Microsoft Word - -F'F H3H3F HDEI23- 35 This is an open access article under the CC BY license: Al-Khwarizmi Engineering Journal Al-Khwarizmi Engineering Journal, Vol. 17, No. 4, December, (2021) P. P. 23- 35 Evaluation of the Main Causes of Diesel Engine Injector Failure using Fault Tree Analysis Hanan B. Matar* Sawsan S. Al-Zubaidi** Luma A. Al-Kindi*** *,**,*** Department of Production Engineering and Metallurgy Engineering / University of Technology/ Iraq pme.19.29@grad.uotechnology.edu.iqEmail: * 70004@uotechnology.edu.iqEmail: ** .iquluma.a.alkindi@uotechnology.edEmail: *** (Received 29 August 2021; Accepted 29 December 2021) 6.0012https://doi.org/10.22153/kej.2021. Abstract The maintenance of the diesel engine parts in any electric power station contains many problems that lead to stopping. Several reasons lead to such problems; these reasons should be analyzed and evaluated in order to eliminate their effects. This paper is based on evaluation of the main causes that lead to diesel engine injector failure as a main part of electric power stations, using fault tree analysis (FTA). The FTA is the most broadly utilized strategies in the industrial area to perform reliability analysis of complex designing frameworks. A fault tree is a logical representation of the relationship of basic events that lead to a given unwanted event (i.e., top event). Starting with introducing the FTA and how it could be utilized in analyzing the reasons of main issues that lead to that the injector stops working, in which the probability of occurrence of each issue is calculated. The application of the root cause analysis principle of diesel engine injector failure in diesels Haditha station is chosen as a case under study. According to the probabilities’ evaluation by using the FTA based on of the causes under the top event its value was (0.80). It is concluded from the analysis that the three largest values of the occurrence of a problem with the diesel engine injector are 0.50 for fuel problems, 0.20 for overloads, and 0.18 for nozzle head corrosion. Plant management can define a specific plan with taking into consideration the calculated value of each cause in order to reduce their impacts and to avoid long downtime hours compared to operating hours. Finally, recommandations are suggested to overcome these causes. Keywords: Fault tree analysis (FTA), Fault tree symbols, injector, qualitative assessment. 1. Introduction Given the importance of maintenance of all kinds, it is recommended implementing it and adhering to the time and operating hours specified for its work. Providing all maintenance requirements and tolerance for one minute under any circumstances, especially during peak work times, is essential in order to avoid a damage that may be caused to parts of the generating units and expose them to a stoppage. This state leads to energy loss, stopped production, and losses in lives and money. Most engineering systems are maintained the mainteance work when they fail, and the maintenance work is performed on them to keep their operation. [1] Fault tree analysis (FTA) is one of the analytical methods for tracking events, which can be contributed to identify the most important parts that cause the malfunction or stoping the work. Accordingly, the sub-causes of the problem is Hanan B. Matar Al-Khwarizmi Engineering Journal, Vol. 17, No. 4, P.P. 23- 35 (2021) 24 displayed and the probability of an undesirable fault event is calculated. In this paper adopting the FTA is considered in the proposed methodology to determine the main and sub causes, and their probabilities of the parts of power plants’ stoppage. The focus on the most influential causes probabilities in order is essential to find the solutions to these causes which significantly reduce the rate of breakdowns. 2. Fault Tree Analysis (FTA)  An FTA begins with a top fault event and continues deductively by inquiring, "How could this event have occurred?". Immediate contributing causes to this top fault event are distinguished and afterward recorded as a component of the following lower level of analysis (sub-faults). These sub-fault events are then associated as affecting contributions to the upper-level event by the same token either "AND" gates or "OR" gates. A definitive outcome is a qualitative fault tree ending in a bunch of basic faults and undeveloped events. [2] The FTA is a technique for planning a fault tree that intelligently examines the causal relationship of a framework disappointment, and discovers the likelihood of a framework disappointment. [3] The FTA is deductive in nature implying that the analysis begins with a top event (system failure) and works in reverse from the highest point of the tree towards the leaves of the tree to decide the root causes of the top event. [4] Failure itself can be characterized as the interruption of an item's capacity, from a component to a complex system, to carry out its function. The failure of a part can be classified into three groups [5]: 1) Primary failure, 2) A component is in a damaged state (non-working state), and 3) Command faults. 