scholarly journals Image Process of Rock Size Distribution Using DexiNed-Based Neural Network

Minerals ◽  
2021 ◽  
Vol 11 (7) ◽  
pp. 736
Author(s):  
Haijie Li ◽  
Gauti Asbjörnsson ◽  
Mats Lindqvist
Keyword(s):  
Neural Network ◽  
Size Distribution ◽  
Estimation Method ◽  
Ground Truth ◽  
Operating Conditions ◽  
Image Process ◽  
Morphological Transformation ◽  
The Right ◽  
Crushing Plant ◽  
And Control

In an aggregate crushing plant, the crusher performances will be affected by the variation from the incoming feed size distribution. Collecting accurate measurements of the size distribution on the conveyors can help both operators and control systems to make the right decisions in order to reduce overall power consumption and avoid undesirable operating conditions. In this work, a particle size distribution estimation method based on a DexiNed edge detection network, followed by the application of contour optimization, is proposed. The proposed framework was carried out in the four main steps. The first step, after image preprocessing, was to utilize a modified DexiNed convolutional neural network to predict the edge map of the rock image. Next, morphological transformation and watershed transformation from the OpenCV library were applied. Then, in the last step, the mass distribution was estimated from the pixel contour area. The accuracy and efficiency of the DexiNed method were demonstrated by comparing it with the ground-truth segmentation. The PSD estimation was validated with the laboratory screened rock samples.

Modelling and control of greenhouse system using neural networks

2016 ◽  
Vol 40 (3) ◽  
pp. 918-929 ◽  
Author(s):  
A Manonmani ◽  
T Thyagarajan ◽  
M Elango ◽  
S Sutha
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Moving Average ◽  
Growth Conditions ◽  
Operating Conditions ◽  
Adverse Weather ◽  
Control Schemes ◽  
Non Linear ◽  
Auto Regressive ◽  
And Control

A greenhouse system (GHS) is a closed structure that facilitates modified growth conditions to crops and provides protection from pests, diseases and adverse weather. However, a GHS exhibits non-linearity due to the interaction between the biological subsystem and the physical subsystem. Non-linear systems are difficult to control, particularly when their characteristics change with time. These systems are best handled with methods of computation intelligence, such as artificial neural networks (ANNs) and fuzzy systems. In the present work, the approximation capability of a neural network is used to model and control sufficient growth conditions of a GHS. An optimal neural network-based non-linear auto regressive with exogenous input (NARX) time series model is developed for a GHS. Based on the NARX model, two intelligent control schemes, namely a neural predictive controller (NPC) and non-linear auto regressive moving average (NARMA-L2) controller are proposed to achieve the desired growth conditions such as humidity and temperature for a better yield. Finally, closed-loop performances of the above two control schemes for servo and regulatory operations are analysed for various operating conditions using performance indices.

scholarly journals Improving the energy efficiency of oil production using identification and prediction of operating modes of production wells based on data analysis methods, machine learning and neural networks

2019 ◽  
Vol 124 ◽  
pp. 05031 ◽  
Author(s):  
A.M. Sagdatullin
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Control System ◽  
Oil And Gas ◽  
Gas Production ◽  
Operating Conditions ◽  
Identification System ◽  
Oil And Gas Production ◽  
The Neural Network ◽  
And Control

Currently, there is a need to improve the systems and control of pumping equipment in the oil and gas production and oil and gas transport industries. Therefore, an adaptive neural network control system for an electric drive of a production well was developed. The task of expanding the functional capabilities of asynchronous electric motors control of the oil and gas production system using the methods of neural networks is solved. We have developed software modules of the well drive control system based on the neural network, an identification system, and a scheme to adapt the control processes to changing load parameters, that is, to dynamic load, to implement the entire system for real-time control of the highspeed process. In this paper, based on a model of an identification block that includes a multilayered neural network of direct propagation, the control of the well system was implemented. The neural network of the proposed system was trained on the basis of the error back-propagation algorithm, and the identification unit works as a forecaster of system operation modes based on the error prediction. In the initial stage of the model adaptation, some fluctuations of the torque are observed at the output of the neural network, which is associated with new operating conditions and underestimated level of learning. However, the identification object and control system is able to maintain an error at minimum values and adapt the control system to a new conditions, which confirms the reliability of the proposed scheme.

scholarly journals Modeling the damages of belt gears

2021 ◽  
Vol 1199 (1) ◽  
pp. 012017
Author(s):  
Gawrońska Elżbieta ◽  
Domek Grzegorz ◽  
Krawiec Piotr ◽  
Kołodziej Andrzej
Keyword(s):  
Coupling Model ◽  
Operating Conditions ◽  
Selection Algorithm ◽  
Control Functions ◽  
Damage Monitoring ◽  
The Right ◽  
Belt Transmission ◽  
And Control ◽  
The Relationship ◽  
Selection Of

