Interval model updating is typically performed when gathering data is expensive, time-consuming, or complex and only a limited amount of data is available to perform non-deterministic model updating. In these situatio...
详细信息
ISBN:
(数字)9781624107115
ISBN:
(纸本)9781624107115
Interval model updating is typically performed when gathering data is expensive, time-consuming, or complex and only a limited amount of data is available to perform non-deterministic model updating. In these situations, the fitted intervals will only provide an estimate of the exact interval bounds. This is because the limited data available is unlikely to include any samples that fall precisely on the interval boundaries. In these situations, an analyst could use a metric to assess the accuracy of identified model uncertainties against unseen missing data. Furthermore, when this metric is able to estimate the required amount of data for accurate uncertainty quantification, data-gathering efforts are minimized. This paper defines this metric as the reliability of a data-enclosing set as the probability that future unseen data will fall within the set. Recently, Crespo et al. [1] presented a scenario optimization approach to determine a lower bound for this reliability without having to characterize the underlying distribution of the data generation mechanism. To calculate the reliability, the scenario optimization approach needs the number of hyper-parameters to fit the data enclosing set, the number of samples, and the dimension of the data enclosing set. Once these are obtained, and a confidence level is determined, the approach calculates the lower bound of the reliability. Additionally, analysts can calculate the number of samples required to fit the data enclosing set with predefined lower bound reliability before the measurement campaign. The goals of this paper are to develop the certified interval model updating based on scenario optimization and to apply this to a dynamical modal analysis of a structural finite element model. A four-level building numerical model is used to illustrate the accuracy and the practical application of the developed methodologies.
A combination of piezoelectric drive and synthetic jet technology is proposed for an ultrasonic synthetic jet piezo pump. The piezo vibrator operates silently and has a resonant frequency exceeding 20 kHz. The inlet c...
详细信息
A combination of piezoelectric drive and synthetic jet technology is proposed for an ultrasonic synthetic jet piezo pump. The piezo vibrator operates silently and has a resonant frequency exceeding 20 kHz. The inlet channel of the proposed piezo pump is equipped with a special Tesla structure to increase the flow resistance difference and reduce flow loss. The results demonstrate that the proposed piezo pump has a significant advantage over a typical pump in flow rate. The optimized piezo pump achieves a flow rate of 1037 mL/min when subjected to a typical square wave signal excitation. The optimized waveform enhances the flow rate to 1343.9 mL/min, resulting in a 29.6% improvement. Additionally, this article also investigates the ultrasonic synthetic jet piezo pump in the form of a dual cavity. The two cavities are positioned on either side of the piezo vibrator and alternate between suction and discharge. When the two cavities were connected in parallel, the flow rate was 1912.68 mL/min. Finally, the proposed piezo pump was explored for impact cooling applications. The impact cooling effect on the heated ceramic is good at an impact distance and angle of 30 mm and 60 degrees, respectively.
The transonic buffet is a critical phenomenon that limits the flight envelope of commercial aircraft. For years, RANS-based criteria like the lift-curve-break method have been applied to predict the buffet onset, but ...
详细信息
ISBN:
(数字)9781624107115
ISBN:
(纸本)9781624107115
The transonic buffet is a critical phenomenon that limits the flight envelope of commercial aircraft. For years, RANS-based criteria like the lift-curve-break method have been applied to predict the buffet onset, but it requires flowfield simulations under multiple angles of attack, which is still too time-consuming for optimization. This paper presents a prior-based neural network model to predict pressure profiles under different angles of attack with reference to the one at cruise condition. The model is utilized to replace the offdesign CFD simulations in the lift-curve-break criterion so that only one simulation is needed to predict the buffet onset of an airfoil. It is then employed in a multi-objective genetic algorithm to optimize the buffet onset and the cruise lift-drag ratio simultaneously. To test the optimization procedure, the model is trained on an airfoil database and applied to optimize four airfoils not similar to the training database. The results show that all the optimizations receive positive gains of buffet onset, which affirm that the proposed model and optimization procedure can be reliably employed in search for airfoils with better buffet performance.