3. The Symbols used in the FTA The essential symbols utilized in the FTA are assembled as events, gates, and transfer symbols. Fig.1 shows them with their definitions [6] [7] [8]: Fig.1. The fault tree symbols. Hanan B. Matar Al-Khwarizmi Engineering Journal, Vol. 17, No. 4, P.P. 23- 35 (2021) 25 4. Research Methodology The research methodology is based on applying the FTA to evaluate and calculate the failure rate of the part under study. To arrange the work steps, a flowchart is used to define processes in sequence. The utilization of standardized symbols for represent the steps of methodology makes the flow chart simpler to decipher. [9] 4.1. Proposed Flow Chart Methodology The main steps in the proposed methodology are as follows: 1. Describing of the whole system, and studying its work, to select the part with frequent failure with the ability to be maintained. 2. Collecting all the data on faults and dividing them into main faults, in which the root causes of the main problems branch off, as to be illustrated in details in the qualitative assessment method followed. 3. After clarifying all the reasons that led to the failure of the part under study, the FTA is drawn to analyze the problems more accurately and clearly. 4. Applying steps (6) (to be illustrated in the following section), according to the type of gate (OR or AND) in order to get the top event and draw the fault tree with related values (quantitative assessment). 5. A bar chart is drawn to illustrate the most influential causes and to choose an appropriate maintenance plan that reduces breakdowns and costs. 6. Suggesting solutions to reduce the influence of the occurrence are given. 7. If the suggestions improve the faults rates, then the aim is reached. But if not, new suggestions are given. Fig.2. represents the steps of the proposed methodology processes. 4.2 The Assessment of Problem: A fault tree is evaluated as Qualitative Assessment or Quantitative Assessment. Qualitative Assessment is the kind of evaluation of an issue that has effectively occurred in a venture, to analyze the causes behind this issue. [10] In Quantitative Analysis, the probability of the top event and other quantitative dependability files, for example, significant measures are numerically determined, given the failure rate or probability of individual system component. [4] Two equations are used according to the type of gate (AND or OR) for quantitative analysis[11]: For AND gate out fault event Eo is given by: … ( 1 ) and for OR gate out fault event Xo is given by : …( 2 ) Concerning analysis procedures, recognizing qualitative the FTA is performed, which thinks about the construction of the FTA; and quantitative FTA, which estimate the numbers, for example, disappointment probabilities for fault trees. [12] Both qualitative and quantitative analyses bring a bunch of basic ways, often named "cut sets" or "min sets." These are group of parts that, when they happen, will cause a top fault event. [2] In the case study of this research, both qualitative and quantitative assessment are shown. Hanan B. Matar Al-Khwarizmi Engineering Journal, Vol. 17, No. 4, P.P. 23- 35 (2021) 26 Fig. 2. The flowchart of steps of the processes flow. Hanan B. Matar Al-Khwarizmi Engineering Journal, Vol. 17, No. 4, P.P. 23- 35 (2021) 27 5. Adopted Case Study The case study adopted in this paper is the diesel engine injector in Haditha diesels station. The engine is a type of four-stroke internal combustion engine. Under such circumstances, the focus will be on the injector as a highly effective part of the diesel engine. It is one of the parts of the fuel system whose function is to inject fuel in the form of a mist with a force (3000bar) that reaches the combustion chamber. The presence of the injector results in more efficiency and lower cost in terms of fuel exchange. The injector is shown in Fig.3. Fig. 3. The Injector. 