Explore the vast range of titles available.

Novel hyperspectral indices, which are the first derivative normalized difference nitrogen index (FD-NDNI) and the first derivative ratio nitrogen vegetation index (FD-SRNI), were developed to estimate the leaf nitrogen content (LNC) of wheat. The field stress experiments were conducted with different nitrogen and water application rates across the growing season of wheat and 190 measurements were collected on canopy spectra and LNC under various treatments. The inversion models were constructed based on the dataset to evaluate the ability of various spectral indices to estimate LNC. A comparative analysis showed that the model accuracies of FD-NDNI and FD-SRNI were higher than those of other commonly used hyperspectral indices including mNDVI705, mSR, and NDVI705, which was indicated by higher R2 and lower root mean square error (RMSE) values. The least squares support vector regression (LS-SVR) and random forest regression (RFR) algorithms were then used to optimize the models constructed by FD-NDNI and FD-SRNI. The p-R2 values of the FD-NDNI_RFR and FD-SRNI_RFR models reached 0.874 and 0.872, respectively, which were higher than those of the exponential and SVR model and indicated that the RFR model was accurate. Using the RFR inversion model, remote sensing mapping for the Operative Modular Imaging Spectrometer (OMIS) image was accomplished. The remote sensing mapping of the OMIS image yielded an accuracy of R2 = 0.721 and RMSE = 0.540 for FD-NDNI and R2 = 0.720 and RMSE = 0.495 for FD-SRNI, which indicates that the similarity between the inversion value and the measured value was high. The results show that the new hyperspectral indices, i.e., FD-NDNI and FD-SRNI, are the optimal hyperspectral indices for estimating LNC and that the RFR algorithm is the preferred modeling method.

Complex subsurface structures exhibit significant anisotropic characteristics, making multi-parameter imaging techniques important for achieving a more comprehensive geological interpretation. Full-waveform inversion (FWI) as a state-of-the-art method for reconstructing subsurface properties based on seismic wavefield modeling and data misfit minimization has been widely applied to isotropic media in both synthetic and field datasets. However, challenges such as crosstalk correlation and inaccuracy of the initial model indicate that further advancements are required to enhance resolution and computational efficiency. We propose an elastic FWI in the frequency domain for two-dimensional (2D) TI media to characterize their physical properties appropriately, as they are common in sedimentary basin environments. Different from traditional inversion schemes, our approach is formulated based on Bayesian inference, which automatically facilitates uncertainty analysis of the inversion results. Seismic data are acquired via the integral equation (IE) method grounded in scattering theory, where the sensitivity kernel is explicitly constructed using Green’s functions, hence facilitating the calculation of gradient and Hessian. A Krylov subspace iterative method provides the approximated solution of the Lippmann-Schwinger (L-S) equation without sacrificing the accuracy. Furthermore, we incorporate the minimum support (MS) stabilizing functional as a model misfit term to regularize the objective function. A randomized singular value decomposition (SVD) approach is used to approximate and decompose the prior preconditioned Hessian. Both the model and covariance are updated through the iterative extended Kalman filter (IEKF) that implemented in the form of the Levenberg–Marquardt (LM) algorithm, thereby enabling practical uncertainty quantification. Numerical tests are conducted on two synthetic TI models with vertical and tilted symmetry axes, respectively, illustrating the precision and robustness of our method.

Amplitude-preserving depth-domain prestack inversion using the point-spread function (PSF) has emerged as a powerful approach for quantitative reservoir characterization. This technique employs the PSF to approximate the Hessian operator, illumination-induced blurring is compensated during imaging, thereby improving amplitude fidelity and spatial resolution. However, most existing depth-domain inversion studies have focused on poststack applications. Systematic investigations of prestack inversion—particularly for quantitative fluid characterization—remain limited. To address this limitation, we develop a depth-domain prestack inversion framework that leverages an angle-domain Gaussian-beam PSF. Under high-frequency asymptotic assumptions, we derive an analytical expression for the angle-domain Gaussian-beam PSF and compute PP-wave PSFs to approximate the local Hessian, which we incorporate directly into the inversion operator as a local-Hessian preconditioner. The framework combines a nonstationary convolution model with the Aki–Richards approximation to simultaneously invert for elastic parameters in the depth domain. Applied to a 2D marine streamer line from the Northern Viking Graben in the North Sea, the method reveals multiple low-Vp/Vs anomalies within Paleocene and Jurassic sandstones that correspond to hydrocarbon-bearing intervals identified from well data. A depth-domain ϕw (water-filled porosity) section is then constructed through a well-log-calibrated Vp/Vs–ϕw relationship established for this study area, which effectively discriminates potential hydrocarbon reservoirs. These results demonstrate that the angle-domain PSF–based depthdomain prestack inversion exhibits robust applicability and geological consistency in structurally complex rift basins, providing a novel pathway for quantitative depth-domain reservoir characterization.

