Search

Existing macroeconomic forecasting methods primarily focus on the characteristics of economic data, but they overlook the energy-related features concealed behind these economic characteristics, which may lead to inaccurate GDP predictions. Therefore, this paper meticulously analyzes the relationship between energy big data and economic data indicators, explores the coupling feature mining of energy big data and economic data, and constructs features coupling economic and energy data. Targeting the nonlinear variation coupling features in China's quarterly GDP data and using the long short-term memory (LSTM) neural network model based on deep learning, we employ wavelet analysis technology (WA) to decompose selected macroeconomic variables and construct a prediction model combining LSTM and WA, which is further compared with multiple benchmark models. The research findings show that, in terms of quarterly GDP data prediction, the combined deep learning model and wavelet analysis significantly outperform other methods. When processing structurally complex, nonlinear, and multi-variable data, the LSTM and WA combined prediction model demonstrate better generalization capabilities, with its prediction accuracy generally surpassing other benchmark models. [ABSTRACT FROM AUTHOR]

Angiogenesis is one of the most essential developmental processes and plays a key role in organogenesis and tumorigenesis in which epithelial cells proliferate and migrate, thus resulting in sprouting and extension of the existing vasculature. The study of angiogenesis in vivo is limited by difficulties related to imaging of the fine structure of vascular sprouting within non-transparent bulk tissue. Thus, many model systems have been proposed in recent years. However, to meet the urgent need for high-throughput studies and screening, further improvements are still required, particularly in terms of scaling-up. In this study, we combined microchip fabrication with the culture of three-dimensional (3D) spheroids, thus providing a platform for 3D multilayer angiogenesis-on-a-chip. Using this platform, we investigated the precise effects of vascular endothelial growth factor (VEGF) on angiogenesis. In comparison with two-dimensional (2D) angiogenesis assays, our 3D angiogenesis platform demonstrated superior sprouting and provided proof of concept that our 3D biomimetic angiogenesis-on-a-chip could serve as a powerful tool for pro- or anti-angiogenesis candidate drug screening. [ABSTRACT FROM AUTHOR]

Aiming at structure of dishwasher pump shell, a set of injection mold with composite demoulding mechanism was designed. The outer cylindrical surface of water outlet pipe of product has nine rings and one buckle ring. Outer cylindrical surface of circular pipe and main body of the shell have buckle and locking positions along the direction of main parting surface, and the core-pulling and demoulding distance of circular pipe is long. The outlet pipe dynamic model cavity is designed as slider driven by inclined guide column with time-delay opening function. The core of outlet pipe is designed as core-pulling mechanism driven by oil cylinder, and composite demoulding mechanism is composed of moving model cavity slider and circular pipe core-pulling mechanism. In order to shorten demoulding distance, cavity slider is split into two, and the two sliders move in different directions. The mold opening process is that circular tube core pulling first completes demoulding, the fixed and moving molds then start parting, and the inclined guide pillar drives sliding block of the moving model cavity to demould. The cooling water path adopts mixed layout of straight-through+ well+circular-arc flume. The mold adopts four-cavity mold. In order to make injection pressure of the product the same, the gating system of hot runner and point gate is adopted. The mold structure is reasonable, which provides reference for mold design of similar products. [ABSTRACT FROM AUTHOR]