Computer Modeling and Calculation of the Ventilation Systems for Workplaces and Premises of Electric Welding Production

2021 ◽  
pp. 13-20
Author(s):  
V.V. Shaptala ◽  
◽  
K.I. Logachev ◽  
E.E. Khukalenko ◽  
◽  
...  
Keyword(s):  
Neural Network ◽  
Computer Modeling ◽  
Mutual Arrangement ◽  
Breathing Zone ◽  
Ansys Fluent ◽  
Electric Welding ◽  
Local Exhaust Ventilation ◽  
Rational Combination ◽  
Important Means ◽  
Gas Environment

The most important means of protecting the labor of electric welders is local exhaust ventilation, if required supplemented by general exchange ventilation of electric welding rooms, which contributes to maintaining the normalized state of the air environment. In real conditions, the dust-gas environment of the welding station is influenced by a combination of factors such as the location of welders breathing zone, the relative location of the welding point, and the suction opening of the exhaust device, and the air mobility associated with the aerodynamic interaction of the supply and exhaust devices. The desire to reduce the productivity and energy consumption of exhaust devices usually leads to a change in their effect on the state of the air environment. Therefore, along with the design features of the air extract, it is of great importance that the designs of the exhaust devices correspond to the nature of the welding work performed, as well as the space-planning solution for the mutual arrangement of welding posts and supply and exhaust devices, especially in the rooms with a limited volume. The article is devoted to the analysis of the effect of these factors on the dust and gas environment of the electric welding stations, and the search for their rational combination by the method of computer modeling using the Ansys Fluent hydrodynamic package. To approximate the complex dependence of the carryover into the room, and the concentration in the breathing zone of harmful impurities on the main influencing factors, it is proposed to use a neural network of direct propagation. The architecture and results of testing a neural network model of one of the types of exhaust devices are presented.

Computer modeling of ventilation when performing electric welding

2020 ◽  
pp. 59-65
Author(s):  
V. V. Shaptala ◽  
K. I. Logachev ◽  
N. N. Severin ◽  
E. E. Khukalenko ◽  
Yu. M. Gusev
Keyword(s):  
Computer Modeling ◽  
Maximum Permissible Concentration ◽  
Ventilation System ◽  
Exchange Effect ◽  
Electric Welding ◽  
Local Exhaust Ventilation ◽  
Exhaust Ventilation ◽  
Maximum Permissible Concentrations ◽  
Harmful Emissions ◽  
Effective Use

The design of an integrated ventilation system, which combines local exhaust ventilation of welders' workplaces and general ventilation of premises, is considered. Thefields of air mobility, its temperature and concentration of harmful impurities are calculated using the method of computer modeling. A comparative analysis of the main types of stationary devices for local exhaust ventilation of electric welding workstations (exhaust hoods, welding panels, as well as blowers) is performed. The conditions for the effective use of lift-and-turn suction pumps have been determined. The factors influencing the efficiency of their work have been identified, and combinations of influencing factors have been established that ensure the maximum permissible concentration of harmful emissions in the working areas of welders. The conditions under which the general exchange effect of local suction is sufficient to maintain the concentration in the room at the level of maximum permissible concentrations (MPC) have been determined. The results obtained can be used to develop engineering methods for calculating and optimizing ventilation devices and systems, as well as to analyze structural and design solutions for the ventilation device of electric welding industries.

scholarly journals Pedestrian Flow Tracking and Statistics of Monocular Camera Based on Convolutional Neural Network and Kalman Filter

2019 ◽  
Vol 9 (8) ◽  
pp. 1624 ◽  
Author(s):  
Miao He ◽  
Haibo Luo ◽  
Bin Hui ◽  
Zheng Chang
Keyword(s):  
Neural Network ◽  
Kalman Filter ◽  
Data Sets ◽  
Pedestrian Flow ◽  
Motion Models ◽  
Monocular Camera ◽  
Public Data ◽  
Urban Safety ◽  
Pedestrian Counting ◽  
Important Means

