23875288-Experimental-Fluid-Dynamics

Views:
 
Category: Education
     
 

Presentation Description

No description available.

Comments

Presentation Transcript

Experimental Fluid Dynamics : 

12/16/2010 1 Experimental Fluid Dynamics M. Muste, S. Ghosh, F. Stern

Table of Contents : 

Table of Contents Definitions EFD Philosophy Types of Experiments EFD Phases Measurement Systems EFD Validation Reporting EFD

Definitions : 

Definitions Experimentation: operation carried out under controlled conditions with a measurement system for determining or demonstrating a principle or effect, for testing a hypothesis, or for acquiring data for specific purposes (design evaluation, validation, calibration) Experimental Fluid Dynamics: scientific method based on dimensional analysis, similarity, and experimentation used for defining behavior of systems and/or processes in Fluid Dynamics (FD) that cannot be satisfactory/completely formulated by Analytical (AFD) or Computational (CFD) approaches Measurement Systems: facilities, instrumentation, operational environment, data-acquisition and data-reduction procedures used in an experiment A pretty experiment is in itself often more valuable than twenty formulae extracted from our minds." - Albert Einstein

EFD Philosophy : 

EFD Philosophy EFD general approach: Establish expected outcome and allowable uncertainties Define needs (identify variable, establish scale) Understand the nature of the FD problem (dimensional analysis) Evaluate physical situation, estimate results, and anticipate instruments Understand flow measurement principles Current trends in EFD methodology: Synergy between AFD, CFD, and EFD (shift from routine tests for global variables to detailed tests for local variables) Implementation of EFD international standard procedures Integration of UA considerations in all phases of EFD

EFD Philosophy : 

EFD Philosophy UA involved in multiple stages of the experiment

EFD Philosophy : 

Use UA in all experimental phases to ensure maximum efficiency (time, effort, and financial resources) EFD Philosophy

Types of Experiments : 

Types of Experiments Grouped by field/purpose: Science & Technology: understand and investigate a phenomenon/process, substantiate and validate a theory (hypothesis) Research & Development: document a process/system, provide benchmark data (standard procedures, validations), calibrate instruments, equipment, and facilities Industry: design optimization and analysis, provide data for direct use, product liability, and acceptance Teaching: Instruction/demonstration

Types of Experiments : 

Types of Experiments Grouped by methodology: Timewise (data collected over a period of time) Sample-to-sample (data analyzed over several realizations/samples) Grouped from UA perspective: Repeated Replicated (repetition carried over in a very specific manner) Replication levels: zeroth order (one experiment, multiple measurements, same instrumentation) first order (multiple experiments, multiple measurements, same instrumentation) N-th order (multiple experiments, multiple measurements, multiple instrumentation of the same type)

EFD Phases : 

EFD Phases Planning: formulate objectives and allowable uncertainties, define needs, identify pertinent process variables (targeted, independent, controlled, extraneous, parameters), evaluate model scale and various MS and experimental approaches Design: understand the nature of the FD problem (dimensional analysis), select MS and UA methodologies, evaluate physical situation (understand the MS interaction with the process under investigation), establish data-acquisition pattern, estimate results, budget, and timeline Construction: assembly of individual components and calibration of instruments (use available standard procedures) Debugging: trial runs Execution (setup experiments, acquire & process data, UA, analysis) Reporting

EFD Process : 

EFD Process

Measurement Systems : 

Measurement Systems MS components: facilities, instrumentation, operational environment, data acquisition, data reduction

Measurement Systems : 

Measurement Systems MS behavior: Zero-order systems: (no inertia or damping) First-order systems: (inertia) Second-order systems: (inertia and damping)

Measurement Systems : 

Measurement Systems Facilities Scales: small-, model-, and full-scale (in-situ experiments) Selection of the model scale: governed by similitude analysis Special considerations: similitude distortion (geometric, kinematic, dynamic) scale-induced effects (test of family of models) facility-induced effects (wall interference, model-induced perturbation, replication of boundary conditions)

Measurement Systems : 

Measurement Systems Instrumentation Components: sensors, transducers, signal conditioning, display Calibration: trace of the instrument accuracy to a primary or secondary standards (end-to-end procedure) Selection: function of the nature of the measured physical quantity (mono, multi-phase, scalar, vector, static, fluctuating, local, field) function of temporal and spatial scales of the flow satisfaction of the UA requirements minimization of flow-sensor and flow-facility interferences

Measurement Systems : 