Abstract This paper looks into the problem of choosing a driving belt for a drive. The previously developed selection of algorithms was subjected to another evaluation that helped us recognize the need for changes indicated in developing new designs of drive belts. The new algorithm will be tested by simulating the operating conditions of the transmission, to which the right belt must be selected. Damage assessment after operation and belt selection allows for the identification of a new coupling model. By presenting the relationship between specific failure cases and the parameters of the coupling model, we can see the functionality of the selection algorithm. There may be multiple belt transmission damages. The feed may be broken; the surface may be damaged; the same applies to the edges. Furthermore, the wheels and bearings may be damaged too. The belt can have many additional functions that affect its operating parameters. Next to the drive function, the belt often performs conveyor and control functions. Thus, additional types of damage occur in belts with additional functions. The number of causes of their occurrence is also growing. For example, any damage to the sling in the passenger elevator can endanger the life of the passengers. Intensive research is being carried out on the real-mode damage monitoring systems. Specific failures are being monitored, and appropriate systems are being designed for them. Therefore, it is important to investigate the damages to belt transmissions, modeling their course of progression and causes.

scholarly journals Visibility Estimation via Deep Label Distribution Learning

2021 ◽  
Author(s):  
Mofei Song ◽  
Han Xu ◽  
Xiao Fan Liu ◽  
Qian Li
Keyword(s):  
Neural Network ◽  
Feature Extraction ◽  
Conditional Probability ◽  
Estimation Method ◽  
Ground Truth ◽  
Learning Performance ◽  
Label Distribution Learning ◽  
Training Examples ◽  
Label Distribution ◽  
Visibility Estimation

This paper proposes an image-based visibility estimation method with deep label distribution learning. To train an accurate model for visibility estimation, it is important to obtain the precise ground truth for every image. However, the ground-truth visibility is difficult to be labeled due to its high ambiguity. To solve this problem, we associate a label distribution to each image. The label distribution contains all the possible visibilities with their probabilities. To learn from such annotation, we employ a CNN-RNN model for visibility-aware feature extraction and a conditional probability neural network for distribution prediction. Our experiment shows that labeling the image with visibility distribution can not only overcome the inaccurate annotation problem, but also boost the learning performance without the increase of training examples.

scholarly journals Simulation Software Accelerates Development of Neural Network Control Applications in the Water Industry

2001 ◽  
Vol 34 (7) ◽  
pp. 202-206
Author(s):  
Adam Adgar ◽  
Ian Fletcher
Keyword(s):  
Neural Network ◽  
Simulation Software ◽  
Network Control ◽  
Operating Conditions ◽  
Data Sets ◽  
Application Development ◽  
Water Treatment Process ◽  
Plant Data ◽  
Artificial Neural Network Ann ◽  
And Control

Artificial neural network (ANN) applications have become popular in the field of water treatment process for modelling and control purposes. This is due to the fact that the processes are not well understood and also due to the large amounts of data available which are recorded for regulatory purposes. Development of such ANN applications are made more difficult by the data handling issues. It is often advantageous to try techniques developed on different plant data sets, under different operating conditions etc, but this is difficult to achieve efficiently in software. In this research, however, we have described several different applications of ANNs to demonstrate the significant results which may be achieved. But more importantly we have shown how the ANN application development, modification, comprehensive testing and benchmarking can be efficiently completed. To simplify the presentation, we have omitted a section on neural network architectures and learning processes since this is adequately covered in the paper by Renotte et al included as part of this Special Feature.

scholarly journals Visibility Estimation via Deep Label Distribution Learning

2021 ◽  
Author(s):  
Mofei Song ◽  
Han Xu ◽  
Xiao Fan Liu ◽  
Qian Li
Keyword(s):  
Neural Network ◽  
Feature Extraction ◽  
Conditional Probability ◽  
Estimation Method ◽  
Ground Truth ◽  
Learning Performance ◽  
Label Distribution Learning ◽  
Training Examples ◽  
Label Distribution ◽  
Visibility Estimation

This paper proposes an image-based visibility estimation method with deep label distribution learning. To train an accurate model for visibility estimation, it is important to obtain the precise ground truth for every image. However, the ground-truth visibility is difficult to be labeled due to its high ambiguity. To solve this problem, we associate a label distribution to each image. The label distribution contains all the possible visibilities with their probabilities. To learn from such annotation, we employ a CNN-RNN model for visibility-aware feature extraction and a conditional probability neural network for distribution prediction. Our experiment shows that labeling the image with visibility distribution can not only overcome the inaccurate annotation problem, but also boost the learning performance without the increase of training examples.

