Construction Technology and Quality Control of Power and Electrical Engineering Based on Convolutional Neural Network

In the context of the Internet era, more and more parties have begun to store, process, and analyze data, but the accompanying question is whether people are reasonable about the data under the impact of massive data, effective and efficient analysis, especially the problems faced in this project. This article aims to study the quality control problems faced by electric power and electrical engineering in the construction process through the use of convolutional neural networks. Under this idea, this article proposes a multilayer convolution method. The experimental results show that the use of the improved multilayer convolution method for the convolution method of the convolutional neural network can effectively improve the multiple analysis problems of small datasets in the construction of electric power and electrical engineering; in this way, the relevant data are analyzed; by controlling the quality of construction, the quality problem has been greatly improved. After comparison, it is concluded that the overall construction quality has increased by 35%.

Prediction of Plasma Drug Concentration Profiles and Pharmacokinetic Parameters of Nifedipine Commercial Tablets using the Convolution Method

The aim of this study was to predict the blood/plasma drug concentration profiles for five brand of nifedipine present on the Saudi Arabia market by using the numerical convolution method and to estimate the pharmacokinetic parameters (Cmax, Tmax, Ka, K and Vd) by the application of the residual method to the predicted plasma drug concentration profiles. Results showed that the higher Cmax was 118.95ng/ml for brand A2 and the lower Cmax was 72.29ng/ml for brand A3. The Tmax was ranged from 2.3 hr to 4.9 hr for brands A2 and A3 respectively. The total area under plasma drug concentration curve (AUCinf.) was in lower value equal to 585.59 ng x hr/ml for brand A2 and the higher value was for brand A5 equal to 743.52ng x hr/ml. The volume of distribution was also increased from 52.5 L for free nifidipine to 72 L for brand A1. The predicted first order elimination rate constant was decreased from 0.34 hr-1 for free nifedipine to 0.17 hr-1 for brand A3. The half-life of nifedapine was increased from 2 hours for free drug to 3.93 hours for brand A3. From this study it can be concluded that brands present in the market that shows similarity in accordance to the Dissimilarity factor f1 are not always guaranty that they will be bioequivalent in vivo and vice versa. Also, this study indicates that the method of convolution is a useful tool for prediction of bioequivalence of different brands present on the market.

Exploiting Activation Sparsity for Fast CNN Inference on Mobile GPUs

Over the past several years, the need for on-device deep learning has been rapidly increasing, and efficient CNN inference on mobile platforms has been actively researched. Sparsity exploitation has been one of the most active research themes, but the studies mostly focus on weight sparsity by weight pruning. Activation sparsity, on the contrary, requires compression at runtime for every input tensor. Hence, the research on activation sparsity mainly targets NPUs that can efficiently process this with their own hardware logic. In this paper, we observe that it is difficult to accelerate CNN inference on mobile GPUs with natural activation sparsity and that the widely used CSR-based sparse convolution is not sufficiently effective due to the compression overhead. We propose several novel sparsification methods that can boost activation sparsity without harming accuracy. In particular, we selectively sparsify some layers with an extremely high sparsity and adopt sparse convolution or dense convolution depending on the layers. Further, we present an efficient sparse convolution method without compression and demonstrate that it can be faster than the CSR implementation. With ResNet-50, we achieved 1.88 speedup compared to TFLite on a Mali-G76 GPU.

About Liquid Filtration Under a Point Dam with a Vertical Weakly Permeable Film

The problem of groundwater filtration under a point dam in a piecewise homogeneous porous medium in the presence of a weakly permeable film under the dam is considered. The filtration area is considered in the form of a vertical half-plane with a horizontal line of water courses. A weakly permeable film divides the filtration area into two quadrants with different constant permeability. By the convolution method of Fourier expansions, the solution of the problem is obtained explicitly. The influence of a weakly permeable film on the filtration process is investigated. It is shown that the presence of a weakly permeable film reduces the filtration rates in the downstream.