Pedestrian flow statistics and analysis in public places is an important means to ensure urban safety. However, in recent years, a video-based pedestrian flow statistics algorithm mainly relies on binocular vision or a vertical downward camera, which has serious limitations on the application scene and counting area, and cannot make use of the large number of monocular cameras in the city. To solve this problem, we propose a pedestrian flow statistics algorithm based on monocular camera. Firstly, a convolution neural network is used to detect the pedestrian targets. Then, with a Kalman filter, the motion models for the targets are established. Based on these motion models, data association algorithm completes target tracking. Finally, the pedestrian flow is counted by the pedestrian counting method based on virtual blocks. The algorithm is tested on real scenes and public data sets. The experimental results show that the algorithm has high accuracy and strong real-time performance, which verifies the reliability of the algorithm.

scholarly journals Local Exhaust Ventilation to Control Dental Aerosols and Droplets

2021 ◽  
Author(s):  
James R Allison ◽  
Christopher Dowson ◽  
Kimberley Pickering ◽  
Greta Červinskytė ◽  
Justin Durham ◽  
...  
Keyword(s):  
Infectious Disease ◽  
Disease Transmission ◽  
High Efficiency ◽  
Breathing Zone ◽  
Local Exhaust Ventilation ◽  
Exhaust Ventilation ◽  
Ultrasonic Scaler ◽  
Dental Procedures ◽  
Air Turbine ◽  
Open Plan

Dental procedures produce aerosols which may remain suspended and travel significant distances from the source. Dental aerosols and droplets contain oral microbes and there is potential for infectious disease transmission and major disruption to dental services during infectious disease outbreaks. One method to control hazardous aerosols often used in industry is Local Exhaust Ventilation (LEV). The aim of this study was to investigate the effect of LEV on aerosols and droplets produced during dental procedures. Experiments were conducted on dental mannequins in an 825.4 m3 open plan clinic, and a 49.3 m3 single surgery. 10-minute crown preparations were performed with an air-turbine handpiece in the open plan clinic, and 10-minute full mouth ultrasonic scaling in the single surgery. Fluorescein was added to instrument irrigation reservoirs as a tracer. In both settings, Optical Particle Counters (OPCs) were used to measure aerosol particles between 0.3 - 10.0 μm and liquid cyclone air samplers were used to capture aerosolised fluorescein tracer. Additionally, in the open plan setting fluorescein tracer was captured by passive settling onto filter papers in the environment. Tracer was quantified fluorometrically. An LEV device with High Efficiency Particulate Air (HEPA) filtration and a flow rate of 5,000 L/min was used. LEV reduced aerosol production from the air-turbine handpiece by 90% within 0.5 m, and this was 99% for the ultrasonic scaler. OPC particle counts were substantially reduced for both procedures, and air-turbine settled droplet detection reduced by 95% within 0.5 m. The effect of LEV was substantially greater than suction alone for the air-turbine and was similar to the effect of suction for the ultrasonic scaler. LEV reduces aerosol and droplet contamination from dental procedures by at least 90% in the breathing zone of the operator and it is therefore a valuable tool to reduce the dispersion of dental aerosols.

scholarly journals TO THE QUESTION OF LEARNING MODELING A NEURAL NETWORK IN SCHOOL COURSE OF INFORMATICS

2020 ◽  
Vol 72 (4) ◽  
pp. 180-187
Author(s):  
D.N. Issabayeva ◽  
◽  
G.A. Abdulkarimova ◽  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Data Mining ◽  
Neural Networks ◽  
Computer Modeling ◽  
Machine Learning Algorithms ◽  
Distinctive Features ◽  
Design Task ◽  
Recognition Of Objects ◽  
Simulated Object

The article describes an approach to computer modeling of a neural network. To simulate a neural network, it is necessary to collect a large amount of data about the simulated object, identify its features, create an algorithm that is a set of operations performed on a computer. The function of neural networks is based on classification (sorting), prediction and recognition of objects or events. Modeling neural networks can be carried out by collecting data about an object, highlighting its distinctive features and creating an algorithm using the example of Iris-Fisher flowers. The article also includes a description of the execution of a neural network design task and the use of machine learning algorithms for data mining in an MS Excel spreadsheet processor for teaching lessons.

