Publications

This paper investigates the journal asymmetric temperature-induced thermal bow vibration of a rotor, as supported by a flexure pivot journal bearing (FPJB). Thermal bow-induced vibration, known as the Morton effect (ME), is caused by non-uniform viscous heating of the journal, and the resulting thermal bow often causes increasing vibration amplitudes with the time-varying phase. Full FPJB’s structural and thermal finite element models are developed and integrated into the flexible rotor model. The model is validated by comparing its predicted ME response with experimental results. An FPJB model, which uses predicted “equivalent” radial and tilting stiffness of the bearing, is compared with the full finite element method (FEM)-based model. The impact of FPJB’s design parameters such as web thickness, bearing material, and housing thicknesses are investigated with parametric studies. The results show that FPJB parameter values may have a major effect on the speed range of ME vibration, and its severity.

 

Parametric Study of Flexure Pivot Bearing Induced Thermal Bow-Rotor Instability (Morton effect)

Hirth coupling transmits high torques in the rotating assemblies of compressors and turbines. Their mating surface contacts cause local changes in lateral shaft stiffness. This is affected by the teeth geometry, contact surface area, coupling preload, and surface finish at the contact faces. Industry practice ignores localized lateral flexibility from the Hirth coupling, or is guided by limited experience-based rules of thumb. The authors provide a novel modeling approach utilizing 3D solid finite elements which accounts for contact deformations, intricate interface teeth geometries, stress concentration, and surface finish. This provides an increased accuracy localized stiffness model for the Hirth coupling, to improve rotordynamic response predictions. Free–free natural frequencies of a test rotor including a Hirth coupling are experimentally measured. The rotor is instrumented with strain gauges for preload force measurements, and the Hirth coupling contacting surface profiles are measured with a stylus type surface profiler. A GW contact model is obtained from the measured surface profiles. An iterative computation algorithm is utilized to calculate Hirth coupling contact stiffness and contact pressure at the complex-shaped contact surfaces. Predicted and measured natural frequencies are compared versus preload.

Thanks to the unique advantages such as long life cycles, high power density, minimal environmental impact, and high power quality such as fast response and voltage stability, the flywheel/kinetic energy storage system (FESS) is gaining attention recently. There is noticeable progress in FESS, especially in utility, large-scale deployment for the electrical grid, and renewable energy applications. This paper gives a review of the recent developments in FESS technologies. Due to the highly interdisciplinary nature of FESSs, we survey different design approaches, choices of subsystems, and the effects on performance, cost, and applications. This review focuses on the state of the art of FESS technologies, especially those commissioned or prototyped. We also highlighted the opportunities and potential directions for the future development of FESS technologies.

 

A review of flywheel energy storage systems state of the art and opportunities

Reynolds based thermo-elasto-hydrodynamic (TEHD) simulations of tilting pad journal bearings (TPJBs) generally provide accurate results; however, the uncertainty of the pad’s leading edge thermal boundary conditions causes uncertainty of the results. The highly complex thermal-flow mixing action between pads (BPs) results from the oil supply nozzle jets and geometric features. The conventional Reynolds approach employs mixing coefficients (MCs), estimated from experience, to approximate a uniform inlet temperature for each pad. Part I utilized complex computational fluid dynamics (CFD) flow modeling to illustrate that temperature distributions at the pad inlets may deviate strongly from being uniform. The present work retains the uniform MC model but obtains the MC from detailed three-dimensional CFD modeling and machine learning, which could be extended to the radially and axially varying MC case. The steps for implementing an artificial neural network (ANN) approach for MC regression are provided as follows: (1) utilize a design of experiment step for obtaining an adaptable training set, (2) conduct CFD simulations on the BP to obtain the outputs of the training set, (3) apply an ANN learning process by Levenverg–Mardquart backpropagation with the Bayesian regularization, and (4) couple the ANN MC results with conventional TEHD Reynolds models. An approximate log fitting method provides a simplified approach for MC regression. The effectiveness of the Reynolds TEHD TPJB model with ANN regression-based MC distributions is confirmed by comparison with CFD based TEHD TPJB model results. The method obtains an accuracy nearly the same as the complete CFD model, but with the computational economy of a Reynolds approach.

 

CFD Based Mixing Prediction for Tilt Pad Journal Bearing TEHD Modeling- Part II Implementation with Machine Learning

The core contributions of Part I (1) present a computational fluid dynamics (CFD)-based approach for tilting pad journal bearing (TPJB) modeling including thermo-elasto hydrodynamic (TEHD) effects with multi-mode pad flexibility, (2) validate the model by comparison with experimental work, and (3) investigate the limitations of the conventional approach by contrasting it with the new approach. The modeling technique is advanced from the author’s previous work by including pad flexibility. The results demonstrate that the conventional approach of disregarding the three-dimensional flow physics between pads (BP) can generate significantly different pressure, temperature, heat flux, dynamic viscosity, and film thickness distributions, relative to the high-fidelity CFD model. The uncertainty of the assumed mixing coefficient (MC) may be a serious weakness when using a conventional, TPJB Reynolds model, leading to prediction errors in static and dynamic performance. The advanced mixing prediction method for “BP” thermal flow developed in Part I will be implemented with machine learning techniques in Part II to provide a means to enhance the accuracy of conventional Reynolds based TPJB models.

