Simulation Tool Research Articles

Spatially Resolved Uncertainties for Machine Learning Potentials.

Machine learning potentials have become an essential tool for atomistic simulations, yielding results close to ab initio simulations at a fraction of computational cost. With recent improvements on the achievable accuracies, the focus has now shifted on the data set composition itself. The reliable identification of erroneously predicted configurations to extend a given data set is therefore of high priority. Yet, uncertainty estimation techniques have achieved mixed results for machine learning potentials. Consequently, a general and versatile method to correlate energy or atomic force uncertainties with the model error has remained elusive to date. In the current work, we show that epistemic uncertainty cannot correlate with model error by definition but can be aggregated over groups of atoms to yield a strong correlation. We demonstrate that our method correctly estimates prediction errors both globally per structure and locally resolved per atom. The direct correlation of local uncertainty and local error is used to design an active learning framework based on identifying local subregions of a large simulation cell and performing ab initio calculations only for the subregion subsequently. We successfully utilized this method to perform active learning in the low-data regime for liquid water.

Kiphynet: an online network simulation tool connecting cellular kinetics and physiological transport.

Human metabolism is sustained by functional networks that operate at diverse scales. Capturing local and global dynamics in the human body by hierarchically bridging multi-scale functional networks is a major challenge in physiological modeling. To develop an interactive, user-friendly web application that facilitates the simulation and visualization of advection-dispersion transport in three-dimensional (3D) microvascular networks, biochemical exchange, and metabolic reactions in the tissue layer surrounding the vasculature. To help modelers combine and simulate biochemical processes occurring at multiple scales, KiPhyNet deploys our discrete graph-based modeling framework that bridges functional networks existing at diverse scales. KiPhyNet is implemented in Python based on Apache web server using MATLAB as the simulator engine. KiPhyNet provides the functionality to assimilate multi-omics data from clinical and experimental studies as well as vascular data from imaging studies to investigate the role of structural changes in vascular topology on the functional response of the tissue. With the network topology, its biophysical attributes, values of initial and boundary conditions, parameterized kinetic constants, biochemical species-specific transport properties such as diffusivity as inputs, a user can use our application to simulate and view the simulation results. The results of steady-state velocity and pressure fields and dynamic concentration fields can be interactively examined. KiPhyNet provides barrier-free access to perform time-course simulation experiments by building multi-scale models of microvascular networks in physiology, using a discrete modeling framework. KiPhyNet is freely accessible at http://pallab.cds.iisc.ac.in/kiphynet/ and the documentation is available at https://deepamahm.github.io/kiphynet_docs/ .

FAST (fast analytical simulator of tracer)-PET: an accurate and efficient PET analytical simulation tool

Objective. Simulation of positron emission tomography (PET) images is an essential tool in the development and validation of quantitative imaging workflows and advanced image processing pipelines. Existing Monte Carlo or analytical PET simulators often compromise on either efficiency or accuracy. We aim to develop and validate fast analytical simulator of tracer (FAST)-PET, a novel analytical framework, to simulate PET images accurately and efficiently. Approach. FAST-PET simulates PET images by performing precise forward projection, scatter, and random estimation that match the scanner geometry and statistics. Although the same process should be applicable to other scanner models, we focus on the Siemens Biograph Vision-600 in this work. Calibration and validation of FAST-PET were performed through comparison with an experimental scan of a National Electrical Manufacturers Association (NEMA) Image Quality (IQ) phantom. Further validation was conducted between FAST-PET and Geant4 Application for Tomographic Emission (GATE) quantitatively in clinical image simulations in terms of intensity-based and texture-based features and task-based tumor segmentation. Main results. According to the NEMA IQ phantom simulation, FAST-PET’s simulated images exhibited partial volume effects and noise levels comparable to experimental images, with a relative bias of the recovery coefficient RC within 10% for all spheres and a coefficient of variation for the background region within 6% across various acquisition times. FAST-PET generated clinical PET images exhibit high quantitative accuracy and texture comparable to GATE (correlation coefficients of all features over 0.95) but with ∼100-fold lower computation time. The tumor segmentation masks comparison between both methods exhibited significant overlap and shape similarity with high concordance CCC > 0.97 across measures. Significance. FAST-PET generated PET images with high quantitative accuracy comparable to GATE, making it ideal for applications requiring extensive PET image simulations such as virtual imaging trials, and the development and validation of image processing pipelines.

