Order Polynomial Regression Model Research Articles

Optimized Conditions for the Production of Yogurt from Soya Bean (Glycine max) and Cinnamon (Cinnamomum verum)

This study aimed at producing functional soy bean yoghurt by optimizing the production conditions using response surface methodology. The quantity of ferment (LYOFAST Y439A) (50–100 g) and quantity of cinnamon (10-40 g) were optimized using central composite design. Responses (global acceptability and scavenging activity) obtained from experimental runs were fitted into second order polynomial regression model. Also, the multiple optimization technic was used to obtain the compromised optimum condition. The optimized yoghurt was evaluated for its pH and proximate composition using standard methods. The optimum conditions for the production of this yoghurt was as thus: 69.64 g for quantity of ferment and 10 g for quantity of cinnamon, 100 g soya beans, 100 g of sugar and 1 L of water soybeans. Soy bean yoghurt made from optimized conditions had a global acceptability of 6.8, scavenging activity of 34 %. Also, protein, fat, carbohydrate, calcium, sodium magnesium and energy value was respectively 17.1%, 2.91%, 17.93%, 144 mg/100 g, 159.91 mg/100g, 159.91 mg/100g, 63.18 mg/100g and 166.31 kcal/100g. Thus, acceptable yoghurt with functional properties can be obtained from soya beans and cinnamon.

Hot Deformation Behavior of Inconel 718 Alloy Using Regression-based Dynamic Material Model and Applications in Ring Rolling Manufacturing

Inconel 718 nickel-based alloy is extensively used in the aerospace industry (e.g., gas turbine engine components) because of its excellent corrosion resistance and high mechanical properties at elevated temperatures. However, there is a certain limit to manufacturing the alloy through plastic deformation due to its high deformation resistance and complicated deformation behaviors. In this study, the hot deformation behavior of Inconel 718 alloy was investigated to establish how processing conditions of flow stress-strain, at strain rates from 0.001 to 10 s<sup>-1</sup>, and temperatures from 850 to 1200<sup>o</sup>C, affected dynamic recrystallization. The regression-based material model was utilized to calculate the strain-rate sensitivity, and subsequently depict the efficiency of the power dissipation and instability criterion of hot deformation. The processing map and instability criterion predicted by the developed 3<sup>rd</sup>-order polynomial regression model corresponded with the experimental results and in particular, showed a better prediction for instability regime compared to the existing discrete derivative approach. Predicting the strain-rate sensitivity values on a continuous scale with regression analysis covered the additional instability region of the high strain rate near 10 s<sup>-1</sup>. The dynamic recrystallization deformation was also characterized by microstructural analysis along with the processing map. Consequently, ring-rolled aviation parts were manufactured with the optimum processing parameters, which conform to the AMS 5663 standard (Aerospace material specifications for Inconel 718).

Differences in lower extremity joint stiffness during drop jump between healthy males and females

The primary purpose of this study was to examine sex differences in lower extremity joint stiffness during vertical drop jump performance. A secondary purpose was to examine the potential influence of sex on the relationship between joint stiffness and jump performance. Thirty healthy and active individuals performed 15-drop jumps from 30 and 60 cm boxes. Hip, knee, and ankle joint stiffnesses were calculated for subphases of landing using a 2nd order polynomial regression model. Males had greater hip stiffness during the loading phase in drop jumps from both box heights than females’ drop jump from 60 cm box. Also, males had a greater ground reaction force at the end of eccentric phase, net jump impulse, and jump height regardless of box height. The 60 cm box height increased knee stiffness during the loading phase, but reduced hip stiffness during the loading phase and knee and ankle stiffness during the absorption phase regardless of sex. Joint stiffnesses significantly predicted drop jump height for females (p < .001, r2 = 0.579), but not for males (p = .609, r2 = −0.053). These results suggest that females may have different strategies to maximize drop jump height as compared to males.

Predicting tensile strength of cemented paste backfill with aid of second order polynomial regression

The materials left behind after the process of separating an ore's valuable fraction from the unprofitable fraction are known as tailings in the mining industry. Mixing tailing, cement and water can create a new material called Cemented paste backfill (CPB). Research and solve the problem of predicting the tensile strength of cement paste backfill based on a polynomial model combined with the Monte Carlo Simulation method. Three models were built to evaluate performance. The optimal performance model is then used to predict the tensile strength of cement paste backfill. The results indicate that using the polynomial regression model has a satisfactory result for predicting the tensile strength of cement paste backfill. The best performance of second order polynomial regression model is evaluated by three metrics such as R2=0.958 RMSE=33.211 kPa MAE=29.097 kPa for testing part in predicting the tensile strength of cemented paste backfill. Finally, the influence of Cement/Tailings ratio and Solid content on the tensile strength on tensile strength and importance is also evaluated with aid of the best performance of second order polynomial regression model.

