Complete Simulation Research Articles

Sensitivity Analysis of a Stator Current-based MRAS Estimator for Sensorless Induction Motor Drives

The sensitivity of speed estimators for parameter variations presents a significant challenge for sensorless Induction Motor (IM) drives, particularly at very low speeds. This paper examines the impact of parameter variations and the PI adaptation mechanism on the stator current-based Model Reference Adaptive System (MRAS). In contrast to the estimation of rotor flux, the MRAS method uses the observed stator current and the stator current estimate error within the adjustable IM model. The stability analysis for changes in machine parameters and PI controller gains is examined using small-signal perturbation. Additionally, a sensitivity analysis of stator and rotor resistance changes is included. A complete simulation using MATLAB/Simulink and experimental validation using a laboratory prototype based on the DSP-DS1103 are provided. The analytical, modeling, and measurement results reveal that the suggested observer responds well and provides precise speed estimation in all four quadrants of operation.

Effect of double-layer formwork curing on the early temperature field and properties of precast concrete components in environments with large diurnal temperature variations

An efficient double-layer formwork curing method using an “inner supporting formwork + outer insulation formwork” was proposed in this study to address the early cracking of precast concrete components in high altitude regions. Steel and plywood formwork were designed as inner support formwork, while polystyrene (PS) and polyurethane (PU) were used as outer insulation formwork. Indoor experiments and two finite element methods (The complete simulation method focuses on computational accuracy, and the equivalent simulation method emphasizes computational efficiency) were employed to analyze the evolution of the concrete temperature field under different double-layer formwork curing methods throughout the curing period, combined with compressive strength and pore structures testing. The results show that steel + 5-mm-thick PU insulation formwork curing method can significantly mitigate the impact of large diurnal temperature variations on the internal temperature of concrete. Unlike traditional steam-curing, this method does not deteriorate the pore structure or compressive strength of the concrete. This study is of great significance in addressing the problem of early cracking of precast concrete components exposed to large diurnal temperature vriations in high altitude regions.

Strategies for the alignment of electronic states in quantum-dot tunnel-injection lasers

In quantum-dot tunnel-injection lasers, the excited charge carriers are efficiently captured from the bulk states via an injector quantum well and then transferred into the quantum dots via a tunnel barrier. The alignment of the electronic levels is crucial for the high efficiency of these processes and especially for the fast modulation dynamics of these lasers. In particular, the quantum mechanical nature of the tunneling process must be taken into account in the transition from two-dimensional quantum well states to zero-dimensional quantum-dot states. This results in hybrid states, from which the scattering into the quantum-dot ground states takes place. We combine electronic state calculations of the tunnel-injection structures with many-body calculations of the scattering processes and insert this into a complete laser simulator. This allows us to study the influence of the structural design and the resulting electronic states as well as limitations due to inhomogeneous quantum-dot distributions. We find that the optimal electronic state alignment deviates from a simple picture in which the quantum-dot ground state energies are one LO-phonon energy below the injector quantum well ground state.

A Physically Constrained Deep-Learning Fusion Method for Estimating Surface NO2 Concentration from Satellite and Ground Monitors.

Accurate estimation of atmospheric chemical concentrations from multiple observations is crucial for assessing the health effects of air pollution. However, existing methods are limited by imbalanced samples from observations. Here, we introduce a novel deep-learning model-measurement fusion method (DeepMMF) constrained by physical laws inferred from a chemical transport model (CTM) to estimate NO2 concentrations over the Continental United States (CONUS). By pretraining with spatiotemporally complete CTM simulations, fine-tuning with satellite and ground measurements, and employing a novel optimization strategy for selecting proper prior emission, DeepMMF delivers improved NO2 estimates, showing greater consistency and daily variation alignment with observations (with NMB reduced from -0.3 to -0.1 compared to original CTM simulations). More importantly, DeepMMF effectively addressed the sample imbalance issue that causes overestimation (by over 100%) of downwind or rural concentrations in other methods. It achieves a higher R2 of 0.98 and a lower RMSE of 1.45 ppb compared to surface NO2 observations, overperforming other approaches, which show R2 values of 0.4-0.7 and RMSEs of 3-6 ppb. The method also offers a synergistic advantage by adjusting corresponding emissions, in agreement with changes (-10% to -20%) reported in the NEI between 2019 and 2020. Our results demonstrate the great potential of DeepMMF in data fusion to better support air pollution exposure estimation and forecasting.