In situ simulation in the intensive care unit: A phenomenological study of staff experiences.

To explore the experiences of clinical and non-clinical staff in an intensive care unit regarding the perceived benefits and drawbacks of using in situ simulation as a training tool. A descriptive phenomenological qualitative study was conducted among clinical and nonclinical ICU personnel. Simulations and interviews were conducted until data saturation was achieved. The interviews were recorded, transcribed verbatim for analysis, and interpreted using the Colaizzi method. Ten participant interviews generated data saturation. ISS was found to be feasible and beneficial in the ICU, facilitating experiential and emotion-based learning in real-world environments. Eight result categories were identified: simulation benefits, simulation benefits in real conditions, scenario authenticity, interference with usual work, ISS sessions, high-fidelity generating affective bonding, ISS as knowledge reinforcement, and recommendations for improvement. The fundamental structure revealed that ISS is perceived as an authentic and emotionally impactful team simulation modality that promotes experiential learning, reflection, and care improvement opportunities within the complex sociotechnical system of the ICU. All interviewees considered ISS to be a feasible simulation tool that should be implemented in the ICU to improve knowledge and skills, thereby enhancing teamwork.

A Systematic Literature Review on AC Microgrids

The objective of this work is to analyze and compare AC microgrid (ACMG) solutions to introduce the topic to new researchers. The methodology used to achieve this goal is a systematic literature review using five questions: (1) How have ACMGs evolved in five years? (2) What are the standards for ACMGs? (3) What are the different schemes for connecting MGs to the utility grid? (4) What are the different control schemes in ACMGs? (5) What is an appropriate way to compare results when working with ACMGs? The articles were published in Q1/Q2 journals as based on either the Scimago Journal Rank (SJR) and/or the Journal Citation Report (JCR) between 2018 and 2022 and were from three databases: (1) Web of Science (WoS), (2) Scopus, and (3) IEEE Xplore. Publications not describing pure ACMGs, review papers, publications not related to the questions, and papers describing work that did not meet a quality assessment were excluded, resulting in 34 articles being included in this review. Results show: (1) the energy sources and AC bus nature of microgrids over five years, (2) the identification and quantification of cited standards for microgrids, (3) the pros and cons of different schemes for connecting an AC microgrid to the main grid, (4) the control schemes, classified in a hierarchical control structure, and (5) the simulation tools and experimental benches used in microgrids. Most studies considered a generic energy source and a low-voltage three-phase AC bus, 16 standards were found, and the most cited standard was IEEE Standard 1547. The most common connection scheme to the utility grid was a direct connection, most of the works proposed a modification to a hierarchical control system scheme, and the most common simulation tool was MATLAB. The preferred experimental setup consisted of parallel inverters for testing a control scheme, a prototype when proposing a power electronic system, and a laboratory microgrid for testing fault detection methods.

Empirically measuring online social influence

Social influence pervades our everyday lives and lays the foundation for complex social phenomena, such as the spread of misinformation and the polarization of communities. A disconnect appears between psychology approaches, generally performed and tested in controlled lab experiments, and quantitative methods, which are usually data-driven and rely on network and event analysis. The former are slow, expensive to deploy, and typically do not generalize well to topical issues; the latter often oversimplify the complexities of social influence and ignore psychosocial literature. This work bridges this gap by introducing a human-in-the-loop active learning method that empirically quantifies social influence by crowdsourcing pairwise influence comparisons. We develop simulation and fitting tools, allowing us to estimate the required budget based on the design features and the worker’s decision accuracy. We perform a series of pilot studies to quantify the impact of design features on worker accuracy. We deploy our method to estimate the influence ranking of 500 X/Twitter users. We validate our measure by showing that the obtained empirical influence is tightly linked with agency and communion, the Big Two of social cognition, with agency being the most important dimension for influence formation.

