Search

Prognostic and health management (PHM) has become increasingly popular due to the requirement of improved maintenance techniques in the industry. Remaining useful life (RUL) estimation is an important parameter through which PHM can be utilized to implement timely and cost-effective maintenance. Due to recent advancements in sensor-based and other Industry 4.0 related technologies, data-driven methods for RUL estimation have become more prevalent and effective. In this paper, a novel data-driven method for sensor-based RUL estimation using a combination of multi-scale convolutional neural network (MS-CNN) and long short-term memory (LSTM) is proposed. The proposed hybrid multi-scale convolutional LSTM (HMCL) model is capable of extracting both spatial features of various scales and temporal features from the input data to provide accurate RUL predictions. L2 regularization and dropout techniques are used to reduce overfitting. The performance of the proposed model is evaluated using the C-MAPSS dataset. It achieves excellent performance as compared to other state-of-the-art methods making it a promising approach for sensor-based RUL prediction. Additionally, to discern the cause for occurrence of offsets, i.e., deviations in the model’s predictions with the true RUL value, an offset analysis is carried out. Through the analysis, an estimate on the location and cause of offsets is established and based on the sensory input data, offsets are identified using an SVM classification model. Despite being a simple classification model, it is able to achieve a decent performance in classifying the offsets.

Background: The approach of teaching has been modified over the years with incorporation of E-learning tools consisting of adaptive tutorials, audiovisual clips, and virtual models. These educational media possess several distinct advantages over traditional didactic models of instruction, including the ability to update material in a timely manner to ensure delivery of the latest evidence-based content to trainees. The study was aimed to assess the effectiveness of the mentioned tools of E-learning in the field of medical science. Aims and Objectives: The aim of the study was to improve understanding and learning of metabolism of biochemistry for first MBBS students. Materials and Methods: Participants were taught the subject of metabolism of carbohydrates by the conventional methods following with a test and then were made to understand the topic by the means of audio-visual presentation and were tested again to achieve data for objective assessment of the E-learning methods of education. Results: The scores of the participants significantly increased after incorporation of E-learning tools in teaching methods. Conclusion: E-learning tools help increase the quality of education and compliance of students, even when dealing with difficult topics. Students become more interested with the subject and improves performance overall.

From recent few years, need of information security is realized by society amd researchers specially in multi-path, unstructured networks as Mobile Ad-hoc Network. Devices connected in such network are self-configuring and small in size and can communicate in infra less environment. Architecture is very much dynamic and absence of central controlling authority puts challenges to the network by making more vulnerable for various threats and attacks in order to exploit the function of the network. The paper proposes, TCP analysis against very popular attack i.e. blackhole attack. Under different circumstance, reliable transport layer protocol TCP is analyzed for the effects of the attack on adhoc network. Performance has been measured using metrics of average throughput, normalized routing load and end to end delay and conclusions have been drawn based on that.

In this paper, we proposed a data cleaning technique, for memory space optimization. We are using sdhash techniques for effective, fast and efficient duplicate files detection and removing in memory. The correct identification of duplicate files is the first critical step in data cleaning process. The fast growth of the data targets demands new automated methods for removing data duplication quickly, accurately, and reliably. Sdhash tool is used for calculation of similarity score of a data files, store, and compare its similarity hashes referred to as similarity digests (sdhash). In contrast, compare whole file, to brute force method, our method compares only the finger prints of all files and is able to efficiently distinguish among duplicate files. In addition, our evaluation data which contains hundreds of files, provides insights into the typical levels of content similarity across related Files. The proposed method is excellent in metric of time and space complexity.

Massive computing on cloud infrastructure had achieved remarkable consideration and popularity in recent years. However, security and privacy issues are also emerging as cloud infrastructures are deployed and maintained by third parties. In this paper, we propose a simple method to secure an image for privacy-preserving storage over cloud infrastructure. One-dimensional chaotic logistic map is used to divide the secret image into two shares such that no image information is revealed. Secure shares are generated using chaotic key stream derived from one-dimensional chaotic map. For successful decryption, both the shares are gathered and secret image is revealed. The proposed method is secure and simple, and involves less complex operations. Different experimental tests have been performed to demonstrate security aspects of the proposed scheme.

Drilling, Completion and Well Intervention operations generate waste water. With the increasing volume of well fluids being flowed back due to aggressive Well Intervention activities, there was a need to handle and dispose the waste water efficiently. This was required to maintain continuity of well flow back and stimulation operations in an efficient and sustainable manner without incurring additional cost keeping in mind the current Oil and Gas market scenario. This paper presents the "Best out of Waste" methodology adopted by the Well Services Department of Cairn India Limited to manage the huge waste water volumes by pooling in-house resources for developing Waste Water Treatment Project. This project as it stands today was brought together by first, evaluating various treatment methods such as EC (Electro Coagulation) unit, Chemical methods, Particle settling and Disposal methods such as natural evaporation, evaporators (Mechanical and Solar) and injection and further combining features of the above stated methods to gain maximum benefit. These evaluations led to selection of EC Unit combined with the effectiveness of Mechanical evaporators and Natural evaporation for decreasing the waste water footprint. Second, identifying a disposal well for treated water injection, established by various integrity and injectivity tests to confirm that the well was ready to accept the water with parameters and properties similar to that of EC unit output. The encouraging results in the injectivity tests proved to be the kick-off point of the project. Continuous monitoring of the treated water was also done to ensure the TSS and oil in water remained within the acceptable limits for injection, as per prevailing norms. Third, developing disposal pits and tank farm to facilitate the treatment and dumping of the treated waste water. Lastly, decreasing current waste water inventory along with reducing the OPEX involved with handling, transportation and disposal of the waste water. The key feature of this project was the utilization of above mentioned equipment's from internal resources of the company, which were either working stand-alone or were nonfunctional. Thus, the CAPEX for the project was brought to a bare minimum. The objective of this paper is to present the various technical, administrative and commercial aspects associated with this project and conclude with plans of further utilizing the treated water for future Fracturing, Stimulation and even completion jobs.

