Scribd
Upload
98 views

Assignment 2 - Report of Failure Probability of Design

1. The research proposes a new methodology to design complex systems for reliability when failure data is limited. 2. The methodology evaluates specific test and field failure data scarcity for horizontal drilling equipment. 3. Key calculations presented include failure rate correction factors, typical sampling for Monte Carlo simulation, reliability models for series, parallel and k-out-of-n systems, load sharing systems, renewal equations, and determining importance measures and variance/covariance for maximum likelihood estimation of parameters.

Uploaded by

norerny shuhada
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
98 views

Assignment 2 - Report of Failure Probability of Design

1. The research proposes a new methodology to design complex systems for reliability when failure data is limited. 2. The methodology evaluates specific test and field failure data scarcity for horizontal drilling equipment. 3. Key calculations presented include failure rate correction factors, typical sampling for Monte Carlo simulation, reliability models for series, parallel and k-out-of-n systems, load sharing systems, renewal equations, and determining importance measures and variance/covariance for maximum likelihood estimation of parameters.

Uploaded by

norerny shuhada
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 11

MKAR 1043 RISK ANALYSIS

Assignment 2 (Group)

FAILURE PROBABILITY OF DESIGN

Name Matrix No. No. H.p


AL’AZHARINO BIN AHMAD MKA191021 013-3462894
CHAN KAH WAI MKA191009 011-12257099
NORERNY SHUHADA BINTI HAMDAN MKA191014 013-2772135
Research Article: Design for Reliability of Complex System: Case Study of Horizontal
Drilling Equipment with Limited Failure Data

Morteza Suleiman, Mohammad Pourgol-Mohammad, Ali Rostami, and Ahmad Ghanbari


1
Tabriz University, Tabriz 5166616471, Iran
2
Sahand University of Technology, Tabriz 51335-1996, Iran
Received 11 August 2014; Accepted 30 October 2014; Published 30 November 2014
Academic Editor: Antoine Grall

1. Objective of the research: Propose new methodology for complex systems’ design for
reliability.

2. Methodology: Evaluate specific test and field failure data scarcity.

3. Summarize of research:

In this article, industries involved in drilling activities often face failures of their own
drilling systems and equipment. During the drilling process, sometimes the drill bit is
broken. This causes the downtime take too long. It happens whether to repatriate broken
drill points and reinstall new drill points. In addition, due to excessive drilling, the
electrical components of the drilling system will fail. All of these contribute to waste of
time and result in losses in operating and production costs. Therefore, they want to
evaluate a new design that is intended to be created whether it is reliable or not. The
objective is to predict when damage and failure will occur, making replacement of the
equipment and components involved before failure and thus reducing downtime.
4. Summarize the Calculation

a. Failure Rate Correction (for mechanical relay) :

𝜆𝑝 = 𝜆𝑏 ⋅ 𝜋𝐿 ⋅ 𝜋𝐶 ⋅ 𝜋CYC ⋅ 𝜋𝐹 ⋅ 𝜋𝑄 ⋅ 𝜋𝐸 Failure per 106 hour, (1)

i. where base failure rate (𝜆𝑏) is

𝜆𝑏 = 0.0054 × exp [(𝑇𝐴 + 273) / 377 ]10.4 for 125ºC rated temperature,

where 𝑇𝐴 is ambient temperature (°C).

ii. Load stress factor (𝜋𝐿) is

𝜋𝐿 = exp (𝑆/0.8)2 for resistive load type,


𝑆 = (Operating Load Current / Rated Resistive Load Current).

iii. Contact form factor (𝜋𝐶) is

𝜋𝐶 = 1.00 for SPST contact form.

iv. Cycling factor (𝜋CYC) is

𝜋CYC = (cycle per hour / 10) for 10 − 1000 cycle rate.

v. Application and construction factor (𝜋𝐹) is

𝜋𝐹 = 12 for General purpose Application and Solenoid Construction Type.


vi. Quality factor (𝜋𝑄) is

𝜋𝑄 = 2.9 for commercial quality.

vii. Environment factor (𝜋𝐸) is

𝜋𝐸 = 44 for 𝐺𝑀 (ground, mobile) Environment.

b. Typical sampling for 𝑘 elements in 𝑛 iterations: (Monte Carlo Simulation Method).