Monte Carlo simulation of initial and volume ion recombination correction factor in plane parallel ionization chambers exposed to ion beams

Objective. Accurate reference dosimetry with ionization chambers (ICs) relies on correcting for various influencing factors, including ion recombination. Theoretical frameworks, such as the Boag and Jaffe theories, are conventionally used to describe the ion recombination correction factors. Experimental methods are time consuming, the applicability may be limited and, in some cases, impractical to be used in clinical routine. The development of simulation tools becomes necessary to enhance the understanding of recombination under circumstances that may differ from conventional use. Before progressing, it is crucial to benchmark novel approaches to calculate ion recombination losses under known conditions. In this study, we introduce and validate a versatile simulation tool based on a Monte Carlo scheme for calculating initial and volume ion recombination correction factors in air-filled ICs exposed to ion beams with clinical dose rates. Approach. The simulation includes gaussian distribution of ion positions to model the distribution of charge carriers along the chamber volume. It accounts for various physical transport effects, including drift, diffusion, space charge screening and free electron fraction. To compute ion recombination, a Monte Carlo scheme is used due to its versatility in multiple geometries, without exhibiting convergence problems associated with numerically solved procedures. Main results. The code is validated in conventional dose rates against Jaffe’s theory for initial recombination and Boag’s theory for volume recombination based on parameters derived from experimental data including proton, helium and carbon ion beams measured with a plane parallel IC. Significance. The simulation demonstrates excellent agreement, typically 0.05% or less relative difference with the theoretical and experimental data. The current code successfully predicts ion recombination correction factors, in a large variety of ion beams, including different temporal beam structures.

DNA tetrahedral nanocages as a promising nanocarrier for dopamine delivery in neurological disorders.

Dopamine is a neurotransmitter in the central nervous system that is essential for many bodily and mental processes, and a lack of it can cause Parkinson's disease. DNA tetrahedral (TD) nanocages are promising in bio-nanotechnology, especially as a nanocarrier. TD is highly programmable, biocompatible, and capable of cell differentiation and proliferation. It also has tissue and blood-brain barrier permeability, making it a powerful tool that could overcome potential barriers in treating neurological disorders. In this study, we used DNA TD as a carrier for dopamine to cells and zebrafish embryos. We investigated the mechanism of complexation between TD and dopamine hydrochloride using gel electrophoresis, fluorescence and circular dichroism (CD) spectroscopy, atomic force microscopy (AFM), and molecular dynamic (MD) simulation tools. Further, we demonstrate that these dopamine-loaded DNA TD nanostructures enhanced cellular uptake and differentiation ability in SH-SY5Y neuroblastoma cells. Furthermore, we extended the study to zebrafish embryos as a model organism to examine survival and uptake. The research provides valuable insights into the complexation mechanism and cellular uptake of dopamine-loaded DNA tetrahedral nanostructures, paving the way for further advancements in nanomedicine for Parkinson's disease and other neurological disorders.

The Addition of Pore Holes in Paving Blocks in an Effort to Increase Infiltration

The problem of waterlogging and surface runoff on impermeable pavement surfaces requires finding new ways to manage water flow, especially from rainwater. The type of permeable pavement that is often used is paving block. The use of paving blocks for pavement materials is expected to reduce puddles that occur after rain. One way is to modify the paving block to be greater in seeping water into the soil. This study aims to determine the extent to which the effect of adding pore holes in paving block can increase water infiltration in surrounding soil. This research was conducted through testing in the Laboratory. The type of research is quantitative with experimental method using a rain simulator tool with variations of paving block. From the research results obtained that the addition of pores in the paving block affects the infiltration that occurs. This is evidenced by the increae in the value of infiltration, the infiltration rate value of normal paving block is 182.92-10.43 mm/hour, 2% ratio is 183.75-10.75 mm/hour, 5% ratio is 190.00-11.30 mm/hour, 7% ratio is 231.33-11.46 mm/hour, 10% ratio is 236.88-17.61 mm/hour. From the test results between pore holes with infiltration rate has a significant relationship of 0.037 (<0,05).

