Computational Time Requirements Research Articles

Reducing neural network complexity via optimization algorithms for fault diagnosis in renewable energy systems

This article discusses the challenges involved in diagnosing faults in renewable energy systems. A significant challenge is the high computational demands of the artificial intelligence algorithms needed for grid-connected photovoltaic (GCPV) and wind energy conversion (WEC) systems in fault diagnosis. To address this issue, several methods are proposed to reduce computation time, minimize memory requirements, and improve efficiency. First, optimization algorithms such as the Salp Swarm Algorithm (SSA), Genetic Algorithm (GA), and Particle Swarm Optimization (PSO), combined with machine learning classifiers for feature selection, are suggested to reduce computational time and memory space requirements. This approach notably decreases computation time but has a limited impact on memory space. Secondly, further reduction in memory space requirements is recommended by using the variogram method for data reduction. This method can leverage the outputs of preceding algorithms to optimize renewable energy systems, making their operations cost-effective and efficient. Finally, the outputs of the variogram are used to train neural network (NN), recurrent neural network (RNN), and long short-term memory (LSTM) classifiers to differentiate between various modes of operation in GCPV and WEC systems. Experimental results demonstrate the effectiveness and robustness of the proposed methods, showing that memory space can be reduced by 1.3 to 5.6 times and CPU time by 1.1 to 11 times while maintaining accuracy and improving efficiency compared to conventional algorithms.

Simultaneous design optimisation methodology for floating offshore wind turbine substructure and feedback-based control strategy

This research article explores the application of control co-design methodologies for optimising floating offshore wind turbine systems concurrently. The primary objective is to offer insights into concurrent design approaches employing an advanced genetic optimisation algorithm. To achieve this, a reduced-order dynamic model is employed to minimise computational time requirements, complemented by a modified version of the levelized cost of energy equation serving as the cost function. Furthermore, various optimisation scenarios are investigated under diverse wind and wave conditions to assess the advantages and drawbacks of increasing the complexity of dynamic cases used in evaluating the cost function. The optimised system designs are then compared against baseline floating system designs to underscore the advantages of employing this approach to floating wind turbine design.

Dual Domain Decomposition Method for High-Resolution 3D Simulation of Groundwater Flow and Transport

The high-resolution 3D groundwater flow and transport simulation problem requires massive discrete linear systems to be solved, leading to significant computational time and memory requirements. The domain decomposition method is a promising technique that facilitates the parallelization of problems with minimal communication overhead by dividing the computation domain into multiple subdomains. However, directly utilizing a domain decomposition scheme to solve massive linear systems becomes impractical due to the bottleneck in algebraic operations required to coordinate the results of subdomains. In this paper, we propose a two-level domain decomposition method, named dual-domain decomposition, to efficiently solve the massive discrete linear systems in high-resolution 3D groundwater simulations. The first level of domain decomposition partitions the linear system problem into independent linear sub-problems across multiple subdomains, enabling parallel solutions with significantly reduced complexity. The second level introduces a domain decomposition preconditioner to solve the linear system, known as the Schur system, used to coordinate results from subdomains across their boundaries. This additional level of decomposition parallelizes the preconditioning of the Schur system, addressing the bottleneck of the Schur system solution while improving its convergence rates. The dual-domain decomposition method facilitates the partition and distribution of the computation to be solved into independent finely grained computational subdomains, substantially reducing both computational and memory complexity. We demonstrate the scalability of our proposed method through its application to a high-resolution 3D simulation of chromium contaminant transport in groundwater. Our results indicate that our method outperforms both the vanilla domain decomposition method and the algebraic multigrid preconditioned method in terms of runtime, achieving up to 8.617× and 5.515× speedups, respectively, in solving massive problems with approximately 108 million degrees of freedom. Therefore, we recommend its effectiveness and reliability for high-resolution 3D simulations of groundwater flow and transport.

