Actual Sensor Research Articles

Artificial neural network-based ground reaction force estimation and learning for dynamic-legged robot systems.

Legged robots have become popular in recent years due to their ability to locomote on rough terrains; these robots are able to walk on narrow stepping-stones, go upstairs, and explore soft ground such as sand. Ground reaction force (GRF) is the force exerted on the body by the ground when they are in contact. This is a key element and is widely used for programming the locomotion of the legged robots. Being capable of estimating the GRF is advantageous over measuring it with the actual sensor system. Estimating allows one to simplify the system, and it is meant to be capable of prediction, and so on. In this article, we present a neural network approach for GRF estimation for the legged robot system. In order to fundamentally study the GRF estimation of the robot leg, we demonstrate our approach for a single-legged robot with a degree of freedom (DoF) of two with hip and knee joints on a flat-surface. The first joint is directly driven from the actuator, and another joint is belt-pulley driven from the second actuator to take advantage of the long range of motion. The neural network is designed to estimate GRF without attaching force sensors such as load cells, and the encoder is the only sensor used for the estimation. We propose a two-staged multi-layer perceptron (MLP) solution based on supervised learning to estimate GRF in the physical-world. The first stage of the MLP model is trained using datasets from the simulation, enabling it to estimate the simulation-staged GRF. The second stage of the MLP model is trained in the physical world using the simulation-staged GRF obtained from the first stage MLP as the input. This approach enables the second stage MLP to bridge the simulation to the physical world. The root mean squared error (RMSE) is 0.9949 N on the validation datasets in the best case. The performance of the trained network is evaluated when the robot follows trajectories that are not used in training the two-stage GRF estimation network.

A symmetrical dual-mass block fiber Bragg grating vibration sensor based on a V-shaped flexible hinge

Real-time monitoring of vibration properties in micro-seismic applications, such as petroleum exploration, physical oceanography, and perimeter security, is important for deep exploration and the prediction of geological disasters and other fields. The fiber Bragg grating (FBG) vibration sensor with symmetrical V-shaped flexible hinges and a dual mass block is proposed. Utilizing formula derivation and simulation software, the sensor's structural parameters are optimized and subjected to analysis. These simulations inform the fabrication of the actual sensor, and subsequently, the sensor's performance is empirically tested. The results show an average sensitivity of 73.86 pm/G and a natural frequency of 800 Hz, with a working range of 20–340 Hz. Its exhibits anti-interference in the transverse direction is less than 5 % during vibration measurement. The research outcomes offer a point of guidance for the progress of FBG accelerometer development within applications that involve measuring low- to medium-frequency accelerations.

A Virtual System and Method for Autonomous Navigation Performance Testing of Unmanned Surface Vehicles

An overall framework of the virtual testing system has been established based on the analysis of the virtual testing requirements for autonomous navigation performance of unmanned surface vehicles (USVs). This system consists of several modules, including the environment module, motion module, sensor module, and 3D visualization module. Firstly, within the robot operating system (ROS) environment, a three-dimensional navigation environment was generated by combining actual wave spectra with Gerstner waves. By designing a power plugin for USV navigation, the system was made to reflects the coupled motion model of USVs in wind, waves and currents, along with predictive results. Regarding the four typical sensor information on USVs, the actual sensors were virtualized, and a simulation approach for virtual sensor information is provided. The three-dimensional visualization of USV’s motion enables the intuitive display and analysis of the virtual testing process. Based on the prediction of coupled motion characteristics in wind, waves and currents, the interaction between USVs and the virtual testing system has been realized. A platform for virtual testing experiments to determine the autonomous navigation performance of USVs was established, and the effectiveness of the platform was verified in terms of perception and environmental interference. In virtual environmental interference validation, the average amplitude deviation of the heave motion of USVs under sea state 3 reaches 0.74 m, and the average amplitude deviation of the pitch motion reaches 0.25 rad, showing the gradually increasing disturbance of the sea state. Finally, virtual testing experiments were conducted on a specific USV to evaluate its autonomous navigation perception performance, trajectory tracking performance, and autonomous obstacle avoidance. The evaluation results indicate that the platform can achieve the functionality of virtual testing for the autonomous navigation performance of USVs from the perspective of cost function, taking the reaction distance, regression distance, and obstacle avoidance time into consideration. A representative example is that the cost function deviation rates of overtaking obstacle avoidance between static and dynamic seas reach 5.11%, 8.98% and 18.43%, respectively. The gradually increasing data shows that the virtual simulating method matches the drifting-off-course tendency of boats in rough seas. This includes acquiring perception information of navigation and simulating the motion and navigation processes for visualization. The platform provides new means for testing and evaluating the autonomous navigation performance of USVs.