An adaptive numerical method for multi–cellular simulations of tissue development and maintenance

In recent years, multi–cellular models, where cells are represented as individual interacting entities, are becoming ever popular. This has led to a proliferation of novel methods and simulation tools. The first aim of this paper is to review the numerical methods utilised by multi–cellular modelling tools and to demonstrate which numerical methods are appropriate for simulations of tissue and organ development and disease. The second aim is to introduce an adaptive time–stepping algorithm and to demonstrate it’s efficiency and accuracy. We focus on off–lattice, mechanics based, models where cell movement is defined by a series of first order ordinary differential equations, derived by assuming over–damped motion and balancing forces. We see that many numerical methods have been used, ranging from simple Forward Euler approaches through to higher order single–step methods like Runge–Kutta 4 and multi–step methods like Adams–Bashforth 2. Through a series of exemplar multi–cellular simulations, we see that if: care is taken to have events (births deaths and re–meshing/re–arrangements) occur on common time–steps; and boundaries are imposed on all sub–steps of numerical methods or implemented using forces, then all numerical methods can converge with the correct order. We introduce an adaptive time–stepping method and demonstrate that the best compromise between L∞ error and run–time is to use Runge–Kutta 4 with an increased time–step and moderate adaptivity. We see that a judicious choice of numerical method can speed the simulation up by a factor of 10–60 from the Forward Euler methods seen in Osborne et al. (2017), and a further speed up by a factor of 4 can be achieved by using an adaptive time–step.

Enhancing the streamflow simulation of a process-based hydrological model using machine learning and multi-source data

Streamflow simulation is crucial for flood mitigation, ecological protection, and water resource planning. Process-based hydrological models and machine learning algorithms are the mainstream tools for streamflow simulation. However, their inherent limitations, such as time-consuming and large data requirements, make achieving high-precision simulations challenging. This study developed a hybrid approach to simultaneously improve the accuracy and computational efficiency of streamflow simulation, which integrates Block-wise use of the TOPMODEL (BTOP) model into the eXtreme Gradient Boosting (XGBoost), i.e., BTOP_XGB. In this approach, BTOP generates simulated streamflow using the Latin hypercube sampling algorithm instead of the time-consuming calibration algorithms to reduce computational costs. Then, XGBoost combines BTOP simulated streamflow with multi-source data to reduce simulation errors. In which, serval input variable selection algorithms are employed to choose relevant inputs and remove redundant information for model. The hybrid approach is validated and compared with a standalone model at three hydrological stations in the Jialing River basin, China. The results show that the performance of BTOP_XGB is significantly better than the BTOP and XGBoost models. The NSE of BTOP_XGB at Beibei, Xiaoheba, and Luoduxi stations increases by 54%, 21%, and 83%, respectively. Meanwhile, the computational time of BTOP_XGB is saved by >90% compared to the original calibrated BTOP. BTOP_XGB is less affected by parameter sample sizes and data amounts, demonstrating the robustness of the hybrid model. This study simplifies the complexity of the hydrological model and enhances the stability of machine learning, jointly improving the reliability of streamflow simulation. The hybrid approach provides a potential shortcut for streamflow simulation over basins with large areas or limited observed data.