The mechanical performance of nanocrystalline membranes plays a critical role in determining the reliability and stability of advanced integrated circuit devices and nano‐electromechanical systems. Conventional characterization techniques such as nano‐indentation and micro‐scale mechanical testing, while widely used, are generally destructive and incapable of resolving three‐dimensional (3D) anisotropic properties. Targeting nanocrystalline membranes, we introduce an ultrafast optoacoustics‐based approach for characterizing their 3D anisotropic elastic constants and thickness simultaneously. Gigahertz Lamb waves are thermoelastically generated to propagate in nanocrystalline copper membranes. Both non‐propagating zero‐group‐velocity resonances and propagating modes are captured in high spatial and temporal resolution. The resonance and dispersion characteristics are employed to determine the thickness and anisotropic elastic constants of the membrane via a nontrivial multi‐parameter inversion algorithm. Accordingly, substantially different elastic properties are observed by changing the substrate, especially the shear stiffness, which could be a result of the underlying microstructural mechanisms. The developed ultrafast optoacoustic approach provides a fully non‐destructive and in situ metrology tool for characterizing parameters such as 3D anisotropic elastic constants and thickness of freestanding membranes. This capability not only enables a deeper understanding of the mechanisms of nanocrystalline materials but also facilitates improvements in the design and fabrication of reliable, high‐performance next‐generation micro‐ and nano‐devices.

The finite element numerical simulation results of deep pit deformation are greatly influenced by soil layer parameters, which are crucial in determining the accuracy of deformation prediction results. This study employs the orthogonal experimental design to determine the combinations of various soil layer parameters in deep pits. Displacement values at specific measurement points were calculated using PLAXIS 3D under these varying parameter combinations to generate training samples. The nonlinear mapping ability of the Back Propagation (BP) neural network and Particle Swarm Optimization (PSO) were used for sample global optimization. Combining these with actual onsite measurements, we inversely calculate soil layer parameter values to update the input parameters for PLAXIS 3D. This allows us to conduct dynamic deformation prediction studies throughout the entire excavation process of deep pits. The results indicate that the use of the PSO-BP neural network for inverting soil layer parameters effectively enhances the convergence speed of the BP neural network model and avoids the issue of easily falling into local optimal solutions. The use of PLAXIS 3D to simulate the excavation process of the pit accurately reflects the dynamic changes in the displacement of the retaining structure, and the numerical simulation results show good agreement with the measured values. By updating the model parameters in real-time and calculating the pile displacement under different working conditions, the absolute errors between the measured and simulated values of pile top vertical displacement and pile body maximum horizontal displacement can be effectively reduced. This suggests that inverting soil layer parameters using measured values from working conditions is a feasible method for dynamically predicting the excavation process of the pit. The research results have some reference value for the selection of soil layer parameters in similar areas.

In this paper, improved physics-informed neural networks (PINNs) with hard constraints (PINNs-H) are introduced to simulate the variation of the excess pore water pressure in the nonlinear consolidation problems with continuous drainage boundary conditions. In the PINNs-H, we modify the network architecture to automatically satisfy the corresponding initial and boundary conditions accurately, and obtain high-precision soil consolidation behaviors. The accuracy and effectiveness of the presented PINNs-H are demonstrated on two examples of the nonlinear consolidation models. Specifically, the results indicate that based on less training data, we may better predict the consolidation behaviors through the PINNs-H. Furthermore, the training data required by the PINNs-H is significantly less than the grid point data of the finite difference method (FDM), and the PINNs-H exhibits a better memory advantage. For the inverse problem, we find that on the basis of less observed data of the excess pore water pressure, the PINNs can provide a great estimate to the interface parameters of the continuous drainage boundary conditions, and effectively resist the noise interference. We also use the PINNs and PINNs-H to identify the nonlinear factor, and reveal that PINNs-H can provide high-precision predicted results, whereas the PINNs fail.

The river pollutant traceability problem represents a critical challenge in environmental monitoring and water resource management. In this work, we propose an approach based on a physics‐informed neural network (PINN) for identifying key parameters of pollutant sources, including the release intensity and location, on the basis of cross‐sectional observations. The accuracy of the proposed method was validated through experiments conducted on steady, unsteady, and noisy unsteady flows, with numerical simulations of real‐world river systems as test cases. The results demonstrate that the method can not only accurately identify source parameters beyond the gauging river reach but also effectively capture the spatial distribution of pollutants across the entire computational domain. This approach provides a novel solution for addressing the challenges of tracing pollutant sources in river channels.

