8/17/2019 h5. PBSD_Presentation.pdf
1/49
Performance Performance - - Based Seismic Design Based Seismic Design
8/17/2019 h5. PBSD_Presentation.pdf
2/49
2
Outline
• General concepts of PBD
• FEMA-350 in building design: Step-by Step
1. Performance definition2. Evaluation approach
3. Analysis4. Mathematical modeling 5. Acceptance criteria
8/17/2019 h5. PBSD_Presentation.pdf
3/49
3
Basic Design Approach
• Prescriptive design• Prescribed
strength• Prescriptive
details.
8/17/2019 h5. PBSD_Presentation.pdf
4/49
4
Basic Design Approach
• Structural system type and frame
configuration• Frame member sizes
• Design force levels• Perform a structural analysis
• Revise specifications• Complete the design (detailing requirements)
8/17/2019 h5. PBSD_Presentation.pdf
5/49
5
Expected Performance
8/17/2019 h5. PBSD_Presentation.pdf
6/49
6
Why Performance-Based Design?
• Evaluate the probable seismic performance of
steel moment-frame buildings.
• Design steel moment-frame buildings foralternative performance capabilities and also toquantify the ability of a specific design to achieve
desired performance objectives.
8/17/2019 h5. PBSD_Presentation.pdf
7/49
7
PB methodology in simple words…
Design for the achievement of specifiedresults rather than adherence to
prescribed means.
8/17/2019 h5. PBSD_Presentation.pdf
8/49
8
PB Methodology
Earthquake professionals quantify seismic risk
in terms that are meaningful to the decisionmakers.
Decision-makers make informed decisions thatdefine a rational course of action for theearthquake professionals.
8/17/2019 h5. PBSD_Presentation.pdf
9/49
9
Applying FEMA-350 in Design
A Step by Step Description
8/17/2019 h5. PBSD_Presentation.pdf
10/49
10
PBSD Steps:
1. Performance definition2. Evaluation approach3. Analysis4. Mathematical modeling 5. Acceptance criteria
8/17/2019 h5. PBSD_Presentation.pdf
11/49
11
I. Performance Definition
A Performance Objective has two
components:
• Hazard Level• Performance Level (limiting damage state)
Association of a performance level (damagestate) to a hazard level is called a performance- objective .
8/17/2019 h5. PBSD_Presentation.pdf
12/49
12
Hazard Level
• Characterized by a hazard curve, whichindicates the probability that a given value ofa ground motion parameter, for examplepeak ground acceleration, will be exceededover a certain period of time.
• For example: – 2% POE in 50 years or 2500 years return period – 50% POE in 50 years or 100 years return period
8/17/2019 h5. PBSD_Presentation.pdf
13/49
13
Performance Level
Performance levels are discrete damage states
selected among all possible damage statesthat a building could experience as a result ofearthquake response.
• For example: – Collapse Prevention (CP) – Immediate Occupancy (IO)
8/17/2019 h5. PBSD_Presentation.pdf
14/49
14
Performance Level
8/17/2019 h5. PBSD_Presentation.pdf
15/49
8/17/2019 h5. PBSD_Presentation.pdf
16/49
16
Uncertainties
• Behavior and response of the structure• Accuracy of analysis procedures
• Ground motions
Therefore the performance objective isexpressed in terms of confidence levels.
8/17/2019 h5. PBSD_Presentation.pdf
17/49
17
Performance Objective
Therefore a performance objective for a design
should more correctly be stated as:
• A certain level of confidence (i.e. 95%)that thestructure will provide Collapse Prevention orbetter performance for earthquake hazards
with 2% probability of exceedance in 50 years
8/17/2019 h5. PBSD_Presentation.pdf
18/49
18
PBSD Steps:
Performance definition2. Evaluation approach3. Analysis4. Mathematical modeling 5. Acceptance criteria
8/17/2019 h5. PBSD_Presentation.pdf
19/49
19
II. Evaluation
• The structural analysis is used to predict the valueof various structural response parameters: – Interstory drift: – Axial force
• Predicted demands are adjusted by two factors: – Analytical uncertainty factor, γa – Demand variability factor, γ
• Structural capacities are adjusted by resistantfactors, φ• Level of confidence is calculated based on the ratio
of factored demand to factored capacity.