Optimal sensor placement for permanent magnet synchronous motor condition monitoring using a digital twin-assisted fault diagnosis approach

Efficient health monitoring for identifying and quantifying damages can substantially improve the performance and structural integrity of engineered systems. Specifically, new advances in sensing technologies have pushed the research of large sensor networks to monitor complex mechanical structures. Given the need for health state monitoring, designing an optimal sensor framework with accurate detectability of failure modes has great significance. However, there is often little to no experimental data available for newly proposed mechanical systems; so a digital-twin method would make fault detection feasible for this applications. In this paper, a data-driven reliability-based design optimization (RBDO) approach is employed for sensor placement and fault detection of a permanent magnet synchronous motor (PMSM), which is a relatively new system for high power engineering applications. This system suffers from inter-turn and inter-phase short-winding faults, which can cause catastrophic failure of the whole structure. For PMSMs, current sensing and magnetic field sensing can be utilized for the detection of faults, but actual sensor placement has not been considered in recent literature. In this study, the first step is to create an FEA model of the PMSM for the simulation of faults, which serves as the digital twin. Next, a data-driven approach is implemented for sensor placement and classification of faults. The proposed method utilizes distance clustering for identification of various failure modes, which is suitable for many applications due to its high accuracy and computational efficiency. In addition, a genetic algorithm is implemented to determine the minimum number and optimal placement of sensors. This framework simultaneously searches for the optimal placement of sensors while training the classifier for detectability of system health states. Ultimately, the proposed methodology shows convergence to a solution with high accuracy for detection of faults, and is demonstrated on the novel system of a PMSM with magnetic field sensors.

Self-Attention-Based Deep Convolution LSTM Framework for Sensor-Based Badminton Activity Recognition.

Sensor-based human activity recognition aims to classify human activities or behaviors according to the data from wearable or embedded sensors, leading to a new direction in the field of Artificial Intelligence. When the activities become high-level and sophisticated, such as in the multiple technical skills of playing badminton, it is usually a challenging task due to the difficulty of feature extraction from the sensor data. As a kind of end-to-end approach, deep neural networks have the capacity of automatic feature learning and extracting. However, most current studies on sensor-based badminton activity recognition adopt CNN-based architectures, which lack the ability of capturing temporal information and global signal comprehension. To overcome these shortcomings, we propose a deep learning framework which combines the convolutional layers, LSTM structure, and self-attention mechanism together. Specifically, this framework can automatically extract the local features of the sensor signals in time domain, take the LSTM structure for processing the badminton activity data, and focus attention on the information that is essential to the badminton activity recognition task. It is demonstrated by the experimental results on an actual badminton single sensor dataset that our proposed framework has obtained a badminton activity recognition (37 classes) accuracy of 97.83%, which outperforms the existing methods, and also has the advantages of lower training time and faster convergence.

Using machine learning for cutting tool condition monitoring and prediction during machining of tungsten