Simulation of Ozone Distribution in an Innovative Drying and Sanitising Cabinet Chamber

Common designs of workwear drying units require not only energy efficiency but also effective disinfection. One possibility of sanitising clothes during drying is to use the ozone generated inside the drying chamber. This process requires precise management of airflow and a uniform distribution of ozone in the chamber. Therefore, optimising the shape of the drying chamber must include not only the correct flow of drying air but also the correct distribution of ozone. This paper addresses the difficult problem of modelling the flow of sanitising ozone in an innovative drying chamber. The innovative shape of the chamber is shown in this article. Due to the low percentage of ozone in the air (up to 10 ppm), CFD simulation models of the usual mixture type are too inaccurate; therefore, special models have to be used. Therefore, this paper presents an experimentally verified methodology to simulate ozone flow in an innovative drying and sanitising cabinet using two methods: Discrete Phase Model (DPM) and Species Transport (ST). The DPM method uses a Euler–Lagrange approach to qualitatively assess the spread of ozone particles, treated with a description of the movement of the particles and not as a continuous gaseous substance. On the other hand, this already allows the verification of ozone concentrations, with an appropriate conversion of the measured quantities. The ANSYS/FLUENT 2023R1 program was used for the simulations after careful selection of the mesh, closing models, boundary conditions, etc. Simulations made it possible to analyse the distribution of ozone in the workspace and assess the effectiveness of the sanitisation process. The results of the simulations were verified on the basis of empirical tests, which showed the correctness of the model and the correct distribution of the sanitising ozone in the entire volume of the drying chamber in the innovative drying–sanitising chamber. The complete simulation of the air and ozone distribution using the presented models allowed for the optimisation of the opening and shapes, which contributed to improving the energy efficiency of the unit and increasing the efficiency of the sanitisation processes, making the described methodology very effective for optimising the chambers of various types of dryers.

A plaintext-related and ciphertext feedback mechanism for medical image encryption based on a new one-dimensional chaotic system

Abstract In recent years, Plaintext-Related Image Encryption (PRIE) algorithms have been introduced, demonstrating a commendable level of plaintext sensitivity to resist chosen plaintext attack (CPA). However, these approaches suffer from several drawbacks, including inability to fully reconstruct the original image, limited practical value and excessive computational demands etc.. Moreover, the exponential expansion of medical data necessitates the formulation of more secure and efficient encryption algorithms. In this paper, firstly, a novel one-dimensional chaotic map, designated as 1D-SAM, which strikes an excellent balance between structural complexity and chaotic performance is proposed. The 1D-SAM achieve a larger chaotic range and an elevated Lyapunov exponent, signifying enhanced dynamical complexity. Subsequently, we devise a lightweight medical image encryption system leveraging the 1D-SAM and an innovative diffusion architecture, termed the plaintext-related and ciphertext feedback mechanism(PRCFM). This encryption system is a symmetric-key cryptosystem, eliminating the need for transmitting supplementary data beyond the secret keys to the recipient. Notably, the encrypted image maintains identical dimensions to its original counterpart and is fully recoverable. Complete simulation experiments were conducted on a personal computer equipped with MATLAB R2021a, OS Windows 11, 2.60GHz CPU and 16GB RAM. The experimental results indicate that our encryption system, employing a single permutation-diffusion round, efficiently encrypts a 512×512 image in approximately 0.2854 seconds. Leveraging the advantages of the PRCFM, our approach demonstrates superior plaintext sensitivity, achieving an average number of pixels changing rate (NPCR) of 99.6051 and a unified average changed intensity (UACI) of 33.4452. In summary, our work addresses key limitations of contemporary encryption frameworks, exhibiting acceptable performance in both encryption speed and security strength.

