Rajasthan Technical University

About: Rajasthan Technical University is a education organization based out in Kota, Rajasthan, India. It is known for research contribution in the topics: Photovoltaic system & PID controller. The organization has 716 authors who have published 1084 publications receiving 4530 citations. The organization is also known as: RTU.

Genetic algorithm based cascaded H-bridge multilevel inverters for PV system with MPPT technique

Abstract: This paper presents a single-phase seven-level cascade H-Bridge multilevel inverter for photovoltaic systems making use of maximum power point tracking technique with genetic algorithm to reduce total harmonic distortion. Cascaded H-Bridge multilevel inverters are very popular in the power conversion systems for high power and power quality demanding applications. This configuration of multilevel level inverter reduces the value of total harmonic distortion. The proposed system consists of three H-Bridge inverters with individual photovoltaic array with separate perturbs & observer based maximum power point tracking and DC-DC boost converters. Genetic algorithm is used to control the switches of inverter to minimize harmonics in MLIs which is a complex optimization problem involving non-linear transcendental equation. The work presented in this paper is implemented using MATLAB software based simulation with different irradiation and temperature conditions.

Prediction of Ultimate Bearing Capacity of Aggregate Pier Reinforced Clay Using Machine Learning

Abstract: Aggregate piers are extensively in use for increasing bearing pressure and diminish settlement under the footing. The ultimate bearing capacity of aggregate pier reinforced clay is majorly affected by soil strength (cu), area replacement ratio (ar) of piles, geometry, and slenderness ratio (λ) of piles. Various prediction models have been proposed to predict the ultimate bearing capacity of aggregate piers. However, existing models have shown a broad range of bias, variation, errors, and as such they are unsuitable for practical design. In this study, machine learning algorithms (linear and non-linear regression) and Artificial neural networks (ANNs) were performed using field loading test results to predict the ultimate bearing capacity of ground reinforced by aggregate piers. Sensitivity analysis was conducted to identify the influence of input variables. To fulfil this objective, 37 test results were used for training and testing of different models and compared with each other based on statistical parameters (mean absolute error, root mean squared error, and r2-score). Random Forest Regression model came out to be the best suitable for prediction of ultimate bearing capacity with minimum mean absolute error (MAE = 38.93 kPa) and r2-score equal to 0.98.

Pest Detection on Leaf using Image Processing

Abstract: A survey report showed that 70% Indian population depends on agriculture sector. Numerous heterogeneous diseases and various kind of pests affect the production of crops which leads to quality and quantitative loss. Automatic in-field pest detection using computer vision technique is an important topic in modern intelligent agriculture but suffers from serious challenges including complexity of wild environment, detection of tiny size pest and classification into multiple classes of pests. In the literature attention is mostly focused on machine learning based techniques and image processing has not received equal attention. This paper illustrates an automatic approach for pest detection using Wavelet transformation and Oriented FAST and rotated BRIEF (ORB). The core objective of the research is to enhance feature extraction phase to improve the detection efficiency. The proposed approach is implemented on images of fluffy caterpillar pests on mustard crop and fava bean collected from farms in Rajasthan. The experimental results affirm the efficiency of the proposed approach.

Machine Learning Algorithms for Big Data Analytics

Abstract: A machine learning algorithm (MLA) is an approach or tool to help in big data analytics (BDA) of applications. This tool is suitable to analyze a large amount of amount generated by an application for effective and efficient utilization of the data. Machine learning algorithms considered to find out meaningful data and information for industrial applications. It is one of the services under big data analytics (BDA). Big data analytics (BDA) is suitable for identifying risk management, cause of failure, identifying the customer based on their procurement detail records, detection of fraud, etc. So, this paper deals with the work done in this field to analyze the importance of machine learning tools and techniques, identify the field where it is suitable to use including industries such as marketing, human resource, healthcare, insurance, banking, automobile, etc. This paper identifies different challenges of machine learning tools and technologies including the current status of adoption in industries.

Inventory control for nonperishable and perishable goods based on model predictive control

Abstract: Dynamics of inventory systems involve, stock filling by issuing orders which are fulfilled after some time called lead time and a time-varying demand which is not known a priori. The objectives fro...