Soft Sensor Research Articles

An Automatic Implementation of Oropharyngeal Swab Sampling for Diagnosing Respiratory Infectious Diseases via Soft Robotic End-Effectors

The most widely adopted method for diagnosing respiratory infectious diseases is to conduct polymerase chain reaction (PCR) assays on patients’ respiratory specimens, which are collected through either nasal or oropharyngeal swabs. The manual swab sampling process poses a high risk to the examiner and may cause false-negative results owing to improper sampling. In this paper, we propose a pneumatically actuated soft end-effector specifically designed to achieve all of the tasks involved in swab sampling. The soft end-effector utilizes circumferential instability to ensure grasping stability, and exhibits several key properties, including high load-to-weight ratio, error tolerance, and variable swab-tip stiffness, leading to successful automatic robotic oropharyngeal swab sampling, from loosening and tightening the transport medium tube cap, holding the swab, and conducting sampling, to snapping off the swab tail and sterilizing itself. Using an industrial collaborative robotic arm, we integrated the soft end-effector, force sensor, camera, lights, and remote-control stick, and developed a robotic oropharyngeal swab sampling system. Using this swab sampling system, we conducted oropharyngeal swab-sampling tests on 20 volunteers. Our Digital PCR assay results (RNase P RNA gene absolute copy numbers for the samples) revealed that our system successfully collected sufficient numbers of cells from the pharyngeal wall for respiratory disease diagnosis. In summary, we have developed a pharyngeal swab-sampling system based on an “enveloping” soft actuator, studied the sampling process, and implemented whole-process robotic oropharyngeal swab-sampling.

Dielectric elastomers with large actuation strain and high output force

AbstractDielectric elastomer actuator (DEA) is considered as an ideal artificial muscle with soft and light in nature, widespread application in soft robotics and sensors. Dielectric elastomer (DE), the core component of DEA, with high output force and large actuation strain under relatively low driving electric field is critical for high‐performance DEA. However, the widely used VHB series (3M) DE suffered from very large pre‐strain and high driving voltage to get their high actuation performance, limiting their scope of application. In this work, the DEs with good mechanical and dielectric properties were prepared via UV curing method consisting of acrylic acid (AA) with strong polar groups as functional monomer, n‐butyl acrylate (BA) as soft monomer and urethane acrylate as cross‐linker (c‐p(BA‐AA)). The c‐p(BA‐AA) DE films obtained the simultaneous increasing of dielectric constant and breakdown strength with the increasing content of AA, which guaranteed the good actuation performance. The c‐p(BA‐AA) film exhibited excellent actuating performance (44.11% actuation strain at 45 kV/mm) without pre‐strain. Besides, a simple linear actuator based on the c‐p(BA‐AA) film could easily output weight that is tens of times of its own active weight (654.1 mN/g at 30 kV/mm).

Structure-Property Relationships of Granular Hybrid Hydrogels Formed through Polyelectrolyte Complexation.

Hybrid hydrogels are hydrogels that exhibit heterogeneity in the network architecture by means of chemical composition and/or microstructure. The different types of interactions, together with structural heterogeneity, which can be created on different length scales, determine the mechanical properties of the final material to a large extent. In this work, the microstructure-mechanical property relationships for a hybrid hydrogel that contains both electrostatic and covalent interactions are investigated. The hybrid hydrogel is composed of a microphase-separated polyelectrolyte complex network (PEC) made of poly(4-styrenesulfonate) and poly(diallyldimethylammonium chloride) within a soft and elastic polyacrylamide hydrogel network. The system exhibits a granular structure, which is attributed to the liquid-liquid phase separation into complex coacervate droplets induced by the polymerization and the subsequent crowding effect of the polyacrylamide chains. The coacervate droplets are further hardened into PEC granules upon desalting the hydrogel. The structure formation is confirmed by a combination of electron microscopic imaging and molecular dynamics simulations. The interpenetration of both networks is shown to enhance the toughness of the resulting hydrogels due to the dissipative behavior of the PEC through the rupture of electrostatic interactions. Upon cyclic loading-unloading, the hydrogels show recovery of up to 80% of their original dissipative behavior in less than 300 s of rest with limited plasticity. The granular architecture and the tough and self-recoverable properties of the designed hybrid networks make them good candidates for applications, such as shape-memory materials, actuators, biological tissue mimics, and elastic substrates for soft sensors.

