Monitoring ships and their motion states is vital for the marine economy. Traditional methods focus on directly detecting ship hulls but face significant challenges with smaller vessels or those with camouflage patterns on their surfaces. In contrast, detecting ship wakes offers a promising alternative. Ship wakes contain essential information about a ship’s heading and velocity, persist for extended periods, and are easier to observe. By examining the characteristics of wakes, we effectively overcome the limitations of direct detection methods. Kelvin wakes, characterized by well-established theoretical models, are frequently employed in this article. In addition, the polarization imaging technology provides advantages such as antiscattering capabilities, reduced glare on the water surface, and enhanced detection of weak targets, making it particularly effective for wake detection applications. This article illustrates the benefits of polarization imaging in analyzing ship wakes, using the polarimetric bidirectional reflection distribution function (pBRDF) as a framework. We model Kelvin wakes and simulate images of wakes for various ship motion states. Furthermore, we capture polarized visible light images of ship wakes, creating a novel dataset. We develop the KelvinPointNet model, based on key point detection of wake skeletons, to extract critical features of Kelvin wakes and facilitate the inversion of ship velocity and heading. Experimental results show that our algorithm achieves a key point detection accuracy of 92.7%. By employing the Kelvin transverse wave inversion method, we attain a velocity inversion error of less than 10% and a heading inversion error of less than 2.5%, underscoring the substantial implications of our research.

Investigating process monitoring techniques is required to reduce the loss of property and life caused by industrial processes accidents. Fault diagnosis, which attempts to determine the fault type, is a vital step in process monitoring because it can help operators respond to abnormal situations appropriately. Adequate data labels to train supervised fault diagnosis models are difficult to acquire in practice; however, semisupervised methods, which are attracting increasing attention, can use unlabeled data. Self-labeled algorithms are an effective paradigm of semisupervised methods, but their applications in industrial process fault diagnosis do not meet expectations, because they are prone to performance deterioration when handling industrial process data. To address this issue, in this article, a self-training algorithm with a modified confidence measure is proposed. The confidence measure is temporal-spatial with temporal identities of data introduced to its definition and calculation, which makes the algorithm adaptable to industrial processes. The proposed algorithm is also self-adaptive to avoid time-consuming hyperparameter tuning processes. The benchmark Tennessee Eastman process data were used to evaluate the proposed algorithm, and the experiment results demonstrate its superiority compared to competing semisupervised methods.

In this paper, we present a finite difference method (FDM) based on layered and adaptive meshing to extract substrate resistance. To allow adaptive meshing, a novel synchronization method is introduced which expresses the potential at other points than the centroids of panels by modifying the finite difference equations. This method makes it possible to overcome the high computational cost of the FDM due to tight uniform meshing requirements while enjoying a straightforward implementation and easy handling of irregular/arbitrary substrate structures to extract the substrate coupling of integrated circuits.

While research work on fast integral equation solver has resulted in several algorithms of similar linear equation solving performance, it has been well observed that the convergence of capacitance versus discretization is rather slow due to the most commonly used first-order collocation or Galerkin methods. This paper reports a new high-order scheme, Quadrature-evaluated Method of Moments Solver (QMMS), that uses high-order weighting and Gaussian quadrature to optimally handle the singularities at the edges and corners. For practical interconnect extraction problems, singularities at edges are analytically known and the corresponding Gaussian nodes allocation affords a near optimal discretization scheme. The new formulation avoids the special position-dependent quadrature rules, and surprisingly, provides a well-behaved matrix that converges rapidly without any pre-conditioning. Combining the high-order scheme with a kernel-independent fast solver yields an efficient algorithm for 3-D capacitance extraction.

In tegral equation approaches based on layered media Green's functions are often used to extract models of integrated circuit structures. The primary advan tage of these approaches over equiv alen t-sourcebased schemes is the dramatic reduction in problem size. When combined with an SVD-accelerated sc heme for the solution of the associated dense linear system, this leads to a substantial speedup.

The Method of Moments (MoM) is often effectively used inthe extraction of passive components in modeling integratedcircuits and MCM packaging. MoM extraction, however,involves solving a dense system of linear equations, and using direct factorization methods can be prohibitive for large problems. In this paper, we present a Fast Method of Moments Solver (FMMS) for the rapid solution of such linear systems. Our algorithm exploits the fact that the integral equation kernels are locally smooth and can be dramatically compressed via the singular value decomposition(SVD). This greatly speeds up the matrix-vector products ina Krylov-subspace iterative algorithm (e.g., GMRES). Wedemonstrate the efficiency and exibility of our scheme forthe modeling of embedded inductors in MCM-D. Results arepresented to show that the method isaccurate and can betwo orders of magnitude faster than Gaussian eliminationand one order of magnitude faster than standard iterative schemes.