ABSTRACT Machining of single-phase tungsten, used as a plasma facing material in fusion energy reactors, is commonly associated with rapid tool wear and short tool life. Conventional methods of monitoring tool wear or changing cutting tools after a predetermined period are inefficient and can lead to unnecessary tool change or risk damaging the workpiece. Tool wear can adversely affect the surface finish and dimensional tolerances of machined parts. Predicting its onset can avoid this critical damage whilst ensuring maximum tool life is utilised. In this paper, firstly the tool life results in end milling single-phase tungsten using different cutting tool geometries and cutting speeds are provided for the first time. A novel method is proposed by combining sensor signal prediction and classification machine learning models. It works by forecasting the cutting tool bending moment signal which is then used for predicting future cutting tool condition in end milling of pure dense tungsten. A series of machining experiments, covering the whole life of a cutting tool, were performed to collect the sensor signals. The current time series signal from the sensory tool holder is employed to forecast the future signal by training a 1D convolutional neural network (1D CNN) and an artificial neural network (ANN). The forecasted signal is then used to predict the state of the cutting tool in the future. Machine learning classifiers namely, random forest (RF), support vector machine (SVM) and extreme gradient boosting (XGBoost) supervised learning models were trained and validated on actual sensor signals to correlate the tool conditions with specific sensor signal features. The investigations revealed that the 1D CNN performed best in forecasting the time series sensor signal whilst achieving a mean absolute error of 3.37. In addition, the RF, when trained on Wavelet Scattering features, resulted in the most accurate classification of sensor signals for tool condition detection. The analysis showed that the combination of 1D CNN signal forecasting, feature extraction through statistical analyses and RF classifier performs best in predicting the state of a cutting tool in near future. Using this method allows for decision making for changing the tool whilst ensuring that the maximum useful life of a cutting tool is utilised. It also enables preventing undesired damage to the machined surface due to late detection of tool wear or delays in taking appropriate actions. The application of this method can reliably reduce the manufacturing costs and resource consumption associated with cutting tools for machining tungsten and minimise tool wear induced damage to the workpiece.

Realistic sensor simulations for the digital twin

Digital twins use actual sensor data to replicate the current state of a plant in a virtual model. They can be used to evaluate the current state, predict future behavior, and thus allow to refine control or optimize operation, enable predictive maintenance as well as detection of anomalies and failures.The model of a digital twin includes models of the components, behaviors and dynamics of a system. With the ability to simulate real scenarios, such models can therefore also be used before a plant is actually implemented, e.g., to predict the actual performance, identify potential issues for the implementation and to develop optimal operation strategy and algorithms. Furthermore, interfaces may be defined, implemented, and tested with such models allowing fast and easy commissioning of the physical implementation.Accurate digital twins therefore also need to include realistic sensor models, considering adverse effects that impact their output signals. The proposed work presents approaches for accurate sensor simulations allowing researchers and industries to assess sensor performance, optimize algorithms, and evaluate system-level integration. We address Frequency Modulated Continuous Wave (FMCW) radar sensors and time-of-flight cameras as examples for far-field sensors and capacitive sensors as an example for near-field sensors. The approaches can be transferred to other sensors, e.g., ultrasound sensors, LiDAR sensors and inductive or magnetic sensors so that a wide range of industrial sensors can be covered.The proposed simulations are benchmarked with different tests, including real-world experiments and compared with the corresponding real sensors.

Optimal Deployment for Hybrid Sensor Networks Based on Efficient Node Configuration

Hybrid sensor networks, which contain mobile nodes and stationary nodes, are being used more and more widely. The second deployment of mobile nodes is a key problem to be solved, and the deployment performance of the network directly affects the monitoring effect of the network. Optimizing the configuration ratio of the two nodes can effectively reduce the network cost. In this paper, under the premise of knowing the coverage of the required monitoring area, the impact of sensor devices on node configuration is studied through parameter analysis, and the number and types of sensors that should be deployed in the hybrid sensor network are deduced, which can be conveniently and accurately used to design the actual hybrid sensor network. At the same time, for the secondary deployment of mobile nodes, this paper proposes a new mobile coverage method BS-CCP (box search and concentric circle positioning) to improve the coverage of the hybrid sensor network and maximize the coverage of the target area with the specified sensor types and numbers. Compared with existing work, the method in this paper reduces the number of iterations and reduces the number of required nodes. Comparing BS-CCP with the existing network mobile coverage algorithm, the experimental results show that the coverage obtained by this method is larger and more efficient.

Simulation of Laser Profilometer Measurements in the Presence of Speckle Using Perlin Noise.