Modeling of Hydrogen Combustion from a 0D/1D Analysis to Complete 3D-CFD Engine Simulations

Hydrogen and its unique properties pose major challenges to the development of innovative combustion engines, while it represents a viable alternative when it is based on renewable energy sources. The present paper deals with the holistic approach of hydrogen combustion modeling from a 0D/1D reactor evaluation with Cantera up to complete engine simulations in the 3D-CFD tool QuickSim. The obtained results are referenced to the current literature and calibrated with experimental data. In particular, the engine simulations are validated against measurements of a single-cylinder research engine, which was specifically adapted for lean hydrogen operation and equipped with port fuel injection and a passive pre-chamber system. Special attention is hereby given to the influence of different engine loads and varying lambda operation. The focus of this work is the complementary numerical investigation of the hydrogen flame speed and its self-ignition resistance under the consideration of various reaction mechanisms. A detailed transfer from laminar propagation under laboratory conditions to turbulent flame development within the single-cylinder engine is hereby carried out. It is found that the relatively simple reaction kinetics of hydrogen can lead to acceptable results for all mechanisms, but there are particular effects with regard to the engine behavior. The laminar flame speed and induction time vary greatly with the inner cylinder conditions and significantly affect the entire engine’s operation. The 3D-CFD environment offers the opportunity to analyze the interactions between mixture formation and combustion progress, which are indispensable to evaluate advanced operating strategies and optimize the performance and efficiency, as well as the reliability, of the engine.

Research on the new control strategy of the three-phase VIENNA rectifier

Abstract The VIENNA rectifier has long been used in applications such as charging piles, and the theoretical research on its control strategy is enumerated. Compared with other circuit components, the VIENNA rectifier has fewer power components and a higher power factor, so it is applied to more application scenarios. The traditional PI control has a long start-up response time and a high overshoot, while the sliding mode control (SMC) has excellent control performance for nonlinear systems. In my article, the control mode of the VIENNA rectifier is analyzed mathematically, SMC takes controls of the current inner loop, and a new exponential reaching rate is put forward the at first improve the control process. Finally, a complete system simulation model is built to prove the superiority of the VIENNA rectifier based on an optimized SMC control strategy for each control target and the corresponding control strategy. The results show that the control effect is obviously optimized by using this method.

Optimization of energy storage based on floating charge lithium iron phosphate battery

Abstract Lithium iron phosphate batteries are often used as power supplies, power batteries and energy storage batteries for electronic equipment, and their charge and discharge cycle characteristics have been studied extensively. However, when it is used as a new battery power source in a substation, the battery pack must be hung on the charger in a fully charged state for a long time and run for float charging. In this operating mode, the choice of the float billing voltage of the lithium iron phosphate battery, the size of the float charging existing, and features of unexpected discharge and battery life under lasting float charging still need to be studied. For that reason, a simulation model of the lithium battery power storage space system was developed, and the control criteria were adapted to make the particular curve constant with the examination information. Based on the real control logic of the energy storage inverter, a simulation model of the power storage inverter is built. With simulation screening and comparison with the actual outside characteristics of the inverter, the control parameters of the simulation model are continuously optimized to make them regular with the outside attributes of the actual inverter. The actual battery access point substation operating mode data is collected and sorted out, an appropriate equivalence on the battery system is performed, and an equivalence model in MATLAB/SIMULINK and PSCAD/EMTDC is built, which forms a complete simulation model together with the battery system model[1].

Polymer semiconductor films and bacteria hybrid artificial bio-leaves.

Bio-artificial photosynthetic systems can reduce CO2 into multicarbon compounds by simulating natural photosynthesis. Here, inspired by organic photovoltaic structures, we demonstrate a bio-artificial photosynthetic system based on the hybridization of polymer semiconductor films and bacteria. The study suggests that the polymer-based semiconductor film can efficiently drive the non-photosynthetic bacteria to convert CO2 to acetate. By systematically characterizing the charge transport behavior of the bio-artificial photosynthetic system, the bulk-heterojunction structure and charge transport layers are proven to enhance the system performance markedly. The scalable floating artificial bio-leaf system can produce acetate to gram scale in a week. Notably, the semiconductor film is easy to recycle and maintains stable performance, showing good sustainable production capability of the system. A quasi-solid-state artificial bio-leaf is successfully prepared using agar to simulate the morphology and function of natural leaves. Last, the acetate production converted from CO2 was used to grow yeast for food production, thus achieving a complete simulation of natural photosynthesis.

