MVFOSMAnalysis.cppGo to the documentation of this file.00001 /* ****************************************************************** ** 00002 ** OpenSees - Open System for Earthquake Engineering Simulation ** 00003 ** Pacific Earthquake Engineering Research Center ** 00004 ** ** 00005 ** ** 00006 ** (C) Copyright 2001, The Regents of the University of California ** 00007 ** All Rights Reserved. ** 00008 ** ** 00009 ** Commercial use of this program without express permission of the ** 00010 ** University of California, Berkeley, is strictly prohibited. See ** 00011 ** file 'COPYRIGHT' in main directory for information on usage and ** 00012 ** redistribution, and for a DISCLAIMER OF ALL WARRANTIES. ** 00013 ** ** 00014 ** Developed by: ** 00015 ** Frank McKenna (fmckenna@ce.berkeley.edu) ** 00016 ** Gregory L. Fenves (fenves@ce.berkeley.edu) ** 00017 ** Filip C. Filippou (filippou@ce.berkeley.edu) ** 00018 ** ** 00019 ** Reliability module developed by: ** 00020 ** Terje Haukaas (haukaas@ce.berkeley.edu) ** 00021 ** Armen Der Kiureghian (adk@ce.berkeley.edu) ** 00022 ** ** 00023 ** ****************************************************************** */ 00024 00025 // $Revision: 1.2 $ 00026 // $Date: 2003/04/28 20:51:25 $ 00027 // $Source: /usr/local/cvs/OpenSees/SRC/reliability/analysis/analysis/MVFOSMAnalysis.cpp,v $ 00028 00029 00030 // 00031 // Written by Terje Haukaas (haukaas@ce.berkeley.edu) 00032 // 00033 00034 #include <MVFOSMAnalysis.h> 00035 #include <ReliabilityAnalysis.h> 00036 #include <ReliabilityDomain.h> 00037 #include <RandomVariablePositioner.h> 00038 #include <GFunEvaluator.h> 00039 #include <GradGEvaluator.h> 00040 #include <Matrix.h> 00041 #include <Vector.h> 00042 #include <tcl.h> 00043 00044 #include <fstream> 00045 #include <iomanip> 00046 #include <iostream> 00047 using std::ifstream; 00048 using std::ios; 00049 using std::setw; 00050 using std::setprecision; 00051 using std::setiosflags; 00052 00053 00054 MVFOSMAnalysis::MVFOSMAnalysis(ReliabilityDomain *passedReliabilityDomain, 00055 GFunEvaluator *passedGFunEvaluator, 00056 GradGEvaluator *passedGradGEvaluator, 00057 Tcl_Interp *passedTclInterp, 00058 const char *passedFileName) 00059 :ReliabilityAnalysis() 00060 { 00061 theReliabilityDomain = passedReliabilityDomain; 00062 theGFunEvaluator = passedGFunEvaluator; 00063 theGradGEvaluator = passedGradGEvaluator; 00064 theTclInterp = passedTclInterp; 00065 fileName = new char[256]; 00066 strcpy(fileName,passedFileName); 00067 } 00068 00069 00070 MVFOSMAnalysis::~MVFOSMAnalysis() 00071 { 00072 if (fileName != 0) 00073 delete [] fileName; 00074 } 00075 00076 00077 00078 int 00079 MVFOSMAnalysis::analyze(void) 00080 { 00081 00082 // Alert the user that the FORM analysis has started 00083 opserr << "MVFOSM Analysis is running ... " << endln; 00084 00085 00086 // Initial declarations 00087 int i,j,k; 00088 00089 00090 // Open output file 00091 ofstream outputFile( fileName, ios::out ); 00092 00093 00094 // Get number of random variables 00095 int nrv = theReliabilityDomain->getNumberOfRandomVariables(); 00096 00097 00098 // Get mean point 00099 RandomVariable *aRandomVariable; 00100 Vector meanVector(nrv); 00101 for (i=1; i<=nrv; i++) 00102 { 00103 aRandomVariable = theReliabilityDomain->getRandomVariablePtr(i); 00104 if (aRandomVariable == 0) { 00105 opserr << "MVFOSMAnalysis::analyze() -- Could not find" << endln 00106 << " random variable with tag #" << i << "." << endln; 00107 return -1; 00108 } 00109 meanVector(i-1) = aRandomVariable->getMean(); 