Current airframe health monitoring generally relies on deterministic physics models and ground inspections. This paper uses the concept of a dynamic Bayesian network to build versatile probabilistic model for diagnosis prognosis in order realize digital twin vision, it illustrates proposed method by an aircraft wing fatigue crack growth example. The integrates various aleatory (random) epistemic (lack knowledge) uncertainty sources prediction. In diagnosis, is used track evolution...

Automatic modulation recognition (AMR) is a widely used technique in various communication systems. In this work, we propose complex-valued transformer (CV-TRN) network for AMR. Considering the in-phase (I) and quadrature (Q) components of signal are two consistent data with only phase difference π/2, they can teach independently which disguise augment training data, but I/Q collectively needed to measure similarity multi-head self-attention (MHSA). We input individually into shared...

This paper presents a methodology to quantify the uncertainty in fatigue crack growth prognosis, applied structures with complicated geometry and subjected variable amplitude multi-axial loading. Finite element analysis is used address calculate stress intensity factors. Multi-modal factors due loading are combined an equivalent factor using characteristic plane approach. Crack under modeled modified Paris law that includes retardation effects. During cycle-by-cycle integration of law,...

Detection of high frequency (HF) signal in the wideband is challenging since HF environment chaotic. Recent works adopt deep learning-based object detectors to capture signals spectrogram, but task detection exhibits different characteristics from that generic detection, which causes classical have defects such as limited receptive field and prior anchor mismatch. Based on analysis, this letter proposes a learning framework extracts features along time axis at each bin, predicts multiple...

Fatigue damage to asphalt pavements due continuous loading occurs mainly at the binder–aggregate interface or within binder. The mechanical response of binder under variable stress conditions was comprehensively analyzed by repeated tests. viscoelastic intervals three binders (Pen70–80, Pen60–70, and SBS) were determined scanning tests, two different sizes stresses selected for constant time inside outside based on experimental thresholds, provide a reference selection load combinations...

Adaptive sampling methods have been used to build accurate meta-models across large design spaces from which engineers can explore data trends, investigate optimal designs, study the sensitivity of objectives on modeling features, etc. For global optimization applications, adaptive need be extended sample more efficiently near domains space (i.e., Pareto front/frontier in multi-objective optimization). Expected Improvement (EI) shown efficient solve problems using by incorporating prediction...

Wideband signal detection is an important problem in wireless communication. With the rapid development of deep learning (DL) technology, some DL-based methods are applied to communication and have shown great potential. In this paper, we present a novel neural network for detecting signals classifying types wideband spectrograms. Our utilizes key point estimation locate rough centerline region recognize its class. Then, several regressions carried out obtain properties, including local...

Many engineering design and industrial manufacturing applications are tasked with finding optimum designs while dealing uncertainty in the parameters. The performance or quality of may be sensitive to input variation, making it difficult optimize. Probabilistic robust optimization methods used these scenarios find that will perform best under presence known uncertainty. Robust algorithms often require a two-level problem (double-loop) solution. outer-loop repeatedly calls series inner loops...