KUANTIFIKASI KETIDAKPASTIAN PADA ANALISIS POHON KEGAGALAN DENGAN PENDEKATAN FUZZY

Analisis pohon kegagalan dipakai untuk mengevaluasi kinerja sistem keselamatan pembangkit listrik tenaga nuklir. Analisis ini memerlukan ketersediaan data kegagalan komponen. Karena keandalan komponen dipengaruhi oleh lingkungan kerjanya maka perlu digunakan data kegagalan komponen yang berasal dari sistem yang sedang dievaluasi. Namun kenyataannya, data ini sangat sulit diperoleh sehingga penggunaan data jenerik menjadi tak terhindarkan. Penggunaan data jenerik tentunya akan menyebabkan ketidakpastian pada hasil analisis. Simulasi Monte Carlo sering dipakai untuk mengkuantifikasi ketidakpastian ini. Namun sebenarnya metode ini kurang tepat untuk mengevaluasi ketidakpastian apabila jumlah data yang dimiliki sangat terbatas. Tujuan dari penelitian ini adalah pengembangan sebuah metode analisis pohon kegagalan baru yang menerapkan konsep fuzzy untuk kuantifikasi ketidakpastian. Dalam metode baru ini, probabilitas fuzzy dipakai untuk merepresentasikan probabilitas kejadian dasar, antara serta puncak dan hukum kombinasi fuzzy dipakai untuk mengevaluasi ketidakpastian hasil analisis. Kebolehjadian gagalnya sistem injeksi akumulator AP1000 telah dievaluasi dengan menggunakan metode baru ini dan diperoleh ketidakpastian kegagalan pada interval 8,87E-12 – 8,87E-8 dengan nilai titik tengah 8,87E-10. Hasil ini membuktikan bahwa analisis pohon kegagalan dengan pendekatan fuzzy ini layak dipakai apabila yang menjadi fokus evaluasi adalah ketidakpastian karena keterbatasan data kegagalan yang dimiliki. Kata kunci : Analisis pohon kegagalan, analisis ketidakpastian, probabilitas fuzz y , hukum kombinasi fuzzy Fault tree analysis has been applied to evaluate nuclear power plant safety systems. To perform this analysis, component reliabilities need to be provided well in advance. Since working environment can affect component reliability, it is necessary to directly collect such data from the safety system being evaluated. However , due to lack of resources , such data may be unattainable . Hence, the use of generic data cannot be avoided. Unfortunately, generic data will add uncertainty to the analysis. Monte Carlo simulation has been performed to evaluate such uncertainty. However, this method is not appropriate when components do not have probability distribution s of their lifetime to failures. The aim of this study is to propose a new fault tree analysis method which implements fuzzy concepts for quantifying such uncertainty. In the proposed method, fuzzy probabilities represent basic, intermediate as well as top event probabilities and fuzzy combination rules are used to evaluate the overall uncertainty of the fault tree. The proposed method has been performed to evaluate failure probability of the AP1000 accumulator injection system and generate a probability distribution between 8.87E-12 and 8.87E-8 with the point median value of 8.87E-10. This result confirms that the proposed method is feasible to evaluate system fault tree when uncertainty raised by the lack of reliability data is the main focus of the analysis. Keywords : F ault tree analysis, uncertainty analysis, fuzzy probabilitie s , fuzzy combination rules

[1]  Jhp Julwan Hendry Purba,et al.  A fuzzy-based reliability approach to evaluate basic events of fault tree analysis for nuclear power plant probabilistic safety assessment , 2014 .

[2]  Jhp Julwan Hendry Purba,et al.  Fuzzy probability on reliability study of nuclear power plant probabilistic safety assessment: A review , 2014 .

[3]  Ümit Sami Sakalli,et al.  An application of investment decision with random fuzzy outcomes , 2010, Expert Syst. Appl..

[4]  M Sam Mannan,et al.  Fuzzy risk matrix. , 2008, Journal of hazardous materials.

[5]  Takehisa Onisawa.  Fuzzy theory in reliability analysis , 1989 .