In the manufacturing industry, inspection systems play a crucial role in ensuring product quality. High-resolution profilometric sensors have become increasingly popular for inspection due to their ability to provide detailed surface information. However, the development and testing of inspection systems can be costly and time-consuming. This paper presents the development of a simulation of an inspection system using a high-resolution profilometric sensor. A geometrical and noise model is proposed to simulate the readings of any actual profilometric sensor. The model replicates the sensor's movement on the CAD model of the inspected part. The model incorporates the physical properties of the sensor and combines noise sources from sensor uncertainty and speckle noise induced by the roughness of the material. Our contribution lies in noise modeling. This work proposes a combination of Perlin noise to simulate the speckle noise and Gaussian noise for the uncertainty-related noise. Perlin noise is generated based on the surface roughness parameters of the inspected part. The accuracy of the simulation system is evaluated by comparing the simulated scans with real scans. The results highlight the ability to simulate real scans of different parts, using commercial sensor specifications and the CAD model of the inspected part.

Instrumentation of Field-Testing Sites for Dynamic Characterization of the Temperature-Dependent Stiffness of Pavements and Their Layers

Falling weight deflectometer (FWD) tests are performed worldwide for assessing the health of pavement structures. Interpretation of FWD-measured surface deflections turns out to be challenging because the behavior of pavement structures is temperature-dependent. In order to investigate the influence of temperature on the overall pavement performance and on the stiffness of individual layers, temperature sensors, asphalt strain gauges, and accelerometers were installed into one rigid (concrete) and two flexible (asphalt) pavement structures, mostly at layer interfaces. Three different methods for installation of the strain gauges are compared. From correspondingly gained experience, it is recommended to install a steel dummy as a place-holder into the surface of hot asphalt layers, immediately after their construction and right before their compaction, and to replace the dummy with the actual sensor right before the installation of the next layer. Concerning the first data obtained from dynamic testing at the field-testing sites, FWD tests performed at different temperatures deliver, as expected, different surface deflections. As for the rigid pavement, sledgehammer strokes onto a metal plate, transmitted to the pavement via a rubber pad, yield accelerometer readings that allow for detection of curling (=temperature-gradient-induced partial loss of contact of the concrete slab from lower layers). In the absence of curling, the here-proposed sledgehammer tests yield accelerometer readings that allow for quantification of the runtime of longitudinal waves through asphalt, cement-stabilized, and unbound layers, such that their stiffness can be quantified using the theory of elastic wave propagation through isotropic media.

Laser frequency noise correction in LFM-based interferometric fiber-optic hydrophone array.

In this paper, we propose a novel time-division multiplexed (TDM) array for a large-scale interferometric fiber-optic hydrophone system, in which we introduce a power-optimized reference probe and effectively reduce the additional white noise while correcting for light source frequency noise. Laser frequency noise usually introduces appreciable phase noise during demodulation of interferometric fiber-optic hydrophones. In the previous means, one would introduce an additional probe isolated from the environment in sensor array, and use it as a reference to calibrate the demodulation results of the other actual sensors. However, while correcting, the reference probe also introduces a large white noise. In our array, the echo of the reference probe is higher than the other sensors, thus solving this problem. The novel array design is applied to our previously proposed fiber-optic hydrophone based on a linear frequency modulated (LFM) light source. Experiments show that the deterioration of phase noise floor caused by additional white noise is improved from at least 3 dB originally to within 1 dB. This paper analyzes the factors that need to be concerned for the successful implementation of correction algorithms in hydrophone systems based on LFM sources. Particular focus is given to the impact of the power optimization of reference probe on the white noise and the corrected phase noise. Our proposal allows a significant relaxation of the demanding linewidth requirement for interferometric hydrophone. It is shown that laser with linewidth of 338.06 MHz can replace that with 1.417 kHz in our new system, while achieves the same demodulation noise floor.

