In this paper, an Active Fault-Tolerant Control System (AFTCS) based on a Fuzzy Logic Controller (FLC) is suggested to improve the reliability of the Air-Fuel Ratio (AFR) control system of an IC engine. For analytical redundancy, a nonlinear Fuzzy Logic (FL) based observer is implemented in the proposed system for the Fault Detection and Isolation (FDI) unit for nonlinear sensors of the AFR system. Lyapunov stability analysis was used for designing a stable system in both faulty and normal conditions. To evaluate its performance, this system was developed in the MATLAB/Simulink platform. The simulation results show that the developed system is robust under sensor fault conditions, retaining stability with a minimum decrease of AFR. This study's comparison with the existing literature demonstrates that the proposed system is effective for maintaining the AFR in IC engines during sensor faulty conditions thus reducing shutdown of engine and production loss for increased profits.

Modeling nonlinear high-pressure sensors based on degenerate four-wave mixing in photonic crystal fibers.

We propose a high-pressure sensing mechanism in solid-core silica photonic crystal fibers (PCF) that is based on the nonlinear optical process of degenerate four-wave mixing. A quite simple configuration for the pressure sensor is given and is theoretically investigated by obtaining signal gain spectra. We focus on the Stokes and anti-Stokes sidebands that are generated during propagation of the field along the PCF due to the interplay of dispersion and nonlinear optical effects. We have considered wave propagation in the normal dispersion regime with the inclusion of negative fourth-order dispersion. We have also considered a pumping field with wavelength of 1076 nm and peak power of 3000 W that propagates along a PCF with 40 cm of length. We have optimized sensitivity, varying lattice pitch and air-hole diameter, and we have obtained high sensitivities of the Stokes and anti-Stokes lines of 3.672 and -0.146nm/MPa, respectively. The proposed sensors have potential applicability in high-pressure environments.

An adaptive calibration technique for thermistor with varying temperature coefficient and reference resistance

Background: A thermistor is a nonlinear sensor requiring a precise calibration technique to achieve accurate temperature measurements. This paper attempts to design a calibration technique employing artificial neural network (ANN) algorithms. The present work fulfills the following objectives: (i) to cover 100% input range in the linearity range measurement; (ii) to make the measurement technique adaptive to variations in reference resistance and thermistor temperature coefficient using a calibration technique. Methods: An ANN-based calibration circuit is cascaded to the data conversion circuit. Optimized ANN is trained with linear data independent of reference resistance and temperature coefficient effects on thermistor output. ANN optimization is performed by comparing various schemes, algorithms, and numbers of hidden layers to achieve a minimum mean square error and a regression close to 1. Results: The proposed technique provides a linear relationship for the system over the entire input range and avoids the requirement of repeated calibrations each time the thermistor is replaced. Practical data are used to validate the proposed measurement technique. Conclusions: The objectives and proposed technique have been demonstrated by results with a root mean square percentage error of 1.8%.

A Compact Optical MEMS Pressure Sensor Based on Fabry-Pérot Interference.

Pressure sensors have important prospects in wind pressure monitoring of transmission line towers. Optical pressure sensors are more suitable for transmission line towers due to its anti-electromagnetic interference. However, the fiber pressure sensor is not a suitable choice due to expensive and bulky. In this paper, a compact optical Fabry–Pérot (FP) pressure sensor for wind pressure measurement was developed by MEMS technology. The pressure sensor consists of a MEMS sensing chip, a vertical-cavity surface-emitting laser (Vcsel), and a photodiode (PD). The sensing chip is combined with an FP cavity and a pressure sensing diaphragm which adopts the square film and is fabricated by Silicon on Insulator (SOI) wafer. To calibrate the pressure sensor, the experimental platform which consists of a digital pressure gauge, a pressure loading machine, a digital multimeter, and a laser driver was set up. The experimental results show that the sensitivity of the diaphragm is 117.5 nm/kPa. The measurement range and sensitivity of the pressure sensor are 0–700 Pa and 115 nA/kPa, respectively. The nonlinearity, repeatability, and hysteresis of the pressure sensor are 1.48%FS, 2.23%FS, and 1.59%FS, respectively, which lead to the pressure accuracy of 3.12%FS.

An adaptive gain based approach for event-triggered state estimation with unknown parameters and sensor nonlinearities over wireless sensor networks

The distributed state and parameter estimation problem is investigated in this paper for discrete-time nonlinear systems subject to sensor nonlinearities and stochastic disturbances over a wireless sensor network. A novel architecture for distributed state estimator is introduced that incorporates adaptive coupling gains to govern the information exchange between the sensor nodes under event-triggering mechanism. The aim of this paper is to provide a scalable structure for unknown parameter identification independent of sensor networks’ complexity. The boundedness of estimation error is ensured in the framework of uniformly ultimately bounded stability by developing an algebraic connectivity based criterion. The estimator gains including proposed coupling gains are then presented as solution to matrix inequalities. Finally, two simulation examples are presented to demonstrate the effectiveness of proposed estimation architecture.

