Soft Sensor Research Articles (Page 1)

Monitoring bottom-hole pressure (BHP) is critical for reservoir management and flow assurance, especially in offshore fields where challenging conditions and production losses are more impactful. However, reliability issues and high installation costs of Permanent Downhole Gauges (PDGs) often limit access to this vital data. Soft sensors offer a cost-effective and reliable alternative, serving as backups or replacements for physical sensors. This study proposes a novel data-driven methodology for estimating flowing BHP using wellhead and topside measurements from plant monitoring systems. The framework employs ensemble methods combined with clustering techniques to partition datasets, enabling tailored supervised training for diverse production conditions. Aggregating results from sub-models enhances performance, even with simpler machine learning algorithms. We evaluated Linear Regression, Neural Networks, and Gradient Boosting (XGBoost and LightGBM) as base models. A case study of a Brazilian Pre-Salt offshore oilfield, using data from 60 wells across nine platforms, demonstrated the methodology’s effectiveness. Error metrics remained consistently below 2% across varying production conditions and reservoir lifecycle stages, confirming its reliability. This solution provides a practical, economical alternative for studies and monitoring in wells lacking PDG data, improving operational efficiency and supporting reservoir management decisions.

Abstract Developing soft materials that integrate mechanical compliance, functional responsiveness, and environmental sustainability is key for next‐generation wearable and implantable electronics. Here, a sustainable, fully bio‐based organohydrogel sensor made entirely from food‐grade and biodegradable components, including gelatin, microbial transglutaminase (TG), and glycerol, prepared via a simple one‐pot process under mild thermal conditions, is reported. In this system, TG enzymatically crosslinks gelatin chains into a robust covalent network, while glycerol enhances flexibility, stabilizes hydration, and facilitates proton conduction. The multicomponent system reveals a tunable network morphology governed by enzymatic crosslinking density. The resulting gels exhibit remarkable stretchability (up to 450%), linear strain sensitivity up to 300%, and a high gauge factor of 2.86—placing them among the top‐performing hydrogel‐based strain sensors to date. In addition to strain sensing, the material shows strong thermal responsivity (0.26 °C −1 in the 20–45 °C range) without being affected by variations in environmental humidity. Long‐term electromechanical stability is demonstrated over 5000 cycles. Unlike conventional soft sensors that rely on synthetic polymers, fillers, or dopants, this platform entirely uses food‐safe components and a simple one‐pot process—offering a scalable and sustainable route to soft electronics. These findings establish enzyme‐guided polymer engineering as a powerful tool for functional material design.

Soft paw sensor for tactile and force sensing in legged robots

Melt viscosity control in polymer extrusion using nonlinear model predictive control with neural state space modelling and soft sensor feedback

Corrigendum to “Low-cost, data-efficient, on-device soft sensors for sewer flow monitoring—learning from adjacent water level sensors” [Water Research X 29 (2025) 100415

TFF-TL: Transformer based on temporal feature fusion and LSTM for dynamic soft sensor modeling of industrial processes

Enhancing reliability of data-driven soft sensors with stable loss function and sample graph

Multivariable soft sensor with a predictor of mutually dependent errors applied to an industrial fractionator

A soft sensor net based on the symplectic decomposition-global attention reconstruction architecture for biopharmaceutical industry

Resformer: Time-token transformer with residual compensation for quality prediction in industrial processes.

Flexible pressure sensors are essential for wearable electronics, human–machine interfaces, and soft robotics. However, conventional Polydimethylsiloxane (PDMS)-based sensors often suffer from limited conductivity, poor filler dispersion, and low structural integration with textile substrates. In this work, we present a robotic drop-coating approach for fabricating graphite–copper nanoparticle (G-CuNP)/PDMS composite pressure sensors with textile-integrated electrodes. By precisely controlling droplet deposition, a three-layer sandwiched structure was realized that ensures uniformity and scalability while avoiding the drawbacks of conventional full-line coating. The effects of filler loading and graphite nanoparticle (GNP) and copper nanoparticle (CuNP) ratios were systematically investigated, and the optimized sensor was obtained at 40 wt% total fillers with a graphite content of 55 wt%. The fabricated device exhibited high sensitivity in the low-pressure region, stable performance in the medium- and high-pressure ranges, and an exponential saturation fitting with R2 = 0.998. The average hysteresis was 7.42%, with excellent cyclic stability over 1000 loading cycles. Furthermore, a hand-shaped sensor matrix composed of five distributed sensing units successfully distinguished grasping behaviors of lightweight and heavyweight objects, demonstrating multipoint force mapping capability. This study highlights the advantages of robotic drop-coating for scalable fabrication and provides a promising pathway toward low-cost, reliable, and wearable soft pressure sensors.