Improved Henry gas solubility optimization algorithm is a better way to use an existing method (HGSO) inspired by gas dissolving in liquids, to find optimal solutions. It leverages the influence of pressure and temper...
详细信息
ISBN:
(纸本)9798350365740;9798350365757
Improved Henry gas solubility optimization algorithm is a better way to use an existing method (HGSO) inspired by gas dissolving in liquids, to find optimal solutions. It leverages the influence of pressure and temperature on gas solubility to guide their search for optimal solutions in optimization problems. It aims to overcome the limitations of the original Henry gas solubility optimization algorithm and enhance its performance in solving various optimization problems. It can be used in many applications, such as feature extraction, task scheduling, joint mining, and parameter optimization. This paper proposes a new approach called the beta-Hill operator, which builds upon the traditional hill climbing method. to improve the balance between finding better solutions (exploitation) and exploring new possibilities (exploration).
Effective management of construction project portfolios demands informed decisions driven by data and mathematical models, aiming to enhance decision-making and address complex decision problems. This article introduc...
详细信息
ISBN:
(纸本)9783031782374;9783031782381
Effective management of construction project portfolios demands informed decisions driven by data and mathematical models, aiming to enhance decision-making and address complex decision problems. This article introduces an innovative approach that prioritizes sustainability criteria selection, utilizing data-driven insights and mathematical optimization to optimize project portfolio management. Through multi-objective optimization, our study targets improved resource efficiency, time reduction, and minimized environmental impact, offering portfolio managers a powerful tool for decision-making.
作者：
Nimura, NaruhikoOyama, AkiraUniv Tokyo
Dept Aeronaut & Astronaut 3-1-1 Yoshinodai Sagamihara Kanagawa 2525210 Japan JAXA
Inst Space & Astronaut Sci Dept Space Flight Syst 3-1-1 Yoshinodai Sagamihara Kanagawa 2525210 Japan
A global multiobjective design optimization method is introduced for optimizing three-dimensional geometries based on topology optimization. To validate the method's effectiveness in aerodynamic design optimizatio...
详细信息
ISBN:
(数字)9781624107115
ISBN:
(纸本)9781624107115
A global multiobjective design optimization method is introduced for optimizing three-dimensional geometries based on topology optimization. To validate the method's effectiveness in aerodynamic design optimization problems, design optimization of wingtip design for micro aerial vehicles is conducted The proposed method employs a technique for compressing three-dimensional images using octree encoding to reduce the number of design variables and increase optimization efficiency. The octree structure is optimized through genetic programming. It is confirmed that the proposed method can reproduce a variety of geometry. Compared to the baseline SD7003 rectangular wing, the obtained non-dominated solutions show improvements of up to about 50% in lift coefficient and a reduction of up to about 2% in drag coefficient. The geometry, aerodynamic performance, and surface pressure distribution of the obtained non-dominated solutions revealed that volume, surface area, and positions of cant-up and cant-down geometry play a significant role in designing the wingtip geometry.
Machine learning (ML) algorithms have been widely used in big data prediction and analysis in terms of their excellent data regression ability. However, the prediction accuracy of different ML algorithms varies betwee...
详细信息
Machine learning (ML) algorithms have been widely used in big data prediction and analysis in terms of their excellent data regression ability. However, the prediction accuracy of different ML algorithms varies between different regression problems and data sets. In order to construct a prediction model with optimal accuracy for fly ash concrete (FAC), ML algorithms such as genetic programming (GP), support vector regression (SVR), random forest (RF), extremely gradient boost (XGBoost), backpropagation artificial neural network (BP-ANN) and adaptive network-based fuzzy inference system (ANFIS) were selected as regression and prediction algorithms in this study;the particle swarm optimization (PSO) algorithm was also used to optimize the structure and hyperparameters of each algorithm. The statistical results show that the performance of the assembled algorithms is better than that of an NN-based algorithm. In addition, PSO can effectively improve the prediction accuracy of the ML algorithms. The comprehensive performance of each model is analyzed using a Taylor diagram, and the PSO-XGBoost model has the best comprehensive performance, with R2 and MSE equal to 0.9072 and 11.4546, respectively.