6. Drawing the Fault Tree and Calculations The fault tree is drawn from the data that the researcher collects (values and events) and were recorded through his attendance at the station and with the help of the engineer who is supervising the work and the workers. According to the data recorded in the power station foles, main and sub- causes are named. Table (1) lists the main problems and sub-causes events that cause the injector to stop working in order to draw a fault tree which represente the qualitative assessment as shown in Fig.4. Table1, The nomenclature of the main problems of the occurrence of malfunctions and their sub-causes. Name of Event Symbol of Event No maintenance order E1 High fuel temperature E2 Work for long periods E3 Fuel laden with water E4 Water leakage from the heat sink E5 Fuel laden with impurities E6 Fuel filter has stopped working E7 Accumulation of previous fuel impurities E8 Neglect of maintenance manger E9 Urgent need for E10 Increased pressure E11 High temperature E12 Other reasons E13 Excessive pressure at the attached nozzle E14 Over load E15 Poor Maintenance E16 Water in fuel E17 Viscosity of the fuel E18 Impurities in the fuel E19 Injector vibration E20 Other reasons E21 Water pressure E22 Defect in the washer E23 Fuel heat E24 Overloads E25 Delay maintenance time E26 Nozzle head corrosion E27 Nozzle head does not pump fuel E28 Fracture of the injector hull E29 Fuel problems E30 O-ring I E31 O-ring II E32 Diesel engine injector mal functioning E33 Hanan B. Matar Al-Khwarizmi Engineering Journal, Vol. 17, No. 4, P.P. 23- 35 (2021) 28 Fig.4. Fault tree analysis of diesel engine injector. Hanan B. Matar Al-Khwarizmi Engineering Journal, Vol. 17, No. 4, P.P. 23- 35 (2021) 29 The presence in the station and data collection were done based on the realized events (previous faults) and calculating the probabilities of the basic events is shown in Table (2). Table 2, The probability of occurrence of basic events. Using the data provided by Haditha diesel station company, according to their records as shown in Table (2) and the FTA the following calculators are performed, by apply 'OR' or 'AND' equation, as given: 1.The probability of event occurring E32 is (O- ring II = 0.06), which means that the injector has stopped working due to a malfunction or breakage in the O-rings of the refrigeration system, as given (A very small percentage) 2.The probability of event occurring E31 is (O- ring I = 0.05), meaning that the injector stops are due to a malfunction or breakage in the fuel system's O-ring. (A very small percentage) 3. The probability of event occurring E19 is (Impurities in the fuel = 0.20) caused by (E6 =Fuel laden with impurities, E7 =Fuel filter has stopped working, and E8 =Accumulation of previous fuel impurities) as given: In addition, the probability of an event occurring E17 is (Water in fuel = 0.10) caused by (E4 =Fuel laden with water, and E5 =Water leakage from the heat sink) as: With the probability of an event occurring E18 is (Viscosity of the fuel = 0.30), giving the probability of an event occurring E30 as: (Half the amount of failure) 4. The probability of an event occurring E15 is (Over load = 0.50) caused by (E2 =High fuel temperature, E3 =Work for long periods) as follows: Events Number E1 0 .1 E 2 0.55 E 3 0.89 E4 0.0 5 E5 0.05 E6 0.04 E7 0.10 E8 0.06 E10 0 .05 E11 0.30 E12 0.60 E 13 0 .05 E14 0. 05 E16 0.20 E18 0 .30 E20 0. 20 E21 0.50 E22 0.50 E23 0.01 E24 0.05 E25 0.20 Hanan B. Matar Al-Khwarizmi Engineering Journal, Vol. 17, No. 4, P.P. 23- 35 (2021) 30 The probability of an event occurring E15 (Over load) with probability of an event occurring E16 (Poor maintenance) give the probability of an event occurring E29 as follows: (Small percentage) 5. The probability of an event occurring E28 is (Nozzle head does not pump fuel =0.10) as: (Small percentage) 6. The probability of an event occurring E27 is (Nozzle head wear =0.18) as: (Medium rate of faults) 7. The probability of an event occurring E9 is (Neglect of maintenance manager =0.10) caused by one event (E1 =No maintenance order) as: With the probability of an event, occurring E10 is (Urgent need for operation =0.05), giving the probability of an event occurring E26 as follows: (Medium rate of faults) 8. The probability of an event occurring E25 is (Over Loads =0.20) 9. Calculation the probability of occurrence of the top event E33 is (Diesel engine injector mal functioning =0.80) as follows: (Top event) Fig.5. shows the FTA with the values of the probability indicating the given error event and their occurrence with the calculated probabilities. (quantitative assessment). Hanan B. Matar Al-Khwarizmi Engineering Journal, Vol. 17, No. 4, P.P. 23- 35 (2021) 31 Fig. 5. Fault tree analysis with the values. Hanan B. Matar Al-Khwarizmi