International Journal of Engineering Systems Modelling and Simulation (16 papers in press)
Mathematical Modelling for Prediction of Mechanical properties of Abutilon Indicum fibre reinforced composite using RSM and ANN
by Doni Mohana Krishnudu, D. Sreeramulu, P. Venkateshwar Reddy
Abstract: Major application areas of natural fiber composites are found in packing industries. The focus of the present study is on natural fibre (abutilon indicum) reinforcement composite materials with filler and to predict their mechanical properties of Abutilon Indicum fibre reinforced composites. In the present study fiber content (wt.%) and filler content (wt.%) are taken as the influencing parameters and mechanical properties like tensile, impact and flexural strengths are taken as an output parameter. Various proportions of fiber content and filler content were designed as per L25 orthogonal array and their mechanical properties were found. A regression equation was developed using RSM for each mechanical property in terms of fiber weight and filler weight. Mathematical models are developed by using both RSM and ANN and predictive capability of these two techniques were compared based on its R2 values. From the R2 values it is confirmed that the predictive capability of ANN model is higher than the RSM model for the present study.
Keywords: Abutilon Indicum; Mechanical Properties; RSM; ANN.
Finite-element modelling of magnetic fields for superconducting magnets with magnetic vector and total scalar potentials using COMSOL Multiphysics
by Alexander Chervyakov
Abstract: The finite-element modelling of superconducting magnets is a resource-hungry and
challenging work. For these devices, the high-quality requirements for focusing fields are usually superimposed by complexity of the model geometries and nonlinearity of the magnetic materials. The precise field simulations could result therefore in substantial number of the degrees of freedom and, as a consequence, in significant usage of the computational resources. To achieve the acceptable accuracy with lower number of finite elements, the magnetic field distributions are calculated in terms of the magnetic vector potential (A-formulation) as well as the total scalar potential (V-formulation) with COMSOL Multiphysics and compared. For these calculations, we utilise the model of a superconducting dipole magnet recently designed for operation of the compact isochronous cyclotron SC200. The performance of the both methods is analysed in terms of accuracy and quality of the obtained fields as well as in terms of the computational cost.
Keywords: compact cyclotrons; superconducting magnets; magnetic fields; finite-element method.
Special Issue on: Artificial Intelligence-enabled Computing System Development
Load cell-based PID method controlled Segway system modelling and simulation
by Muhammed Mustafa Kelek, Ugur Fidan, Yuksel Oguz, Ibrahim Celik, Tolga Ozer
Abstract: In the present study, mathematical modelling and simulation of load cell-based
Segway has been done. Four load cells placed on the Segway to provide the control of the system. According to the measured weight information, the dynamic model of the system can be updated instantly. This operation makes the Segway control easier by changing the maximum pitch angle. In order for the system to stay in balance on two wheels, it must move at the appropriate pitching angle and at the desired speed. The control of the suitable pitching angle and speed are controlled by the PID method. As a result of the simulation in Matlab/Simulink environment, Segways speed information, current information of BLDC and pitching angle can be accessed. As a result of the study, it is thought that the load cell-based Segway can be controlled more effectively.
Keywords: Segway; mathematical modelling; load cell; proportional; integral and derivative;
PID; electric vehicle.
Analysing chickpea physical characteristics emphasising on count, shape and size using computer vision
by Ajay Khatri, Shweta Agrawal
Abstract: Chickpeas are the food supplements which are very rich in protein, fibre and minerals. This grain affects a large percentage of Indian economy and India has the largest production and consumption of these grains. The most important quality attribute of chickpeas are size of seed, colour and taste. Based on these quality attributes chickpeas are graded into three main grades 78 mm, 89 mm, 9 mm and above. Determining seed size through sieve analysis in legumes is labour dependent, time consuming and inaccurate method. In general quality assessment of desi chickpea is done by visual inspection of small samples from the lot which is a slow and inaccurate process. The paper proposes a computer vision-based algorithm to assess the quality of chickpea on the basis of their shape, size and count. Experiment is performed for 20 sample images the results present that accuracy achieved through proposed algorithm for width calculation is 97.4%, for height calculation is 98.14%, for aspect ratio accuracy achieved is 97.3% and for chickpea count accuracy achieved is 98.6%. The proposed algorithm used concept of reference object to overcome the problem of dependency on distance of object and camera while capturing the image.