Level of Repair Analysis Oriented on Integrated Materiel Support Based on BP Neural Network

2013 ◽  
Vol 756-759 ◽  
pp. 4706-4709
Author(s):  
Ying Shen ◽  
Zhi Wei Shan ◽  
Jun Hai Cao ◽  
Fu Sheng Liu
Keyword(s):  
Neural Network ◽  
Life Cycle ◽  
Bp Neural Network ◽  
Life Cycle Cost ◽  
Corrective Maintenance ◽  
Design Life ◽  
Operational Readiness ◽  
Important Means ◽  
Level Of Repair Analysis ◽  
Relative Concept

Level of Repair Analysis, for short LORA, is an integrated tradeoff technology and an important means that makes certain corrective maintenance support concept. Although by LORA it determines feasible, best efficient maintenance level or makes disposal decision for materiel repair, and affects design, life cycle cost and operational readiness of materiel, it is not mature for the development and application of LORA. In order to put the analysis technology into practice in deed, particular adapt to the need of integrated materiel support, on the basis of relative concept, principle and application of BP neural network in the paper, it discusses how to go on LORA in the context of integrated materiel support based on BP neural network, better to settle the problem of LORA.

scholarly journals Modelling indoor pollutant distribution via passive scalar and virtual box approach

2021 ◽  
Vol 327 ◽  
pp. 05001
Author(s):  
George Pitchurov ◽  
Detelin Markov ◽  
Iskra Simova ◽  
Rositsa Velichkova ◽  
Peter Stankov ◽  
...  
Keyword(s):  
Mass Flux ◽  
Body Shape ◽  
Passive Scalar ◽  
Thermal Manikin ◽  
Breathing Zone ◽  
Ansys Fluent ◽  
Indoor Pollutant ◽  
Pollutant Source ◽  
Pollutant Distribution ◽  
Occupied Zone

The distribution of PM2.5 around a thermal manikin with realistic female body shape in a naturally ventilated room has been modelled. The health risk (HR) due to inhalation of the PM2.5 has been quantified by integrating the pollutants mass flux over the boundaries of a virtual box around the mannequin’s head (the breathing zone). By the same approach HR is evaluated over the boundaries of another virtual box that surrounds the manikins body and defines the occupied zone. The paper focuses on the peculiarities of creating and meshing a virtual geometry, as well as on the application of user-defined functions (UDF) for defining a pollutant source within the room using Ansys Fluent modelling package.

Solution to China’s GDP Prediction Problem by BP Neural Network

2011 ◽  
Vol 50-51 ◽  
pp. 451-454
Author(s):  
Guan Chen Zhou ◽  
Kai Li Wang ◽  
Tong Cui
Keyword(s):  
Neural Network ◽  
Bp Neural Network ◽  
Original Data ◽  
Practical Significance ◽  
Sigmoid Function ◽  
Economic Management ◽  
Macroeconomic Policies ◽  
Operation Conditions ◽  
Important Means ◽  
Predicted Values

GDP algorithm has become the important means for China’s economic management departments to acquaint with the economic operation conditions and the important basis on which the economic development strategies and planning and various macroeconomic policies are worked out by them. Therefore, research and establishment of GDP model have an important and practical significance. In this paper, the BP neural network is adopted to perform the research and prediction of China’s GDP. Firstly, the original data of larger ones are observed and introduced into the commonly applied function of sigmoid function, but the effect is not ideal and the normalization point concentrates on , which may affect the predicted values, so a simple improvement is made for the sigmoid function to make the function value distribute between 0 and 1 and then the normalized function after improvement is better.

Yarn Quality Prediction and Process Parameters Optimization Based on Genetic Algorithm and Neural Network

2015 ◽  
Vol 742 ◽  
pp. 377-383
Author(s):  
Jian Guo Yang ◽  
Lan Xu ◽  
Zhi Jun Lu ◽  
Qian Xiang ◽  
Bin Liu
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Process Parameters ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Parameters Optimization ◽  
Model Parameters ◽  
Quality Prediction ◽  
Yarn Quality ◽  
Important Means

Quality prediction is an important means of the quality management in modern spinning production. This paper proposed a yarn quality prediction model based on Genetic Algorithm and back propagation neural network to predict the yarn quality and optimize the process parameters. The main identification model parameters were optimized by using genetic algorithm, and the prediction performance of the model has been compared against that of the BP neural network model. The effectiveness and availability of the proposed model are verified with the use of actual production data.