Engineering Journal, Vol. 17, No. 4, P.P. 23- 35 (2021) 32 7. Results and Discussion After calculating the failure rate of each cause that lead to injector failure and plotting the FTA with the values as shown in Fig.5., a bar chart is drawn to show the differences between the rates of faults causes and the most influential failures, which is shown in Fig.6. Fig. 6. The faults that occur to the diesel engine injector. The bar chart is the best choice for comparing the calculated data to be represented in a way that shows the data which divided into nominal, numeric, and even symbolic categories. Here, the vertical bar chart is chosen, so that the longer bar is the greater size of the category. In Fig.6., the sum of rates gives a percentage of 98% and not 100%, this means that 2% is still not calculated and represents “other reasons” affecting the diesel engine injector malfunction, which is unknown factors. Looking at the shape of the chart and based on the columns that represent the probability of faults occurrence, we find that the largest is fuel problems, followed by over loads, and then nozzle head corrosion. By analyzing the main reasons for the occurrence of these faults through the FTA several solutions are suggested to reduce these faults. For discussing these results and according to the presentation and calculations, the followings points are highlighted: 1. In the FTA, the focus of the researcher is to identify the main causes firstly and then to identify their sub-causes, with assuming that no error occurs in identifying the causes. 2. For logical consideration, the two gates OR and AND are used for linking in the plotted fault tree. 3. To solve the root causes, it is possible to significantly reduce the percentage of errors that causes the injector to stop working. 4. The bar chart is used to compare the results of the data, showing the largest failure rates according to the length of the vertical bar. 5. It is concluded from the bar chart that the three largest values of the occurrence of a problem with the diesel engine injector are E30= 0.50 (fuel problems), E25= 0.20 (overloads), and E27= 0.18 (nozzle head corrosion). 8. Conclusions and Recommendations Through the calculations, it is found that the main and sub-causes lead to the clarity of the probability of failure by 80%, which was a high percentage of failure. By studying the ratios of the main causes as shown in the bar chart in Fig.6., we have to focus on the most influential causes in order to find the solutions to these causes which significantly reduce the rate of breakdowns. Several plans and ideas have been proposed to solve and reduce these rates: Hanan B. Matar Al-Khwarizmi Engineering Journal, Vol. 17, No. 4, P.P. 23- 35 (2021) 33 1. New maintenance plan is determined in which not only the rate of failures but also cost is reduced. The new maintenance plan depends on two types of maintenance according to the type of malfunction, these are : a. Preventive maintenance (periodic) is done by supervising an engineer with (4-6) workers. b. Corrective Maintenance (emergency) is done by workers with or without the supervising an engineer who performs quick and sudden. Maintenance such as a leak or breakage in a part of the injector that is replaced or repaired immediately to restart the engine; this type of maintenance does not depend on a specific time . 2. The cause E30 = 0.50 (fuel problems) can be reduced by : a. Replacing the fuel filter every certain period of time to reduce the amount of impurities . b. Checking the fuel before filling it into the engine to ensure its purity. 3. The cause E25 = 0.20 (overload), the rate of failures due to overloading increases with high temperatures, high pressures and electrical energy consumption, especially in summer. This rate can be reduced by : a. Supplying water cooling systems to cool the injector . b. Organizing a schedule for operating the motors to avoid power outages due to excessive loads. 4. The cause E27 = 0.18 (nozzle head corrosion) can be reduced by adjusting (O-ring) the nozzle of the cooling system, to prevent water leakage causing rust in the injector head. Notation P (Eo) = occurrence probability of the AND gate output fault event, Eo n = total number of independent input fault events P (Ei) = probability of occurrence of input fault event Ei, for i=1, 2, 3,… ,n. P (Xo) = occurrence probability of the OR gate output fault event, Xo k = total number of independent input fault events P (Xi) = probability of occurrence of input fault event Xi, for i =1, 2, 3,… ,k . 