Equivalent Simulation Study of Delta-Rotor Engine

The mechanical structure, movement mode, and combustion process of a triangular rotor engine differ from those of a conventional engine with a crank connecting rod mechanism as the core. This makes it impossible to directly use existing simulation software to simulate the performance of the entire engine, rendering it difficult to conduct structural evaluation and performance prediction during the engine development stage. Therefore, using existing performance simulation software to establish a performance-equivalent model for a triangular rotor engine is crucial. To solve the above problems, this study proposes a performance-equivalent simulation modelling method for a triangular rotor engine based on a three-dimensional simulation model and the operating principles of the triangular rotor and four-stroke engines. Compared to various performance indicators obtained from the three-dimensional simulation model, the results show that the equivalent model established in this study has sufficient accuracy for key indicators such as the cylinder pressure, cylinder temperature, and in-cylinder quality under various working conditions. This can satisfy the requirements of the complete machine-performance simulation of a triangular rotor engine.

Realistic Simulation of Dissolution Process on Rock Surface

Hydraulic dissolution, driven by carbon dioxide-rich precipitation and runoff, leads to the gradual breakdown and removal of soluble rock materials, creating unique surface and subsurface features. Dissolution is a complex process that is related to numerous factors, and the complete simulation of its process is a challenging problem. On the basis of deep investigation of the theories of geology and rock geomorphology, this paper puts forward a method for simulating the dissolution phenomenon on a rock surface. Around the movement of water, this method carries out dissolution calculations, including processes such as droplet dissolution, water flow, dissolution, deposition, and evaporation. It also considers the lateral dissolution effect of centrifugal force when water flows through bends, achieving a comprehensive simulation of the dissolution process. This method can realistically simulate various typical karst landforms such as karst pits, karst ditches, and stone forests, with interactive simulation efficiency.

Transformation and development of simulation-based education and child first aid skills training

Background: Simulation-based medical education is a complex learning methodology in different fields. Exposing children to this teaching method is uncommon as it is designed for adult learning. This study aimed to develop and implement simulation-based education in first aid training of children and investigate the emotions of children in post-simulation scenarios that replicate emergency situations. Methods: This was a phenomenological qualitative research study. The participants attended the modified “Little Doctor” course that aims to train children in first aid and, subsequently, completed simulation scenarios. The children attended focus groups and were asked about their experiences of the course and how they felt during the simulation scenarios. Results: 12 children (Age 8–11 years old) attended the course, and 10 completed the simulation scenarios and focus groups. The major theme derived from was the simulation experience’s effect, which was divided into two subthemes: the emotion caused by—and the behavioral response to—the simulation. The analysis revealed shock and surprise toward the environment of the simulation event and the victim. The behaviors expressed during the simulation scenarios ranged from skill application and empathy to recall and teamwork. Conclusions: Simulation scenarios were successfully implemented during the first-aid training course. Although participants reported mixed feelings regarding the experience, they expressed confidence in their ability to perform real-life skills.

Implementation of frozen density embedding in CP2K and OpenMolcas: CASSCF wavefunctions embedded in a Gaussian and plane wave DFT environment.

Most chemical processes happen at a local scale where only a subset of molecular orbitals is directly involved and only a subset of covalent bonds may be rearranged. To model such reactions, Density Functional Theory (DFT) is often inadequate, and the use of computationally more expensive correlated wavefunction (WF) methods is required for accurate results. Mixed-resolution approaches backed by quantum embedding theory have been used extensively to approach this imbalance. Based on the frozen density embedding freeze-and-thaw algorithm, we describe an approach to embed complete active space self-consistent field simulations run in the OpenMolcas code in a DFT environment calculated in CP2K without requiring any external tools. This makes it possible to study a local, active part of a chemical system in a larger and relatively static environment with a computational cost balanced between the accuracy of a WF method and the efficiency of DFT, which we test on environment-subsystem pairs. Finally, we apply the implementation to an oxygen molecule leaving an aluminum (111) surface and a ruthenium(IV) oxide (110) surface.