An Algorithm for Soft Sensor Development for a Class of Processes with Distinct Operating Conditions.

Soft sensors are increasingly being used to provide important information about production processes that is otherwise only available through off-line laboratory analysis. However, usually, they are developed for a specific application, for which thorough process analysis is performed to provide information for the appropriate selection of model type and model structure. Wide industrial application of soft sensors, however, requires a method for soft sensor development that has a high level of automatism and is applicable to a significant number of industrial processes. A class of processes that is very common in the industry are processes with distinct operating conditions. In this paper, an algorithm that is suitable for the development of soft sensors for this class of processes is presented. The algorithm possesses a high level of automatism, as it requires minimal user engagement regarding the structure of the model, which makes it suitable for implementation as a customary industrial solution. The algorithm is based on a radial basis function artificial neural network, and it enables the automatic selection of the model structure and the determination of model parameters, only based on the training data set. The testing of the presented algorithm is done on the cement production process, since it represents a process with distinct operating conditions. The results of the test show that, besides providing a high level of automatism in model development, the presented algorithm generates a soft sensor with high estimation performance.

Stretchable and recyclable In3SnBi/PU nanofiber composite as flexible sensors for human motion monitoring

Abstract Low-melting-point metal alloys (LMMAs) have great potential for application in soft sensors. However, the high cost limits the application of gallium-based liquid metals. In this paper, LMMA In3SnBi has been designed and prepared through a facile melt-mixing process followed by an ultrasonic dispersion technique. To overcome the fragility of low-melting-point alloys, flexible electrospun polyurethane (PU) matrix is used as structural support. When used as a sensor, the composite film exhibits a response time of 54.37 ms within a strain range of 15%. In addition, the composite film exhibits self-healing properties, and cracks generated by cyclic stretching can be repaired through a hot-pressing process. The current work presents a versatile fabrication strategy for LMMA-based flexible sensors, which holds great promise in the field of wearable electronic devices.

Photo- and Humidity-Responsive Liquid Crystal Copolymer Actuators Fabricated via Vapor-Assisted Alignment.

Deformable liquid crystal polymers (LCPs) driven by more than one external stimulus have received extensive attention in fields ranging from multifunctional soft robots to bionic actuators. Combining responsive liquid crystal with nonmesogenic responsive groups within polymer offers a versatile way to obtain multiresponsive LCPs. However, the incorporation of nonmesogenic responsive groups causes interruption in the assembly of mesogens and brings a challenge to the alignment of LCPs. Herein, a new method is put forward to facilitate uniform mesogen alignment by exerting water vapor in the film preparation process. Using this method, vapor-assisted alignment, the homeotropic alignment of azobenzene mesogens is achieved in a copolymer containing nonmesogenic poly(ethylene glycol) (PEG). The obtained copolymer films present photodeformation brought by azobenzene isomerization and humidity-responsive deformation resulting from the asymmetric swelling of film surfaces. The dual-responsive smart "blinds" and bionic flower actuators are fabricated to demonstrate the integration of the two different stimuli. This work is anticipated to provide a feasible alignment method for multiresponsive LCPs, showing the potential applications in soft robots, sensors, and biomimetic devices.

Novel method for temperature prediction in rotary kiln process through machine learning and CFD

