Applications of machine learning to machine fault diagnosis: A review and roadmap

Intelligent fault diagnosis techniques have replaced time-consuming and unreliable human analysis, increasing the efficiency of fault diagnosis. Deep learning models can improve the accuracy of intelligent fault diagnosis with the help of their multilayer nonlinear mapping ability. This paper proposes a novel method named Deep Convolutional Neural Networks with Wide First-layer Kernels (WDCNN). The proposed method uses raw vibration signals as input (data augmentation is used to generate more inputs), and uses the wide kernels in the first convolutional layer for extracting features and suppressing high frequency noise. Small convolutional kernels in the preceding layers are used for multilayer nonlinear mapping. AdaBN is implemented to improve the domain adaptation ability of the model. The proposed model addresses the problem that currently, the accuracy of CNN applied to fault diagnosis is not very high. WDCNN can not only achieve 100% classification accuracy on normal signals, but also outperform the state-of-the-art DNN model which is based on frequency features under different working load and noisy environment conditions.

A New Deep Learning Model for Fault Diagnosis with Good Anti-Noise and Domain Adaptation Ability on Raw Vibration Signals

Arthur Chickering is Distinguished Professor of Higher Education at Memphis State University. On leave from the Directorship of the Center for the Study of Higher Education at Memphis State, he is Visiting Professor at George Mason University. Zelda Gamson is a sociologist who holds appointments at the John W. McCormack Institute of Public Affairs at the University of Massachusetts-Boston and in the Center for the Study of Higher and Postsecondary Education at the University of Michigan.

Seven Principles for Good Practice in Undergraduate Education: Faculty Inventory. Institutional Inventory.

The vibration signals measured from a rolling bearing are usually affected by the variable operating conditions and background noise which lead to the diversity and complexity of the vibration signal characteristics, and it is a challenge to effectively identify the rolling bearing faults from such vibration signals with no further fault information. In this paper, a novel optimization deep belief network (DBN) is proposed for rolling bearing fault diagnosis. Stochastic gradient descent is used to efficiently fine-tune all the connection weights after the pre-training of restricted Boltzmann machines (RBMs) based on the energy functions, and the classification accuracy of the DBN is improved. Particle swarm is further used to decide the optimal structure of the trained DBN, and the optimization DBN is designed. The proposed method is applied to analyze the simulation signal and experimental signal of a rolling bearing. The results confirm that the proposed method is more accurate and robust than other intelligent methods.

https://iopscience.iop.org/article/10.1088/0957-0233/26/11/115002/pdf

Rolling bearing fault diagnosis using an optimization deep belief network

Artificial Neural Networks (ANNs)

A review of drillstring vibration modeling and suppression methods

In this paper, a system based on artificial neural networks (ANNs) was designed to diagnose different types of fault in a gearbox. An experimental set of data was used to verify the effectiveness and accuracy of the proposed method. The system was optimized by eliminating unimportant features using a feature selection method (UTA method). Consequently, the fault detection system operates faster while the classification error decreases or remains constant in some other cases. This method of feature selection is compared with Genetic Algorithm (GA) results. The findings verify that the results of the UTA method are as accurate as GA, despite its simple algorithm.

http://bme2.aut.ac.ir/mhmoradi/EN.Journal%20Articles/Application%20and%20comparison%20of%20an%20ANN-based%20feature%20selection%20method%20and.pdf

Application and comparison of an ANN-based feature selection method and the genetic algorithm in gearbox fault diagnosis

Surgical Skill Assessment Using Motion Quality and Smoothness

Vibration Analysis of a Drillstring in Vibration-Assisted Rotary Drilling: Finite Element Modeling With Analytical Validation

The ability to exert an appropriate amount of force on brain tissue during surgery is an important component of instrument handling. It allows surgeons to achieve the surgical objective effectively while maintaining a safe level of force in tool–tissue interaction. At the present time, this knowledge, and hence skill, is acquired through experience and is qualitatively conveyed from an expert surgeon to trainees. These forces can be assessed quantitatively by retrofitting surgical tools with sensors, thus providing a mechanism for improved performance and safety of surgery, and enhanced surgical training. This paper presents the development of a force-sensing bipolar forceps, with installation of a sensory system, that is able to measure and record interaction forces between the forceps tips and brain tissue in real time. This research is an extension of a previous research where a bipolar forceps was instrumented to measure dissection and coagulation forces applied in a single direction. Here, a planar forceps with two sets of strain gauges in two orthogonal directions was developed to enable measuring the forces with a higher accuracy. Implementation of two strain gauges allowed compensation of strain values due to deformations of the forceps in other directions (axial stiffening) and provided more accurate forces during microsurgery. An experienced neurosurgeon performed five neurosurgical tasks using the axial setup and repeated the same tasks using the planar device. The experiments were performed on cadaveric brains. Both setups were shown to be capable of measuring real-time interaction forces. Comparing the two setups, under the same experimental condition, indicated that the peak and mean forces quantified by planar forceps were at least 7% and 10% less than those of axial tool, respectively; therefore, utilizing readings of all strain gauges in planar forceps provides more accurate values of both peak and mean forces than axial forceps. Cross-correlation analysis between the two force signals obtained, one from each cadaveric practice, showed a high similarity between the two force signals.

Ahmad Ghasemloonia

Papers

A review of drillstring vibration modeling and suppression methods

Application and comparison of an ANN-based feature selection method and the genetic algorithm in gearbox fault diagnosis

Surgical Skill Assessment Using Motion Quality and Smoothness

Vibration Analysis of a Drillstring in Vibration-Assisted Rotary Drilling: Finite Element Modeling With Analytical Validation

A Force-Sensing Bipolar Forceps to Quantify Tool–Tissue Interaction Forces in Microsurgery