The development of smart materials capable of rapid and reversible responses to ambient humidity is essential for next-generation sensors, monitoring systems, and adaptive devices. Covalent organic frameworks (COFs), with their tunable porosity and designable architectures, represent a promising class of materials for such stimuli-responsive systems, yet practical implementation remains limited. In this study, we report a self-standing ionic COF membrane, synthesized by integrating hydrogen-bonding ionic functionalities into the framework backbone. This rational design endows the membrane with exceptional moisture-driven actuation and sensing behavior. The unique hydrogen bonding interactions within the framework facilitate rapid water uptake and release, enabling a rapid response time of1s. The membrane demonstrates excellent mechanical flexibility, high water sorption capacity, and robust cycling durability. Results from DFT calculations and MD simulations revealed that water molecules could strongly adsorb onto the membrane via hydrogen bonding to modulate its micropore structure and facilitate its responsive behavior. Furthermore, its responsive behavior to subtle humidity changes makes it suitable for applications such as human-interfacing soft actuators, smart switches, and soil moisture sensors. This study highlights the utility of ionic COF membranes as a versatile platform for creating next-generation intelligent materials.

Abstract The development of smart materials capable of rapid and reversible responses to ambient humidity is essential for next‐generation sensors, monitoring systems, and adaptive devices. Covalent organic frameworks (COFs), with their tunable porosity and designable architectures, represent a promising class of materials for such stimuli‐responsive systems, yet practical implementation remains limited. In this study, we report a self‐standing ionic COF membrane, synthesized by integrating hydrogen‐bonding ionic functionalities into the framework backbone. This rational design endows the membrane with exceptional moisture‐driven actuation and sensing behavior. The unique hydrogen bonding interactions within the framework facilitate rapid water uptake and release, enabling a rapid response time of 1 s. The membrane demonstrates excellent mechanical flexibility, high water sorption capacity, and robust cycling durability. Results from DFT calculations and MD simulations revealed that water molecules could strongly adsorb onto the membrane via hydrogen bonding to modulate its micropore structure and facilitate its responsive behavior. Furthermore, its responsive behavior to subtle humidity changes makes it suitable for applications such as human‐interfacing soft actuators, smart switches, and soil moisture sensors. This study highlights the utility of ionic COF membranes as a versatile platform for creating next‐generation intelligent materials.

Ultrathin wearable sensors have emerged as a transformative platform for next-generation health and performance monitoring, offering intimate integration with human skin for real-time physiological and biochemical sensing based upon electrophysical and electrochemical response at human-sensor interfaces. These sensory systems, often composed of micrometer-to-nanometer-thick functional responsive materials, achieve seamless integration at skin-electronics interfaces by mimicking the dynamic and mechanical properties of human skin. Recent advances in material synthesis, nanoscale engineering, and structural design have enabled novel sensors that are not only stretchable and breathable but also robust, biocompatible, and highly responsive. This review highlights the fundamental materials principles governing skin-conformal interactions, explores various material systems, including planar, porous, and hybrid architectures, and outlines state-of-the-art developments in smart adhesives, wearable ultrathin sensors, responsive behavior, true conformability, and printed sensors. We discuss the broad spectrum of current and prospective applications, from tactile and electrophysiological sensing to biochemical and multimodal wearable devices, as well as key challenges, existing trends, and future prospects.

Soft sensors to predict critical quality attributes and monitor crystallinity and polymorphism change in solid oral dosage manufacturing: case studies

Mechanical properties are critical for characterizing and fabricating advanced materials. While current characterization methods are well-established for the nanoscale and larger millimeter-scale, a significant gap exists in automated testing at the microscale. To address this, we propose an automated, rapid characterization method based on a microrobotic system. We first develop a 6-degree-of-freedom (DOF) microrobotic system for sample alignment and testing. An image processing method is then designed for real-time sample recognition, supplying essential feedback for both alignment and testing procedures. Furthermore, a soft force sensor is fabricated and calibrated to ensure precise force measurement. Experiments on copper wires and graphite films demonstrate the method’s high precision and reliability. This work provides a robust solution for microscale mechanical property characterization, offering significant potential for advanced material development.