Vision-based virtual vibration sensor using error calibration convolutional neural network with signal augmentation

In this study, we propose a methodology for using image-based virtual vibration sensors to replace acceleration sensors in the vibration monitoring of targets. Most image-based vibration-monitoring techniques typically involve the installation of a non-contact-capture camera at a location unaffected by disturbances to photograph the outside of a target. However, these methods have several drawbacks, including the additional cost of installing sensors and the need to select appropriate installation locations. The proposed virtual vibration sensor addresses these limitations by combining signal extraction based on a contact-capture camera installed inside the target with a convolutional neural network for error calibration. By employing a camera using the contact-capture method, the installed cameras can be reused to integrate the functions of other sensors and reduce the total number of sensors required. Moreover, in the proposed method, a calibration convolutional neural network (CCNN) improves the limited accuracy of contact-capture cameras. Furthermore, the calibration performance of the CCNN was improved via signal augmentation even with the availability of limited data. The experimental results obtained using a quadcopter platform demonstrated the improved accuracy and effectiveness of the proposed virtual sensor compared to that of the actual sensors.

A digital twin architecture for real-time and offline high granularity analysis in smart buildings

Smart buildings aim to create a safe, comfortable and sustainable environment for the occupants while increasing the energy performance of the building to reduce the environmental footprint and operational cost. Towards this direction, the use of model-based digital twins able to simulate key aspects of the building is important. Despite the utilization of various digital twin approaches for smart buildings in recent years, most of these approaches mainly focus on one aspect of the building, such as indoor environmental quality or energy consumption. This observation necessitates the development of a virtual but realistic and non-invasive environment able to simulate all the key domains of the building, including indoor air quality, thermal comfort, and energy consumption. Towards this direction, this work proposes an architecture for smart buildings investigations, that emulates the thermal comfort, air-quality, and electricity consumption of the building, while considering the activities and behavior of occupants. For simulating the building operations, ordinary differential equations and power balance modeling are incorporated to facilitate real-time or faster than real-time digital twin approaches, while computational fluid dynamics modeling is used complementarily, to enable higher granularity simulations for offline analysis. Both models have been validated in a real building achieving less than 1% average zone temperature error between the simulation models and the actual building sensor measurements. Furthermore, a user-friendly software platform has been developed and integrated with the proposed architecture, where multiple important features and characteristics have been included to enable various investigations. In particular, the developed platform enables physical and simulated building integration, model calibration using field measurements, experimental configuration and management, execution of novel what-if investigations, and evaluation of novel algorithms using hardware-in-the-loop configurations. The proposed architecture is applied in a physical building, allowing the demonstration of several investigations and the execution of interesting use-cases.

Актуальные проблемы сенсоров и аппаратного обеспечения коммерческих беспилотных авиационных систем

Использование беспилотных транспортных средств, особенно беспилотных авиационных систем, для решения широкого спектра задач коммерческих и государственных организаций в последние годы интенсифицировалось. Преднамеренное или непреднамеренное искажение данных БАС потенциально может привести к опасным инцидентам техногенного характера с угрозой жизни и здоровью людей. Соответственно, возрастает потребность в обеспечении кибербезопасности таких систем в целом и защиты данных, необходимых для их корректного функционирования, в частности. Традиционные методы обеспечения безопасности не всегда подходят для сенсоров и аппаратного обеспечения систем беспилотного транспорта из-за целого ряда ограничений и особенностей функционирования. Настоящая статья открывает цикл публикаций, посвященных информационной безопасности БАС. Целью настоящей статьи является анализ векторов атак на сенсоры и их каналы, а также изучение аппаратного обеспечения БАС. Проведен обзор методов защиты от подобных атак. В работе применялись методы контент-анализа и эксперименты с коммерческими беспилотными авиационными системами, доступными для гражданских пользователей. В результате был разработан систематизированный перечень атак на сенсоры и аппаратное обеспечение, оформленное в виде таблиц в данной статье. The use of unmanned vehicles, especially unmanned aerial systems (UAS) for a wide range of tasks of commercial and government organizations has intensified in recent years. Intentional or unintentional distortion of UAS data can potentially lead to dangerous man-made incidents with a threat to human life and health. Accordingly, there is an increasing need for ensuring the cybersecurity of such systems in general and protecting the data necessary for their correct functioning, in particular. Traditional security methods are not always suitable for sensors and hardware of unmanned vehicle systems due to a number of limitations and features of operation. With this article, the authors open a series of articles on UAS information security. The purpose of this article is to analyse attack vectors on sensors and their channels, as well as UAS hardware. In addition, the authors reviewed methods of protection against such attacks. In their work, the authors applied content analysis methods and experiments with real commercial unmanned aerial systems available to civilian users. As a result, the authors have developed a systematic list of attacks on sensors and hardware, presented in the form of tables in this article.

