Search

В этой работе рассматривается применение метода согласования мод для исследования неоднородности и цилиндрических штырей в прямоугольном волноводе. Этот метод позволяет уменьшить время моделирования по сравнению с методами создания 3D моделей

Time-series analysis has long been a challenging problem and has been studied extensively over the past decades. In fact, several phenomena possess the dynamic nature of time, with related data collected and expressed in the form of time-series data. While the current development of hardware and software infrastructures provides us a tremendous amount of data to build and validate our models, the noisy and stochastic nature observed in many data modalities still prevent us from having definite solutions. This is especially true for chaotic systems such as the stock market in which the involvement of actors with different goals in the feedback loop leads to complex behaviors. In this thesis, the author proposes a neural network layer design that incorporates the intuitive idea of bilinear mapping to multivariate time-series, as well as an attention module that enables the layer to automatically calculate and focus on important temporal instances. The contribution of the new design is two-fold. Firstly, the proposed layer is highly interpretable thanks to its ability to quantify the contribution of different instances encoding temporal information. In the post-training and inference phase, the attention quantities can be visualized to highlight the time instances of interest, opening up the opportunity for further analysis. Secondly, the new layer design requires both lower memory and fewer computations compared to the popular attention-based Long-Short-Term-Memory design, which is the state-of-the-art solution. In order to validate the proposed architecture, the author has conducted experiments on the problem of stock mid-price movement prediction using information available in Limit Order Book. In the algorithmic trading regime, an automated forecasting system is required to be both accurate and efficient since the market operates on nanosecond resolution. Our experimental results demonstrate that the proposed architecture establishes new state-of-the-art forecasting performances in the problem of interest while running much faster than previously proposed solutions.

Multi-motor drive systems are nonlinear, multi-input multi-output (MIMO) and strong-coupling complicated system, including the effect of friction and elastic, backlash. The control law for this dive system much depend on the determining of the tension. However, it is hard to obtain this tension in practice by using a load cell or a pressure meter due to the accuracy of sensors or external disturbance. In order to solve this problem, a high-gain observer is proposed to estimate the state variables in this drive system, such as speeds, tension. An emerging proposed technique in the control law is the use of high-gain observers together with state feedback control to obtain a separation principle property for the stabilization of whole system. The theory analysis and simulation results point out the good performance of the proposed output feedback for the drive system.

In this study, we investigate and improve the testing methods for more accurate measurement of the early-age autogenous shrinkage (AS) in concrete. During the first 24 h after casting, the temperature of concrete rises due to hydration heat and causes the simultaneous development of thermal deformation (TD) and AS. Because there is no general agreement on evolution of coefficient of thermal expansion (CTE) in the literature, the most commonly adopted approach of subtracting the TD based on an assumed constant value of CTE results in incorrect values of early-age AS. To avoid such errors, we propose improvements in measuring rig and control of hydration-induced temperature, which serves as an aid in achieving isothermal conditions in concrete specimens right after casting. The validity of the proposed improvements is investigated by performing a series of tests under semi-adiabatic (including TD) and artificially controlled isothermal conditions (excluding TD). Results indicate that the AS calculated from ...