Papers by Keyword: Regression

Abstract: In the electroless nickel-boron coating process, surfactant helps to minimize the surface tension between the substrate and the electrolyte in the bath. Despite, its high cost and the formation of micelles from monomeric surfactant molecules at its critical micelle concentration (CMC), it is essential to optimize the concentration while using in the bath. In this study, to solve this problem, mathematical models are developed using regression and artificial neural network (ANN) techniques to relate the concentration of amphoteric surfactant (0-0.162 g/L) as an independent variable and microhardness as a dependent variable. Then, the developed model was used to optimize microhardness at CMC using a genetic algorithm (GA). The goodness of fit of the models was evaluated using the coefficient of determination (R²). The ANN model was found to be the best fit with R² = 0.99. The maximum microhardness of 852 HV was achieved at the CMC of 0.064 g/L, from the GA using the validated model as a fitness function.

Abstract: Today’s markets are rather matured and arbitrage opportunities remain for a very short time. The main objective of the paper is to devise a stock market ontology-based novel trading strategy employing machine learning to obtain maximum stock return with the highest stock ratio. The paper aims to create a dynamic portfolio to obtain high returns. In this work, the impact of the applied machine learning techniques on the Chinese market was studied. The problem of investing a particular total amount in a large universe of stocks is considered. The Chinese stocks traded on Shanghai Stock Exchange and Shenzhen Stock Exchange are chosen to be the entire universe. The inputs that are considered are fundamental data and company-specific technical indicators unlike the macroscopic factors considered in the existing systems. In the stock market document repository, ontological constructs with Word Sense Disambiguation (WSD) algorithm improve the conceptual relationships and reduce the ambiguities in Ontological construction. The machine learning techniques Kernel Regression and Recurrent Neural Networks are used to start the analysis. The predicted values of stock prices from the Artificial Neural Network provided quite accurate results with an accuracy level of 97.55%. In this study, the number of nodes will be selected based on Variance-Bias plots by tracking the error on the in-sample data set and the validation data set.

Abstract: Stock analysis and forecasting is a very challenging study due to the unpredictable and volatile database environment. However, their patterns are often unique as they are influenced by many uncertainties, such as financial results of companies (Earnings per share), risk transactions, market sentiment, government policies, and conditions such as epidemics. Even though they are challenging our goal is to predict the accurate values within a shorter span of a dataset. In this paper we have compared and analyzed the best ML model that predicts the exact closing amount of the next few days, using three to four months of nifty50 Indian stock from Yahoo Finance. Five regression models are involved in this analysis, Linear Regression (LR), Decision Tree (DT), Support Vector Regression (SVR), SARIMAX (Integrated Seasonal Integrated Season with EXogenous features), Gated Recurrent Unit (GRU – deep learning). The performance metrics like RMSE (Root Mean Squared Error), MAE (Mean Absolute Error) and MAPE (Mean Absolute Percentage Error) are used. On the basis of our comparison, we would like to conclude that GRU provides a low error value in all three performance metrics and also gives accurate predictions compared to the other five regression models used.

Abstract: Photochemical machining (PCM) is an emerging method for machining of very thin and difficult-to-cut material with complex geometrical profile. PCM is one of recommended method for machining of aerospace components, biomedical appliances, electronics part and decorative items. High corrosion resistance, better life, good appearance and strength recommend SS-430 as suitable material for various applications. In the current investigation, the parametric investigations of process parameters in photochemical machining for concentration and temperature of etchant, time of etching is done through ANOVA analysis. Grey Relational Analysis is performed to estimate the optimum machining parameters during PCM of SS-430. Formulation of mathematical model is done for prediction of results. Taguchi (L₂₇) experimental array is used for Design of Experiments (DoE). The significance process parameters are estimated to govern the process with F-Values. Confirmatory test is conducted to observe the improvement in the responses. ANN predictive model is built up for investigation of error between predictive and experimental values. The obtained optimum set is used for manufacturing of micromesh typically used in smoke detector to safeguard human life.

