5 results
Filters
Settings
Now showing1 - 5 of 5
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Reconstruction-free EIT for injection-pattern classification and superficial gas velocity regression as proxies for local gas holdup in bubble columns(American Chemical Society (ACS), 2026-09-10); ; ; ; Electrical impedance tomography (EIT) as a noninvasive tomographic technique is increasingly applied to multiphase reactor monitoring; however, conventional image reconstruction is ill-posed and regularization-dependent and may be redundant in applications where the primary objective is operating-state identification rather than explicit spatial conductivity field reconstruction. Here, we present a reconstruction-free, measurement-domain framework for bubble-column monitoring that maps raw complex boundary impedance data directly to two reactor-relevant inference tasks: (i) gas injection pattern classification and (ii) superficial gas velocities regression. Together, these two quantities ─ the spatial injection distribution and the total volumetric flow ─ constitute the primary process-state information from which gas holdup can subsequently be inferred and are therefore reported as proxies for local gas holdup monitoring. Experiments were conducted in an acrylic bubble column (600 mm height, 104 mm inner diameter) equipped with a 256-electrode array distributed over eight axial rings and operated at four excitation frequencies (1 kHz-1 MHz). Experiments covered gas flow rates between 1.0 and 6.5 L min–1 (Ug = 1.96 → 12.75 mm s–1), within which near-perfect gas injection pattern classification was achieved with accuracies of 93–100% for excitation frequencies between 1 and 100 kHz using the full 256-electrode configuration. For quantitative superficial gas velocity estimation, increasing calibration density along Ug reduced the mean absolute error from 0.388 to 0.105 L min–1 (MAE[Ug] = 0.76 → 0.205 mm s–1, i.e. 7.0% → 1.9% of the operating range) for localized injection and from 0.298 to 0.157 L min–1 (MAE[Ug] = 0.585 → 0.307 mm s–1, i.e. 5.4% → 2.8% of the operating range) for distributed injection conditions. These results demonstrate that direct inference from raw EIT boundary measurements enables accurate, real-time monitoring of bubble-column operation without tomographic reconstruction and provide quantitative guidance on excitation frequency selection, axial sensing placement, and calibration resolution.Publicationtype: Journal Article23 23 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , A cost-efficient electrical impedance tomography system for adaptive process analytics(American Chemical Society (ACS), 2026-08-02); ; ; ; Electrical impedance tomography (EIT) has significant potential as a noninvasive tool for spatially resolved process monitoring, yet most reported systems remain laboratory prototypes and are rarely suitable for integration into chemical engineering environments. We present a fully embedded, low-cost EIT platform designed as a deployable tomographic sensor for in situ monitoring of reactive media in laboratory-scale reactors. The platform can be realized for approximately $442 in prototype quantities and below $310 in small-batch production. The system enables spatially resolved conductivity mapping with 32 measurement channels and is designed for straightforward integration into laboratory and reactor environments. Its compact USB-powered hardware is complemented by microcontroller firmware in C and an object-oriented Python interface for experiment control and data analysis. On-board impedance calculation allows stand-alone operation and supports real-time monitoring of dynamic conductive processes. Hardware characterization includes measurements of total harmonic distortion plus noise for signal excitation and voltage measurement with mean values up to 0.0651%, a signal frequency of 105 kHz and optimized analogue amplification across all gain settings, a variable analogue filter gain ranging from −1.35 to 34.81 dB, and characterization of the current source at excitation currents of 100 and 350 μA. Additionally, the system’s performance is validated for the adjacent and opposed pair pattern, achieving a mean signal-to-noise ratio of up to 76.59 dB. The imaging performance is quantified on circular phantoms in terms of position error, spatial resolution, and detection limit, and the long-term stability is verified over a 20-h measurement with a relative drift below 0.03%. The system achieves frame rates of 7.36 to 13.18 fps depending on the measurement pattern. The system is validated in a chemically relevant scenario by monitoring the dissolution of 5 mol L–1 H2SO4(aq). The resulting spatial conductivity maps resolve transient concentration gradients associated with proton activity, illustrating the potential of the platform for the detection of local process states in adaptive reactor environments. Together, the characterized hardware performance and validation under chemically relevant conditions establish the platform as a scalable in situ imaging sensor suitable for integration into adaptive, data-driven reactor systems.Publicationtype: Journal Article26 28 