Optimized Dissolved Oxygen Prediction Using Genetic Algorithm and Bagging Ensemble Learning for Smart Fish Farm

The field of aquaculture is one of the numerous scientific disciplines that benefit greatly from machine learning (ML). The amount of dissolved oxygen (DO), an important indicator of water quality in sustainable fish farming, affects the yield of aquatic production. It is essential to make DO projections in fishing ponds to carry out the process of artificial aeration. We present DO forecasts utilizing time series analysis based on data obtained from Hanwha Aqua Planet Jeju, located in South Korea. This information could form the basis of a data foundation for an early detection system and improved aquaculture farm management. This research presents a unique genetic algorithm called GA-XGCBXT bagging ensemble model based on genetic algorithms. This model is built on XGBoost, CatBoost, and Extra Trees. In order to select the most outstanding features, various methodologies that exhibit a strong association with the primary data were applied. The performance of the proposed model was evaluated by comparing it to actual sensor data that had been observed, both in the training and validation sets. The precise evaluation accuracy of the anticipated results of the recommended GA-XGCBXT model was determined using various performance indices. By utilizing the strategy we suggested, we acquired a root mean square error of 0.310. Our objective is to enhance the machine learning model for aquaculture so that academics and practitioners can employ applications for smart fish farming with complete reliability.

New Array Designs for DoA Estimation of Non-Circular Signals With Reduced Mutual Coupling

Recently, many sparse arrays are proposed to achieve increased uniform degrees of freedom (uDoFs) for the direction of arrival (DoA) estimation of non-circular sources. In light of this recent attention, the high number of uDoFs is the primary motivation for designing those sparse array structures. Although both the number of uDoF and mutual coupling can influence the DoA estimation performance. However, mutual coupling is not considered in many of those sparse array configurations. In this paper, two optimized shifted coprime arrays with sum-difference coarrays, called OSCAwSDCa-I and OSCAwSDCa-II, are proposed to reduce mutual coupling. The OSCAwSDCa-I and OSCAwSDCa-II are obtained by rearranging actual sensors of the augmented coprime array. The proposed array designs have closed-form expressions for their sensor locations and achievable uDoFs. Additionally, it is proven that the first three weights of OSCAwSDCa-II are independent of the array size. Moreover, the number of virtual sensors co-locating at the same lag is reduced. Consequently, smaller weight values are obtained, resulting in more robustness against mutual coupling. Finally, extensive simulations show that OSCAwSDCa-I and OSCAwSDCa-II can achieve comparable uDoFs concerning the existing array structures. In the presence of strong mutual coupling, simulation results demonstrate the robustness of OSCAwSDCa-II more than other compared configurations.

Highly selective capacitive micromachined ultrasonic transducer–based miniature gravimetric CO2 sensor with in-situ calibration for relative humidity

The prototype gravimetric microelectromechanical imine functionalized CO2 sensor equipped with custom electronics and significantly increased stability under sub-percent concentrations is investigated. Instabilities caused by the relative air humidity were eliminated by two in-situ calibration methods: two-channel readings correction and single-channel correction by the reference disturbance. Capacitive micromachined ultrasound transducers (CMUT) were designed for 15 MHz nominal resonance frequency, while actual sensor readings were lower (over 12 MHz) due to the functionalization. The resonance frequency of polyethyleneimine (PEI) functionalized CMUT during sub-percent CO2 detection experienced instabilities of several hundred Hz while the relative humidity of the air was changed between 10% and 35%. Readings of graphene oxide (GOx) functionalized CMUT working in parallel with PEI functionalized CMUT were used for two-channel correction, resulting in highly stable sensor output, characterized by 3.2 Hz standard deviation during almost one hour experiment, while relative humidity varied between 10% and 25%. Single channel reading correction method resulted in larger residual instability, characterized by a standard deviation of up to 36 Hz over a two-hour experiment, however, it was found to be more effective for longer humidity-induced drifts, since reference disturbance of known relative humidity of 75% maintained by saturated NaCl solution technique was used.

