Scale-Fractal Detrended Fluctuation Analysis for Fault Diagnosis of a Centrifugal Pump and a Reciprocating Compressor.

Reciprocating compressors and centrifugal pumps are rotating machines used in industry, where fault detection is crucial for avoiding unnecessary and costly downtime. A novel method for fault classification in reciprocating compressors and multi-stage centrifugal pumps is proposed. In the feature extraction stage, raw vibration signals are processed using multi-fractal detrended fluctuation analysis (MFDFA) to extract features indicative of different types of faults. Such MFDFA features enable the training of machine learning models for classifying faults. Several classical machine learning models and a deep learning model corresponding to the convolutional neural network (CNN) are compared with respect to their classification accuracy. The cross-validation results show that all models are highly accurate for classifying the 13 types of faults in the centrifugal pump, the 17 valve faults, and the 13 multi-faults in the reciprocating compressor. The random forest subspace discriminant (RFSD) and the CNN model achieved the best results using MFDFA features calculated with quadratic approximations. The proposed method is a promising approach for fault classification in reciprocating compressors and multi-stage centrifugal pumps.

From fault detection to one-class severity discrimination of 3D printers with one-class support vector machine.

The lack of faulty condition data reduces the feasibility of supervised learning for fault detection or fault severity discrimination in new manufacturing technologies. To deal with this issue, one-class learning arises for building binary discriminative models using only healthy condition data. However, these models have not been extrapolated to severity discrimination. This paper proposes to extend OCSVM, which is typically used for fault detection, to 3D printer fault severity discrimination. First, a set of features is extracted from a set of normal signals. An optimized OCSVM model is obtained by tuning the kernel and model hyperparameters. The resulting models are evaluated for fault detection and fault severity discrimination using a proposed performance evaluation approach. Experimental comparisons for belt-based faults in 3D printers show that the distance to the hyperplane has the information to discriminate the severity level, and its use is feasible. The proposed hyperparameter optimization technique improves the OCSVM for fault detection and severity discrimination compared to some other methods.

Measurements and automated mechanism generation modeling of OH production in photolytically initiated oxidation of the neopentyl radical.

Production of OH in the reaction of the neopentyl radical with O2 has been measured by a laser photolysis/cw absorption method for various pressures and oxygen concentrations at 673, 700, and 725 K. The MIT Reaction Mechanism Generator (RMG) was used to automatically generate a model for this system, and the predicted OH concentration profiles are compared to present and literature experimental results. Several reactions significantly affect the OH profile. The experimental data provide useful constraints on the rate coefficient for the formally direct chemical activation reaction of neopentyl radical with O2 to form OH (CH3)3CCH2 + O2 --> OH + 3,3-dimethyloxetane (Rxn 1) At 673 K and 60 Torr, log k(1) (cm(3) molecule(-1) s(-1)) = -13.7 +/- 0.5. Absolute absorbance measurements on OH and I indicate that the branching ratio for R + O2 to OH is about 0.03 under these conditions. The data suggest that the ab initio neopentyl + O2 potential energy surface of Sun and Bozzelli is accurate to within 2 kcal mol(-1).

Measurements and modeling of DO2 formation in the reactions of C2D5 and C3D7 radicals with O2.

Time-resolved production of HO2 and DO2 from the reactions of nondeuterated and deuterated ethyl and propyl radicals with O2 are measured as a function of temperature and pressure in the "transition region" between 623 and 748 K using the technique of laser photolysis/long path frequency modulation spectroscopy. Experimental measurements, using both pulsed-photolytic Cl-atom-initiated oxidation of ethane and propane and direct photolysis of ethyl, n-propyl, and isopropyl iodides, are compared to kinetic models based on the results of time-dependent master equation calculations with ab initio characterization of stationary points. The formation of DO2 and HO2 from the subsequent reaction of the alkyl radicals with O2 is followed by infrared frequency modulation spectroscopy. The concentration of I atoms is simultaneously monitored by direct absorption of a second laser probe on the spin-orbit transition. The kinetic models accurately describe the time scale and amplitude of the DO2 and HO2 formation resulting from C2D5 + O2, n-C3D7 + O2, i-C3D7 + O2, and i-C3H7 + O2. Overall, a very good level of agreement is found between theory and experiments over a wide range of temperatures, pressures, and O2 concentrations. Good agreement is also found between previous literature studies and the theory presented in this work except in the case of the high-temperature rate coefficients for the reaction of i-C3H7 + O2 to form propene. A reinvestigation of the high-temperature kinetics of the i-C3H7 + O2 reaction appears warranted. The results from the present work suggest that the theory for formation of HO2 from the reactions of ethyl and both isomeric forms of propyl radicals with O2 are very well established at this time. It is hoped that these reactions can now form the groundwork for the study and interpretation of larger and more complex R + O2 systems.

Measurements and modeling of HO2 formation in the reactions of n-C3H7 and i-C3H7 radicals with O2.

The formation of HO(2) in the reactions of C(2)H(5), n-C(3)H(7), and i-C(3)H(7) radicals with O(2) is investigated using the technique of laser photolysis/long-path frequency-modulation spectroscopy. The alkyl radicals are formed by 266 nm photolysis of alkyl iodides. The formation of HO(2) from the subsequent reaction of the alkyl radicals with O(2) is followed by infrared frequency-modulation spectroscopy. The concentration of I atoms is simultaneously monitored by direct absorption of a second laser probe on the spin-orbit transition. The measured profiles are compared to a kinetic model taken from time-resolved master-equation results based on previously published ab initio characterizations of the relevant stationary points on the potential-energy surface. The ab initio energies are adjusted to produce agreement with the present experimental data and with available literature studies. The isomer specificity of the present results enables refinement of the model for i-C(3)H(7) + O(2) and improved agreement with experimental measurements of HO(2) production in propane oxidation.