Keywords: computer vision; chickpea; image processing; grain analysis; accuracy.
Prediction of Euclidean distance between existing and target product for software product line testing using FeatureIDE
by Ashish Saini, Raj Kumar, Satendra Kumar, Mohit Mittal
Abstract: Software product line (SPL) is a paradigm that consists a family of products, all these products have some common features. SPL contains enormous products due to variable features, to test all of them is unfeasible. For this reason, several methods have been introduced to test the product line. These methods are used to prioritise products because they are based on feature interaction and do not provide information regarding products validity. To check the validity of products, we introduce a method based on Euclidean distance, which verifies the product based on the calculated distance between the real and desired product features. In addition, we compare the proposed method with the existing interaction-based method for product lines of different sizes. The results show that the proposed method takes less time to test the effectiveness of the product and increases the impact of the method in terms of time.
Keywords: software product line; SPL; feature model; software product line testing; testing;
software product line engineering.
High speed energy efficient multiplier for signal processing
by S. Karthick, C. Kamalanathan, P. Sunita, S. Ananthakumaran, E. Prabhu
Abstract: In digital signal processors, computation intensive arithmetic functions such as image smoothing, convolution and filtering frequently involve multiplication-based operations like inner-product generation and accumulation. Multiplication time is the predominate element in determining the execution time of any digital signal processing chip. Switching activity of the functional units in the multiplier contributes to significant amount of power dissipation. This paper presents high speed energy efficient multiplier. By reducing the switching activity and number of computations, the proposed multiplier achieves a better performance in terms of delay and PDP. The proposed high speed energy efficient multiplier is designed using Verilog-HDL and synthesised using Cadence RTL compiler with respect to 180 nm and 90 nm technological libraries. The proposed multiplier shows the delay reduction of 8.95% to 31.40%. The potential benefit of reducing the delay realises a PDP reduction of 13.66% to 26.95%. The performance of the proposed multiplier is verified by implementing it in 16 tap 16-bit coefficient band pass finite impulse response filter. The multiplier used here can be used in signal processing application to obtain energy efficient hardware.
Keywords: energy efficient; high speed; switching activity.
Real time prediction of solar radiation of Indore region using machine learning algorithms
by Sanjiv Kumar Jain, Kaustubh Yawalkar, Prakhar Singh, Advait Apte
Abstract: The most crucial data requirement for all solar energy researches is solar radiation. As solar radiation is the quantity which is dependent on time, the desired power output of any solar power plant is also dependent on time. The objective of this paper is to utilise machine learning models to estimate the solar radiations on daily data of Indore region (22.7196 N, 75.8577 E). The speed of wind, temperature, pressure, humidity along with solar radiation are used and applied in the process of prediction. The evaluation of the model is done in terms of prediction efficiency. In the work, boosted decision tree algorithm is used for the solar radiation prediction, which gives an accuracy of 96.9%. Also, the multiple linear regression algorithm is utilised in the work for the real time estimation of hourly solar radiations. The method gives an accuracy of 91.8%.
Keywords: solar radiation; linear regression; boosted decision tree; machine learning.
An empirical study on user buying behaviour in fashion industry using logistic regression
by Shaifali Chauhan, Richa Banerjee, Mohit Mittal, Sher Singh Bhakar
Abstract: The potential growth in technology is very high from past two decades. Due to this, every field of life is connected to internet. Nowadays, millions of people use internet and do shopping online. The fashion industry has opened new gates for online users by providing various offers. In this paper, a model is proposed for analysing impulse buying (IB) towards apparels based on consumers'; shopping behaviour. The data of 569 responses is collected and evaluated by using partial least square-structural equation modelling framework. Further, for analysing and identification of exact parameters that is highly important which affect user buying behaviour has used statistical approaches such as logistic regression. Based on results, hedonic and positive emotions (PE) have a significant impact on impulse buying, whereas involvement and sales promotion have an insignificant relationship.