9. References [1] Patrick, D.T. O’Connor and Ander Kleyner, "Practical Reliability Engineering", Fifth Edition, WILEY, A John and sons, Ltd., Publication, this edition first published 2012 in John Wiley & Sons, Ltd, pp. 263. [2] S. Iverson, J. C. Kerkering, and P. Coleman, "Using Fault Tree Analysis to Focus Mine Safety Research ", Spokane Research Laboratory, National Institute for Occupational Safety and Health, Spokane, WA, Sep 21, 2012, pp. 2. [3] Feel-Soon Kang, and Sung-Geun Song, Life- Cycle Expectation Using Fault-Tree Analysis for Improved Hybrid Submodule in HVDC System, Article in Electronics, pp.5, Published in 9 January 2021 https://doi.org/10.3390/electronics10020133. [4] Sohag Kabir, ‘’An overview of Fault Tree Analysis and its application in model based dependability analysis’’, School of Engineering and Computer Science, University of Hull, Hull, HU6 7RX, UK, 1 July 2017, pp. 1. [5] Etika Nur’Aini, Rachmawan Budiarto, Bakti Setiawan, and Alfian Ma’arif, Reliability Analysis and Maintainability for the Design of Grid and Hybrid Solar Power Plant Systems in Wonogiri Regency, Jurnal Teknik Elektro, Vol. 13 No.1, pp. 78, April 2021. [6] Pallavi Sharma1, and Dr. Alok Singh Assistant Professor2, ‘’Overview of Fault Tree Analysis’’, 1. Student Department of Mechanical Engineering Maulana Azad National Institute of Technology Bhopal, India, 2. Department of Mechanical Engineering, Maulana Azad National Institute of Technology, Bhopal, India, 03, March, 2015, pp. 1 & pp. 2. [7] Sohag Kabir, An overview of Fault Tree Analysis and its application in model based dependability analysis, School of Engineering and Computer Science, University of Hull, Hull, HU6 7RX, UK, pp.3, ©2018, Elsevier. This manuscript version is made available under the CC-BY-NC-ND4.0license http://creativecommons.org/licenses/by-nc-nd/4.0/ [8] L. Y. Waghmode and Rajkumar B. Patil,"An Overview Of Fault Tree Analysis (FTA) Method For Reliability Analysis", Department of Mechanical Engineering, Annasaheb Dange College of Engineering and Technology, Ashta, Shivaji University, Kolhapur, Maharashtra (India), March 2013, pp. 3. Hanan B. Matar Al-Khwarizmi Engineering Journal, Vol. 17, No. 4, P.P. 23- 35 (2021) 34 [9] Christine M. Walsh-Kelly, Flow Chart/Process Flow Diagram ,MD,8-26-2015,p.p1,2, http://www.ihi.org/education/IHIOpenSchool/resou rces/Pages/BobLloydWhiteboard.aspx#FC [10] Abbas M. Burhan, "Fault Tree Analysis as a Modern Technique for Investigating Causes of Some Construction Project Problems ", Baghdad University, College of Engineering, June 2010, pp. 4. [11] Dhillon, B.S., "Reliability, Quality, and Safety for Engineers", CRC, Boca Raton, London, New York, Washington, D.C, 2005, pp. 92. [12] Enno Ruijters and Marielle Stoelinga, "Fault Tree Analysis: A survey of the state-of- the-art in modeling", analysis and tools, Formal Methods and Tools, University of Twente, The Netherlands, 2014,pp.3. ) 2021( 23- 35، صفحة 4العدد، 17مجلة الخوارزمي الهندسية المجلد حنان بشار مطر 35    تقييم قابلية الصيانة لمكونات منتج للتخطيط االمثل للصيانة **سوسن صبيح عبد* حنان بشار مطر *** لمى عدنان حميد العراق هندسة االنتاج والمعادن /الجامعة التكنلوجيا/*,**,***قسم pme.19.29@grad.uotechnology.edu.iq*البريد االلكتروني: uotechnology.edu.iq@70004***البريد االلكتروني: luma.a.alkindi@uotechnology.edu.iq***البريد االلكتروني: الخالصة ؛ من المشاكل التي تؤدي إلى التوقف. عدة أسباب تؤدي إلى مثل هذه المشاكل محرك الديزل في أي محطة كهرباء العديد يجب تتضمن صيانة أجزاء لرئيسية التي تؤدي إلى فشل حاقن محرك الديزل كجزء رئيسي من األسباب ا لى تقييم تحليل هذه األسباب وتقييمها من أجل إزالة آثارها. تستند هذه الورقة إ ) األعطال شجرة تحليل أن األعطال. شجرة تحليل باستخدام ، الكهربائية الطاقة في FTAمحطات واسع نطاق على استخداًما االستراتيجيات أكثر هو ( التي تؤدي وليةالقة األحداث األساسية / األة. شجرة األعطال هي تمثيل منطقي لعالتصميم المعقدر أطمن خالل الصناعية إلجراء تحليل موثوقية مجاالت ال ). رئيسيإلى حدث معين غير مرغوب فيه (على سبيل المثال ، الحدث ال ى توقف الحاقن عن العمل من إلي ) وكيف يمكن استخدامه في تحليل أسباب المشكالت الرئيسية التي تؤد FTA(تحليل شجرة االعطال البدء بتقديم تم مشكلة. كل حدوث محرك الديزل في محطة الديزل خالل حساب احتمال لفشل حاقن مبدأ تحليل السبب الجذري تطبيق كحالة في تم اختيار دراسية حديثة من التحليل بان اكبر تم االستنتاج ).٠٫٨٠( متهقيت ، كان الرئيسيحدث للالمؤدية سباب تحليل شجرة االعطال لال. وفقًا لتقييم االحتماالت باستخدام تطبيقية يمكن إلدارة المصنع تحديد لتأكل راس النوزل ٠٫١٨الي و الع للحمل ٠٫٢لمشاكل الوقود، ٠٫٥ت القيم المحتملة السباب العطل لحاقن محرك الديزل كان آثارها لتقليل سبب لكل المحسوبة القيمة االعتبار في األخذ مع محددة التوقأ منخطة ساعات تجنب التشغيلجل بساعات مقارنة الطويلة اقتراح .ف تم توصيات للتغلب على هذه المسببات.