Predicting tensile strength of cemented paste backfill with aid of second order polynomial regression

The materials left behind after the process of separating an ore's valuable fraction from the unprofitable fraction are known as tailings in the mining industry. Mixing tailing, cement and water can create a new material called Cemented paste backfill (CPB). Research and solve the problem of predicting the tensile strength of cement paste backfill based on a polynomial model combined with the Monte Carlo Simulation method. Three models were built to evaluate performance. The optimal performance model is then used to predict the tensile strength of cement paste backfill. The results indicate that using the polynomial regression model has a satisfactory result for predicting the tensile strength of cement paste backfill. The best performance of second order polynomial regression model is evaluated by three metrics such as R2=0.958 RMSE=33.211 kPa MAE=29.097 kPa for testing part in predicting the tensile strength of cemented paste backfill. Finally, the influence of Cement/Tailings ratio and Solid content on the tensile strength on tensile strength and importance is also evaluated with aid of the best performance of second order polynomial regression model.

Predicting tensile strength of cemented paste backfill with aid of second order polynomial regression

The materials left behind after the process of separating an ore's valuable fraction from the unprofitable fraction are known as tailings in the mining industry. Mixing tailing, cement and water can create a new material called Cemented paste backfill (CPB). Research and solve the problem of predicting the tensile strength of cement paste backfill based on a polynomial model combined with the Monte Carlo Simulation method. Three models were built to evaluate performance. The optimal performance model is then used to predict the tensile strength of cement paste backfill. The results indicate that using the polynomial regression model has a satisfactory result for predicting the tensile strength of cement paste backfill. The best performance of second order polynomial regression model is evaluated by three metrics such as R2=0.958 RMSE=33.211 kPa MAE=29.097 kPa for testing part in predicting the tensile strength of cemented paste backfill. Finally, the influence of Cement/Tailings ratio and Solid content on the tensile strength on tensile strength and importance is also evaluated with aid of the best performance of second order polynomial regression model.

Predicting tensile strength of cemented paste backfill with aid of second order polynomial regression

The materials left behind after the process of separating an ore's valuable fraction from the unprofitable fraction are known as tailings in the mining industry. Mixing tailing, cement and water can create a new material called Cemented paste backfill (CPB). Research and solve the problem of predicting the tensile strength of cement paste backfill based on a polynomial model combined with the Monte Carlo Simulation method. Three models were built to evaluate performance. The optimal performance model is then used to predict the tensile strength of cement paste backfill. The results indicate that using the polynomial regression model has a satisfactory result for predicting the tensile strength of cement paste backfill. The best performance of second order polynomial regression model is evaluated by three metrics such as R2=0.958 RMSE=33.211 kPa MAE=29.097 kPa for testing part in predicting the tensile strength of cemented paste backfill. Finally, the influence of Cement/Tailings ratio and Solid content on the tensile strength on tensile strength and importance is also evaluated with aid of the best performance of second order polynomial regression model.

Structural and compressibility properties of weft-knitted rib fabrics from glass yarn

The structural and compressibility properties of the weft-knitted glass yarn fabrics from 1x1, 2x2, English, and fisherman rib architectures were investigated in this study. Due to their tight structures; 2x2 and fisherman rib fabric architectures exhibited higher loop density, and shorter loop length than 1x1 and English rib fabric architectures. English and fisherman rib fabric architectures displayed higher fiber volume fraction than 1x1 and 2x2 rib architectures in multi-layer compaction and recovery tests where the pressure was varied between 2 and 200 kPa. Number of layer increased the fiber content that pointed the nesting between the fabric layers. As a result of lack of complete recovery from compression; the fabrics exhibited lower thicknesses (i.e. higher fiber volume fractions) during the recovery periods than they did during the compression periods. A second order polynomial regression model with 0,89 R2 (coefficient of determination) was developed to estimate the fiber volume fraction by means of knit architecture, number of fabric layers, pressure, and measurement period.