Blast furnace high-proportion pellet smelting stands as a pivotal pathway towards achieving low-carbon and environmentally friendly ironmaking. However, challenges in pellet ore production, particularly through the rotary kiln method, have surfaced due to the scarcity of iron ore resources, leading to issues like “ringing” that hamper production efficiency and energy conservation. Effectively predicting and optimally controlling the temperature field within rotary kilns emerges as a crucial strategy to address this pressing challenge. In a pioneering effort, this paper introduces a data-driven machine learning methodology designed to expedite the prediction of temperature field distributions within rotary kilns. Through a detailed analysis of field data from the factory and the insights of experts, the paper has discussed on the key factors influencing the temperature field dynamics in rotary kilns. Leveraging Computational Fluid Dynamics (CFD), a numerical simulation model was established to replicate rotary kiln oxy-coal combustion scenarios. A comprehensive dataset was generated using high-throughput computing technology, encompassing a variety of conditions such as coal injection rates, return air temperature, return airflow volume, fixed carbon content of pulverized coal, and volatile fraction of pulverized coal. This dataset included a total of 625 distinct cases. Subsequently, employing machine learning techniques including random forest (RF) and genetic algorithm-optimized Random forest (GA-RF), predictive models were developed for rotary kiln temperature field dynamics. A rigorous comparative analysis was conducted to explore the impact of different machine learning dataset sizes on the predictive accuracy of the models. The results of this analysis demonstrate that, enhanced by the genetic algorithm of RF, it surpasses the performance of the standard RF approach. This enhancement resulted in a significant improvement in prediction accuracy, as evidenced by the reduction of the Root Mean Square Error (RMSE) from 25.0879 to 17.5330 and the Mean Absolute Error (MAE) from an initial value of 6.5600 to 4.7661. Furthermore, the novel soft-measurement method introduced in this paper significantly reduces simulation time by several orders of magnitude. It has diminished the response time from the traditional industrial methods, which took two to three weeks, to approximately 10 s, as compared to conventional CFD approaches. Remarkably, this approach successfully preserves a high level of accuracy. The implications of these research findings are far-reaching,providing novel insights into the acceleration of CFD simulation calculations and the design of soft measurement sensors customized for black-box reactors in industrial processes.

Model-level Weight Update Domain Adaptive Dynamic CNN Soft Sensor for Free Calcium Ion Concentration in Cement Clinker

The paramount determinant of clinker quality within the cement clinker production process resides in the precise prediction of F-CAO (free calcium ion) content. Given the intricate nature of cement clinker production, a profound coupling and temporal progression are inherently intertwined between the process data, as working conditions undergo gradual transformations over time. This research suggests a Model-level Weight Update Domain Adaptive Dynamic CNN Soft Sensor for Free Calcium Ion Concentration in Cement Clinker to overcome the aforementioned issues. We introduce a graph-structured two-dimensional methodology aimed at transforming both labeled and unlabeled data within the original process dataset into two-dimensional process data blocks. This innovative approach serves to enhance the spatial and temporal attributes of the labeled samples within the comprehensive sample set. Importantly, this methodology effectively addresses the intricate challenges posed by robust data coupling and the substantial presence of unlabeled samples. In response to the challenge posed by the frequent temporal fluctuations in working conditions, we introduce the Model-level Weight Update Domain Adaptive (MWUDA) approach. MWUDA leverages the hidden layer outputs from both the source and target domains to optimize the weight updates of the fully-connected layer within the CNN model. This strategic weight optimization is aimed at ameliorating the performance degradation observed in static models when confronted with shifting patterns in sample distributions. Subsequently, a limited subset of source domain samples exhibiting analogous regularities to those in the test samples is identified, enabling the acquisition of refined weights through the MWUDA method. Employing this strategy safeguards the precision of model updates throughout the MWUDA computation procedure, ensuring a robust and accurate adaptation process. The effectiveness of the proposed method is verified in an experimental study with real cement production data.

Towards an AI-driven soft toy for automatically detecting and classifying infant-toy interactions using optical force sensors.

Introduction: It is crucial to identify neurodevelopmental disorders in infants early on for timely intervention to improve their long-term outcomes. Combining natural play with quantitative measurements of developmental milestones can be an effective way to swiftly and efficiently detect infants who are at risk of neurodevelopmental delays. Clinical studies have established differences in toy interaction behaviors between full-term infants and pre-term infants who are at risk for cerebral palsy and other developmental disorders. Methods: The proposed toy aims to improve the quantitative assessment of infant-toy interactions and fully automate the process of detecting those infants at risk of developing motor delays. This paper describes the design and development of a toy that uniquely utilizes a collection of soft lossy force sensors which are developed using optical fibers to gather play interaction data from infants laying supine in a gym. An example interaction database was created by having 15 adults complete a total of 2480 interactions with the toy consisting of 620 touches, 620 punches-"kick substitute," 620 weak grasps and 620 strong grasps. Results: The data is analyzed for patterns of interaction with the toy face using a machine learning model developed to classify the four interactions present in the database. Results indicate that the configuration of 6 soft force sensors on the face created unique activation patterns. Discussion: The machine learning algorithm was able to identify the distinct action types from the data, suggesting the potential usability of the toy. Next steps involve sensorizing the entire toy and testing with infants.

Bringing a scalable adaptive hybrid modeling framework closer to industrial use: Application on a multiscale fungal fermentation.

