Research On The Failure Rate On Pump Recovery Well In Oil Field Based On Weibull Distribution

Taking pump recovery well failure rate obey the tub curve as the researchbackground, the failure and failure rate of pump recovery well were emphaticallyanalyzed, the best preventive pump inspection cycle and inspection strategy weredetermined according to the study of failure rate of pump recovery well. This, notreduces the pump inspection costs but improves the reliability and the pumpefficiency of pump recovery well, is of great important practical significance to thepump inspection maintenance of pump recovery well in oil field. Firstly the failure ofpump recovery well was analyzed. Then the correlation coefficient of parametersestimation method based on FPSO-SA(Fuzzy Particle Swarm Optimization-SimulatedAnnealing) was presented, and the optimization method was used to estimatethe parameters of the Weibull distribution model. Finally the failure rate of pumprecovery well was proved to obey three parameter Weibull distribution, and the failurerate function of pump recovery well was determined according to this parameterestimation method used in the study of failure rate. The concrete content is as follows.First of all, using failure mode and effect analysis to analyze and study failure ofpump recovery well. Classify the collected information and data of failure of pumprecovery well, and analyze the failure in failure mode, failure location and failurecause, and at last the corresponding maintenance and repair measures of pumprecovery well are given according to the results.Second, FPSO-SA was presented. The fuzzy theory is introduced into PSO. Inorder to improve the convergence rate of the PSO and solve particle "together" in theprocess of optimization, it was turned by adopting fuzzy logic reasoning and the fuzzyrule according to the characteristic of particle’s motion trajectory in longitudinaldirection and lateral direction. In order to improve the ability of the global optimalvalue of the particles and to solve the particle swarm "premature" problem in theoptimization process, combined the FPSO(Fuzzy Particle Swarm Optimization) andSA(Simulated Annealing), used the update formulas of FPSO as the state generatingfunction of SA, and then used Metropolis algorithm to accept the new state withprobability and then eventually find the global best position by cooling strategy anditerative steps of SA. And the convergence of the FPSO algorithm and FPSO-SAalgorithm was proved, through the benchmark function test, and comparing with otherimproved intelligent algorithms’performances.Third, FPSO-SA was used to optimize the correlation coefficient of parametersestimation method, and then the optimization method was applied to estimate theparameters of the Weibull distribution model. Assume the Weibull distribution location parameter is given, using the least square method to the Weibull distributionfunction for linear regression, and get the linearly dependent coefficient function ofthe Weibull distribution, and FPSO-SA was applied to optimize the best locationparameter estimate, finally calculated estimates of the other two parameters. Theeffect of the method was verified through simulation.Finally, correlation coefficient of parameters estimation method based onFPSO-SA was applied to determine the parameters of pump recovery well failure ratefunction. Analyzed the data of pump inspection cycle, assumed the failure frequencyof pump recovery well obey the Weibull distribution, established the Weibulldistribution model, used the method to estimate the parameters of the Weibulldistribution model, used k-s test method test hypothesis and the assumption turned tocorrect, and then failure rate function of pump recovery well was determined by thereliability function. According to the different element, preventive pump inspectioncycle of pump recovery well was analyzed qualitatively, and considering differentactual situation, the best preventive pump inspection cycle of pump recovery well wasanalysed and determined by using fuzzy decision method. Finally, according to thebest preventive pump inspection cycle and the inspection strategy existed in oilfield,the best preventive pump inspection strategy was resreached and determined.