Sampling of CDF of each xi (sampling size to estimate availability values)

𝑥11, 𝑥21,..., 𝑥𝑘1 = 𝐹 (𝑥𝑖1),


𝑥12, 𝑥22,...,2 = 𝐹 (𝑥𝑖2),
. (2)
.
.
𝑥1𝑛, 𝑥2𝑛,..., 𝑥𝑘𝑛 = 𝐹 (𝑥𝑖𝑛),

where 𝑥𝑘𝑛 is the 𝑛th iteration of 𝑘th parameter and 𝐴(𝑡) is the availability value.

c. The respective reliability models :

The cumulative distribution function: (𝛿)


The probability density function of the component strength: (𝛿),
The load probability density function is (𝑠).
Reliability of the series system

(3)

Reliability of the parallel system

(4)

Reliability of the 𝑘-out-of-𝑛 system

(5)

Where :
d. Load- sharing system, with two same items, initially both units share the load, with
times to failure distribution being ƒh(t).

(6)

e. Renewal equation:

(7)

Where W(t) is CIF and F(t) is CDF function.

f. Let An be the virtual age of the system has time to the (n+1)th failure Xn+1 which is
distributed according to the following CDF. Conditional CDF of system at age y
shown below:

(8)

where F(X) is the CDF of the TTFF distribution of the system when it was new
(underlying) distribution.
h. For the GRP, the expected number of failures in (0, 𝑡), that is, CIF 𝑊(𝑋), is given by
a solution of the so-called 𝐺- renewal equation:

(9)

where

(𝑡), and 𝑓(𝑡) are the CDF and PDF of the TTFF (underlying) distribution.

i. Average availability is formulated as follows:

(10)

h. Importance measure IRi is:

(11)

Where Rs(t) is reliability of the system and Ri(t) is reliability of the component i.

j. From mathematical point of view, the reliability allo- cation problem is a nonlinear
programming problem. It is shown as follows:

Maximize
(12)
where
 Rs is system reliability, 0 ≤ Rs ≤ 1, Fs unrealibility of system
 Rj is componenet reliability of stage j, 0 ≤ Rs ≤ 1
 Rj min is lower limit on Rs
 Rj max is upper limit on Rs, bi is resources allocated to ith type of
constraint,
 f (∙) is the system reliability function
 gi(∙) is the ith constraint function
 n is number of subsystem in the system
 m is the number of resources
k. In this method, weighting factors for a subsystem are equal to the division of the
failure rate of the subsystem to the sum of all subsystems failure rates of a system.
Equation shows below:

(13)

where
n is the number of subsystems
λi is the failure rate of ith subsystems
λ* is the required failure rate of system
λi* is the allocated failure rate for ith subsystems
wi is the weighting factors

l. Determination of variance and covariance of the MLE parameters matrix is obtained


by the inverse of Fisher matrix:

(14)
5. Calculation Report

Equation No. Discussions / Opinion


1 The limit state function of the operation failure probability. This equation
represent the various function of operation which can contributes to the
failure rate of the electromechanical. Hence this equation can give results
on how much the electromechanical relay for every 106 hour needed.
2 The failure rate of a system or component fails which expressed failures
per unit of time. This function is in the normal distribution form which can
give the exact time where the mechanical failure predicted to happen.
3 Load stress factor is expressed in Rayleigh distribution whereby it has
equal variance with zero mean. In this case, the load stress is represented
with operating load current and the resistive load current at the same time
during the drilling work.
4 Function with a constant value contact form factor 1.00 for single pole
single throw (SPST) in one circuit which has the single ON position.
5 Factor which represented with the number of drilling equipment cycles
operate for an hour
6 Constant value = 12, for application and construction factor for general
purpose application and Solenoid construction type
7 Constant value = 2.9 for common quality of commercial mechanical driller
8 Constant value = 44 which represent the environment factor based on the
ground condition (fixed environment disturbance) or any mobile (moving
disturbance) from surrounding
9 Monte Carlo Simulation Method that can simulate million values of
sampling value, x at different parameters. It can give the area of
probability value at different range of samples required.
10
11
12
13
14

You might also like