Development of non-linear models to evaluate the NiTi SMA spring actuator

The shape memory alloy (SMA) based actuators are replacing huge and bulky actuators because of its ability to provide high work per unit mass and serves as active prismatic joint to develop precise robotic manipulators. It also helps in developing a light weight manipulators with a simple actuation process. Here, the study presents the relationship between the contraction of Nitinol SMA prings and the input variables such as current and time to effectively interpret the behavioural complexity. In addition, the response of two Shape memory alloy springs in series combination has been discussed. The finite element analysis (FEA) of the SMA wire and spring has also been carried out to predict the fatigue life of SMA wire and spring. The result showed that the contraction rate of SMA spring increases with increase in current and vice-versa. Moreover, the range of current is classified based on its significance. The relationship of current, displacement and time parameters, during the actuation of SMA spring, is a second order polynomial regression model. The applied current and time have positive impact on the contractiong rate of SMA. Several polynomial regression models were developed in order to predict the precise amount of spring actuation. This study also predicts the range of current suitable for its actuation based on its application as actuator. The FEA result shows that the SMA springs can have high endurance stress limit which makes it unique as compared to other commercially available actuators. This study enables to predict the rate of contraction and the deflection trend of SMA based actuators for precise positioning applications.

Combining Global Geopotential Models, Digital Elevation Models, and GNSS/Leveling for Precise Local Geoid Determination in Some Mexico Urban Areas: Case Study

This work shows improvements of geoid undulation values obtained from a high-resolution Global Geopotential Model (GGM), applied to local urban areas. The methodology employed made use of a Residual Terrain Model (RTM) to account for the topographic masses effect on the geoid. This effect was computed applying the spherical tesseroids approach for mass discretization. The required numerical integration was performed by 2-D integration with 1DFFT technique that combines DFT along parallels with direct numerical integration along meridians. In order to eliminate the GGM commission error, independent geoid undulations values obtained from a set of GNSS/leveling stations are employed. A corrector surface from the associated geoid undulation differences at the stations was generated through a polynomial regression model. The corrector surface, in addition to the GGM commission error, also absorbs the GNSS/leveling errors as well as datum inconsistencies and systematic errors of the data. The procedure was applied to five Mexican urban areas that have a geodetic network of GNSS/leveling points, which range from 166 to 811. Two GGM were evaluated: EGM2008 and XGM2019e_2159. EGM2008 was the model that showed relatively better agreement with the GNSS/leveling stations having differences with RMSE values in the range of 8–60 cm and standard deviations of 5–8 cm in four of the networks and 17 cm in one of them. The computed topographic masses contribution to the geoid were relatively small, having standard deviations on the range 1–24 mm. With respect to corrector surface estimations, they turned out to be fairly smooth yielding similar residuals values for two geoid models. This was also the case for the most recent Mexican gravity geoid GGM10. For the three geoid models, the second order polynomial regression model performed slightly better than the first order with differences up to 1 cm. These two models produced geoid correction residuals with a standard deviation in one test area of 14 cm while for the others it was of about 4–7 cm. However, the kriging method that was applied for comparison purposes produced slightly smaller values: 8 cm for one area and 4–6 cm for the others.

Utilizing Design of Experiments Approach to Assess Kinetic Parameters for a Mn Homogeneous Hydrogenation Catalyst

Homogeneous hydrogenation catalysts based on metal complexes provide a diverse and highly tunable tool for the fine chemical industry. To fully unleash their potential, fast and effective methods for the evaluation of catalytic properties are needed. In turn, this requires changes in the experimental approaches to test and evaluate the performance of the catalytic processes. Design of experiment combined with statistical analysis can enable time‐ and resource‐efficient experimentation. In this work, we employ a set of different statistical models to obtain the detailed kinetic description of a highly active homogeneous Mn (I) ketone hydrogenation catalyst as a representative model system. The reaction kinetics were analyzed using a full second order polynomial regression model, two models with eliminated parameters and finally a model which implements “chemical logic”. The coefficients obtained are compared with the corresponding high‐quality kinetic parameters acquired using conventional kinetic experiments. We demonstrate that various kinetic effects can be well captured using different statistical models, providing important insights into the reaction kinetics and mechanism of a complex catalytic reaction.