Digitalization has paved the way for new paradigms such as digital shadows and digital twins for fermentation processes, opening the door for real-time process monitoring, control, and optimization. With a digital shadow, real-time model adaptation to accommodate complex metabolic phenomena such as metabolic shifts of a process can be monitored. Despite the many benefits of digitalization, the potential has not been fully reached in the industry. This study investigates the development of a digital shadow for a very complex fungal fermentation process in terms of microbial physiology and fermentation operation on pilot-scale at Novonesis and the challenges thereof. The process has historically been difficult to optimize and control due to a lack of offline measurements and an absence of biomass measurements. Pilot-scale and lab-scale fermentations were conducted for model development and validation. With all available pilot-scale data, a data-driven soft sensor was developed to estimate the main substrate concentration (glucose) with anormalized root mean squared error(N-RMSE)of 2%. This robust data-driven soft sensor was able to estimate accurately in lab-scale (volume < 20× pilot) with a N-RMSE of 7.8%. A hybrid soft sensor was developed by combining the data-driven soft sensor with a mass balance to estimate the glycerol and biomass concentrations on pilot-scale data with N-RMSEs of 11% and 21%, respectively. A digital shadow modeling framework was developed by coupling a mechanistic model (MM) with the hybrid soft sensor. The digital shadow modeling framework significantly improved the predictability compared with the MM. The contribution of this study brings the application of digital shadows closer to industrial implementation. It demonstrates the high potential of using this type of modeling framework for scale-up and leads the way to a new generation of in silico-based process development.

Carbon black silicone composite in surgical simulation—A detailed electromechanical characterization

The integration of advanced materials into surgical simulators has revolutionized medical training. In this study, a detailed characterization of carbon black silicone composites specifically designed as soft sensor material in hybrid surgical simulators including a hyper visco plasto elastic model is performed. Carbon black-silicone composites offer unique properties such as flexibility, softness and electrical conductivity, making them ideal for integration into synthetic tissue imitations in simulators. These sensors provide real-time feedback on tissue deformation, improving the realism and evaluation capability of surgical simulations. The comprehensive, closed-form material model enables detailed analysis of the composite’s mechanical and electrical behavior, accounting for viscoelasticity, plasticity and hyperelasticity. The application of this model provides insights into the behavior of the conductive composite in different sensory scenarios, which facilitates the development of more realistic simulators.

Measurement of 3D Wrist Angles by Combining Textile Stretch Sensors and AI Algorithm.

The wrist is one of the most complex joints in our body, composed of eight bones. Therefore, measuring the angles of this intricate wrist movement can prove valuable in various fields such as sports analysis and rehabilitation. Textile stretch sensors can be easily produced by immersing an E-band in a SWCNT solution. The lightweight, cost-effective, and reproducible nature of textile stretch sensors makes them well suited for practical applications in clothing. In this paper, wrist angles were measured by attaching textile stretch sensors to an arm sleeve. Three sensors were utilized to measure all three axes of the wrist. Additionally, sensor precision was heightened through the utilization of the Multi-Layer Perceptron (MLP) technique, a subtype of deep learning. Rather than fixing the measurement values of each sensor to specific axes, we created an algorithm utilizing the coupling between sensors, allowing the measurement of wrist angles in three dimensions. Using this algorithm, the error angle of wrist angles measured with textile stretch sensors could be measured at less than 4.5°. This demonstrated higher accuracy compared to other soft sensors available for measuring wrist angles.

Multi-mode industrial soft sensor method based on mixture Laplace variational auto-encoder

The industrially collected process data usually exhibit non-Gaussian and multi-mode characteristics. Due to sensor failures, irregular disturbances, and transmission problems, there are unavoidable outliers that make the data exhibit heavy-tailed characteristics. To this end, a variational auto-encoder regression method based on the mixture Laplacian distribution (MLVAER) is proposed, by introducing a type-II multivariate Laplacian distribution in the latent variable space for robust modeling, and further extending it to the mixture form to accommodate multi-mode processes, the corresponding reparameterization trick is finally proposed for the mixture form of this distribution for neural network gradient descent training. The model based on this distribution assumption has higher degrees of freedom than the model based on the traditional multivariate Laplace distribution assumption when the network structure is the same. Numerical simulation and experiments on two industrial examples demonstrate that the proposed algorithm reduces the root mean square error by over 15% compared to other algorithms.