This paper will describe how good project management and communications between the various project stake holders resulted in the successful completion of the Raageshwari 15 wells fracturing campaign. The 93 fracturing stages were completed under budget and in a shorter time frame than planned. The management of the project included multiple and diverse operations and equipment including perforating, wireline, well testing, hydraulic fracturing, HSE, and waste water management. The Raageshwari deep gas field (RDG) is a deep, tight, high condensate gas reservoir located in Rajasthan, India. The 15 wells were located on 3 separate pads within 2 km radius. Their location, in a sparsely populated arid part of India adds to the logistical challenges. Some of the key challenges during the project were: Key operations such as fracturing and perforating cannot be performed at night.Economic handling and disposal of waste water.Continuous supply of water suitable for fracturing and other wellsite operations.Limited space available on the well pads.Requirement for simultaneous operations involving high risk activities such as perforating and high pressure pumping.High CT cleanout frequency due to the requirement of underflushing the frac stages, close proximity of the adjacent stages, and the small volume in the 3 ½ inch monobore completions.Logistics and supply chain management in a remote location. Solutions for these obstacles included improved procedures, workflows and key technology introductions. These included: Simultaneous rig up to multiple wells.Simultaneous operations covered through Risk assessment and controls in place.Selective perforation technology to reduce perforating time.Mechanical evaporators and a robust CSR team for waste water management. The various optimizations resulted in a reduction in days per frac stage from 4 in the previous campaigns to 2 in this campaign. The project was delivered with 0 LTI (Lost Time Injuries), and a phenomenal HSE record due to robust safety procedures, frequent audits, safety drills etc. This paper will detail how the challenges in this project were overcome resulting allowing a high speed fracturing campaign to be successfully executed in a remote location.

Kinetics of batch fermentation of sugarcane juice for lactic acid production by Lactobacillus plantarum NCIM 2912 under anaerobic conditions has been analyzed. Lactic acid fermentation with this bacterium is found to be homolactic. Kinetics of cell growth and multiplication fits reasonably into Monod's kinetic model, and the Luedeking–Piret equation has been found to reliably predict kinetics of lactic acid production. The kinetic parameters have been determined on the basis of experimental data using a program developed in Microsoft Excel and matlab. The value of the key kinetic constants estimated is the maximum specific growth rate (μmax), 0.92/h; substrate limitation constant (Ks), 44.1 g/L; growth-associated constant (α), 10.08 g/g; and nongrowth-associated constant (β), 0.92 g/g h for lactic acid production. Maximum lactic acid concentration and productivity were found to be 89.5 g/L and 1.243 g/L h, respectively. Kinetic analysis thus revealed that lactic acid production is predominantly growth associated. Effect of substrate inhibition has been observed to be relatively small. Stoichiometric parameters have been estimated to be the maximum product yield coefficient YPS, 0.83 g, lactic acid/g sucrose and average cell mass yield coefficient Yxs, 0.47. © 2014 Curtin University of Technology and John Wiley & Sons, Ltd.

In this chapter, the impact of rising cost of fuel and inability of government to supply electricity round the clock leading to spurt in frugal innovations utilizing alternative sources of energy has been analyzed. Rice is one of the prime crops of India, Rice husk, perceived as waste by many, being abundant and available at low cost can be major contributor in alternate energy. Due to lack of awareness of the potential of rice husk it is discarded as waste and results in waste disposal problem and methane emission. The use of rice husk for electricity generation in efficient manner has potential to transform agricultural waste into a valuable fuel and could help in boosting farm economy. This study explores the usage of rice husk and its untapped potential as a major fuel source. The target markets are villages in India with 400-500 households having ample biomass feedstock.

In recent years there has been a significant increase in the use of AI based techniques in the field of industrial control. Different Intelligent techniques like Fuzzy Logic, Artificial Neural Network and other Hybrids have been successfully implemented for various control problems. Waste gases generated in refineries from different processes like the Claus process, cannot be directly released into the atmosphere as it has sulfur compounds and other harmful compounds. These waste gases are incinerated before discharging to make them harmless. The incineration process uses fuel gas for oxidizing waste gases, thus incinerator needs to be optimized to use minimum fuel gas and maximum oxidization. In this paper a neural network is modeled for optimal control of incinerator in Sulfur Recovery Block of refineries. Optimal control is achieved by means of an Artificial Neural Network based Inverse Plant model. The Neural Network model was developed by using the neural network tool box in MATLAB.