8/17/2019 h5. PBSD_Presentation.pdf
20/49
8/17/2019 h5. PBSD_Presentation.pdf
21/49
21
III. Analysis Method
1. Linear static procedure
2. Linear dynamic procedure
3. Nonlinear static procedure
4. Nonlinear dynamic procedure
8/17/2019 h5. PBSD_Presentation.pdf
22/49
22
III. Analysis Method
8/17/2019 h5. PBSD_Presentation.pdf
23/49
23
Linear Static Procedure
• It is an equivalent lateral force technique
• Has inherently more uncertainty because it accountsless accurately the dynamic characteristics of thestructures
• Applied lateral forces follow a pattern thatresembles the distribution of inertial forces in aregular structure responding linearly to the ground
excitations• It is assumed that the structure’s response isdominated by the fundamental mode
8/17/2019 h5. PBSD_Presentation.pdf
24/49
24
LSP Steps
• Calculate the fundamental period for each of
the two orthogonal directions, from: – Eigenvalue analysis of the mathematical model
– Apoximate period formula (4-1):
• Calculate Pseudo Lateral Load for each ofthe two orthogonal directions, from:
( )2 0n n K M ω ϕ − =0.8
t nT C h=
1 2 3 aV C C C S W =
8/17/2019 h5. PBSD_Presentation.pdf
25/49
25
C 1
A modification factor to relate expected maximum inelasticdisplacements to displacements calculated for linear elasticresponse:
8/17/2019 h5. PBSD_Presentation.pdf
26/49
26
C 1
S a ,g
T, sec
1 1.0C =1 1.5C =
T s0.2T s
8/17/2019 h5. PBSD_Presentation.pdf
27/49
27
C 2
• A modification factor to represent the effect
of hysteretic pinching on the maximumdisplacement response, for steel momentframes:
2 1.0C =
8/17/2019 h5. PBSD_Presentation.pdf
28/49
28
C 3
• A modification factor to represent increased
dynamic displacements due to P-delta effects andstiffness degradation:
8/17/2019 h5. PBSD_Presentation.pdf
29/49
29
Base Shear- 1 2 3 aV C C C S W =
• S a : Response spectrum acceleration at the
fundamental period and damping ratio of thebuilding in the direction under consideration, for thehazard level corresponding to the performanceobjective being evaluated (I.e., probability ofexceedance.)
• W: Total dead load and anticipated live load
8/17/2019 h5. PBSD_Presentation.pdf
30/49
30
Nonlinear Dynamic Procedure
Response calculations are carried out usingnonlinear response-history analysis
An estimate of median response parameterof interest will be obtained by performingthe analyses over a set of minimum sevenpairs of strong ground motion records
The mathematical model of the buildingstructures should incorporate the inelasticmaterial and nonlinear geometric behavior
8/17/2019 h5. PBSD_Presentation.pdf
31/49
8/17/2019 h5. PBSD_Presentation.pdf
32/49
32
IV. Mathematical Modeling
1. Frame configuration2. Connection modeling
• Fully-restrained (FR)• Partially-restrained (PR)• Simple shear tab
• Panel zone stiffness3. Horizontal torsion4. Foundation5. Diaphragms6. P-Delta effects7. Multidimensional excitation effects8. Vertical ground motion
8/17/2019 h5. PBSD_Presentation.pdf
33/49
33
PBSD Steps:
Performance definitionEvaluation approach
AnalysisMathematical modeling
5. Acceptance criteria
8/17/2019 h5. PBSD_Presentation.pdf
34/49
34
V. Acceptance Criteria
Acceptability of building performance should
be evaluated by determining a level ofconfidence in the building’s ability to meetthe desired performance objective(s).