Energy efficient improved content addressable memory using quantum-dot cellular automata

Quantum-dot cellular automata (QCA) is an emerging technology with high integration density, low power consumption, and high operating speed. This study introduces a QCA-based modified content addressable memory (CAM) cell employing a five-input minority gate. The functionality, temperature sensitivity, and heat distribution of this modified CAM cell are comprehensively analyzed using QCADesigner E and QCA Pro simulation tools. The results reveal significant advancements over existing designs, with a remarkable 8.33% reduction in area and a substantial 63.7% decrease in energy consumption. Additionally, this modified CAM cell exhibits a notable 5% enhancement in temperature tolerance. These findings emphasize the QCA-based modified CAM cell is more efficient and thermally robust.

CPOpt: A modular framework for genetic algorithm optimization and post-optimization analysis in complex charged particle optical design

This article reports on the development of an optimization framework for charged particle optics design (CPOpt). The optimization part of CPOpt involves a custom-developed Python multi-objective genetic algorithm (MOGA), which incorporates NSGA-II to optimize for multiple objectives, such as particle transmission and beam spot size. The simultaneous tuning of different types of parameters, such as electrode voltages and geometrical parameters, is facilitated by creating isolated groups of parameters. Modeling of parameter configurations is done with the charged particle optics simulation tool SIMION and a ZeroMQ-based communication interface enables a fully automated interaction with the MOGA. The post-optimization analysis part of CPOpt provides standardized and broadly applicable methods to identify suitable solutions, based on the outcome of the optimization part. Global sensitivity analysis, based on the Delta Moment-Independent-Measure in combination with dedicated Latin hypercube sampling, allows higher-level information about parameter sensitivity and robustness to be extracted, as well as effective randomized local optimization. Demonstrated on a two-lens system with intermediate aperture, the CPOpt framework effectively optimized the two conflicting objectives of reduced beam spot size and increased transmission at the detector plane for 12 parameters, covering both electrode voltages and geometrical parameters. The optimization process was performed within less than 11h on a workstation, while all the simulations required for the post-optimization analysis took another 7h. The additional computational effort for the post-optimization analysis can contribute to deliver solutions with high performance and stability.

Optimizing EV charging stations: a simulation-based approach to performance and grid integration

This study addresses the optimization of electric vehicle (EV) charging stations, focusing on enhancing performance and grid integration through a comprehensive simulation approach. By employing advanced simulation tools in Simulink® and MATLAB®, alongside electrical installation planning with SIMARIS®, we meticulously analyze the charging process, infrastructure requirements, and their implications on the power grid. Our results demonstrate significant improvements in charging station efficiency and reliability, highlighting the effectiveness of our proposed control strategies and harmonic mitigation techniques. Notably, the integration of renewable energy sources emerges as a pivotal factor in reducing operational costs and carbon emissions, furthering the sustainability of EV charging solutions. The research delineates the environmental benefits, emphasizing the reduction of greenhouse gas emissions and enhancement of urban air quality, pivotal in the global shift towards cleaner transportation modes. This work contributes valuable insights into the design and grid integration of EV charging stations, offering a scalable model for future infrastructure development. It serves as a critical resource for engineers, policymakers, and stakeholders in the realm of electric mobility, advocating for a strategic transition to EVs supported by robust and efficient charging infrastructure.

A new miniaturized Split-Ring Resonator-Based inverse double Sigma-Shaped artificial material for Penta-Band wireless communication