Measurement Systems Environmental conditions Control MS parameters and variables Document MS parameters, operating conditions, and test observations in chronological order Quantify characteristics for MS noise, interference, drift (e.g., temperature increase during the experiments) Use checks to guard against unnoticed, unwanted, and hazardous changes in the instrumentation and operating conditions Armfield table-top experiment (armfield.co.uk) Lawrence Berkeley National Laboratory

Measurement Systems : 

Measurement Systems Data Acquisition (digital) General scheme (one channel): Current trends: multi-channel, microprocessor-controlled Special considerations: Correlate sampling type, sampling frequency (Nyquist criterion), and sampling time with the dynamic content of the signal Correlate the resolution for the A/D converters (bias error = ± ½ LSB) with the magnitude of the signal Document through calibration or manufacturer’s specs: system-sensor, quantization, saturation, and conversion (hysteresis, gain, linearity, zero) errors

Measurement Systems : 

Data acquisition optimization: Select sufficient number of variables to be measured (at least two concomitant methods for estimating the result) Minimize and control experimental parameters Ensure that the measured variable is the only dependent process variable Ensure that instrument sensor responds only to the variable to be measured Minimize probe-to-output path of the signal Randomize the effect of extraneous variables Measurement Systems

Measurement Systems : 

Measurement Systems Procedures: Adoption of international standard EFD procedures (e.g., wind-tunnel AIAA standard S-071-1995) Use concomitant methods for measurement of the main variables Adopt a comprehensive scheme for controlling the MS operation and environmental conditions Establish chronological sequence for MS operation (data acquisition, reduction, storage) Establish appropriate number of test replications and density for measured data points to fulfill EFD and UA requirements Adopt a random sequence for the data acquisition Set appropriate test scheme to counteract MS noise (e.g., time between replications longer than the period of variation in the extraneous variables), interference, drift in facility and instrument operation

Measurement Systems : 

Measurement Systems Data Reduction Components: data-reduction equations, curve fitting, interpolations, data visualization Special considerations: Eliminate conceptual bias errors: verify data-reduction equations and algorithms for parameter estimation Use spectral analysis (discrete Fourier transform) to reconstruct the amplitude and frequency of the measured signals; check for appropriateness of the selection of the instrument characteristics Check for stability of the statistics (first and second order) and establish the appropriate sample size for reporting the data Use least-square method for curvefitting and determine regression uncertainty

Measurement Systems : 

Measurement Systems Real-Time Data Acquisition Systems Labview is a programming software used for data acquisition, instrument control, data reduction and visualization Graphical programming language that uses icons instead of text Allows to build user interfaces with a set of tools and objects The user interface is called the front panel and a block diagram controls the front panel The program is written on the block diagram and the front panel is used to control and run the program.

Measurement Systems : 

Measurement Systems LabVIEW - objects

Measurement Systems : 

Front panel Block diagram Measurement Systems LabVIEWassembled objects

EFD Validation : 

EFD Validation Uncertainty Analysis (UA)= Data Quality= Confidence in the reported results Standardized UA: rigorous methodology for uncertainty assessment using statistical and engineering concepts

EFD Validation : 

EFD Validation EFD result: A ±UA Benchmark or EFD data: B ± UB Define E = B-A UE2 = UA2+UB2 Validation: |E| < UE Conduct uncertainty analysis for the results: Identify and estimate errors considering all the steps of the measurement process and the environmental factors Use 95% confidence large-sample (multiple tests) uncertainty methodology

Reporting Results : 

Reporting Results Report types: oral, written, exec summary, lab report, formal report, journal articles, tutorial reports Report format: tabular, graphic, text, mathematical expression Report general outline: Abstract Introduction (purpose, background, theoretical considerations) Experimental program (test design, measurement systems – facilities, instrumentation, operational environment, data acquisition, data reduction- experimental procedure) Analysis (results, uncertainty analysis, comparison, interpretation) Discussion (conclusions, recommendations) Acknowledgements References Appendix materials

Reporting Results : 

Hints for graphs: Use non-dimensional coordinates for reporting results (for generalization purposes) Independent variables on x axis, dependent variables on y axis Select proper type for coordinates (linear, semilog, log-log) Use symbols for the EFD data; use lines for benchmark data (CFD, reference, analytical, theoretical) Plot results, total uncertainties and confidence level Graphs should be stand-alone presentation elements: Label plots, axes, and provide units for the plotted variables Insert legends for graphs (direct on graph, indirect & explanations) Use visualization software packages for display (including animation) of multi-variable dependencies Reporting Results

authorStream Live Help