Rancang Bangun Sistem Penerimaan Dan Mutasi Barang Pada PDAM Tirta Kerta Raharja

Sensi Journal ◽  
2020 ◽  
Vol 6 (2) ◽  
pp. 236-246
Author(s):  
Ilamsyah Ilamsyah ◽  
Yulianto Yulianto ◽  
Tri Vita Febriani
Keyword(s):  
System Analysis ◽  
Object Oriented ◽  
Analysis Method ◽  
Web Based ◽  
Visual Modeling ◽  
Long Time ◽  
The Right ◽  
And Control ◽  
Web Based System ◽  
The Web

The right and appropriate system of receiving and transferring goods is needed by the company. In the process of receiving and transferring goods from the central warehouse to the branch warehouse at PDAM Tirta Kerta Raharja, Tangerang Regency, which is currently done manually is still ineffective and inaccurate because the Head of Subdivision uses receipt documents, namely PPBP and mutation of goods, namely MPPW in the form of paper as a submission media. The Head of Subdivision enters the data of receipt and mutation of goods manually and requires a relatively long time because at the time of demand for the transfer of goods the Head of Subdivision must check the inventory of goods in the central warehouse first. Therefore, it is necessary to hold a design of information systems for the receipt and transfer of goods from the central warehouse to a web-based branch warehouse that is already database so that it is more effective, efficient and accurate. With the web-based system of receiving and transferring goods that are already datatabed, it can facilitate the Head of Subdivision in inputing data on the receipt and transfer of goods and control of stock inventory so that the Sub Head of Subdivision can do it periodically to make it more effective, efficient and accurate. The method of data collection is done by observing, interviewing and studying literature from various previous studies, while the system analysis method uses the Waterfall method which aims to solve a problem and uses design methods with visual modeling that is object oriented with UML while programming using PHP and MySQL as a database.

EFEKTIVITAS METODE DEMONSTRASI DALAM MENINGKATKAN HASIL BELAJAR SISWA SEKOLAH MENENGAH KEJURUAN PADA PENGAJARAN SISTEM KOLOID

2013 ◽  
Vol 19 (1) ◽  
pp. 28
Author(s):  
Hamda Situmorang ◽  
Manihar Situmorang
Keyword(s):  
High School ◽  
High School Students ◽  
Abstract Concept ◽  
Vocational High School ◽  
School Students ◽  
Lecture Method ◽  
Colloid System ◽  
Before And After ◽  
The Right ◽  
And Control

Abstract Implementation of demonstration method in the teaching of chemistry is assigned as the right strategy to improve students’ achievement as it is proved that the method can bring an abstract concept to reality in the class. The study is conducted to vocational high school students in SMKN1 Pargetteng getteng Sengkut Pakfak Barat at accademic year 2013. The teaching has been carried out three cycles on the teaching of chemistry topic of colloid system. In the study, the class is divided into two class, experiment class and control class. The demontration method is used to teach students in experimental class while the teaching in control class is conducted with lecture method. Both are evaluated by using multiple choise tests before and after the teaching procedures, and the ability of students to answer the problems are assigned as students’ achievements. The results showed that demonstration method improved students’ achievement in chemistry. The students in experimental class who are taughed with demonstration method (M=19.08±0.74) have higher achievements compare with control class (M=12.91±2.52), and both are significantly different (tcalculation 22.85 > ttable 1.66). The effectivity of demostration method in experimental class (97%) is found higer compare to conventional method in control class (91%).

Adsorption Isotherm Predictions for Multiple Molecules in MOFs Using the Same Deep Learning Model

2019 ◽  
Author(s):  
Ryther Anderson ◽  
Achay Biong ◽  
Diego Gómez-Gualdrón
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Molecular Simulation ◽  
Large Scale ◽  
Learning Model ◽  
Operating Conditions ◽  
Small Subset ◽  
Screening Methods ◽  
Large Set ◽  
Metal Organic

<div>Tailoring the structure and chemistry of metal-organic frameworks (MOFs) enables the manipulation of their adsorption properties to suit specific energy and environmental applications. As there are millions of possible MOFs (with tens of thousands already synthesized), molecular simulation, such as grand canonical Monte Carlo (GCMC), has frequently been used to rapidly evaluate the adsorption performance of a large set of MOFs. This allows subsequent experiments to focus only on a small subset of the most promising MOFs. In many instances, however, even molecular simulation becomes prohibitively time consuming, underscoring the need for alternative screening methods, such as machine learning, to precede molecular simulation efforts. In this study, as a proof of concept, we trained a neural network as the first example of a machine learning model capable of predicting full adsorption isotherms of different molecules not included in the training of the model. To achieve this, we trained our neural network only on alchemical species, represented only by their geometry and force field parameters, and used this neural network to predict the loadings of real adsorbates. We focused on predicting room temperature adsorption of small (one- and two-atom) molecules relevant to chemical separations. Namely, argon, krypton, xenon, methane, ethane, and nitrogen. However, we also observed surprisingly promising predictions for more complex molecules, whose properties are outside the range spanned by the alchemical adsorbates. Prediction accuracies suitable for large-scale screening were achieved using simple MOF (e.g. geometric properties and chemical moieties), and adsorbate (e.g. forcefield parameters and geometry) descriptors. Our results illustrate a new philosophy of training that opens the path towards development of machine learning models that can predict the adsorption loading of any new adsorbate at any new operating conditions in any new MOF.</div>

Sign in / Sign up

Export Citation Format

Share Document