This paper discusses a study on creating and confirming the authenticity of an artificial material with a SNG (Singular Negative) attribute. The artificial material was constructed using a metamaterial structured around split-ring resonators forming an inverse double sigma shape. Advanced electromagnetic simulation tools like CST (Computer Simulation Technology) Microwave Studio, Ansys HFSS (High Frequency Structure Simulator) and ADS (Advanced Design System) were employed to extensively analyze the electromagnetic characteristics of the artificial material. Notably, the proposed artificial material exhibits five distinct resonance frequencies spanning L-, S-, C-, X-, and Ku-band operations, specifically at 1.560 GHz, 2.232 GHz, 4.350 GHz, 8.095 GHz, and 17.370 GHz. Achieving an optimal EMR (Effective Medium Ratio) of 23.45, with a compact unit cell size of 8.2 mm × 8.2 mm and a substrate thickness of 1.905 mm, underscores its innovative design attributes. Of significance, the compact size, SNG characteristics, high EMR, and resonant frequencies render this structure particularly noteworthy. The utilization of the L-band for GPS (Global Positioning System), S-band for Wireless LAN (Local Area Network), C-band for Satellite communication, X-band radar system, and the Ku-band for VSAT (Very Small Aperture Terminal) communication applications is well-documented. This article elucidates the intricacies of the design process and conducts various parametric analyses. Remarkably, the simulation results obtained from Ansys HFSS 3D software closely align with those from CST. Furthermore, a comprehensive examination of surface current, E-field, and H-field distributions is provided. Importantly, comparative assessments indicate the superior performance of the proposed structure across multiple metrics, including EMR, SNG frequency range, and compactness, positioning it as an ideal candidate cellular phone, satellite, and radar-based applications.

A Comprehensive Survey On Machine Learning-Based Intrusion Detection System for Vehicular Area Network Architectures

Today, the development of communication technologies causes changes in many different areas. One of these areas is VANET (Vehicular Area Network) application area. With the increase in usage areas in the VANET field, ensuring VANET network security has become more critical. Many different systems have been developed to detect attacks on VANET networks. Machine learning-based systems are one of the most widely used methods in developing these intrusion detection systems (IDS). In this article, research on machine learning-based VANET IDS, which has been done recently in the literature, has been carried out. First, VANET architecture and security requirements are presented, then a comprehensive literature summary is given, and comparisons are made on different parameters. As a result, it has been determined that many different machine learning models are used in IDSs and perform high-performance detection. In addition to the machine learning algorithm used in the performance of IDSs, it has been shown that many different parameters play a critical role in the performance. The paper aims to guide new studies in this field with the gains that will increase the performance of intrusion detection systems because of the literature comparison (considering criteria such as machine learning model, simulation tools, dataset, machine learning algorithm, and performance criteria).

Boosting the efficiency of an inverted structure halide perovskite solar cell: a numerical investigation

Abstract 
Inverted solar cells have attracted significant attention because they have low hysteresis and are resistant to environmental variables, such as oxygen and humidity, making them more stable and long-lasting. Herein, we investigate the performance optimisation of an inverted design based on MAPbI3-xClx with the structure ITO/PTAA/MAPbI3-xClx/PC60BM/BCP/Ag by utilising SCAPS-1D, a simulation tool.
Accordingly, a substantial improvement in efficiency can be achieved by optimising several factors linked to each layer's performance in a perovskite solar cell (PSC).
Total defect density, work function, thickness, and electron affinity have a significant impact on the photovoltaic response. Specifically, the impact of optimisation of the charge transport layers and the perovskite layer on the device's performance parameters was discussed, resulting in a milestone within a remarkable increase in PCE of 21.59%, whereas the original structure's efficiency was 6.9 %.
Additionally, it has been shown that aluminium can substitute silver in the top electrode of a solar cell without affecting its efficiency, allowing the development of cost-effective solar cells.
 The present study provides an insight in the creation of a highly stable
low-cost and higher-efficiency perovskite solar cell.

Positron Emission Projection Imaging: A technique for concentration field measurements in opaque industrial systems

A novel Positron Emission Projection Imaging (PEPI) algorithm designed to compute the plane-projected spatial distribution of radiolabelled materials without the need for collimation is introduced. By leveraging improved data efficiency, we have achieved a technique with enhanced spatial resolution and temporal resolution compared to previous PEPI algorithms. Validation of this algorithm was conducted using synthetic data generated from a digital twin of a PET scanner, demonstrating its accuracy for practical applications. The industrial advantage of this novel algorithm was applied in the imaging of laminar flow mixing within a ploughshare mixer, with the experimental results compared against those obtained from validated computational fluid dynamics (CFD) models. This comparison highlights an important use case for PEPI as a robust validation tool for CFD simulations, crucial for enhancing industrial processes. PEPI, which uses deeply penetrating gamma-photons, is now capable of imaging opaque fluids and solids in industrial casing. Future directions for this work include further algorithmic refinements and expanding its application across various industrial systems, establishing PEPI as a robust tool for in-depth industrial process analysis. The advancements presented here allow for optimized mixer design and enhanced process efficiency, extending the frontiers of tomographic imaging in industrial applications.