00110 } 00111 00112 00113 // Establish vector of standard deviations 00114 Vector stdvVector(nrv); 00115 for (i=1; i<=nrv; i++) 00116 { 00117 aRandomVariable = theReliabilityDomain->getRandomVariablePtr(i); 00118 stdvVector(i-1) = aRandomVariable->getStdv(); 00119 } 00120 00121 00122 // Evaluate limit-state function 00123 int result; 00124 result = theGFunEvaluator->runGFunAnalysis(meanVector); 00125 if (result < 0) { 00126 opserr << "SearchWithStepSizeAndStepDirection::doTheActualSearch() - " << endln 00127 << " could not run analysis to evaluate limit-state function. " << endln; 00128 return -1; 00129 } 00130 00131 00132 // Establish covariance matrix 00133 Matrix covMatrix(nrv,nrv); 00134 for (i=1; i<=nrv; i++) { 00135 covMatrix(i-1,i-1) = stdvVector(i-1)*stdvVector(i-1); 00136 } 00137 int ncorr = theReliabilityDomain->getNumberOfCorrelationCoefficients(); 00138 CorrelationCoefficient *theCorrelationCoefficient; 00139 double covariance, correlation; 00140 int rv1, rv2; 00141 for (i=1 ; i<=ncorr ; i++) { 00142 theCorrelationCoefficient = theReliabilityDomain->getCorrelationCoefficientPtr(i); 00143 correlation = theCorrelationCoefficient->getCorrelation(); 00144 rv1 = theCorrelationCoefficient->getRv1(); 00145 rv2 = theCorrelationCoefficient->getRv2(); 00146 covariance = correlation*stdvVector(rv1-1)*stdvVector(rv2-1); 00147 covMatrix(rv1-1,rv2-1) = covariance; 00148 covMatrix(rv2-1,rv1-1) = covariance; 00149 } 00150 00151 00152 // 'Before loop' declarations 00153 int numLsf = theReliabilityDomain->getNumberOfLimitStateFunctions(); 00154 Vector gradient(nrv); 00155 Matrix matrixOfGradientVectors(nrv,numLsf); 00156 Vector meanEstimates(numLsf); 00157 Vector responseStdv(numLsf); 00158 double responseVariance; 00159 00160 00161 // Loop over limit-state functions 00162 for (int lsf=1; lsf<=numLsf; lsf++ ) { 00163 00164 // Inform the user which limit-state function is being evaluated 00165 opserr << "Limit-state function number: " << lsf << endln; 00166 00167 00168 // Set tag of active limit-state function 00169 theReliabilityDomain->setTagOfActiveLimitStateFunction(lsf); 00170 00171 00172 // Get limit-state function value (=estimation of the mean) 00173 result = theGFunEvaluator->evaluateG(meanVector); 00174 if (result < 0) { 00175 opserr << "SearchWithStepSizeAndStepDirection::doTheActualSearch() - " << endln 00176 << " could not tokenize limit-state function. " << endln; 00177 return -1; 00178 } 00179 meanEstimates(lsf-1) = theGFunEvaluator->getG(); 00180 00181 00182 // Evaluate (and store) gradient of limit-state function 00183 result = theGradGEvaluator->evaluateGradG(meanEstimates(lsf-1),meanVector); 00184 if (result < 0) { 00185 opserr << "MVFOSMAnalysis::analyze() -- could not" << endln 00186 << " compute gradients of the limit-state function. " << endln; 00187 return -1; 00188 } 00189 gradient = theGradGEvaluator->getGradG(); 00190 for (i=1 ; i<=nrv ; i++) { 00191 matrixOfGradientVectors(i-1,lsf-1) = gradient(i-1); 00192 } 00193 00194 00195 // Estimate of standard deviation 00196 responseVariance = (covMatrix^gradient)^gradient; 00197 if (responseVariance <= 0.0) { 00198 opserr << "ERROR: Response variance of limit-state function number "<< lsf << endln 00199 << " is zero! " << endln; 00200 } 00201 else { 00202 responseStdv(lsf-1) = sqrt(responseVariance); 00203 } 00204 00205 00206 // Print MVFOSM results to the output file 00207 outputFile << "#######################################################################" << endln; 