Keywords: buying behaviour; logistic regression; support vector machine; fashion apparels.
Emotional intelligence creating a new roadmap for artificial intelligence
by Sharmistha Dey, Chinmay Chakraborty
Abstract: Artificial intelligence (AI) changing the world with the power of creating intelligent solutions capable of autonomous decision-making and self-diagnostic abilities. Recent research on AI tends towards the use of emotional intelligence in artificial systems. A clear understanding of human emotion or cognitive behaviour can help an artificial system to become more rational and unbiased while making any decision. If a machine can think or feel like a human, it can be converted into a better decision-making system. This paper has performed a thorough review of state-of-the-art technologies and researches going in this particular field and tries to find out the current roadmap as well as the future trends in this area. This paper presents several types of research that are going on in this area and future trends.
Keywords: affective computing; backpropagation; chatbot; deep learning; emotional artificial intelligence; feature extraction; neural network; psychological sensor; social robot.
Design optimisation and development of thresher machine using artificial intelligence and machine learning
by Rahul S. Warghane, Rajkumar E. Pillai
Abstract: The design validation of thresher mechanism is done with artificial neural network (ANN). The supervised and unsupervised learning models are developed through design test data and experimental test results. The ANN model is developed and trained using back propagation algorithm with seven epoches and data set of 700 test trail results. The trained model gives minimum RSME 0.0057. The model obtained is compared through correlation analysis and average correlation coefficient 0.9623. The parametric design model obtained from ANN is implied through Arduino sketch developed for real-time controlling of thresher parameters in machine. The designed sketch with an interfacing of IR speed sensor is used to address the crop configuration as a function of crop strength. The real-time monitoring of crop configuration is noted and processed for controlling thresher encoder motor speed. The designed ANN model prevents application of single failure model and real-time controlling of threshing parameter commit highest efficiency.
Keywords: design optimisation; artificial neural network; ANN; machine learning; real-time control; thresher machine.
Machine learning in SDN networks for secure industrial cyber physical systems: a case of detecting link flooding attack
by Priyanshi Deliwala, Rutvij H. Jhaveri, Sagar Ramani
Abstract: Industrial cyber physical systems (ICPSs) are seen as a key promoter for a new age of internet-based real-time conveyance and cooperation an emerging network architecture approach known as software-defined networking (SDN) opens a gateway to different network attacks weaknesses one of them is link flooding attack (LFA). This paper illustrates how SDNs control layer is endangered to LFA while the proposed approach (Iwendi et al., 2020a) can appear suitable on the surface, some weaknesses and anomalies are discovered when deliberated deeper. In the current paper, we point out these anomalies and the limitations of those anomalies by applying two machine learning algorithms, namely ANN-MLP and random forest to correctly classifying the virulent traffic during congestion. We carry out experiments on Mininet network emulator and use the WEKA tool to assess the metrics by utilising the available datasets. The results exhibit that random forest can virulent traffic with higher accuracy and mollify them effectively during congestion.
Keywords: software-defined networking; SDN; random forest; OpenFlow; machine learning; link flooding attacks; LFA; industrial cyber physical systems; ICPSs.
Sensor based vehicle detection and classification - a systematic review
by Nikita Singhal, Lalji Prasad
Abstract: Traffic management has become a major problem in every country due to day by day increase of vehicles on road. With this enhancement, sometimes it becomes difficult to keep track of vehicles for the purpose of traffic monitoring and law enforcement. We need an intelligent transportation system (ITS) which will help in traffic management. In this study, we presented a review of the smart transportation system that focuses on vehicle detection and classification (VDC) that is generally used in applications like congestion prediction, future road infrastructure requirement prediction, automated parking, and security enforcement. We have reviewed more than 130 papers that are published between 2010 and 2021 and found that various sensor technologies, machine learning, computer vision and deep learning techniques have been applied for the detection and classification of vehicles by many researchers. This study will provide useful directions to the researchers in selecting appropriate technologies for VDC.
Keywords: vehicle detection; vehicle classification; intelligent transportation system; sensor; machine learning; deep learning.