Runtime integration of machine learning and simulation for business processes: Time and decision mining predictions

Recent research in Computer Science has investigated the use of Deep Learning (DL) techniques to complement outcomes or decisions within a Discrete Event Simulation (DES) model. The main idea of this combination is to maintain a white box simulation model complement it with information provided by DL models to overcome the unrealistic or oversimplified assumptions of traditional DESs. State-of-the-art techniques in BPM combine Deep Learning and Discrete Event Simulation in a post-integration fashion: first an entire simulation is performed, and then a DL model is used to add waiting times and processing times to the events produced by the simulation model.In this paper, we aim at taking a step further by introducing Rims (Runtime Integration of Machine Learning and Simulation). Instead of complementing the outcome of a complete simulation with the results of predictions a posteriori, Rims provides a tight integration of the predictions of the DL model at runtime during the simulation. This runtime-integration enables us to fully exploit the specific predictions while respecting simulation execution, thus enhancing the performance of the overall system both w.r.t. the single techniques (Business Process Simulation and DL) separately and the post-integration approach. In particular, the runtime integration ensures the accuracy of intercase features for time prediction, such as the number of ongoing traces at a given time, by calculating them during directly the simulation, where all traces are executed in parallel. Additionally, it allows for the incorporation of online queue information in the DL model and enables the integration of other predictive models into the simulator to enhance decision point management within the process model. These enhancements improve the performance of Rims in accurately simulating the real process in terms of control flow, as well as in terms of time and congestion dimensions. Especially in process scenarios with significant congestion – when a limited availability of resources leads to significant event queues for their allocation – the ability of Rims to use queue features to predict waiting times allows it to surpass the state-of-the-art. We evaluated our approach with real-world and synthetic event logs, using various metrics to assess the simulation model’s quality in terms of control-flow, time, and congestion dimensions.

Reducing Patient Care Delays in Radiation Oncology Via Optimization of Insurance Preauthorization

Purpose/ObjectivesDifficulties and delays in insurance pre-authorization (pre-auth) can negatively impact patient care, resulting in postponing, modifying, or even cancelling radiation therapy for patients. We aimed to perform a root cause analysis for pre-auth delays in our department and implement solutions to optimize our workflow. Our primary objectives were to decrease mean time for clinical treatment plan (CTP) completion, and for number of cases delayed/denied, by 50% each. Materials/MethodsWe performed a root cause analysis for pre-auth delays, and used the PDSA & A3 quality improvement methods. We sampled ∼2 cases per disease site (19 cases from July – Aug 2022) to determine the baseline. Countermeasures included: 1) optimizing our CTP templates per disease site to contain the specific clinical information required for pre-auth, 2) formalizing earlier completion of CTPs in our Care Path®, and 3) formalizing the pre auth workflow in our Care Path®. We tracked various metrics, including mean time for CTP completion, % usage of our Care Path®, % usage of revised CTP templates, mean time until pre-auth initiated & completed, and % of cases delayed/denied. Two-tailed T-tests and Chi-squared tests were used to generate p-values comparing mean values and percentages, respectively. Results495 patients completed CT simulation in our department between October 2022 and February 2023. Mean time for CTP completion (Day 0 = day of CT simulation scheduling) improved from 16 days at baseline to 4 days (p<0.001). Care Path® usage improved from 16% to 97% (p<0.001), as did usage of our revised CTP templates, from 0% to 97% (p<0.001). The mean time from insurance pre-auth initiation to completion improved from 5 days to 1 day. The percent of cases that were delayed/denied was reduced significantly from 32% to 8% (p<0.001). ConclusionsImproving timeliness and details of CTP documentation and pre-auth by using our Care Path® and optimizing CTP templates improved efficiency of insurance pre-auth completion and reduced the number of cases delayed/denied.

Electromechanical processes in electric power steering system based on permanent magnet synchronous motor

Relevance. Electric power steering of vehicles successfully competes with hydraulic boosters of similar purposes, showing high efficiency, energy saving and environmental friendliness. Aim. To study and research electromechanical processes of electric power steering system based on permanent magnet synchronous motor. Methods. Mathematical and numerical modeling using the Matlab/Simulink software package. Results. The article discusses the operating modes of an electric power steering based on an electric motor with permanent magnets. The authors analyzed two classic schemes for constructing steering systems with electric power steering – with the electric motor placed on the steering column and on the steering rack. Block diagrams of the mechanical and electrical parts of the steering system have been developed. Based on mathematical dependencies characterizing the operating modes of the steering system, a complete dynamic simulation model of the electric power amplifier was obtained on the MATLAB/Simulink platform. Based on a preliminary analysis of control methods and algorithms in the system to solve the problem of control stability, PI controllers with low-pass filters are implemented to regulate the current and voltage of the electric motor depending on the speed of the vehicle and the angle of rotation of the steering wheels. The simulation results were used to design a real electric power steering for a specific vehicle. Conclusions. The complete simulation model of an electric booster based on a synchronous motor with permanent magnets in a car steering system obtained by the authors adequately reflects the dynamic control processes and can be used in the design of automotive vehicles.