Accurate simulation of vibration signals is essential for fault detection in power equipment such as gas-insulated switchgear (GIS) and transformers. The Finite Element Method (FEM) is commonly employed for high-fidelity simulations, but its precision heavily depends on exact physical parameters such as the damping coefficient. These parameters are difficult to measure directly, and existing methods based on empirical values are both time-consuming and often inaccurate. This paper proposes a method using Physics-Informed Neural Networks (PINNs) combined with experimental data to accurately invert damping parameters in vibration systems, exemplified by GIS. PINNs integrate physical laws into neural networks, improving accuracy, robustness, and generalization. By combining experimental data with PINNs, high precision and interpretability of key physical parameters in FEM simulations are achieved. The results show that without noise, the waveform similarity between the FEM and the experimental results is high, with an amplitude similarity coefficient of 0.869 and a normalized cross-correlation of 0.926. At the 1% noise level, the inversion error of the damping parameter is only 3%, and the method shows good noise resistance up to the 5% noise level. This approach improves simulation reliability and provides a new path to enhance the transparency and diagnostic capabilities of power equipment.

The thyristor level is the basic unit of ultra-high-voltage and extra-high-voltage direct current (DC) converter valves, and its main-circuit parameters are important indicators for characterizing the health status of converter valves. To meet the demand for efficient detection of converter valve thyristor levels, this paper proposes a parameter inversion method for converter valve thyristor levels by combining the time-frequency-domain features of valve voltage and current, temporal characteristics of feedback signals from the thyristor-level monitoring unit, and a Grey Wolf Optimizer–Backpropagation Neural Network (GWO-BPNN). First, a six-pulse converter valve circuit simulation model is established. Based on this model, the original dataset is generated using the Latin hypercube sampling (LHS) method. Wavelet packet decomposition is then used to extract time-frequency-domain features, and dimensionality reduction is carried out by comparing the coefficient of variation and explained variance ratio so as to obtain input data suitable for neural network training. A BP neural network is then trained, and the network parameters are optimized using the Grey Wolf Optimizer to improve the accuracy and convergence speed of parameter inversion. Simulation comparison results show that the GWO-BP method is more efficient than the state equation method and is suitable for efficient inversion of damping parameters in multi-level thyristor systems. After GWO optimization, the maximum inversion errors of both parameters are reduced to below 5%. Compared with BP, GA-BP, and PSO-BP, the proposed GWO-BP model provides the best overall balance between resistance-inversion accuracy and training efficiency. By further incorporating feedback feature signals, the inversion error can be reduced to 1%. The proposed method provides a new technical route for efficient detection of thyristor converter valves and has broad application prospects.

It is difficult for remolded loess to reproduce the infiltration law of water in natural loess, resulting in inaccurate parameters of the Van-Genuchten model (VG model) obtained by inversion, which subsequently affects the changes in the moisture field during slope stability numerical simulations. To address this gap, a large-scale undisturbed loess soil column with a diameter of 60 cm and a height of 100 cm was artificially excavated for a water infiltration test under ponding conditions. Based on the test results, the VG model parameters were inverted by the Hydrus-1D software. The simulation accuracy was evaluated using the root mean square error (RMSE) and the Nash-Sutcliffe efficiency coefficient (NSE). Based on the inverted VG model parameters, the single-factor perturbation method was employed to calculate different VG model parameters under a disturbance amplitude of ± 20%. The numerical simulations were carried out to investigate the wetting front arrival times at different depths under varying disturbance parameters. Finally, the sensitivity of soil water characteristic parameters was analyzed. The results indicated that the measured water contents in the column closely matched the simulated values, with minimal errors, as evidenced by the RMSE ranging from 0.010 to 0.019. Except for the NSE values at 10 cm and 20 cm depths, which were below 0.900, all other depths exhibited NSE values greater than 0.900, indicating satisfactory simulation performance. The influence of soil water characteristic parameters on simulation results was θs > ks > n > α > θr. The disturbances in θs and α were negatively correlated, while disturbances in ks, n, and θr were positively correlated. Parameter disturbances less influenced the infiltration laws of water in shallow soils. However, as soil column depth increased, sensitivity to parameter changes gradually increased, leading to a greater impact on simulation results. The research findings provide technical support for evaluating and preventing water-induced landslide hazards.