A K-means clustering approach is proposed for the automated diagnosis of defective rolling element bearings. Since K-means clustering is an unsupervised learning procedure, the method can be directly implemented to measured vibration data. Thus, the need for training the method with data measured on the specific machine under defective bearing conditions is eliminated. This fact consists the major advantage of the method, especially in industrial environments. Critical to the success of the method is the feature set used, which consists of a set of appropriately selected frequency-domain parameters, extracted both from the raw signal, as well as from the signal envelope, as a result of the engineering expertise, gained from the understanding of the physical behavior of defective rolling element bearings. Other advantages of the method are its ease of programming, simplicity and robustness. In order to overcome the sensitivity of the method to the choice of the initial cluster centers, the initial centers are selected using features extracted from simulated signals, resulting from a well established model for the dynamic behavior of defective rolling element bearings. Then, the method is implemented as a two-stage procedure. At the first step, the method decides whether a bearing fault exists or not. At the second step, the type of the defect (e.g. inner or outer race) is identified. The effectiveness of the method is tested in one literature established laboratory test case and in three different industrial test cases. Each test case includes successive measurements from bearings under different types of defects. In all cases, the method presents a 100% classification success. Contrarily, a K-means clustering approach, which is based on typical statistical time domain based features, presents an unstable classification behavior.

Rolling element bearing fault detection in industrial environments based on a K-means clustering approach

This two-part review article presents an overview of mechanics of pipes conveying fluid and related problems such as the fluid-elastic instability under conditions of turbulence in nuclear power plants. In the first part, different types of modeling, dynamic analysis, and stability regimes of pipes conveying fluid restrained by elastic or inelastic barriers are described. The dynamic and stability behaviors of pinned-pinned, clamped-clamped, and cantilevered pipes conveying fluid together with curved and articulated pipes will be discussed. Other problems such as pipes made of viscoelastic materials and active control of severe pipe vibrations are considered. This part will be closed by conclusions highlighting resolved and nonresolved controversies reported in literature. The second part will address the problem of fluid-elastic instability in single- and two-phase flows and fretting wear in process equipment such as heat exchangers and steam generators. Connors critical velocity will be discussed as a measure of initiating fluid-elastic instability. Vibro-impact of heat exchanger tubes and the random excitation by the cross-flow can produce a progressive damage at the supports through fretting wear or fatigue. Antivibration bar supports used to limit pipe vibrations are described. An assessment of analytical, numerical, and experimental techniques of fretting wear problem of pipes in heat exchangers will be given. Other topics related to this part include remote impact analysis and parameter identification, pipe damage-induced by pressure elastic waves, the dynamic response and stability of long pipes, marine risers together with pipes aspirating fluid, and carbon nanotubes conveying fluid.

Overview of Mechanics of Pipes Conveying Fluids—Part I: Fundamental Studies

Simulation-driven machine learning: Bearing fault classification

This paper is the second part of the two-part review article presenting an overview of mechanics of pipes conveying fluid and related problems such as the fluid-elastic instability under conditions of turbulence in nuclear power plants. In the first part, different types of modeling, dynamic analysis and stability regimes of pipes conveying fluid restrained by elastic or inelastic barriers were described. The dynamic and stability behaviors of pinnedpinned, clamped-clamped, and cantilevered pipes conveying fluid together with curved and articulated pipes were discussed. Other problems such as pipes made of viscoelastic materials and active control of severe pipe vibrations were considered. The first part was closed by conclusions highlighting resolved and nonresolved controversies reported in the literature. The second part will address the problem of fluidelastic instability in single- and twophase flows and fretting wear in process equipment, such as heat exchangers and steam generators. Connors critical velocity will be discussed as a measure of initiating fluidelastic instability. Vibro-impact of heat exchanger tubes and the random excitation by the cross-flow can produce a progressive damage at the supports through fretting wear or fatigue. Antivibration bar supports used to limit pipe vibrations are described. An assessment of analytical, numerical, and experimental techniques of fretting-wear problem of pipes in heat exchangers will be given. Other topics related to this part include remote impact analysis and parameter identification, pipe damage-induced by pressure elastic waves, the dynamic response and stability of long pipes, marine risers together with pipes aspirating fluid, and carbon nanotubes conveying fluid. DOI: 10.1115/1.4001270

Mechanics of Pipes Conveying Fluids—Part II: Applications and Fluidelastic Problems

Two-phase flow induced vibration in piping systems

Fluidelastic instability study in a rotated triangular tube array subject to two-phase cross-flow. Part I: Fluid force measurements and time delay extraction

On the bifurcations of a rigid rotor response in squeeze-film dampers

Detailed flow and force measurements in a rotated triangular tube bundle subjected to two-phase cross-flow

Fluidelastic instability tests on an array of tubes preferentially flexible in the flow direction

This paper investigates the role of tip clearance flow in the occurrence of nonsynchronous vibrations (NSVs) observed in the first axial rotor of a high-speed high-pressure compressor in an aeroengine. NSV is an aeroelastic phenomenon where the rotor blades vibrate at nonintegral multiples of the shaft rotational frequencies in operating regimes where classical flutter is not known to occur. A physical mechanism to explain the NSV phenomenon is proposed based on the blade tip trailing edge impinging jetlike flow, and a novel theory based on the acoustic feedback in the jet potential core. The theory suggests that the critical jet velocity, which brings a jet impinging on a rigid structure to resonance, is reduced to the velocities observed in the blade tip secondary flow when the jet impinges on a flexible structure. The feedback mechanism is then an acoustic wave traveling backward in the jet potential core, and this is experimentally demonstrated. A model is proposed to predict the critical tip speed at which NSV can occur. The model also addresses several unexplained phenomena, or missing links, which are essential to connect tip clearance flow unsteadiness to NSV. These are the pressure level, the pitch-based reduced frequency, and the observed step changes in blade vibration and mode shape. The model is verified using two different rotors that exhibited NSV.

Njuki W. Mureithi

Papers

Fluidelastic instability study in a rotated triangular tube array subject to two-phase cross-flow. Part I: Fluid force measurements and time delay extraction

On the bifurcations of a rigid rotor response in squeeze-film dampers

Detailed flow and force measurements in a rotated triangular tube bundle subjected to two-phase cross-flow

Fluidelastic instability tests on an array of tubes preferentially flexible in the flow direction

Blade Tip Clearance Flow and Compressor Nonsynchronous Vibrations: The Jet Core Feedback Theory as the Coupling Mechanism