Search

In this work, a data-driven identification method, based on polynomial nonlinear autoregressive models with exogenous inputs (NARX) and the Volterra series, is proposed to describe the dynamic and nonlinear voltage and current characteristics of polymer electrolyte membrane fuel cells (PEMFCs). The structure selection and parameter estimation of the NARX model is performed on broad-band voltage/current data. By transforming the time-domain NARX model into a Volterra series representation using the harmonic probing algorithm, a frequency-domain description of the linear and nonlinear dynamics is obtained. With the Volterra kernels corresponding to different operating conditions, information from existing diagnostic tools in the frequency domain such as electrochemical impedance spectroscopy (EIS) and total harmonic distortion analysis (THDA) are effectively combined. Additionally, the time-domain NARX model can be utilized for fault detection by evaluating the difference between measured and simulated output. To increase the fault detectability, an optimization problem is introduced which maximizes this output residual to obtain proper excitation frequencies. As a possible extension it is shown, that by optimizing the periodic signal shape itself that the fault detectability is further increased. [ABSTRACT FROM AUTHOR]

For an autonomous mobile robot a novel predictive control algorithm is presented, that enables both trajectory tracking and point stabilization. This algorithm is employed in conjunction with a proprioceptive navigation algorithm using either inertial or odometric data for posture estimation according to detected slip conditions. Based on the latter, slip control is made possible. Experimental results demonstrate the performance of the system. [ABSTRACT FROM AUTHOR]

Abstract In this work, an online algorithm to estimate the electrochemical impedance spectrum of polymer electrolyte membrane fuel cells (PEMFCs) is presented. The electrochemical impedance, constituting the linear dynamic relation between current and voltage, is described by a discrete time filter model. The unknown model parameters are estimated from streaming time domain data using a recursive algorithm based on the generalized total least squares (GTLS) technique. This allows for flexible noise assumptions, leading to unbiased parameter estimates in the presence of noisy current and voltage measurements. Therefore, the robustness of the proposed method is significantly increased. Broad-band current excitation signals are utilized as to sufficiently stimulate the fuel cell for the online parameter estimation. The model order is selected using Akaike's information criterion (AIC), leading to a parsimonious model. Equivalent circuits are automatically generated using Foster's method of analog circuit synthesis. The applicability and benefits of the proposed algorithm regarding fuel cell diagnostics are demonstrated using measurement data of a small scale single cell (5 cm2 active area) subjected to a broad-band excitation. Highlights • Electrochemical impedance spectrum is described by a discrete time domain model. • An online estimation algorithm is derived based on generalized total least squares. • Equivalent circuits are synthesized from the discrete time domain model. • Validation is done using experimental measurements from a 5 cm2 single cell. • Time resolved impedance information during transient operations are obtained. [ABSTRACT FROM AUTHOR]

Objective: Conventional mock circulatory loops (MCLs) cannot replicate realistic hemodynamic conditions without inducing blood trauma. This constrains in-vitro hemocompatibility examinations of blood pumps to static test loops that do not mimic clinical scenarios. This study aimed at developing an atraumatic MCL based on a hardware-in-the-loop concept (H-MCL) for realistic hemocompatibility assessment., Methods: The H-MCL was designed for 450 ± 50 ml of blood with the polycarbonate reservoirs, the silicone/polyvinyl-chloride tubing, and the blood pump under investigation as the sole blood-contacting components. To account for inherent coupling effects a decoupling pressure control was derived by feedback linearization, whereas the level control was addressed by an optimization task to overcome periodic loss of controllability. The HeartMate 3 was showcased to evaluate the H-MCL's accuracy at typical hemodynamic conditions. To verify the atraumatic properties of the H-MCL, hemolysis (bovine blood, n = 6) was evaluated using the H-MCL in both inactive (static) and active (minor pulsatility) mode, and compared to results achieved in conventional loops., Results: Typical hemodynamic scenarios were replicated with marginal coupling effects and root mean square error (RMSE) below 1.74 ± 1.37 mmHg while the fluid level remained within ±4% of its target value. The normalized indices of hemolysis (NIH) for the inactive H-MCL showed no significant differences to conventional loops ( ∆NIH = -1.6 mg/100 L). Further, no significant difference was evident between the active and inactive mode in the H-MCL ( ∆NIH = +0.3 mg/100 L)., Conclusion and Significance: Collectively, these findings indicated the H-MCL's potential for in-vitro hemocompatibility assessment of blood pumps within realistic hemodynamic conditions, eliminating inherent setup-related risks for blood trauma.

Effective treatment of heart failure with preserved ejection fraction (HFpEF) remains an unmet medical need. Although left atrial decompression using mechanical circulatory support devices was previously suggested, the heterogeneous HFpEF population and the lack of tailored devices have prevented the translation into clinical practice. This study aimed to evaluate the feasibility of left atrial decompression in HFpEF patients with a HeartMate 3 (HM3, Abbott Inc, Chicago, USA) in silico and in vitro . Anatomic compatibility of the HM3 pump was assessed by virtual device implantation into the left atrium through the left atrial appendage (LAA) and left atrial posterior wall (LAPW) of 10 HFpEF patients. Further, the efficacy of left atrial decompression was investigated experimentally in a hybrid mock loop, replicating the hemodynamics of an HFpEF phenotype at rest and exercise conditions. Virtual implantation without substantial intersection with surrounding tissues was accomplished through the LAA in 90% and 100% through the LAPW. Hemodynamic analysis in resting conditions demonstrated normalization of left atrial pressures without backflow at a pump speed of around 5400 rpm, whereas a range of 6400-7400 rpm was required during exercise. Therefore, left atrial decompression with the HM3 may be feasible in terms of anatomic compatibility and hemodynamic efficacy., Competing Interests: Disclosure: M.G. received personal fees and research grants from BerlinHeart GmbH and research grants from 4Fontan AG. D.Z. received personal fees from Abbott, Medtronic, Abiomed, Edwards, and Daiichi Sankyo and research grants from Abbott, Medtronic, Berlin Heart, Edwards, and Corcym. The other authors have no conflicts of interest to report., (Copyright © ASAIO 2023.)