An iterative approach to grid topology and redispatch optimization in congestion management

The selection of Remedial Actions (RA) to ensure system security is a highly complex task performed by Transmission System Operators (TSOs). Phase shifting transformer (PST) tap changes and active power changes of generation units (redispatch) are some RA available in Security Constrained Optimal Power Flow (SCOPF) simulations within the operational planning processes. Topological RA are not part of these SCOPF yet, as the optimization thereof adds high complexity to the existing optimization problem. To overcome this complexity, this paper introduces an iterative approach that decouples topology optimization from redispatch & PST optimization. Linearized models of RA are used to meet computation time requirements. Exemplary investigations of the presented method were performed based on modified IEEE 39-Bus, 118-Bus and PEGASE-1354-Bus grid models. Within the scope of these investigations, the method shows good results and a great potential for the reduction of congestions and required redispatch through topological RA. In order to eliminate inaccuracies of the approach and to further improve its suitability for use in grid operation, a need for future investigations and possible further developments were identified.

Fitting Monte Carlo simulation results with an empirical model of megavoltage x-ray beams for rapid depth dose calculations in water

Objective. The purpose of this study was to assess a method of accelerating Monte Carlo simulations for modeling depth dose distributions from megavoltage x-ray beams by fitting them to an empirically-derived function. Approach. Using Geant4, multiple simulations of a typical medical linear accelerator beam in water and in water with an air cavity were conducted with varying numbers of initial electrons. The resulting percent depth dose curves were compared to published data from actual linear accelerator measurements. Two methods were employed to reduce computation time for this modeling process. First, an empirical function derived from measurements at a particular linear accelerator energy, source-to-surface distance, and field size was used to directly fit the simulated data. Second, a linear regression was performed to predict the empirical function’s parameters for simulations with more initial electrons. Main results. Fitting simulated depth dose curves with the empirical function yielded significant improvements in either accuracy or computation time, corresponding to the two methods described. When compared to published measurements, the maximum error for the largest simulation was 5.58%, which was reduced to 2.01% with the best fit of the function. Fitting the empirical function around the air cavity heterogeneity resulted in errors less than 2.5% at the interfaces. The linear regression prediction modestly improved the same simulation with a maximum error of 4.22%, while reducing the required computation time from 66.53 h to 43.75 h. Significance. This study demonstrates the effective use of empirical functions to expedite Monte Carlo simulations for a range of applications from radiation protection to food sterilization. These results are particularly impactful in radiation therapy treatment planning, where time and accuracy are especially valuable. Employing these methods may improve patient outcomes by ensuring that dose delivery more accurately matches the prescription or by shortening the preparation time before treatment in Monte Carlo-based treatment planning systems.

Implementation and optimization of Burg’s method for real-time packet loss concealment in networked music performance applications

AbstractIn networked music performance (NMP) applications, which entail real-time audio streaming over the Internet, strict latency requirements are needed to ensure a realistic interaction between geographically dispersed musicians. Thus, NMP applications typically leverage uncompressed audio and unreliable transport protocols to avoid unnecessary processing and re-transmission delays. Given that no guarantee on packet delivery is offered, NMP applications must deal with late/lost audio packets to mitigate the impact of the resulting audio artifacts on the quality of the playback audio stream. This paper explores an audio packet loss concealment (PLC) technique based on autoregressive (AR) models. In particular, it investigates the algorithmic implementation of Burg’s method and the parameters configuration that offers the best trade-off between prediction error and computational time requirements. The purpose is to find the most suitable solution capable of running on a Raspberry Pi 4B within the real-time audio boundaries imposed by NMP applications. Additionally, we analyze the computational time required to fit the model and predict future samples by considering six implementations and various compilation flags. Results confirm that AR models can predict future audio samples more accurately than traditional PLC approaches, which consist of filling audio gaps with silence or repeating the last received audio segment. Furthermore, results demonstrate the effectiveness of the proposed solution in meeting the strict latency requirements when deployed on a Raspberry Pi 4B.