The impact of big data in predictive analytics towards technological development in cloud computing
by Krishna Kumar Mohbey, Sunil Kumar
Abstract: Presently, we are living in a world of data. More than 2.5 quintillion bytes of data are generated everyday. The volume of data from different sources and in various forms can be identified as big data. With the collection of an enormous amount of data, various kinds of predictions would help make decisions. Making intelligent decisions in different situations using massive datasets is known as predictive analysis. Cloud computing aims to be an important way to handle big data. Working on big data in the cloud, however, poses its challenge. Predictive analytics is applied to generate different kinds of patterns that make optimised decisions. This paper introduces big data predictive analytics and its importance. It also gives details of applications and challenges in the present and future scenarios of cloud computing. Besides, we have also included various technologies and frameworks to store, manage, and process big data in cloud platforms.
Keywords: big data; cloud computing; predictive analytics; machine learning approaches; statistical approaches.
A review on smart city IoT and deep learning algorithms, challenges
by Vankadhara Rajyalakshmi, Kuruva Lakshmanna
Abstract: Recent improvements in the IoT are giving rise to the explosion of interconnected devices, empowering many smart applications. IoT devices engender massive data that requires intellectual processing and data analysis. Especially the DL algorithms applied in the SC applications such as SP, SWM, traffic, healthcare, SB, energy, and a lot. Inspired by these plentiful applications, we present the key abilities of DL in IoT-related smart applications. First, we discussed the main motivation behind the SC and reviewed the use of CNNs, RNNs, SAEs, DBMs, and DBNs. We studied, tabulated several DL practices and use cases of SC. Finally, we categorise many research challenges regarding the operative strategy, implementation of DL-IoT, future research directions, and further research challenges.We proposed promising future directions for DL-IoT in SC environments. The overall idea of this survey is to utilise the few available resources more smartly by incorporating DL-IoT.
Keywords: convolutional neural networks; CNNs; deep belief networks; DBNs; deep Boltzmann
machines; DBMs; deep learning; DL; internet of things; IoT; recurrent neural networks; RNNs; smart building; SB; smart city; SC; smart parking; SP; smart waste management; SWM; stacked auto encoders; SAEs.
Dolphin echolocation algorithm for small-signal stability analysis of DFIG-based wind power system
by Ashutosh Kashiv, H.K. Verma
Abstract: In this paper stability examination of a doubly-fed induction generator (DFIG) wind turbine system has been done. DFIG wind turbine system is modelled using the state-space method. The main aim of the paper is to examine the behaviour of the DFIG at the various speeds of the rotor. This paper mentions the application of the proportional-integral controller optimised by dolphin echolocation optimisation algorithm (DEOA) to get efficient as well as stable operation of the DFIG system. For this purpose, damping ratios and eigenvalues are taken as a result. The simulation results were compared with those obtained using the particle swarm optimisation (PSO) approach for the proposed DFIG model. Simulation data show that the DEOA-PI controller works well to limit the transient in voltage and current response for the DFIG model under consideration.
Keywords: wind energy; dolphin echolocation algorithm; DEOA; doubly-fed induction generator; DFIG; stability analysis; proportional-integral; PI; proportional-integral controller.
Real time voltage security assessment using adaptive fuzzified decision tree algorithm
by Sanjiv Kumar Jain, Narayana Pradas Patidar, Yogendra Kumar, Shweta Agrawal
Abstract: This paper presents the adaptive machine learning approach for voltage security classification. The online probabilistic assessment of voltage security is done using decision tree, which are updated periodically. The advantage of fuzzified decision tree support is robust classification of voltage security in the upcoming samples. Offline learning datasets are generated for each N-1 contingency conditions using continuation power flow method. Security classes are defined by threshold value of maximum loadability margins, calculated using the continuation power flow method. The proposed method is tested on two IEEE bus systems. Classification accuracy from a value of 88% to finally 100% is achieved for line outage no. 5 in IEEE-30 bus system and 100% for line outages no. 51 and 172, in IEEE-118 bus system. The result shows the fast and accurate classification for online decisions. This confirms the proposed method validity and suitability for the energy management system in online control decisions.
Keywords: fuzzy decision tree; continuation power flow; machine learning; power system; voltage security.