Detection of Adulteration in Canola Oil by Using GC-IMS and Chemometric Analysis

The aim of the present study was to detect adulteration of canola oil with other vegetable oils such as sunflower, soybean, and peanut oils and to build models for predicting the content of adulterant oil in canola oil. In this work, 147 adulterated samples were detected by gas chromatography-ion mobility spectrometry (GC-IMS) and chemometric analysis, and two methods of feature extraction, histogram of oriented gradient (HOG) and multiway principal component analysis (MPCA), were combined to pretreat the data set. The results evaluated by canonical discriminant analysis (CDA) algorithm indicated that the HOG-MPCA-CDA model was feasible to discriminate the canola oil adulterated with other oils and to precisely classify different levels of each adulterant oil. Partial least square analysis (PLS) was used to build prediction models for adulterant oil level in canola oil. The model built by PLS was proven to be effective and precise for predicting adulteration with good regression (R2>0.95) and low errors (RMSE ≤ 3.23).

Multiway principal component analysis contributions for structural damage localization

In this article, a novel methodology for damage localization is introduced. The approach is based on a multiactuator system. This means that the system itself has the ability of both exciting the specimen and measuring its response at different points in a pitch-catch mode. Once one of its actuators excites the specimen, the damage affects the normal travel of the guided wave, and this change is mainly detected by sensors in the direct route to the excitation point. In previous works by the authors, it can be observed that the progression using data-driven statistical models (multivariable analysis based on principal component analysis) of all recorded signals to determine whether the damage is present. However, the main contribution of this article is the demonstration of the possibility of localizing damages by analyzing the contribution of each sensor to this index which have detected it ( T2-statistic and Q-statistic). The proposed methodology has been applied and validated on an aircraft turbine blade. The results indicate that the presented methodology is able to accurately locate damages, analyzing the record signals from all actuation phases and giving a unique and reliable region.

Process Monitoring and Fault Diagnosis for Piercing Production of Seamless Tube

With the development of modernization, the application of seamless tube becomes widespread. As the first process of seamless tube, piercing is vital for the quality of the tube. The solid round billet will be transformed into a hollow shell after the piercing process. The defects of hollow shell cannot be cleared in the following process, so a monitoring model for the quality of the hollow shell is important. But the piercing process is very complicated, and a mechanism model is difficult to build between the qualities of the hollow shell and measurement variables. Furthermore, an intelligent model is needed. We established two piercing process monitoring and fault diagnosis models based on the multiway principal component analysis (MPCA) model and the multistage MPCA model, respectively, and furthermore we made a comparison between these two concepts. We took three ways to divide the period based on process,K-means, and GA, respectively. Simulation experiments have shown that the multistate MPCA method has advantage over the MPCA method and the model based on the genetic algorithm (GA) can monitor the process effectively and detect the faults.