Predictive model of water stress in tenera oil palm by means of spectral signature methods

Agriculture as a competitive business, seeks to improve productivity within crops with a more sustainable environmental management. It is important that agriculture includes new technologies that allow it to generate differential, precise and real-time information. In Colombia, the current lack of knowledge about techniques that allow early identification of water stress in African palm could generate a loss in the investment made in the fertilization of the crop, cause an increase in diseases, pests, and susceptibility to compaction or abortions in female flowers that would lead to decreases in production. In this work, a predictive model is established to quantify water stress based on spectral, physiological and soil information in African palm plants. To this end, a study was carried out in an oil palm plantation where treatments were established with 3 ranges of humidity. It was found that the indices with the highest correlation with the biophysical variable soil moisture were: NDVI_1 and NDVI_16 for treatment 1, SR_4 for treatment 2 and NDVI_16 and NDVI_20 for treatment 3. Finally, the third order polynomial regression model that obtained higher correlation coefficients of Pearson R^2=0.73 was selected as the most suitable model to estimate soil moisture content for treatments 2 and 3.

Study of mathematics modeling on ginger geometric changes during drying using image analysis

Many studies have examined the drying process, but only a few studies have examined shrinkage as one of the parameters that affect drying. Depreciation of agricultural products could have an impact on damage and reduce quality. This study aimed to examine the shrinkage pattern of product size during drying using the mathematical model approach regression. The method used was experimental with sliced ginger as material. The equipment used was a rack-type hybrid dryer and a set of image capture devices. This study indicates that to dry the sliced ginger from 95.65% to 9.49% moisture content required approximately 8 hours. The results of image processing showed shrinkage that occurs in sliced ginger was 52.49%. The most accurate mathematical model describing the prediction of sliced ginger shrinkage during drying was the 2nd order polynomial regression model with the equation of y = - 131.3.x 2 - 1804.4x + 41941.6 and an R2 value of 0.99.

Machine Learning-based Flu Forecasting Study Using the Official Data from the Centers for Disease Control and Prevention and Twitter Data

Aim/Purpose: In the United States, the Centers for Disease Control and Prevention (CDC) tracks the disease activity using data collected from medical practice's on a weekly basis. Collection of data by CDC from medical practices on a weekly basis leads to a lag time of approximately 2 weeks before any viable action can be planned. The 2-week delay problem was addressed in the study by creating machine learning models to predict flu outbreak. Background: The 2-week delay problem was addressed in the study by correlation of the flu trends identified from Twitter data and official flu data from the Centers for Disease Control and Prevention (CDC) in combination with creating a machine learning model using both data sources to predict flu outbreak. Methodology: A quantitative correlational study was performed using a quasi-experimental design. Flu trends from the CDC portal and tweets with mention of flu and influenza from the state of Georgia were used over a period of 22 weeks from December 29, 2019 to May 30, 2020 for this study. Contribution: This research contributed to the body of knowledge by using a simple bag-of-word method for sentiment analysis followed by the combination of CDC and Twitter data to generate a flu prediction model with higher accuracy than using CDC data only. Findings: The study found that (a) there is no correlation between official flu data from CDC and tweets with mention of flu and (b) there is an improvement in the performance of a flu forecasting model based on a machine learning algorithm using both official flu data from CDC and tweets with mention of flu. Recommendations for Practitioners: In this study, it was found that there was no correlation between the official flu data from the CDC and the count of tweets with mention of flu, which is why tweets alone should be used with caution to predict a flu out-break. Based on the findings of this study, social media data can be used as an additional variable to improve the accuracy of flu prediction models. It is also found that fourth order polynomial and support vector regression models offered the best accuracy of flu prediction models. Recommendations for Researchers: Open-source data, such as Twitter feed, can be mined for useful intelligence benefiting society. Machine learning-based prediction models can be improved by adding open-source data to the primary data set. Impact on Society: Key implication of this study for practitioners in the field were to use social media postings to identify neighborhoods and geographic locations affected by seasonal outbreak, such as influenza, which would help reduce the spread of the disease and ultimately lead to containment. Based on the findings of this study, social media data will help health authorities in detecting seasonal outbreaks earlier than just using official CDC channels of disease and illness reporting from physicians and labs thus, empowering health officials to plan their responses swiftly and allocate their resources optimally for the most affected areas. Future Research: A future researcher could use more complex deep learning algorithms, such as Artificial Neural Networks and Recurrent Neural Networks, to evaluate the accuracy of flu outbreak prediction models as compared to the regression models used in this study. A future researcher could apply other sentiment analysis techniques, such as natural language processing and deep learning techniques, to identify context-sensitive emotion, concept extraction, and sarcasm detection for the identification of self-reporting flu tweets. A future researcher could expand the scope by continuously collecting tweets on a public cloud and applying big data applications, such as Hadoop and MapReduce, to perform predictions using several months of historical data or even years for a larger geographical area.