1. Factored-Demand-to-Capacity Ratio• For interstory drift• For column axial load• For column splice tension
2. Confidence-level computations
8/17/2019 h5. PBSD_Presentation.pdf
35/49
35
V. Acceptance Criteria
8/17/2019 h5. PBSD_Presentation.pdf
36/49
8/17/2019 h5. PBSD_Presentation.pdf
37/49
37
Analysis Uncertainty Factor
8/17/2019 h5. PBSD_Presentation.pdf
38/49
38
Demand Variability Factor
8/17/2019 h5. PBSD_Presentation.pdf
39/49
39
Structural Capacity and Capacity Factor
8/17/2019 h5. PBSD_Presentation.pdf
40/49
40
Uncertainty Coefficient
8/17/2019 h5. PBSD_Presentation.pdf
41/49
41
Confidence Levels
8/17/2019 h5. PBSD_Presentation.pdf
42/49
42
Classroom activity…
• Problem 1: Find how much confidence a designer h
on his/her design against collapse for a 10-storySMF (with post-Northridge partially retrainedconnection) that has a median interstory drift demanof 5% under a set of seven earthquake input motionrecords with 2% probability of exceedance in 50
years? Nonlinear dynamic analysis is used.
8/17/2019 h5. PBSD_Presentation.pdf
43/49
43
Conclusions
1. PBD can be seen as an approach for managingseismic risk, where by:
8/17/2019 h5. PBSD_Presentation.pdf
44/49
44
Conclusions
Infrequent Rare Very Rare Frequency of DesignGround Shaking Level
l
8/17/2019 h5. PBSD_Presentation.pdf
45/49
45
Conclusions
0 25% 50% 100%$, % Replacement
0 1 7 30 180Downtime, days
C l i
8/17/2019 h5. PBSD_Presentation.pdf
46/49
46
Conclusions
2. Simplified analytical performanceevaluation methodology. (Chapter 4)
Applicable only to well-configured,
regular structures.3. Detailed procedures for performance
evaluation. (Appendix A) Applicableto irregular structures.
C l i
8/17/2019 h5. PBSD_Presentation.pdf
47/49
47
Conclusions
4. Increased confidence in a building’s ability
to provide specific performance can beobtained by:• Providing greater earthquake resistance,• Reducing some of the uncertainties in the
performance evaluation process,• Using more exact procedures of Appendix
A.
8/17/2019 h5. PBSD_Presentation.pdf
48/49
48
Thank you!
Any Questions?
C fid L l
8/17/2019 h5. PBSD_Presentation.pdf
49/49
49
Confidence Levels
exp2
UT x UT k
K b
λ β β ⎧ ⎫⎛ ⎞= − −⎨ ⎬
⎜ ⎟⎝ ⎠⎩ ⎭== Uncertainty measure equal to the vector sum (SRSS) of theUncertainty measure equal to the vector sum (SRSS) of the
logarithmic standard deviation of the variations in demand andlogarithmic standard deviation of the variations in demand andcapacity resulting from uncertainty;capacity resulting from uncertainty;
== Slope of the hazard curve inSlope of the hazard curve in lnln--lnln coordinates at the hazard levelcoordinates at the hazard levelof interest;of interest;
== Coefficient relating the incremental change in demand to anCoefficient relating the incremental change in demand to anincremental change in ground motion intensity (usually taken asincremental change in ground motion intensity (usually taken as 1.0).1.0).
== Standard GaussianStandard Gaussian variate variate associated with the probability ofassociated with the probability of x x notnotbeing exceeded as a function of the number of standard deviationbeing exceeded as a function of the number of standard deviation ssabove or below the mean found in standard probability tables.above or below the mean found in standard probability tables.
K K x x
b b kk
UT
Top Related