00208 outputFile << "# MVFOSM ANALYSIS RESULTS, LIMIT-STATE FUNCTION NUMBER " 00209 <<setiosflags(ios::left)<<setprecision(1)<<setw(4)<<lsf <<" #" << endln; 00210 outputFile << "# #" << endln; 00211 00212 outputFile << "# Estimated mean: .................................... " 00213 <<setiosflags(ios::left)<<setprecision(5)<<setw(12)<<meanEstimates(lsf-1) 00214 << " #" << endln; 00215 outputFile << "# Estimated standard deviation: ...................... " 00216 <<setiosflags(ios::left)<<setprecision(5)<<setw(12)<<responseStdv(lsf-1) 00217 << " #" << endln; 00218 outputFile << "# #" << endln; 00219 outputFile << "#######################################################################" << endln << endln << endln; 00220 00221 00222 // Inform the user that we are done with this limit-state function 00223 opserr << "Done analyzing limit-state function " << lsf << endln; 00224 } 00225 00226 00227 // Estimation of response covariance matrix 00228 Matrix responseCovMatrix(numLsf,numLsf); 00229 double responseCovariance; 00230 Vector gradientVector1(nrv), gradientVector2(nrv); 00231 for (i=1; i<=numLsf; i++) { 00232 for (k=0; k<nrv; k++) { 00233 gradientVector1(k) = matrixOfGradientVectors(k,i-1); 00234 } 00235 for (j=i+1; j<=numLsf; j++) { 00236 for (k=0; k<nrv; k++) { 00237 gradientVector2(k) = matrixOfGradientVectors(k,j-1); 00238 } 00239 responseCovariance = (covMatrix^gradientVector1)^gradientVector2; 00240 responseCovMatrix(i-1,j-1) = responseCovariance; 00241 } 00242 } 00243 for (i=1; i<=numLsf; i++) { 00244 for (j=1; j<i; j++) { 00245 responseCovMatrix(i-1,j-1) = responseCovMatrix(j-1,i-1); 00246 } 00247 } 00248 00249 00250 // Corresponding correlation matrix 00251 Matrix correlationMatrix(numLsf,numLsf); 00252 for (i=1; i<=numLsf; i++) { 00253 for (j=i+1; j<=numLsf; j++) { 00254 correlationMatrix(i-1,j-1) = responseCovMatrix(i-1,j-1)/(responseStdv(i-1)*responseStdv(j-1)); 00255 } 00256 } 00257 for (i=1; i<=numLsf; i++) { 00258 for (j=1; j<i; j++) { 00259 correlationMatrix(i-1,j-1) = correlationMatrix(j-1,i-1); 00260 } 00261 } 00262 00263 00264 // Print correlation results 00265 outputFile << "#######################################################################" << endln; 00266 outputFile << "# RESPONSE CORRELATION COEFFICIENTS #" << endln; 00267 outputFile << "# #" << endln; 00268 if (numLsf <=1) { 00269 outputFile << "# Only one limit-state function! #" << endln; 00270 } 00271 else { 00272 outputFile << "# gFun gFun Correlation #" << endln; 00273 outputFile.setf(ios::fixed, ios::floatfield); 00274 for (i=0; i<numLsf; i++) { 00275 for (j=i+1; j<numLsf; j++) { 00276 // outputFile.setf(ios::fixed, ios::floatfield); 00277 outputFile << "# " <<setw(3)<<(i+1)<<" "<<setw(3)<<(j+1)<<" "; 00278 if (correlationMatrix(i,j)<0.0) { outputFile << "-"; } 00279 else { outputFile << " "; } 00280 // outputFile.setf(ios::scientific, ios::floatfield); 00281 outputFile <<setprecision(7)<<setw(11)<<fabs(correlationMatrix(i,j)); 00282 outputFile << " #" << endln; 00283 } 00284 } 00285 } 00286 outputFile << "# #" << endln; 00287 outputFile << "#######################################################################" << endln << endln << endln; 00288 00289 00290 00291 // Print summary of results to screen (more here!!!) 00292 opserr << "MVFOSMAnalysis completed." << endln; 00293 00294 00295 return 0; 00296 } 00297
Generated on Mon Oct 23 15:05:25 2006 for OpenSees by |
opensees-support @ berkeley.edu
Supported by the National Science Foundation
©2006, UC Regents