[6]  Xiliang Zhang,et al.  Nuclear energy development in China: A study of opportunities and challenges , 2010 .

[7]  R. K. Sinha,et al.  Role of passive systems in advanced reactors , 2007 .

[8]  Wen-Chin Chou,et al.  A hybrid fuzzy MCDM approach for evaluating website quality of professional accounting firms , 2012, Expert Syst. Appl..

[9]  Z. Musicki,et al.  Issues and insights of PRA methodology in nuclear and space applications , 2005, 2005 IEEE International Conference on Systems, Man and Cybernetics.

[10]  Brian Veitch,et al.  Fault and Event Tree Analyses for Process Systems Risk Analysis: Uncertainty Handling Formulations , 2011, Risk analysis : an official publication of the Society for Risk Analysis.

[11]  Celso Marcelo Franklin Lapa,et al.  FUZZY METHODOLOGY APPLIED TO PROBABILISTIC SAFETY ASSESSMENT FOR DIGITAL SYSTEM IN NUCLEAR POWER PLANTS , 2011 .

[12]  Jane M. Booker,et al.  Expert Knowledge in Reliability Characterization: A Rigorous Approach to Eliciting, Documenting, and Analyzing Expert Knowledge , 2004 .

[13]  C. W. Chan,et al.  Fuzzy fault tree analysis based on T–S model with application to INS/GPS navigation system , 2008, Soft Comput..

[14]  Michael Hanss,et al.  Identification procedure for epistemic uncertainties using inverse fuzzy arithmetic , 2010 .

[15]  Robert LIN,et al.  NOTE ON FUZZY SETS , 2014 .

[16]  Jian-Bo Yang,et al.  Failure mode and effects analysis using a group-based evidential reasoning approach , 2009, Comput. Oper. Res..

[17]  Celso Marcelo Franklin Lapa,et al.  Fuzzy uncertainty modeling applied to AP1000 nuclear power plant LOCA , 2011 .

[18]  T. Aven,et al.  Probability and Possibility‐Based Representations of Uncertainty in Fault Tree Analysis , 2013, Risk analysis : an official publication of the Society for Risk Analysis.

[19]  Wei Wang,et al.  Simulation and analysis on 10-in. cold leg small break LOCA for AP1000 , 2012 .

[20]  Alessandro Birolini.  Basic Concepts, Quality and Reliability Assurance of Complex Equipment and Systems , 2010 .

[21]  Liping Fang,et al.  Managing construction risks of AP1000 nuclear power plants in China , 2011 .

[22]  Aminah Robinson Fayek,et al.  Quantitative Assessment of Horizontal Directional Drilling Project Risk Using Fuzzy Fault Tree Analysis , 2010 .

[23]  Ayah E. Abouelnaga,et al.  Assessment the safety performance of nuclear power plants using Global Safety Index (GSI) , 2010 .

[24]  Didier Dubois,et al.  Gradualness, uncertainty and bipolarity: Making sense of fuzzy sets , 2012, Fuzzy Sets Syst..

[25]  Jian-Bo Yang,et al.  On the centroids of fuzzy numbers , 2006, Fuzzy Sets Syst..

[26]  Witold Pedrycz,et al.  A fuzzy reliability assessment of basic events of fault trees through qualitative data processing , 2014, Fuzzy Sets Syst..

[27]  M. Asztalos,et al.  AP1000 passive core cooling system pre-operational tests procedure definition and simulation by means of Relap5 Mod. 3.3 computer code ☆ , 2012 .

[28]  Julwan Hendry Purba,et al.  Reliability Study of the AP1000 Passive Safety System by Fuzzy Approach , 2014 .

[29]  Ajit Srividya,et al.  Quantification of epistemic and aleatory uncertainties in level-1 probabilistic safety assessment studies , 2007, Reliab. Eng. Syst. Saf..

[30]  Yun Zhou,et al.  Is China ready for its nuclear expansion , 2011 .

[31]  Seyed Khalil Mousavian,et al.  A study of large break LOCA in the AP1000 reactor containment , 2012 .