Towards efficient control synthesis for nonlinear wave energy conversion systems: impedance-matching meets the spectral-domain

Existing studies within the literature that focus on designing parametric energy-maximizing controllers for Wave Energy Converter (WEC) systems predominantly rely on the impedance-matching (IM) principle, originally developed for linear time-invariant systems. Alternatively, iterative optimization routines are commonly employed for nonlinear WECs. However, these approaches often face a trade-off between effectiveness in maximizing energy extraction and computational efficiency. To address this limitation, this study proposes a computationally efficient controller tuning method for analogous synthesis in the case of nonlinear WECs. The proposed approach combines a statistical linearization technique known as spectral-domain modeling with the IM principle, to synthesize a Proportional–Integrative (PI) controller for a nonlinear WEC. Furthermore, a comparison is performed with two other synthesis methods: one based on a standard (i.e. linear) frequency-domain representation of the WEC that incorporates the IM principle, and the other employing a gradient-free optimization routine applied to the nonlinear time-domain model of the WEC for PI parameter tuning through exhaustive numerical search. A discussion on the effectiveness of each tuning method in maximizing energy absorption is provided, including an appraisal of their associated computational time requirements. Numerical analyses demonstrate that the proposed method, which integrates spectral-domain modeling and IM, can achieve (almost) optimal PI controller design for a nonlinear WEC. Furthermore, this study addresses the inaccuracies inherent in the frequency-domain approach and significantly reduces the computational time compared to the exhaustive search procedure. The findings of this research represent a significant advancement towards the development of simple, effective, and efficient IM-based techniques for synthesis of controllers in nonlinear WEC systems

Comparison of three machine learning algorithms for classification of B-cell neoplasms using clinical flow cytometry data.

Multiparameter flow cytometry data is visually inspected by expert personnel as part of standard clinical disease diagnosis practice. This is a demanding and costly process, and recent research has demonstrated that it is possible to utilize artificial intelligence (AI) algorithms to assist in the interpretive process. Here we report our examination of three previously published machine learning methods for classification of flow cytometry data and apply these to a B-cell neoplasm dataset to obtain predicted disease subtypes. Each of the examined methods classifies samples according to specific disease categories using ungated flow cytometry data. We compare and contrast the three algorithms with respect to their architectures, and we report the multiclass classification accuracies and relative required computation times. Despite different architectures, two of the methods, flowCat and EnsembleCNN, had similarly good accuracies with relatively fast computational times. We note a speed advantage for EnsembleCNN, particularly in the case of addition of training data and retraining of the classifier.

On the Accurate Estimation of Information-Theoretic Quantities from Multi-Dimensional Sample Data.

Using information-theoretic quantities in practical applications with continuous data is often hindered by the fact that probability density functions need to be estimated in higher dimensions, which can become unreliable or even computationally unfeasible. To make these useful quantities more accessible, alternative approaches such as binned frequencies using histograms and k-nearest neighbors (k-NN) have been proposed. However, a systematic comparison of the applicability of these methods has been lacking. We wish to fill this gap by comparing kernel-density-based estimation (KDE) with these two alternatives in carefully designed synthetic test cases. Specifically, we wish to estimate the information-theoretic quantities: entropy, Kullback-Leibler divergence, and mutual information, from sample data. As a reference, the results are compared to closed-form solutions or numerical integrals. We generate samples from distributions of various shapes in dimensions ranging from one to ten. We evaluate the estimators' performance as a function of sample size, distribution characteristics, and chosen hyperparameters. We further compare the required computation time and specific implementation challenges. Notably, k-NN estimation tends to outperform other methods, considering algorithmic implementation, computational efficiency, and estimation accuracy, especially with sufficient data. This study provides valuable insights into the strengths and limitations of the different estimation methods for information-theoretic quantities. It also highlights the significance of considering the characteristics of the data, as well as the targeted information-theoretic quantity when selecting an appropriate estimation technique. These findings will assist scientists and practitioners in choosing the most suitable method, considering their specific application and available data. We have collected the compared estimation methods in a ready-to-use open-source Python 3 toolbox and, thereby, hope to promote the use of information-theoretic quantities by researchers and practitioners to evaluate the information in data and models in various disciplines.

