In recent years, climate change and the corresponding expected extreme weather conditions have been widely recognized as potential problems. The building industry is taking various actions to achieve sustainable development, implement energy conservation strategies, provide mitigation. addition mitigation, it crucial adapt change, investigate possible risks limitations of mitigation strategies. Although importance adaptation well-understood, there are still challenges in understanding...

Climate change can strongly affect renewable energy production. The state of the art in projecting future generation has focused on using regional climate prediction. However, prediction is characterized by inherent uncertainty due to complexity models. This work provides a comprehensive study quantify impact uncertainties potential over five zones Europe. Thirteen scenarios, including global models (GCMs) and three representative concentration pathways (RCPs), are downscaled RCA4 model...

Abstract We propose a new regularisation strategy for the classical ensemble Kalman inversion (EKI) framework. The consists of: (i) an adaptive choice parameter in update formula EKI, and (ii) criteria early stopping of scheme. In contrast to existing approaches, our does not rely on additional tuning parameters which often have severe effects efficiency EKI. motivate approach using interpretation EKI as Gaussian approximation Bayesian tempering setting inverse problems. show that controls...

In this paper, an accurate and computationally implicit 3D finite-difference time-domain (FDTD) method based on the unconditionally stable Crank–Nicolson scheme (3D CN-FDTD) is presented. The source excitation in CN-FDTD described numerical simulation of demonstrated through examples. results method, ADI-FDTD traditional FDTD schemes are compared. A good agreement obtained for with time steps greatly more than Courant–Friedrich–Levy (CFL) limit Yee method. © 2006 Wiley Periodicals, Inc....

Abstract A gradient method for optimization of shape and materials radar cross section (RCS) reduction is derived from Maxwell's equations. The uses the adjoint problem finds derivatives RCS with respect to all design parameters a single solution scattering problem. tested in two‐dimensional test problems minimize specified angular intervals at frequency, configurations strongly reduced small number iterations. In absence other constraints, optimal shapes perfectly electrically conducting...

An unconditionally stable Crank–Nicolson finite-difference time-domain (CN-FDTD) algorithm is presented for three-dimensional microwave circuit analysis. First, Mur's first–order absorbing boundary condition applied this CN-FDTD algorithm. A symmetric successive over relaxation–preconditioned biconjugate-gradient also proposed to solve the large sparse matrix equation obtained in method. Resonant cavity and several planar microstrip circuits are illustrate versatility of technique. Numerical...

Analytical expressions are derived for the derivatives of radar cross sections (RCS) with respect to scatterer shape in two and three dimensions. Using adjoint formulation, RCS, an arbitrary number design parameters, found from a single solution scattering problem. In dimensions, gradient-based optimizations converge small iterations shapes strongly reduced RCS specified angular intervals. The optimal have sharp corners directions where is minimized exhibit corrugations, which can be...

In this paper, a 3D alternating-direction implicit finite-difference time-domain (ADI-FDTD) algorithm is used to analyze planar circuits. The source excitation in ADI-FDTD described and two kinds of absorbing boundary conditions (ABCs) are jointly employed for algorithm: the Gedney's uniaxial PML scheme applied propagation direction, Mur's 1st-order condition (ABC) set on other outer surfaces. Both waveforms S-parameters presented. numerical simulations several examples show that number...

Abstract This paper presents a new combination of the finite‐difference time‐domain (FDTD) method and least‐squares support vector machines (LS‐SVM) technique. The LS‐SVM is statistical‐learning which has self‐contained basis theory excellent learning performance. A short segment an FDTD record used to train predictor in order obtain accurate future realization. Numerical simulations for two typical microwave filters demonstrate that can achieve good forecasting accuracy efficiency be...

Design sensitivities for microwave problems derived in continuum form are applied to the shape and material parameters optimization. In particular, we optimize several scatterers' radar cross section (RCS) a T-junction's reflection coefficient. The optimizations converge, small number of iterations, designs highly improved performance. Both frequency- time-domain electromagnetic field solvers tested compared optimization RCS.

Abstract The increase of the time step size significantly deteriorates property coefficient matrix generated from Crank‐Nicolson finite‐difference time‐domain (CN‐FDTD) method. As a result, convergence classical iterative methods, such as generalized minimal residual method (GMRES) would be substantially slowed down. To address this issue, article mainly concerns efficient computation large sparse linear equations using preconditioned (PGMRES) Some typical preconditioning techniques, Jacobi...

Abstract The combination of Cartesian grid and the adaptive mesh refinement (AMR) technology is an effective way to handle complex geometry solve flow problems. Some high-efficiency Cartesian-based AMR libraries have been developed dynamic changes in parallel but still can not meet unique requirements simulating around objects. In this paper, we propose efficient generation method information transmission approach for wall boundary parallelize implementation ghost-cell (GCM). Also,...

In this paper, a set of parallelised adaptive hierarchical Cartesian-based immersed boundary methodology is developed for high Reynolds number compressible flow. First, robust and efficient grid generation method based on the separation axis theorem arbitrary geometry presented automatic Cartesian generation. Second, an (IBM) coupling with wall model Third, parallel strategy implemented special treatment proposed to guarantee large-scale computing. Finally, cell-based mesh refinement (AMR)...

Several preconditioning techniques, such as Jacobi, SSOR and ILU0, are used to accelerate the convergence of iterative methods, CG GMRES, which solve large system linear equations resulted from time-domain finite-element methods (TDFEM). Convergence properties these conventional techniques compared. The cylindrical cavity partially filled with dielectric rod perfectly matched layer terminated waveguides simulated. Numerical results show that ILU0 preconditioned method converges fastest in...

Abstract Based on the finite‐element approximation and nonoverlapping domain decomposition, an efficient tearing interconnecting algorithm of time‐domain method is presented. Using edge‐based finite element, divides computation into several subdomains computes electric fields in each subdomain by solving vector curl–curl wave equation. Adjacent are related to other Dirichlet Neumann boundary condition. By use repetition quasiperiod structures, further reduces memory requirement CPU time. The...

Fine-tuning pretrained large models to downstream tasks is an important problem, which however suffers from huge memory overhead due large-scale parameters. This work strives reduce in fine-tuning perspectives of activation function and layer normalization. To this end, we propose the Approximate Backpropagation (Approx-BP) theory, provides theoretical feasibility decoupling forward backward passes. We apply our Approx-BP theory backpropagation training derive memory-efficient alternatives...

Designing building form in urban areas is a complicated process that demands considering high number of influencing parameters. On the other hand, there has been an increasing trend to design highly fenestrated envelopes for office buildings induce higher levels natural lighting into workspace. This paper presents novel optimization framework high-performance and fenestration configuration impacts microclimate typical extreme weather conditions during thirty-year period climate data...

Many cities are striving to develop urban transformation strategies, in order transit from traditional city a sustainable city. Improving the energy efficiency of existing buildings is key address climate change mitigation and adaptation. This paper considers different scenarios using series future four European cities, namely, Kiruna, Stockholm, Valencia Madrid. The study adopts TABULA database get access basic construction information. Energy simulation data analysis IDA ICE MATLAB...

A new Impedance Boundary Condition (IBC) for two dimensional Finite Difference Time Domain simulations containing thin, good conductor sheets is presented. The IBC uses a recursive convolution scheme based on approximating the conductor's impedance as sum of exponentials. effects FDTD parameters such grid size and time step simulation accuracy are verification performed by comparing quality factors rectangular resonant structures determined analytical methods. shown to accurately model loss...