Particle Transport and Deposition in a Ventilated Room With a Standing Mannequin

2020 ◽  
Author(s):  
Mehrdad Azhdari ◽  
Mohammad Mehdi Tavakol ◽  
Goodarz Ahmadi
Keyword(s):  
Recirculation Zone ◽  
Spherical Particles ◽  
Computational Domain ◽  
Breathing Zone ◽  
Ansys Fluent ◽  
Second Mode ◽  
Airflow Field ◽  
The Mean ◽  
Simulation Results ◽  
Aspiration Efficiency

Abstract This study presents the results of a series of numerical simulations for airflow field and particle dispersion and deposition around a mannequin standing inside a ventilated room. A 3-D airway model was constructed from the nostril inlet to the end of 4th lung generation and was integrated into the standing mannequin model in the room. The computational domain included the region around the mannequin and inside the respiratory system. The room was ventilated by a mixing air-conditioning system that supplied air with a speed of 3m/s from a diffuser mounted on the top of the sidewall and exited from a damper mounted at the bottom of the side or front walls. In the first mode, the diffuser and damper were located on the wall in front of the mannequin and in the second mode on the wall at the right side of the mannequin. The mean airflow field inside the room was obtained by solving the Navier-Stokes and continuity equations using the Ansys-Fluent software. The k-ω SST transitional model was employed for turbulence modeling. Then, spherical particles with 5, 10, 20, and 40 μm diameter and unit density were released into the room, and their trajectories were tracked by using the Lagrangian trajectory analysis approach. Aspiration efficiency and deposition of particles for inhalation flow rates of 15 and 30 lit/min were analyzed with the improved discrete random walk (DRW) stochastic model using a user-defined function (UDF) coupled into the Ansys-Fluent discrete phase model. Simulation results for the mean airflow showed the formation of a large recirculation zone inside the room. In the first mode, the main recirculation zone formed behind mannequin that carried the flow streamlines toward the mannequin breathing zone. In the second mode, the recirculation formed in front of the mannequin face that led the streamlines out of the breathing zone. The simulation results for particle inhalation showed that the aspiration efficiency of particles is higher in the first ventilation mode compared to the second mode. Results also showed that the total deposition of particles in the airway passage increases as particle size increases.

Transport and trajectory of cough-induced bimodal aerosol in an air-conditioned space

2020 ◽  
pp. 1420326X2094116 ◽  
Author(s):  
Bo Zhang ◽  
Guangyu Guo ◽  
Chao Zhu ◽  
Zhiming Ji ◽  
Chao-Hsin Lin
Keyword(s):  
Ventilation System ◽  
Three Dimensional ◽  
Experimental System ◽  
Ambient Air ◽  
Lagrangian Model ◽  
Breathing Zone ◽  
Cfd Model ◽  
Ansys Fluent ◽  
Liquid Phases ◽  
Air Ventilation

Aerosol generated from a human cough can be a potential major indoor health risk due to the possible transmission of infectious respiratory diseases to surrounding individuals within the same room and even could spread out via air-ventilation/conditioning systems. This study aims to investigate the transport characteristics and trajectory of coughed aerosols under the influence of conditioned air ventilation as well as near-by human breathing zone using computational fluid dynamics (CFD). An experimental system consisting of air-conditioned space with multiple inlets and outlets, a cough simulator and a receiver was built to validate the CFD predictions. The comparison is in good agreement. The CFD model was established as a transient three-dimensional multiphase multicomponent Eulerian–Lagrangian model and numerically solved using commercial software ANSYS Fluent. Both gas and liquid phases were modelled as multicomponent mixtures. With this CFD model, the indoor transport and trajectory of coughed aerosols can be accounted for the distributions of portions inhaled by each manikin, deposited on surfaces of manikins and chamber walls, as well as recirculated back into the ventilation system. Results reveal that the aerosol source location and the ambient air movement can be crucial factors of aerosol trajectory in terms of direct and indirect influence.

Sign in / Sign up

Export Citation Format

Share Document