Abstract: Palm oil fuel ash (POFA) is solid waste from palm oil industries discarded after burning of shell, fiber, kernel and empty fruit bunches to heat boiler and generate electricity. A standard porcelain consisting of clay, feldspar and quartz is produced by sintering at temperature between 1300 °C to 1400 °C for toughness and translucency. This research developed a prediction model for optimum physical and mechanical properties of porcelain by addition of Fe₂O₃ at 5 wt.%. Quartz was replaced with POFA powder at 15 wt.% and Fe₂O₃ was also added at 5 wt.% of POFA mixed with others porcelain composition. Then the powder was dry pressed into pellet at 91 MPa and the samples were sintered at 1150 °C. The bulk density, compressive strength and Vickers microhardness were found to increase by addition of Fe₂O₃ at 5 wt.%. Prediction model was developed and from the predicted values it is revealed that, the model is efficient and good for the purpose of this research.

Abstract: The application range of natural fiber composites are increasing rapidly in several engineering areas. Unidirectional short castor oil fiber reinforced epoxy composites are fabricated using hand layup process with 40 vol% and 5mm length. Dry sliding wear tests are conducted using pin on disc tribometer. Applied loads are 15, 30 & 45N and track diameters were kept at 100, 110 and 120mm to achieve sliding distances of 1000, 2000 and 3000m respectively with adjustment of running times 6.5, 12 and 16 minutes at speed of 500 RPM. Full factorial DoE is employed and influence of each parameter on amount of wear, CoF and temperature are studied using ANOVA. It is found that load is the highly influential factor affecting amount of wear, CoF and temperature followed by sliding distance and other factors. Also, regression models are developed with good fit. The developed models predicted the results with 0-8 % error.

Abstract: The machinability of fiber reinforced composites is emphatically affected not only by the kind of fiber used in the composite but also by its properties. Milling composite materials is a very usual and plays a vital role for the assembly of composite structures. However, milling of composites is a somewhat complicated errand inferable from the heterogeneity of the material and a plenty of different issues, for instance, delamination factor, which show up amid the machining process and are connected with the qualities of the material and the cutting parameters. Present study looks into the influence of spindle speed, feed rate and depth of cut on thrust force, torque and delamination factor in hemp fiber reinforced polyester composites. The composite specimens were formulated using hand lay-up method. The analysis for attaining the optimality condition is performed using ANOVA and regression model. It was observed that the grouping of the inferior value of feed (0.1 mm/rev), lower spindle speed (1500 rpm) as well as the lower depth of cut (1.5 mm) results in minimum delamination factor and torque. The feed was found to be more significant than the depth of cut for thrust force.

Abstract: The aim of this study is to model the distribution patterns of the different mechanical properties of a submerged arc welded pipeline steel API X70 and to investigate the relationship between Vickers hardness and other mechanical properties of API X70. In this study, serial mechanical properties of 70 pipes, formed by spiral submerged arc welding of high strength low alloy steel (HSLA) API X70, were measured in base metal and weldments. Four main statistical distributions: Normal, Log-normal, Weibull and smallest extreme value distributions were chosen to test the goodness of fit to the experimental data. As a result, normal and lognormal distributions can equally model the distribution patterns of the whole experimental data of studied mechanical properties except for the hardness and toughness of the base metal that can be approximated by Weibull and smallest extreme value distributions, respectively. Using the current data, a weak but statistically significant correlation is obtained only between the toughness of the fusion zone and the hardness of both the base metal and the heat affected zone. Consequently, the calculated regression models were unable to estimate impact toughness values based on future measures of Vickers hardness components.

Abstract: The aim of the paper is to study friction force behavior by experimental investigation in the sliding of Inconel 600 alloy. The Inconel alloys are high-temperature Ni-Based alloy which have various applications like gas turbines, heat exchangers, condensers etc. The sliding tests are performed on a pin-on-disc tribometer. The Design of experiments technique is used to find out the number of experiments by using Taguchi’s L27 orthogonal array which is developed in MINTAB 14 software. The experiments were reduced for the various combinations to get the optimal parameters. The results showed that the friction force increases with the increase of load and speed. Also it is found that the speed was most significant parameter affecting frictional force. The regression model helps us to model the values for the responses.

Abstract: The end milling process is most commonly used where the large amount material can be removed to produce almost final shape of component. The present work deals with the experimental study and optimization the machining parameter of AISI 304 stainless steel. The effects of spindle speed, feed rate and depth of cut have been studied on the cutting force and surface roughness using Taguchi’s 27 orthogonal arrays. Regression analyses were used to develop the model of response parameters. The analysis of the result shows, the surface roughness and the cutting force is increased with feed rate and depth of cut but decreased with increased the cutting speed. The ANOVA indicate the feed rate was the most dominate parameter on surface roughness and cutting force than speed and depth of cut.