Cultivating hydrogen-oxidizing bacteria (HOB), such as Cupriavidus necator, using , , and offers a promising route for valorization into chemicals and materials. To enhance cultivation efficiency in a lab-scale gas fermenter lacking a gas recycling system, an automated gas supply strategy based on real-time and monitoring was developed. Fine-tuning gas delivery is essential to ensure an adequate supply for cellular growth while minimizing excess gas, particularly , that leaves the bioreactor unused, to improve process economics. In the absence of ATEX-compliant sensors, a soft sensor was implemented to estimate dissolved concentrations from uptake rates and growth phase identification. Total gas flow was controlled according to the requirements of the cells. This strategy reduced overall gas and consumption by 67%. In addition, a high-cell-density medium was formulated by integrating published recipes with Inductively Coupled Plasma Optical Emission Spectroscopy and nutrient inhibition testing. The optimized medium increased biomass yield from 15 g/L to 53 g/L, with 75% of the dry weight consisting of the bioplastic poly(3-hydroxybutyrate), without requiring nutrient addition or pH control. Together, these strategies improve the scalability, efficiency, and sustainability of -based cultivation of hydrogen-oxidizing bacteria.

Liquid crystal elastomers (LCEs) have garnered significant attention for their potential in applications, such as soft robotics, electronic sensors, and wearable electronics. However, achieving electrical conductivity while preserving their intrinsic stimulus-responsive deformation remains a key challenge. To address this issue, we have designed and developed a kind of conductive LCEs by incorporating polymerizable deep eutectic solvent (PDES) as a conductive component within LCE substrates. The resultant LCEs exhibit ionic conductivity (σ) ranging from 5.36 × 10-4 S·cm-1 ± 1.16 × 10-6 S·cm-1 to 8.19 × 10-4 S·cm-1 ± 1.55 × 10-6 S·cm-1 and demonstrate fully reversible thermotropic deformation with substantial maximum shrinkage strain. Leveraging the exceptional thermally actuated contraction behavior and intrinsic electrical conductivity, we engineer a temperature-responsive electrical switching device capable of executing programmable circuit switching operations through temperature-controlled on/off state transitions. This work overcomes the limitations of traditional flexible materials in smart response and electrical signal interaction, providing an innovative direction for the design of next-generation smart materials.

ABSTRACT Digital twin (DT) technology is gaining attention for effective water quality management by integrating diverse data sources and enabling real-time insights. The practical implementation of DT technology for intelligent river water quality management requires extensive spatiotemporal big data, underscoring the critical need to integrate physical sensors, soft sensors, and remote sensing technologies. Here, we synthesized recent advancements in hybrid physical–soft sensing systems and highlighted their potential to address the inherent limitations of conventional water quality monitoring methods, such as limited spatiotemporal resolution and high operational costs. Soft sensors, driven by machine learning (ML), estimated difficult-to-measure water quality parameters by leveraging easily measurable variables from physical sensors. Therefore, soft sensors significantly expanded the range of measurable parameters and improved data collection frequency. In addition, remote sensing offers broad spatial coverage, enabling large-scale monitoring of optically active constituents, algal blooms, and sediment dynamics. We critically review methodologies and applications that integrate these sensing technologies into DT frameworks, and identify critical knowledge gaps, particularly the lack of a fully unified integration framework combining these technologies for next-generation DT systems. By assessing the strengths and limitations of each approach and proposing integration strategies, this study offers practical guidance and integration recommendations for DT-based river management.

ABSTRACTEfficient real‐time monitoring of directly unmeasurable variables, such as product moisture content and quality attributes, is crucial for optimizing process control in smart dryers. Advanced soft sensing techniques, which integrate analytical hardware with mathematical models, have enabled the development of intelligent drying systems. This review comprehensively evaluates applications of soft sensors for online monitoring in food drying, emphasizing performance characteristics such as accuracy, response time, and robustness for real‐time control. Key obstacles, including data contamination, model selection, and adaptability, are examined, and emerging solutions like adaptive algorithms and hybrid modeling strategies are discussed. The review highlights how soft sensors contribute to improved drying efficiency, energy savings, and product quality retention. Broader implications for related industries are also considered.