Scribd Logo
Upload

Welcome to Scribd!

  • 88 views

    Transformer Lifetime Prediction: Christian Osorio Nandan Sawant

    Uploaded by

    Armel Brissy
    This document discusses a project to predict the lifetime of power transformers using a failure model. The goals are to quantify expected performance, risks, and costs using a transformer's operating history and failure model as inputs. A paper-oil insulation degradation model is presented. The simulation outline generates ideal and history curves, estimates a GEV distribution from the history, and uses Monte Carlo simulation to predict failure probability and associated costs in a given time period. Future work includes modeling multiple failure modes and components, dependent failures, and different cost structures.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Transformer Lifetime Prediction: Christian Osorio Nandan Sawant

    Uploaded by

    Armel Brissy
    88 views17 pages
    This document discusses a project to predict the lifetime of power transformers using a failure model. The goals are to quantify expected performance, risks, and costs using a transformer's operating history and failure model as inputs. A paper-oil insulation degradation model is presented. The simulation outline generates ideal and history curves, estimates a GEV distribution from the history, and uses Monte Carlo simulation to predict failure probability and associated costs in a given time period. Future work includes modeling multiple failure modes and components, dependent failures, and different cost structures.

    Original Title

    ChristianNandan.pdf

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document discusses a project to predict the lifetime of power transformers using a failure model. The goals are to quantify expected performance, risks, and costs using a transformer's operating history and failure model as inputs. A paper-oil insulation degradation model is presented. The simulation outline generates ideal and history curves, estimates a GEV distribution from the history, and uses Monte Carlo simulation to predict failure probability and associated costs in a given time period. Future work includes modeling multiple failure modes and components, dependent failures, and different cost structures.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    88 views17 pages

    Transformer Lifetime Prediction: Christian Osorio Nandan Sawant

    Uploaded by

    Armel Brissy
    This document discusses a project to predict the lifetime of power transformers using a failure model. The goals are to quantify expected performance, risks, and costs using a transformer's operating history and failure model as inputs. A paper-oil insulation degradation model is presented. The simulation outline generates ideal and history curves, estimates a GEV distribution from the history, and uses Monte Carlo simulation to predict failure probability and associated costs in a given time period. Future work includes modeling multiple failure modes and components, dependent failures, and different cost structures.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 17

    Transformer Lifetime Prediction

    Christian Osorio
    Nandan Sawant
    Guide: Prof. Dan O'Neill

    Table of Contents
    Problem Statement
    Project Goals
    Transformer Failure Model
    Simulation Outline and Results
    Future Work
    References

    Introduction
    Cost of a blackout is astronomical. Reliability
    of electric grid is of paramount economic
    importance
    With grid becoming smarter, we can
    effectively monitor the state of the power
    grid and its components
    Goal is to make the expected performance
    quantifiable, make risks and costs
    predictable and controllable

    Reliability of Transformers

    Courtesy: Presentation by Alex Rojas, GE in EE392N

    Project Goals
    Inputs:
    Transformer performance model
    History of unpredictable events
    Outputs:
    Expected lifetime of the transformer
    Probability of failure at any given point in
    time
    Expected cost of failure

    Transformer Failure Modes


    Failure of a transformer is usually a failure of a
    component
    Paper-Oil Insulation Degradation
    Bushing Failure
    Tap Changers:
    Asynchronous operation or carbon layer formation

    Paper Degradation Model


    DP = Degree of Polymerization
    Rg = Universal Gas Constant
    A = Process Constant
    Ea = Activation Energy
    k(t) = Reaction Rate
    t0 = Initial Time

    Simulation Outline
    Generate Input Data
    Generate the ideal transformer curve based on the
    model
    Generate transformer history curve by
    superimposing unpredictable events using a
    Generalized Extreme Value distribution

    Process Input Data


    Estimate GEV distribution looking at the history
    curve and comparing it against model
    Predict probability of failure in some set number of
    days using a monte-carlo simulation

    Ideal Transformer Curve

    DP(t0) = 1200
    Threshold DP = 250
    Ideal transformer lifetime ~ 6000 days ~ 16 years

    Generate History Curve

    Generate History Curve

    Example of a highly unlikely catastrophic event


    expediting transformer failure

    Extract Deviations

    Compute the deviations at each step using


    interpolation techniques

    Estimate The Distribution

    Fitting data to
    the GEV
    distribution,
    mu ~ 0
    sigma ~ 0.05
    shape ~ -3

    Probability of Failure

    Graph produced by a Monte-Carlo simulation


    Computes probability of transformer failure in
    next 30 days

    Cost of Failure
    We assume most systems are N-1 reliable
    In which case, cost of failure is nothing but
    the cost of replacing a transformer
    Cost of a transformer ~ $ 500,000
    Cost of Failure curve is nothing but
    500,000-times scaled version of failure
    probability curve!

    Future Work
    Multiple failure modes
    Bushing failure
    Tap chargers
    Dependence of these modes on one another

    Multiple component failure


    Independent failures
    Dependent failures

    Cost structure
    Cost of component(s) failure
    Cost of a blackout
    Cost of maintenance

    References
    1. The Economic Impacts of the August 2003 Blackout, Electricity Consumers
    Resource Council, February 9, 2004
    2. Prediction of remaining life of power transformers based on left truncated
    and right censored lifetime data, Yili Hong, William Q. Meeker, and James D.
    McCalley, Ann. Appl. Stat. Volume 3, Number 2 (2009), 857-879
    3. Grid Analytics - T&D Power Equipment Diagnosis/Prognosis thru Smart
    Grid Technologies, Alex Rojas, GE
    4. Power Transformer Reliability Modeling - Arjan Schijndel
    5. Extreme Value Theory as a Risk Management Tool, Paul Embrechts, Sidney
    I. Resnick and Gennady Samorodnitsky, North American Actuarial Journal,
    Volume 3, Number 2, April 1999

    You might also like