Self-normalization for a 1-mm3resolution clinical PET system using deep learning.

Normalization in positron emission tomography (PET) corrects for non-uniformity of sensitivity across all system lines of response (LOR). Self-normalization is a framework that aims to estimate normalization components from the emission data without a separate scan of a normalization phantom. In this work, we propose for the first time an image-based end-to-end self-normalization framework using conditional generative adversarial networks (cGAN). We evaluated different approaches by exploring each of the following three methodologies. First, we used images that were either unnormalized or corrected for geometric factors, which encompass all time-invariant factors, as input data types. Second, we set the input tensor shape as either a single axial slice (2-D) or three contiguous axial slices (2.5-D). Third, we chose either Pix2Pix or polarized self-attention (PSA) Pix2Pix, which we developed for this work, as a deep learning network. The targets for all approaches were the axial slices of images normalized using the direct normalization method. We performed Monte Carlo simulations of ten voxelized phantoms with the SimSET simulation tool and produced 26,000 pairs of axial image slices for training and testing. The results showed that 2.5-D PSA Pix2Pix trained with geometric-factors-corrected input images achieved the best performance among all the methods we tested. All approaches improved general image quality figures of merit peak signal to noise ratio (PSNR) and structural similarity index (SSIM) from ~15% to ~55%, and 2.5-D PSA Pix2Pix showed the highest PSNR (28.074) and SSIM (0.921). Lesion detectability, measured with region of interest (ROI) PSNR, SSIM, normalized contrast recovery coefficient (NCRC), and contrast-to-noise ratio (CNR), was generally improved for all approaches, and 2.5-D PSA Pix2Pix trained with geometric-factors-corrected input images achieved the highest ROI PSNR (28.920) and SSIM (0.973).

Advancing Offshore Renewable Energy: Integrative Approaches in Floating Offshore Wind Turbine-Oscillating Water Column Systems Using Artificial Intelligence-Driven Regressive Modeling and Proportional-Integral-Derivative Control

This research investigates the integration of Floating Offshore Wind Turbines (FOWTs) with Oscillating Water Columns (OWCs) to enhance sustainable energy generation, focusing on addressing dynamic complexities and uncertainties inherent in such systems. The novelty of this study lies in its dual approach, which integrates regressive modeling with an aero-hydro-elasto-servo-mooring coupled system with a deep data-driven network and implements a proportional-integral-derivative (PID) control mechanism to improve system stability. By employing Artificial Neural Networks (ANNs), the study circumvents the challenges of real-time closed-loop control on FOWT structures using the OpenFAST simulation tool. Data-driven models, trained on OpenFAST datasets, facilitate real-time predictive behavior analysis and decision-making. Advanced computational learning techniques, particularly ANNs, accurately replicate the dynamics of FOWT-OWC numerical models. An intelligent PID control mechanism is subsequently applied to mitigate structural vibrations, ensuring effective control. A comparative analysis with traditional barge-based FOWT systems underscores the enhanced modeling and control methodologies’ effectiveness. In this sense, the experimental results demonstrate substantial reductions in the mean oscillation amplitude, with reductions from 5% to 35% observed across various scenarios. Specifically, at a wave period from 20 s and a wind speed of 5 m/s, the fore-aft displacement was reduced by 35%, exemplifying the PID control system’s robustness and efficacy under diverse conditions. This study highlights the potential of ANN-driven modeling as an alternative to managing the complex non-linear dynamics of NREL 5 MW FOWT models and underscores the significant improvements in system stability through tailored PID gain scheduling across various operational scenarios.