Federated Learning (FL) is an emerging distributed machine learning framework that enables a large number of devices to train machine learning models collaboratively without sharing local data. Despite the extensive p...
详细信息
Federated Learning (FL) is an emerging distributed machine learning framework that enables a large number of devices to train machine learning models collaboratively without sharing local data. Despite the extensive potential of FL, in practical scenarios, different characteristics of clients lead to the presence of different heterogeneity in resources, data distribution, and data quantity, which poses a challenge for the training of FL. To address this problem, in this paper, we first conduct an exhaustive experimental study on all three kinds of heterogeneity in FL and provide insights into the specific impact of heterogeneity on training performance. Subsequently, we propose GridFL, a 3D-grid-based FL framework, where the three kinds of heterogeneity are defined as three dimensions (i.e., dimensions of training speed, data distribution, and data quantity) independently, and all clients in FL training are assigned to corresponding cells of the 3D grid by a gridding algorithm based on K-means clustering. In addition, we propose a grid scheduling algorithm with a dynamic selection strategy, which can select an optimal subset of clients to participate in FL training per round by adopting different strategies for different dimensions and cells. The simulation experiments show that GridFL exhibits superior performance in heterogeneous environments and outperforms several related state-of-the-art FL algorithms. Thus, the effectiveness of the proposed algorithms and strategies in GridFL are verified.
Medical electromagnetic tracking technology offers significant benefits in puncture and interventional surgeries by effectively mitigating obstacles. Traditional optimization-based algorithms for pose estimation in el...
详细信息
Medical electromagnetic tracking technology offers significant benefits in puncture and interventional surgeries by effectively mitigating obstacles. Traditional optimization-based algorithms for pose estimation in electromagnetic localization often converge to local optima and exhibit slow iterative convergence, which limits their accuracy and efficiency. To address these issues, we propose an enhanced pose estimation algorithm that integrates advanced optimization techniques for receiver sensors. The algorithm successfully converged to the global optimal solution in all test cases and no local optimal solution problem occurred in certain volume, which enables rapid real-time tracking of multiple coils simultaneously, outperforming traditional methods. Furthermore, we have developed a novel calibration method for transmission coils that corrects manufacturing-induced errors in size, position, and orientation. These innovations achieved system positional accuracy of 2.64 mm and directional accuracy of 1.33 degrees within a tracking volume of 350 x 350 x 350 mm3.
Based on thermal radiation, radiative cooling is a widely used technology, and plays an important role in many scenarios such as passive building cooling and infrared camouflage. Currently, dynamic control of thermal ...
详细信息
Based on thermal radiation, radiative cooling is a widely used technology, and plays an important role in many scenarios such as passive building cooling and infrared camouflage. Currently, dynamic control of thermal radiation devices is endowed with great expectations. However, a more efficient systematic strategy for designing smart radiant temperature regulators (SRTRs) is still urgently needed. By leveraging the phase change properties of vanadium dioxide, the proposed SRTR achieves dynamic modulation of thermal emissivity, demonstrating an average total directional emissivity exceeding 0.9 when the temperature is above the threshold temperature, and dropping below 0.1 when the threshold temperature is not been reached, particularly within the atmospheric transparent window band. The optimization results achieved by using Autonomous Particles Groups for Particle Swarm optimization demonstrate better performance metrics compared to those obtained using conventional methods. Notably, the absorption mechanism of layered metastructure is further investigated, and its high angle dependency is effectively addressed. The demonstrates excellent angular stability and polarization insensitivity, with up to 73 degrees in transverse electric mode and 60 degrees in transverse magnetic one. Without taking non-radiative heat transfer into account, the cooling power achieves 214.59 W.m(-2). In this paper, a more systematic strategy for designing SRTRs is exhibited, which offers extensive potential in fields such as surface radiative cooling, infrared camouflage, and spacecraft thermal management.
暂无评论