Machine Learning-Based Predictive Techno-Economic Analysis of Power System

This paper proposes a predictive techno-economic analysis in terms of voltage stability and cost using regression-based machine learning (ML) models and effectiveness of the analysis is validated. Predictive analysis of a power system is proposed to address the need for faster and accurate analyses that would aid in the operation and control of modern power system. Several methods of analyses including metaheuristic optimization algorithms, artificial intelligence techniques and machine learning algorithms are being developed and used. Predictive ML models for two modified IEEE 14-bus and IEEE-30 bus systems, integrated with renewable energy sources (solar and wind) and reactive power compensative device (STATCOM) are proposed and developed with features that include hour of the day, solar irradiation, wind velocity, dynamic grid price and system load. An hour-wise input database for the model development is generated from monthly average data and hour-wise daily curves with normally distributed standard deviations. The data feasibility tests and output database generation is performed using MATLAB. Linear and higher order polynomial regression models are developed for the 8760hr database using Python 3.0 in JupyterLab and a best-fit predictive ML model is identified by analysing the coefficients of determination. The voltage stability and cost predictive ML models were tested for a 24hr input profile. The results obtained and the comparison with the expected values are furnished. Prediction of the outputs for the test data validate the accuracy of the developed model.

Experimental and statistical analysis of strength of glass fiber reinforced polymer composite for different fiber architecture

AbstractIn the present study, the effect of fiber architecture on mechanical properties of E‐glass fiber reinforced polymer composite has been investigated experimentally. The influence of fiber volume fraction on strength of composite has also been predicted using second order polynomial regression models. The glass fibers were arranged in different architecture like unidirectional continuous, unidirectional discontinuous, short randomly oriented, transversely aligned, and woven for preparation of composites. The tensile and flexural tests were performed as per ASTM D3039 and ASTM D790, respectively. The experimental results of tensile and flexural properties of composites have been discussed. Results predicted by the regression models are in good agreement with experimental results which indicate the goodness of fit and the models are statistically significant.

Response Surface Methodology Optimization and Kinetics of Diesel Degradation by a Cold-Adapted Antarctic Bacterium, Arthrobacter sp. Strain AQ5-05

Petroleum hydrocarbons, notably diesel oil, are the main energy source for running amenities in the Antarctic region and are the major cause of pollution in this area. Diesel oil spills are one of the major challenges facing management of the Antarctic environment. Bioremediation using bacteria can be an effective and eco-friendly approach for their remediation. However, since the introduction of non-native organisms, including microorganisms, into the Antarctic or between the distinct biogeographical regions within the continent is not permitted under the Antarctic Treaty, it is crucial to discover native oil-degrading, psychrotolerant microorganisms that can be used in diesel bioremediation. The primary aim of the current study is to optimize the conditions for growth and diesel degradation activity of an Antarctic local bacterium, Arthrobacter sp. strain AQ5-05, using the Plackett-Burman approach and response surface method (RSM) via a central composite design (CCD) approach. Based on this approach, temperature, pH, and salinity were calculated to be optimum at 16.30 °C, pH 7.67 and 1.12% (w/v), respectively. A second order polynomial regression model very accurately represented the experimental figures’ interpretation. These optimized environmental conditions increased diesel degradation from 34.5% (at 10 °C, pH 7.00 and 1.00% (w/v) salinity) to 56.4%. Further investigation of the kinetics of diesel reduction by strain AQ5-05 revealed that the Teissier model had the lowest RMSE and AICC values. The calculated values for the Teissier constants of maximal growth rate, half-saturation rate constant for the maximal growth, and half inhibition constants (μmax, Ks, and Ki), were 0.999 h−1, 1.971% (v/v) and 1.764% (v/v), respectively. The data obtained therefore confirmed the potential application of this cold-tolerant strain in the bioremediation of diesel-contaminated Antarctic soils at low temperature.