Adaptive non-linear soft sensor for quality monitoring in refineries using Just-in-Time Learning—Generalized regression neural network approach

Real time estimation of target quality variables using soft sensor relevant to time varying process conditions will be a significant step forward in effective implementation of Industry 4.0. Generalized Regression neural network (GRNN) has been used as a steady state quality monitoring soft sensor with reasonable estimation accuracy. However, the accurate prediction capability of GRNN has rarely been explored in a time varying environment. This article reports design of adaptive soft sensor using GRNN as a local model in Just-in-Time learning (JITL-GRNN) framework. The JITL-GRNN adaptive soft sensing technique is further investigated in various dimensions such as, the effect of different similarity index criteria and relevant dataset size on model prediction accuracy and model computation time. Performance of the proposed JITL-GRNN soft sensor is investigated by assessing its prediction accuracy on two benchmark industrial datasets. In addition, dynamic Non-linear autoregressive with exogenous inputs (NARX) neural network model is also developed and the performance of NARX model was compared with the proposed JITL-GRNN model. Results show that the JITL-GRNN adaptive soft sensor has at par or better prediction capability than the NARX model and many other models reported in literature.

Novel L1 Regularized Extreme Learning Machine for Soft-Sensing of an Industrial Process

Extreme learning machine (ELM) is suitable for nonlinear soft sensor development. Yet it faces an overfitting problem. To overcome it, this work integrates bound optimization theory with variational Bayesian (VB) inference to derive novel L1 norm-based ELMs. An L1 term is attached to the squared sum cost of prediction errors to formulate an objective function. Considering the nonconvexity and nonsmoothness of the objective function, this article uses bound optimization theory, and constructs a proper surrogate function to equivalently convert a challenging L1 norm-based optimization problem into easy one. Then, VB inference is adopted for optimizing the converted problem. Thus, an L1 norm-based ELM can be efficiently optimized by an alternating optimization algorithm with a proved convergence. Finally, a soft sensor is developed based on the proposed algorithm. An industrial case study is carried out to demonstrate that the proposed soft sensor is competitive against recent ones.

Multi objective design optimization of graphene piezoresistive MEMS pressure sensor using design of experiment

This paper investigates the effect of diaphragm thickness, dimensions of piezoresistors, doping profile and temperature compatibility on sensitivity and non-linearity of graphene MEMS pressure sensor. Taguchi method is used for maximizing the sensitivity and minimizing the nonlinearity of the designed pressure sensor. L27 orthogonal array is utilized for five input factors with three levels. Output voltage is obtained from simulation in COMSOL for different combinations of the input parameters as per L27 orthogonal array. It was found that diaphragm thickness and length of the sensing element shows maximum contribution in increasing the sensitivity of the pressure sensor. Similarly, interaction of diaphragm thickness with piezoresistors thickness and doping concentration shows a major contribution in reducing the non-linearity of the pressure sensor. Other factors such as operating temperature affects both sensitivity and nonlinearity of the pressure sensor with a very low contributing percentage of 0.40% and 2.16%, respectively. Pareto Analysis of variance (ANOVA) was employed to validate the predicated results of the designed pressure sensor. The result indicated that the optimum design shows a sensitivity of 4.10 mV/psi with very low non linearity of 0.1%.

Deep Neural Network Based Linearization and Cold Junction Compensation of Thermocouple

In this proposed work, temperature sensors, namely, a thermocouple and thermistor were linearized using deep neural networks. The deep feedforward neural network (DFNN) technique was proposed to linearize the K-type thermocouple’s output, in the given temperature range −100 °C to 1372 °C, while nonlinearity was reduced from 2.03% to 0.002% full scale span (FSS). Deep layer recurrent neural network (DLR-NN) was used to reduce the nonlinearity of a negative temperature coefficient (NTC) thermistor from 84.63% to 0.13% FSS. The linearized thermistor was used for cold junction compensation (CJC) of the thermocouple. In both the thermocouple and thermistor, linearization was achieved in a single stage for a wide range digitally using deep neural networks alone. There were no analog pre-signal conditioning circuits, unlike the existing neural network-based linearization techniques in literature. A hardware setup of a stand-alone module for linearization was designed using the Raspberry pi microcontroller consisting of two soft modules, one for thermocouple linearization and the other for thermistor linearization. The proposed system was experimentally tested using a K-type thermocouple on a thermal calibrator in a 0 °C–300 °C range. The cold junction compensated output of the thermocouple had a maximum absolute error of 0.34 °C when ambient temperature varied from 0 °C to 40 °C. The results were satisfactory and better than the existing National Institute of Standards and Technology (NIST) standard. This linearization technique can be extended to other thermocouple types as well as other nonlinear sensors.

Remaining Useful Life Interval Prediction for Complex System Based on BiGRU Optimized by Log-Norm

The task of remaining useful life (RUL) uncertainty management is the major challenge in solving the failure of the complex mechanical system. Primary research methods use statistical models or stochastic processes to fit the distribution of historical degradation data. However, it is difficult to accurately capture the degradation information of monitoring big data through statistics in practice. In this paper, the prediction interval (PI) obtained by the proposed feature attention-log-norm bidirectional gated recurrent unit (FA-LBiGRU) model is adopted to quantify the prediction uncertainty of RUL. Initially, the critical feature vectors are extracted from multi-dimensional, nonlinear, and large-scale sensor signals using the feature attention mechanism. Additionally, the BiGRU network is used to model and learn the time-varying characteristics of the attention-weighted features from the forward and backward directions, and the network parameters are trained by the maximum log-likelihood loss function. Ultimately, the probability density function based on the lognormal distribution is calculated to measure the uncertainty of the equipment RUL. The effectiveness of the proposed method is verified through the well-known benchmark data set of the turbofan engines provided by NASA. The experimental results show that the proposed methods can obtain higher point prediction accuracy for the complex system compared with state-of-the-art approaches and high-quality PIs satisfying real-time requirements.