Low fidelity data driven machine learning based optimisation method for box-wing configuration

Wing design optimization traditionally involves computationally expensive high-fidelity simulations, limiting the exploration of design spaces. In this study, we propose a methodology that combines low-fidelity numerical models with machine learning algorithms to efficiently navigate the complex parameter space of box-wing configurations. Through the utilisation of a surrogate model trained on a limited dataset derived from low-fidelity simulations, our method strives to predict results within an acceptable range, significantly curtailing computational costs and time. The effectiveness of this methodology is demonstrated through a series of case studies, involving the Onera M6 and NASA CRM wing as test cases and Bionica box-wing optimization as an application case. The initial application of the proposed methodology to the box-wing case successfully achieved an almost 9.82 % increase in overall aerodynamic efficiency. Its competitive performance compared to conventional optimization methods, along with its substantial reduction in computational time and resource requirements, is evident. This efficient methodology holds promise for enhancing the design optimization process for aviation start-ups by efficiently exploring complex design spaces with reduced computational burden.

A comparison of immersed boundary and body-fitted methods: Transient vortex shedding in elaborate bio-mimetic cylinders

In this study, two contrasting computational fluid dynamics (CFD) methods are used to simulate flow around bio-mimetic cylinders. Cylindrical shapes are proposed to represent a cross-sectional view of complex geometries, such as cylinders with biofouling or cylinders with helical strakes. The non-conformal Cartesian immersed boundary method (IBM) and conformal curvilinear body-fitted method (BFM) are applied to solve the governing equations. A critical assessment of the accuracy and efficiency of IBM is presented by employing an open-source solver with similar discretization, algorithms, and properties. It has been found that IBM can reduce computation time by up to 49% for protrusions with different numbers of sides. Results show that IBM can reduce required computation time up to an outstanding 71% for protrusions with different heights. Results also indicate that transient duration doesn't significantly change the reduction trend in required computation time. For this reason, the onset of vortex shedding and eventually fully-developed flow takes place substantially sooner in IBM simulation than in BFM. It is also found that the height and shape of protrusions affect BFM performance but have no effect on IBM performance. This study proves that non-conformal IBM is a robust numerical approach when simulating vortex shedding in complex geometries.

Finite element boundary element numerical simulation study on acoustic vibration coupling problems

Excessive structural resonance response in the acoustic vibration coupling environment can cause adverse consequences such as structural damage, local instability, and component failure. Accurately predicting the vibration response of instruments under external sound fields helps to avoid structural performance degradation or damage caused by vibration, and ensures the reliability and safety of the system, which is of great significance for the effective load of instrument equipment. This paper constructs a finite element boundary element equation for the acoustic vibration coupling problem, and adopts a series of optimization measures for it. In response to the low computational efficiency of finite element boundary element equations and their inability to meet the computational requirements of complex models, this paper introduces the acoustic vibration reciprocity theorem and the fast directional boundary element method to compensate for the aforementioned shortcomings. In addition, the study used the weighted residual method to construct a data exchange algorithm to ensure energy conservation during the transfer process. The research focuses on typical structures of spacecraft and conducts simulation analysis and application research on acoustic vibration coupling problems. The simulation showcases that the introduction of the reciprocity theorem markedly reduces the calculation time and memory occupied by the equation, with a required calculation time of 1.23 h and a data file size of approximately 299.56 MB generated by the calculation; The overall error of the data exchange process is below 10−4. Scanning frequency analysis is conducted on the surface and local openings of the model structure, and it is found that the displacement of the selected excitation point is in good agreement with the analytical solution. This study effectively reduces simulation time and memory requirements through the proposed strategy, ensures energy conservation data exchange algorithms, improves the reliability and safety of structural design, and has important practical value for evaluating the structural performance of key equipment such as spacecraft.