Development of a soft sensor for fouling prediction in pipe fittings using the example of particulate deposition from suspension flow

Fouling is the unwanted accumulation of material on a processing surface which is an especially problematic issue in the food industry. Characterizing or predicting fouling through traditional methods or models is a challenge due to the complexity of fouling mechanisms. Machine learning (ML) techniques can overcome this challenge by creating models for prediction directly from experimental data. Unfortunately, the results can be hard to interpret depending on the algorithm.Here, a soft sensor is generated from an extensive data set to predict the fouling of a model particle material system. This is performed inside two different pipe fittings, an inaccessible and accessible fitting (e.g., for sensor measurements). Additionally, dimensional analysis (DA) is conducted to identify the correlations responsible for fouling while keeping descriptors with physical meaning. The resulting dimensionless numbers (DNs) are further processed by three ML algorithms: linear regression (LR), symbolic regression (SR), and random forest (RF).The soft sensor generated using a RF outperformed the other two regressors for the dimensional (Q2=0.90±0.08) and for the dimensionless data (Q2=0.88±0.09). The parameter time and particle mass fraction were determined to be most influential. Furthermore, seven DNs were obtained allowing a reduced experimental design.

Deep Learning of Partially Labeled Data for Quality Prediction Based on Stacked Target-Related Laplacian Autoencoder.

Partially labeled data, which is common in industrial processes due to the low sampling rate of quality variables, remains an important challenge in soft sensor applications. In order to exploit the information from partially labeled data, a target-related Laplacian autoencoder (TLapAE) is proposed in this work. In TLapAE, a novel target-related Laplacian regularizer is developed, which aims to extract structure-preserving and quality-related features by preserving the feature-target mapping according to the local geometrical structure of the data. In addition, stacked TLapAE (STLapAE) is further constructed to extract deep feature representations of the data by hierarchically stacking TLapAE blocks. For model training, backward propagation equations are derived based on matrix calculus techniques to update the model parameters of the proposed TLapAE. The effectiveness of the proposed STLapAE is evaluated using the butane content prediction case in a debutanizer column, the silicon content prediction case in a blast furnace (BF) ironmaking process, and the ethane concentration prediction case in an ethylene fractionator. The results show that the proposed TLapAE model has significantly improved prediction accuracy compared to soft sensors using only labeled data and other partially labeled data modeling methods.

Soft Sensor of the Key Effluent Index in the Municipal Wastewater Treatment Process Based on Transformer

Soft Sensor of the Key Effluent Index in the Municipal Wastewater Treatment Process Based on Transformer

Data Mode-Related Generative Adversarial Network for Industrial Soft Sensor Application

Data Mode-Related Generative Adversarial Network for Industrial Soft Sensor Application

Novel Multiattribute Space-Based LSTM for Industrial Soft Sensor Applications

Novel Multiattribute Space-Based LSTM for Industrial Soft Sensor Applications

Improved Bi-LSTM With Distributed Nonlinear Extensions and Parallel Inputs for Soft Sensing

Improved Bi-LSTM With Distributed Nonlinear Extensions and Parallel Inputs for Soft Sensing

A Fault-Tolerant Soft Sensor Algorithm Based on Long Short-Term Memory Network for Uneven Batch Process

Batch processing is a widely utilized technique in the manufacturing of high-value products. Traditional methods for quality assessment in batch processes often lead to productivity and yield losses because of offline measurement of quality variables. The use of soft sensors enhances product quality and increases production efficiency. However, due to the uneven batch data, the variation in processing times presents a significant challenge for building effective soft sensor models. Moreover, sensor failures, exacerbated by the manufacturing environment, complicate the accurate modeling of process variables. Existing soft sensor approaches inadequately address sensor malfunctions, resulting in significant prediction inaccuracies. This study proposes a fault-tolerant soft sensor algorithm that integrates two Long Short-Term Memory (LSTM) networks. The algorithm focuses on modeling process variables and compensating for sensor failures using historical batch quality data. It introduces a novel method for converting quality variables into process rates to align uneven batch data. A case study on simulated penicillin production validates the superiority of the proposed algorithm over conventional methods, showing its capacity for precise endpoint detection and effectiveness in addressing the challenges of batch process quality assurance. This study offers a robust solution to the issues of soft sensor reliability and data variability in industrial manufacturing.