Papers by Keyword: Principal Component Analysis (PCA)

Abstract: Aluminum extrusion plays a critical role in lightweight structural applications and circular economy strategies. However, extrusion process design is constrained by competing objectives: increasing productivity through higher ram speeds or increased die-hole count improves throughput and material utilization yet simultaneously elevates force demand and defect susceptibility. In this work, a numerical statistical framework is proposed to identify circularity-tolerant process windows, defined as multi-objective design regions that balance productivity, product quality, and sustainability performance. A three-factor Taguchi design was employed to systematically vary ram speed, billet temperature, and die-hole count in the extrusion of AA6063. Twenty-seven full 3D thermo-mechanical extrusion simulations were conducted using the DEFORM finite element platform employing an Arrhenius-type constitutive model from literature. Key extrusion responses maximum ram force, local damage indicator, and total displacement were analyzed using Principal Component Analysis (PCA) to reveal correlations and trade-offs between productivity-oriented parameters and quality-related responses. The results demonstrate a clear divergence between productivity drivers (ram speed, die-hole count) and process capability indicators, providing quantitative evidence of the inherent productivity quality trade-off. The proposed framework enables the identification of robust extrusion operating regions suitable for circular manufacturing scenarios in aluminum extrusion. The proposed framework is particularly relevant for extrusion scenarios where process robustness must be ensured under increasing material and operational variability, such as those anticipated with higher recycled content.

Abstract: The activity of antibacterial material is conventionally estimated by using an indirect method – a bacteria suspension is inoculated onto a surface, and then the bacteria are collected from the surface and examined as to whether they can form colonies on the agar plate. In the present study, the presence of bacteria was examined by direct detection. Our study is based on FTIR-PAS with an interferometer cantilever detector. Our work discusses the possibility of identifying and distinguishing the presence of different bacteria (Staphylococcus epidermidis and Pseudomonas aeruginosa) and the possibility to evaluate the crystallization processes on the pressed calcium phosphate surface.

Abstract: Arrhythmia, a common form of heart disease, can be detected from an electrocardiogram (ECG) signal. This research work presents a comparative study between five feature extraction methods applied separately on two window sizes for detecting three ECG pulse types, namely normal and two arrhythmia variations. The library support vector machine (LIBSVM) was used to classify the three classes of the ECG pulses. The ECG signals were obtained from MIT-BIH database. The ECG dataset was normalized and filtered to remove any noise and after that the signals were windowed into two window sizes (long window and short window). Five approaches were used to extract the features from the ECG signals. These approaches are scalar Autoregressive model coefficients, Haar discrete wavelet transform (DWT), Daubechies (db) DWT, Biorthogonal (bior) DWT, and principal components analysis (PCA). Each approach was applied separately on the two window sizes. The results of the classification show that scalar Autoregressive model coefficients, Haar, db, and bior are better approaches to catch the ECG features for short window than the long window. However, PCA gave the closest and highest results for the two window sizes than other approaches. That mean the PCA is the better feature extraction approach for both window sizes.

Abstract: To evaluate the aging behavior of high density polyethylene (HDPE) under an artificial accelerated environment, principal component analysis (PCA) was used to establish a non-dimensional expression Z from a data set of multiple degradation parameters of HDPE. In this study, HDPE samples were exposed to the accelerated thermal oxidative environment for different time intervals up to 64 days. The results showed that the combined evaluating parameter Z was characterized by three-stage changes. The combined evaluating parameter Z increased quickly in the first 16 days of exposure and then leveled off. After 40 days, it began to increase again. Among the 10 degradation parameters, branching degree, carbonyl index and hydroxyl index are strongly associated. The tensile modulus is highly correlated with the impact strength. The tensile strength, tensile modulus and impact strength are negatively correlated with the crystallinity.