The COVID-19 pandemic confronts governments and their health systems with great challenges for disease management. In many countries, hospitalization and in particular ICU occupancy is the primary measure for policy makers to decide on possible non-pharmaceutical interventions. In this paper a combined methodology for the prediction of COVID-19 case numbers, case-specific hospitalization and ICU admission rates as well as hospital and ICU occupancies is proposed. To this end, we employ differential flatness to provide estimates of the states of an epidemiological compartmental model and estimates of the unknown exogenous inputs driving its nonlinear dynamics. A main advantage of this method is that it requires the reported infection cases as the only data source. As vaccination rates and case-specific ICU rates are both strongly age-dependent, specifically an age-structured compartmental model is proposed to estimate and predict the spread of the epidemic across different age groups. By utilizing these predictions, case-specific hospitalization and case-specific ICU rates are subsequently estimated using deconvolution techniques. In an analysis of various countries we demonstrate how the methodology is able to produce real-time state estimates and hospital/ICU occupancy predictions for several weeks thus providing a sound basis for policy makers., Competing Interests: Conflict of interestThe authors declare that they have no conflict of interest., (© The Author(s) 2022.)

The currently ongoing COVID-19 pandemic confronts governments and their health systems with great challenges for disease management. Epidemiological models play a crucial role, thereby assisting policymakers to predict the future course of infections and hospitalizations. One difficulty with current models is the existence of exogenous and unmeasurable variables and their significant effect on the infection dynamics. In this paper, we show how a method from nonlinear control theory can complement common compartmental epidemiological models. As a result, one can estimate and predict these exogenous variables requiring the reported infection cases as the only data source. The method allows to investigate how the estimates of exogenous variables are influenced by non-pharmaceutical interventions and how imminent epidemic waves could already be predicted at an early stage. In this way, the concept can serve as an "epidemometer" and guide the optimal timing of interventions. Analyses of the COVID-19 epidemic in various countries demonstrate the feasibility and potential of the proposed approach. The generic character of the method allows for straightforward extension to different epidemiological models., Competing Interests: Conflicts of interestThe authors declare that they have no conflict of interest., (© The Author(s) 2021.)

The testing and certification of vehicles in succession to their development are usually achieved on chassis dynamometer testbeds. In this context, the execution of test cycles shall be performed as automatised as possible even if vehicles with manual transmission powertrain are under test. A particular challenge is the vehicle driveaway from standstill by actuating clutch and engine in accordance with each other. In this paper a non-linear decoupling multivariate control scheme based on feedback linearisation considering the accelerator and clutch pedal positions as manipulated variables is proposed. To obtain the required vehicle model for control in a time-saving manner, the clutch torque transmissibility and engine torque characteristic are identified incorporating a Wiener model structure from efficiently designed drive manoeuvres. By applying this model structure, the dynamics can be considered separately from the static non-linearity, which is beneficial for control. In addition to the successful demonstration of the presented workflow and an examination of the controller performance on vehicle system simulation software, the robustness with respect to parameter uncertainties is investigated., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2019 ISA. Published by Elsevier Ltd. All rights reserved.)

Local model networks approximate a nonlinear system through multiple local models fitted within a partition space. The main advantage of this approach is that the identification of complex nonlinear processes is alleviated by the integration of structured knowledge about the process. This paper extends these concepts by the integration of quantitative process knowledge into the identification procedure. Quantitative knowledge describes explicit dependences between inputs and outputs and is integrated in the parameter estimation process by means of equality constraints. For this purpose, a constrained generalized total least squares algorithm for local parameter estimation is presented. Furthermore, the problem of proper integration of constraints in the partitioning process is treated where an expectation-maximization procedure is combined with constrained parameter estimation. The benefits and the applicability of the proposed concepts are demonstrated by means of two illustrative examples and a practical application using real measurement data.

In this paper, nonlinear system identification utilizing generalized total least squares (GTLS) methodologies in neurofuzzy systems is addressed. The problem involved with the estimation of the local model parameters of neurofuzzy networks is the presence of noise in measured data. When some or all input channels are subject to noise, the GTLS algorithm yields consistent parameter estimates. In addition to the estimation of the parameters, the main challenge in the design of these local model networks is the determination of the region of validity for the local models. The method presented in this paper is based on an expectation-maximization algorithm that uses a residual from the GTLS parameter estimation for proper partitioning. The performance of the resulting nonlinear model with local parameters estimated by weighted GTLS is a product both of the parameter estimation itself and the associated residual used for the partitioning process. The applicability and benefits of the proposed algorithm are demonstrated by means of illustrative examples and an automotive application.