 

CFD Based Mixing Prediction for Tilt Pad Journal Bearing TEHD Modeling- Part I TEHD-CFD Model Validation and Improvements

This paper investigates the influence of misaligned journal bearing effects on the thermally induced rotor instability (Morton effect “ME”) problem. The Morton effect is caused by uneven viscous heating of the journal in a fluid film bearing, which causes thermal bending, especially in rotors with an overhung disc or coupling weight. The thermally induced bending in the shaft may cause a vibration instability, which results in an excessive level of synchronous vibration. Previous research focused on parametric studies of the rotor and bearing design parameters, including overhung mass, bearing radial clearance, and lubricant viscosity. The present study investigates the influence of journal misalignment on the Morton effect. A coupled fluid-thermal-structural, three-dimensional finite element model (FEM) is developed to simulate fluid film pressures and temperatures, and shaft temperatures and vibrations. Simulations were conducted with different ratios of journal misalignment, and different pad-pivot types to determine their effect on the phenomenon. The simulation results indicate that the amplitude of the misalignment angle affects the instability speed range (ISR) caused by the Morton effect under certain conditions.

 

Tilting pad journal bearing misalignment effect on thermally induced synchronouse instability (Morton effect)

Traditional analysis of journal bearings assumed a constant viscosity which simplified the solutions for static and dynamic characteristics and responses. Today’s high-performance machinery requires more accurate models wherein temperature and viscosity distributions in the film must be calculated. Thermal effects in journal bearings have a strong influence on both static and dynamic properties, and consequently play a critical role in determining rotor-bearing system performance. This paper presents an extensive survey of the thermal modeling methods and effects in journal bearings. The subjects include various bearing types, and recent progress in thermal bearing design and thermal instability problems observed in fluid and gas film hydrodynamic bearings. The extent of the survey ranges from conventional Reynolds equation models to more advanced computational fluid dynamics models.

 

A Review of Journal Bearing Thermal Effects on Rotordynamic Response

Conventional active magnetic bearing (AMB) systems use several separate radial and thrust bearings to provide a five-degree of freedom (DOF) levitation control. This article presents a novel combination 5-DOF AMB (C5AMB) designed for a shaft-less, hub-less, high-strength steel energy storage flywheel (SHFES), which achieves doubled energy density compared to prior technologies. As a single device, the C5AMB provides radial, axial, and tilting levitations simultaneously. In addition, it utilizes low-cost and more available materials to replace silicon steels and laminations, which results in reduced costs and more convenient assemblies. Apart from the unique structure and the use of low magnetic grade material, other design challenges include shared flux paths, large dimensions, and relatively small air gaps. The finite-element method (FEM) is too computationally intensive for early-stage analysis. An equivalent magnetic circuit method (EMCM) is developed for modeling and analysis. Nonlinear FEM is then used for detailed simulations. Both permanent magnets (PMs) and electromagnetic control currents provide the weight-balancing lifting force. During the full-scale prototype testing, the C5AMB successfully levitates a 5440-kg and 2-m-diameter flywheel at an air gap of 1.14 mm. Its current and position stiffnesses are verified experimentally.

 

A Combination 5-dof Active Magnetic Bearing for Energy Storage Flywheels

Conventional rotordynamic analyses generally treat the rotor as a continuous body without considering effect of clamped joints. However, in modern rotating machines, rotors are often assembled with multiple complex-shaped parts and joints, which may significantly affect rotordynamic behavior. Several authors have proposed methods for implementing contact effects in rotordynamic analysis, but a more general modeling method for handling arbitrary contact geometries with various levels of surface roughness is needed. The present paper suggests a new contact model for rotordynamic analysis of an assembled rotor-bearing system with multiple parts connected by multiple joints. A contact element formulation is presented using solid finite elements and statistics-based contact theories. A test arrangement was developed to validate the proposed contact model for varying interface surface roughness and preloads. An iterative computation algorithm is introduced to solve the implicit relation between contact stiffness and stress distribution. Prediction results, using the contact model, are compared with measured natural frequencies for multiple configurations of a test rotor assembly. A case study is performed for an overhung type rotor-bearing system to investigate the effect of contact interfaces, between an overhung impeller and a rotor shaft, on critical speeds.

Tilting Pad Journal Bearing (TPJB) are ubiquitous in turbomachinery from small high speed compressors to large turbine generator sets. The TPJB’s power losses increase with machine size and speed, reaching as high as 1–2 MW loss (equivalent power for about 1000 homes) per bearing, in a large turbine generator. This study presents a novel power loss reduction technique, inserting a pocket and step in the pads, to activate cavitation in the pocket while maintaining pad stability. The benefits are demonstrated via Thermo-Elasto-Hydrodynamic (TEHD) CFD simulations comparing results with and without these modifications. The novel approach shows power loss reductions up to −27% without a load capacity loss, while reducing required supply oil flow, and peak pad temperature.

 

Power Loss Reduction for Tilt Pad Journal Bearings Utilizing Pad Pockets and Steps