Chiral Recognition by Supramolecular Porphyrin–Hemicucurbit[8]uril-Functionalized Gravimetric Sensors

Enantiorecognition of a chiral analyte usually requires the ability to respond with high specificity to one of the two enantiomers of a chiral compound. However, in most cases, chiral sensors have chemical sensitivity toward both enantiomers, showing differences only in the intensity of responses. Furthermore, specific chiral receptors are obtained with high synthetic efforts and have limited structural versatility. These facts hinder the implementation of chiral sensors in many potential applications. Here, we utilize the presence of both enantiomers of each receptor to introduce a novel normalization that allows the enantio-recognition of compounds even when single sensors are not specific for one enantiomer of a target analyte. For this purpose, a novel protocol that permits the fabrication of a large set of enantiomeric receptor pairs with low synthetic efforts by combining metalloporphyrins with (R,R)- and (S,S)-cyclohexanohemicucurbit[8]uril is developed. The potentialities of this approach are investigated by an array of four pairs of enantiomeric sensors fabricated using quartz microbalances since gravimetric sensors are intrinsically non-selective toward the mechanism of interaction of analytes and receptors. Albeit the weak enantioselectivity of single sensors toward limonene and 1-phenylethylamine, the normalization allows the correct identification of these enantiomers in the vapor phase indifferent to their concentration. Remarkably, the achiral metalloporphyrin choice influences the enantioselective properties, opening the way to easily obtain a large library of chiral receptors that can be implemented in actual sensor arrays. These enantioselective electronic noses and tongues may have a potential striking impact in many medical, agrochemical, and environmental fields.

Capturing the effect of biofouling on ships by incremental machine learning

Performance data from ships is subject to distributional shifts, sometimes referred to as concept drift. In this study, synthetic monitoring data is simulated for the KVLCC2, considering publicly available reference data and a semi-empirical simulation framework. Neural networks are trained to predict the required shaft power and to overcome the deterioration in model accuracy due to concept drift, three methods of incremental learning are applied and compared: (1) Layer freezing, (2) L2 regularization, and (3) elastic weight consolidation. Furthermore, an implicit methodology for quantifying the changing hull and propeller performance is presented. In addition, a generic feature engineering framework is used for eliminating insignificant features. In two investigations, sudden and incremental concept drift scenarios are examined, and the effect of different uncertainty categories on model performance is studied in parallel based on three different datasets. As a main finding, it is confirmed that data quality is of great importance for accurate machine learning-driven performance monitoring — even in simulated environments. Furthermore, the study shows that freezing layers during incremental learning proves to be most robust and accurate, but it will be part of future work to examine this on actual sensor data.

Adaptive cascade enhancement broad learning system combined with stacked correlation information autoencoder for soft sensor modeling of industrial process

In the actual soft sensor project, the complex industrial process leads to a large number of monitoring variables that lead to obvious problems of high dimension and data redundancy. To solve these problems, an adaptive cascade enhancement broad learning system combined with stacked correlation information autoencoder, referred to as SCIAE-ACEBLS, is proposed in this study. The latter is based on the broad learning system (BLS) and it includes two parts: feature node and enhancement node. In the feature node part, the correlation coefficient and the dominant variable are introduced into the stacked autoencoder (SAE), and all the hidden nodes are used as feature nodes. Both the correlation coefficient and the dominant variable are introduced to ensure that the information of feature nodes not only include the relevant information of the dominant variable in the auxiliary variables, but also in the dominant variable, and thus all the information related to dominant variables can be effectively used. In the enhancement node part, an adaptive cascade enhancement node algorithm is used to reduce the redundancy of effective information and solve the problems of node information redundancy and node number uncertainty caused by the randomness of the node parameters in BLS. Finally, two industrial examples show that the proposed model is effective and outperforms existing methods.