Advancements and Future Directions in the Application of Machine Learning to AC Optimal Power Flow: A Critical Review

Optimal power flow (OPF) is a crucial tool in the operation and planning of modern power systems. However, as power system optimization shifts towards larger-scale frameworks, and with the growing integration of distributed generations, the computational time and memory requirements of solving the alternating current (AC) OPF problems can increase exponentially with system size, posing computational challenges. In recent years, machine learning (ML) has demonstrated notable advantages in efficient computation and has been extensively applied to tackle OPF challenges. This paper presents five commonly employed OPF transformation techniques that leverage ML, offering a critical overview of the latest applications of advanced ML in solving OPF problems. The future directions in the application of machine learning to AC OPF are also discussed.

Gaussian process regression approach for predicting wave attenuation through rigid vegetation

Numerical modelling in the coastal environment often requires highly skilled users and can be hindered by high computation costs and time requirements. Machine Learning (ML) techniques have the potential to overcome these limitations and complement existing methods. This is an exploratory investigation utilising a Gaussian Process (GP) data-driven modelling approach that can reproduce, for the given range of conditions in this study, the results of a widely used process-based model, XBeachX, when applied to the challenging problem of wave attenuation through vegetation. This study utilises efficient sampling strategies for data exploration, providing a valuable framework for future studies. The GP model was trained on a synthetic dataset generated using the numerical model XBeachX, which was calibrated using laboratory measurements. Our findings indicate that well-trained ML models can strongly complement traditional modelling approaches, especially in an environment where data sources are increasingly available. We have also explored the underlying interactions of the GP model's input features and their relationship to the model's output through a sensitivity analysis.

An efficient quantum algorithm for ensemble classification using bagging

AbstractEnsemble methods aggregate predictions from multiple models, typically demonstrating improved accuracy and reduced variance compared to individual classifiers. However, they often come with significant memory usage and computational time requirements. A novel quantum algorithm that leverages quantum superposition, entanglement, and interference to construct an ensemble of classification models using bagging as an aggregation strategy is introduced. Through the generation of numerous quantum trajectories in superposition, the authors achieve B transformations of the training set with only operations, allowing an exponential enlargement of the ensemble size while linearly increasing the depth of the corresponding circuit. Moreover, when assessing the algorithm's overall cost, the authors demonstrate that the training of a single weak classifier contributes additively to the overall time complexity, as opposed to the multiplicative impact commonly observed in classical ensemble methods. To illustrate the efficacy of the authors’ approach, experiments on reduced real‐world datasets utilising the IBM qiskit environment to demonstrate the functionality and performance of the proposed algorithm are introduced.

Ab Initio Path Integral Monte Carlo Simulations of the Uniform Electron Gas on Large Length Scales.

The accurate description of non-ideal quantum many-body systems is of prime importance for a host of applications within physics, quantum chemistry, materials science, and related disciplines. At finite temperatures, the gold standard is given by ab initio path integral Monte Carlo (PIMC) simulations, which do not require any empirical input but exhibit an exponential increase in the required computation time for Fermionic systems with an increase in system size N. Very recently, computing Fermionic properties without this bottleneck based on PIMC simulations of fictitious identical particles has been suggested. In our work, we use this technique to perform very large (N ≤ 1000) PIMC simulations of the warm dense electron gas and demonstrate that it is capable of providing a highly accurate description of the investigated properties, i.e., the static structure factor, the static density response function, and the local field correction, over the entire range of length scales.

Population genetic simulation: Benchmarking frameworks for non-standard models of natural selection.