Fabrication of nanostructured mesoporous starch encapsulating soy-derived phytoestrogen (genistein) by well-tuned solvent exchange method

The present research was concerned with preparation of mesoporous starch (MPS) as a carrier for genistein, a model of poorly water-soluble phytoestrogen isoflavone; and exploration of the impact of different fabrication parameters on structural and loading properties. MPS is considered as a highly porous biomaterial which typically possesses nanometer-sized porous microstructure and low density, providing a large effective specific surface area (SSA) and hydrophilic surface to improve solubility, stability and bioavailability of poorly water-soluble active agents. To fabricate MPS, various concentrations (8–14% w/v) of starch from different sources (corn, potato and tapioca) was used for gel formation and the successive solvent exchange process was performed with use of various ethanol concentrations (40–70% v/v), which were then dried by different techniques (rotary vacuum evaporation, microwave and freeze drying). MPS quality attributes such as SSA, total porous volume, BJH pore diameter and swelling ratio were determined and effects of the fabrication parameters were investigated using L9-Taguchi orthogonal array design. The results indicate that second order polynomial regression models were well fitted for all response variables. Interestingly, the starch components greatly influenced physical properties of MPS. Also, the drying type and ethanol concentration altered significantly the model equations. The overall best fabrication condition (14% corn starch, 100% ethanol concentration in aging step and rotary vacuum drying) resulted in favorable MPS preparation with mean size of 105.4 nm and unimodal distribution. In the next step, genistein was encapsulated in MPS microstructure at different ratios, resulting in high loading capacity and efficiency (44.71% and 79.9%, respectively) at 1:1 weight ratio. Equilibrium adsorption isotherm of genistein was evaluated also by four different kinetics models including Langmuir, Freundlich, Dubinin-Radushkevich, and Temkin isotherms. The experimental data were found to be fitted well to the Langmuir model (R2 = 0.989). According to the electron microscopy and XRD analysis, the degree of genistein crystallinity lowered remarkably after the impregnation in to MPS, indicating improved solubility. In-vitro release profile of genistein from MPS in the simulated gastrointestinal buffer solutions (pH 1.2 and 6.8) demonstrated that incorporating genistein into the MPS enhanced the dissolution rate compared with genistein powder. Release kinetic data were fitted to the Higuchi model (R2 = 0.98), indicating diffusion-controlled release mechanism. Altogether, well-tuned MPS fabrication method can be utilized for an efficient encapsulation and dissolution enhancement of poorly soluble phytochemicals, such as genistein.

Abstract TP146: Rate of Functional Improvement for Stroke Patients Plateaus During Inpatient Rehabilitation

Introduction: Functional outcomes and length of stay (LOS) are common outcome measures in stroke patients at Inpatient Rehabilitation Facilities (IRF). Aim: To examine trends in functional outcomes among patients with hemorrhagic versus ischemic stroke in IRF. Methods: Using a prospectively collected database of stroke patients admitted in our health system of 5 IRFs in Houston, we reviewed patients with either ischemic or hemorrhagic stroke between 1/18 to 6/19. The main outcome measure was the FIM scores. The relationship between LOS and FIM score improvement were analyzed using a third order polynomial regression model. Results: Among 88 patients, 43% were female and 60% had an ischemic stroke. The median LOS was 21 days (IQR14, 26) for all patients; 22 (IQR16, 30) in patients with hemorrhagic stroke; 20.5 (IQR14, 25) in patients with ischemic stroke (P: 0.24). Based on change in FIM, stroke patients benefited the most from IRF in the first 17 days (P&lt;.0001) while the improvement rate declined significantly afterwards with the highest FIM score change of 32. The FIM score changes were not significantly different when adjusting for CMI. There were no significant differences in FIM changes and extent of temporal benefit between ischemic and hemorrhagic strokes. Moreover, bathing, dressing, tub transfer and walking were the areas of greatest improvement (Median change: 2 for all variables). Conclusion: Stroke patients in our IRF network benefit the most on FIM in the first 17 days. Further studies are needed to identify factors that could maximize functional gains in IRF.

Modelling and prediction of strength of ultrasonically welded electrical contact joints using Artificial Neural Network

Abstract Ultrasonic welding is an advanced green solid state welding process used to weld similar or dissimilar materials of same or dissimilar shapes. Ultrasonic welding is used to join both conductive and non-conductive materials. Copper is an excellent conductive non-ferrous metal widely used in wide variety of electrical and electronic components manufacturing applications. Many electrical components have a need to join copper wire to copper sheet and these applications demands inevitably for superior weld strength. In this paper, ultrasonic welding is carried out on copper wire-sheet to produce quality electrical contact joints. The present study deals with the formulation of mathematical model between the predominant input parameters such as clamping force, amplitude of vibration of sonotrode and weld time and the output parameter such as weld strength using Artificial neural network and regression technique. Based on the results from experiments and prediction models, the ANN model is found to be effective when compared with the developed third order polynomial regression model