Systemic comparison of molecular characteristics in different skin fibroblast senescent models.

Senescent human skin primary fibroblast (FB) models have been established for studying aging-related, proliferative, and inflammatory skin diseases. The aim of this study was to compare the transcriptome characteristics of human primary dermal FBs from children and the elderly with four senescence models. Human skin primary FBs were obtained from healthy children (FB-C) and elderly donors (FB-E). Senescence models were generated by ultraviolet B irradiation (FB-UVB), D-galactose stimulation (FB-D-gal), atazanavir treatment (FB-ATV), and replication exhaustion induction (FB-P30). Flow cytometry, immunofluorescence staining, real-time quantitative polymerase chain reaction, co-culturing with immune cells, and bulk RNA sequencing were used for systematic comparisons of the models. In comparison with FB-C, FB-E showed elevated expression of senescence-related genes related to the skin barrier and extracellular matrix, proinflammatory factors, chemokines, oxidative stress, and complement factors. In comparison with FB-E, FB-UVB and FB-ATV showed higher levels of senescence and expression of the genes related to the senescence-associated secretory phenotype (SASP), and their shaped immune microenvironment highly facilitated the activation of downstream immune cells, including T cells, macrophages, and natural killer cells. FB-P30 was most similar to FB-E in terms of general transcriptome features, such as FB migration and proliferation, and aging-related characteristics. FB-D-gal showed the lowest expression levels of senescence-related genes. In comparisons with the single-cell RNA sequencing results, FB-E showed almost complete simulation of the transcriptional spectrum of FBs in elderly patients with atopic dermatitis, followed by FB-P30 and FB-UVB. FB-E and FB-P30showed higher similarity with the FBs in keloids. Each senescent FB model exhibited different characteristics. In addition to showing upregulated expression of natural senescence features, FB-UVB and FB-ATV showed high expression levels of senescence-related genes, including those involved in the SASP, and FB-P30showed the greatest similarity with FB-E. However, D-galactose-stimulated FBs did not clearly present aging characteristics.

Hard diffraction in SHERPA

We present the first complete simulation framework for the simulation of jet production in diffractive events at next-to leading order in QCD, matched to the parton shower. We validate the implementation in the SHERPA event generator with data from the H1 and ZEUS experiments for diffractive DIS and diffractive photoproduction. For the latter, we review different models aiming to explain the observed factorisation breaking and we argue that at NLO the direct component must also be suppressed. We provide predictions for diffractive jet production both in DIS and in photoproduction events for the upcoming EIC.

Predicting sessile droplet evaporation kinetics via cascaded deep networks and tree-based machine learning approach

The intricate nature of sessile droplet evaporation phenomenon makes detailed experimental studies time consuming and requires sophisticated apparatus; while complete numerical simulations are computationally expensive and also time-intensive. In this article, for the first time, we explore the applicability of machine learning (ML) approaches to predict the evaporation kinetics of generalized sessile droplets under various conditions. An in-house dataset, obtained using an experimentally well validated numerical model, is used to develop the ML models: deep artificial neural network (ANN) and decision tree algorithms such as Random Forest (RF) and extreme gradient boosting (XGB). The structures of these models are modified by cascading the output features according to the physics involved. This distinctive approach results in better prediction of the evaporation kinetics than basic ML models. The models are trained by a large set of input parameters for the target variables: viz. droplet evaporation rate, velocity scale, and temperature drop in the solid and liquid domain. Finally, the performance of these ML models is assessed by comparing their predictions with that of physics-based, experimentally validated, numerical models. Results show that the inclusion of additional features obtained using feature engineering significantly improve the prediction performance of ML algorithms, and consistently accurate predictions of droplet evaporation kinetics are obtained. Among various algorithms considered here, the ANN outperforms in term of various error matrices for most of the cases, followed by XGB, and RF models. Also, the highest mean average error (MAE) yield by the ML models for evaporation rate, velocity scale and temperature drop in liquid remains within ∼12.5%. In the case of temperature drop for the solid, the MAE is considerably higher due to large variability of the same target variable. Overall, the work clearly shows that ML algorithms can be used to obtain physically consistent predictions for sessile droplet evaporation parameters.