Abstract: This work addresses the problem of online fault detection of an advanced wind turbine benchmark under actuators (pitch and torque) and sensors (pitch angle measurement) faults of different type. The fault detection scheme starts by computing the baseline principal component analysis (PCA) model from the healthy wind turbine. Subsequently, when the structure is inspected or supervised, new measurements are obtained and projected into the baseline PCA model. When both sets of data are compared, a statistical hypothesis testing is used to make a decision on whether or not the wind turbine presents some fault. The effectiveness of the proposed fault-detection scheme is illustrated by numerical simulations on a well-known large wind turbine in the presence of wind turbulence and realistic fault scenarios.

Abstract: Applications of electronic noses to classify the freshness of food and beverages by mimicking the olfactory perception are becoming widely recognized in food industries. For pasteurized orange juice, packaging and shelf-life are key factors for the quality control, which are generally inspected by the sensory stability and quality (odor, color, texture and taste) of the orange juice. An electronic nose based on five different commercial metal oxide gas sensors, a temperature sensor and a humidity sensor has been designed and constructed to examine the quality of orange juice as subjected to the fermentation process. The duration for a single measurement from an orange juice sample was approximately two minutes. The data acquisition of the voltage responses of the gas sensors were achieved via a microcontroller unit. The data classification was statistically analyzed by the “Principal Component Analysis (PCA)”. The Euclidean distance between two PCA groups was used as an indicator of ethanol concentration. The orange juice was laced with various concentrations of ethanol from 0.1 to 1.0% ethanol to simulate fermented orange juice at different stages. The objective was to characterize the freshness of orange juice by means of the ethanol level from the fermentation process. The results show a distinctive classification of the orange juice for an alcohol concentration lower than 0.1%. Thus the electronic nose offers a rapid, highly sensitive alternative for the quality control process.

Abstract: This paper proposed a theoretically efficient approach for face recognition based on principal component analysis (PCA) and rotation invariant uniform local binary pattern texture features in order to weaken the effects of varying illumination conditions and facial expressions. Firstly, the rotation invariant uniform LBP operator was adopted to extract the local texture feature of the face images. Then PCA method was used to reduce the dimensionality of the extracted feature and get the eigenfaces. Finally, the nearest distance classification was used to distinguish each face. The method has been accessed on Yale and ATR-Jaffe face databases. Results demonstrate that the proposed method is superior to standard PCA and its recognition rate is higher than the traditional PCA. And the proposed algorithm has strong robustness against the illumination changes, pose, rotation and expressions.

Abstract: Based on MMW-1A vertical multifunctional friction and wear tester for the study,taking steel 45 as the research object, randomly changing the experiment load, speed, sliding distance and the size of the contact area, then the data we collect are processed and analyzed by principal component analysis, and obtained linear regression models by principal component regression, regression model has been tested with good fitting effect. The results showed that the principal component analysis method is also suitable for experimental study of friction and wear, explore new methods in the analysis of tribology. It shows that load, speed and sliding distance have a weakening effect on the friction coefficient, on the contrary, the contact area has a promoting role to the friction coefficient.

Abstract: The principal component analysis (PCA) is applied in this paper, since the existing power consumption prediction models of cement manufacturing influenced by many factors are quite complex and have low accuracy. In this way, four new key factors affecting the power consumption of cement manufacturing are obtained instead of the eleven original ones, with the complexity of the computing model simplified. Built upon this is the power consumption prediction model of cement manufacturing based on an improved multiple non-linear regression algorithm. Then the efficiency of the model, obviously improved the forecasting precision, is verified in Pingyi Zhonglian Cement Plant. In other words, a theoretical basis for cement plants power consumption forecasting management is provided in this paper.

Abstract: In the paper, by designing evaluation system, capital construction investment efficiency of 13 coal mines in Shanxi province is analyzed through using principal component analysis. Meanwhile comprehensive scores of capital construction efficiency are calculated. The highest score and the lowest score are 3.813 and-1.141, respectively. Finally, 13 coal mines are divided into three levels, sufficiently distinguishing capital construction efficiency of those coal mines.