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Impedance-based estimation of process parameters in electrolytic systems via circuit-embedded neural network (CENN)(American Chemical Society (ACS), 2026-07-09); ; ; ; We present a physics-aware machine learning framework for real-time estimation of electrolyte concentration and temperature from electrochemical impedance spectroscopy (EIS) measurements. Using nanoporous gold electrodes in aqueous H₂SO₄ as a model system, impedance spectra were acquired across concentrations of 1−20 mM and temperatures of 26−50 °C. The spectra were interpreted using a ZARC + transmission-line equivalent circuit and globally fitted over the full (c, T) domain, revealing two characteristic time scales associated with interfacial charge transfer and porous transport. The resulting physically consistent parameter maps were embedded into a circuit-informed neural network, enabling high-fidelity forward prediction of full impedance spectra directly from state variables (RMSE = 0.06 Ω). Inversion of this model allowed accurate estimation of electrolyte properties from measured spectra, achieving mean absolute errors of 0.10 mM in concentration and 1.0 °C in temperature. To further ccelerate sensing, Jacobian-based sensitivity analysis identified three informative frequencies that preserved essential spectral information. This reduced acquisition time by more than 95%-from minutes to under 6 s-while maintaining high accuracy (0.12mM, 1.1 °C). Although demonstrated for a specific electrochemical system, the proposed framework provides an interpretable and transferable workflow for rapid multiparameter sensing, provided that the equivalent-circuit structure and calibration domain are adapted to the target system.Publicationtype: Journal Article37 64 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Simulation-driven Electrical Impedance Tomography for buoyancy-driven electrolyte mixing: A hybrid hydrodynamic–circuit modeling framework(Elsevier, 2026-02-17); ; ; ; We present a modular, end-to-end simulation benchmark for Electrical Impedance Tomography (EIT) in chemical reactor monitoring. The goal is to enable controlled and repeatable evaluation of EIT reconstruction methods in a realistic, physics-based setting. The framework integrates the full pipeline—from hydrodynamic transport and electrochemical modeling to hardware emulation and image reconstruction. For this purpose, aqueous HCl mixing is modeled using Stokes–Boussinesq flow coupled with advection–diffusion; local conductivity is computed via the Debye–Hückel–Onsager approximation and mapped to resistor networks simulated in SPICE with a 32-electrode EIT analog front-end. Boundary voltages are generated and processed using the Complete Electrode Model (CEM) in EIDORS and reconstructed with NOSER and Total Variation (TV). The modular design enables independent refinement of each module for task-specific front-ends and reconstructions. A comparison of FEM/CEM and SPICE simulations under matched hardware configurations shows close agreement in voltage measurements, validating circuit translation. For homogeneous measurements, simulated voltages were further validated against data from a self-built EIT device, confirming consistency across both simulation paths. Moreover, by modifying hardware-specific parameters in SPICE, we demonstrate how design choices influence measurement behavior, supporting targeted front-end optimization. This simulative benchmark thus provides a validated, reusable testbed for studying physics–algorithm–hardware interactions and for standardized, quantitative evaluation of EIT methods in reactor-relevant settings. It integrates electrical modeling, reconstruction algorithms, and engineering aspects into a coherent cross-domain framework that highlights how modeling and reconstruction choices shape interpretation across applications.Publicationtype: Journal Article95 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Simulation framework for electrical impedance tomography systemsElectrical Impedance Tomography (EIT) is a promising approach for monitoring chemical processes and deriving process quantities. These insights then can be used for optimal control of the process. The EIT hardware must be optimized for the specific requirements of the chemical reaction and its environment. Further influences come from the image reconstruction algorithms, which in turn are optimized for the process itself. This requires a large amount of data to be captured. We employ a hybrid approach using SPICE and COMSOL. Combining these methods we built a framework that is capable of generating simulated EIT data of a real world system enabling hardware optimization as well as algorithm optimization. COMSOL is utilized to fit equivalent circuit models to the object under study, while SPICE simulates the output signal with noise introduced by discrete hardware components. The simulation serves as a digital twin of a real system with which the generated signals are cross-validated. With this framework, we are able to simulate data, reconstruct realistic images, and optimize for any custom-made system effectively.Publicationtype: Journal Article99