Population genetic simulation has emerged as a common tool for investigating increasingly complex evolutionary and demographic models. Software capable of handling high-level model complexity has recently been developed, and the advancement of tree sequence recording now allows simulations to merge the efficiency and genealogical insight of coalescent simulations with the flexibility of forward simulations. However, frameworks utilizing these features have not yet been compared and benchmarked. Here, we evaluate various simulation workflows using the coalescent simulator msprime and the forward simulator SLiM, to assess resource efficiency and determine an optimal simulation framework. Three aspects were evaluated: (1) the burn-in, to establish an equilibrium level of neutral diversity in the population; (2) the forward simulation, in which temporally fluctuating selection is acting; and (3) the final computation of summary statistics. We provide typical memory and computation time requirements for each step. We find that the fastest framework, a combination of coalescent and forward simulation with tree sequence recording, increases simulation speed by over twenty times compared to classical forward simulations without tree sequence recording, although it does require six times more memory. Overall, using efficient simulation workflows can lead to a substantial improvement when modelling complex evolutionary scenarios-although the optimal framework ultimately depends on the available computational resources.

An efficient adaptive h‐refinement for the RWG‐based electric field integral equation applied to antennas with waveport

AbstractAn approximation scheme that combines approximate solutions with error estimators to accelerate the adaptive h‐refinement of the RWG‐based method of moments (MoM) with the electric field integral equation (EFIE) is presented. The scheme is effectively implemented by using the near‐field matrix to solve the approximate solution. Several conventional error estimators incorporated with the scheme are investigated. Compared to using the full MoM system, it significantly reduces the computational cost while maintaining the refinement effect. Numerical results indicate that the scheme could accelerate the adaptive refinement, yielding significant reductions in computational time and memory requirements.

A proposed PMU-based voltage stability and critical bus detection method using artificial neural network

Voltage stability detection is currently still becoming the main issue in the modern integrated renewable energy power systems. To assess the voltage stability, the classical methods based on continuation power flow (CPF) technique were used to show nose curve. However, the classical methods require complete model of power system and long computation time. Data driven analysis and synchronized real time measurement technologies currently are developing in power systems monitoring, including the stability detection. The detection method is built based on the historical event model and uses the real time measurement as an input. For that reason, the algorithm to detect the voltage instability and critical bus is proposed using the artificial neural network (ANN) technique to represent the historical event model using the PMU measurement data. The ANN model architecture for this application is developed by creating seven hidden layers consisting of one normalization, four rectifier linear unit, one softmax and one sigmoid layer. To warrant the accuracy, the k-fold cross-validation is used. The algorithm is simulated on modified IEEE 14 test system which consider different loading scenario, line contingency, number of PMU and Photovoltaic (PV) integration. To mimic the actual historical data, the synthetic data is generated and labelled. The result shows that the proposed method can represent the complete power system model by giving high accuracy which for voltage stability detection is > 97% and critical buses detection is > 96% for all scenarios. Moreover, the required computation time is between 16 and 18 s per detection which makes the scalability to the real time detection is reasonable.

Convolutional neural network hyperparameters for face emotion recognition using genetic algorithm

The development of artificial intelligence in facial emotion recognition (FER) is rapidly growing and has been widely applied in various fields. Deep learning (DL) techniques with evolutionary algorithms have become the preferred choice for solving various security, health, gaming, and other related problems. This research proposes the use of a genetic algorithm (GA) as the main method to optimize hyperparameters in the convolutional neural network (CNN) model for FER. The required computation time is approximately 37 hours 57 minutes 55 seconds, with generation 3 taking the longest time at around 16 hours 45 minutes 4 seconds. However, generation 3 achieved an accuracy of 76.11%, which is the highest compared to other generations. The results indicate that the more generations are involved, the higher the achievable accuracy. Furthermore, the proposed CNN-GA model in this study outperforms previous models that have been examined. Thus, this study makes a significant contribution to improving the understanding of using GAs to optimize the performance of CNN models for FER.