CC BY
12
УДК 629.4.069
Sandag Galbadrakh1, Jamyan-Osor Purevsuren2, Natsag Odgerel1
1 Ulaanbaatar Railway Institute, Ulaanbaatar, Mongolia;
2Mongolian University of science and technology, Ulaanbaatar, Mongolia
DEFINITION OF DIAGNOSTIC PARAMETERS FOR LOCOMOTIVE DIESEL ENGINES
Abstract. The paper aims to determine the diagnostic parameters for monitoring the technical condition of diesel engine 16ChN26/26 of diesel locomotive 2TE116Um used in the transportation of UB Railway JVC in Mongolia by the concentration (g/ton) of wear elements in used oil. The study and experiments were analyzed based on the results of measurements in used oil of 20 diesel engines during a period of maintenance within 5 - 6 years. The study was carried out in three main directions: to identify the concentration of wear elements and contaminants in used oil by spectral analysis, to study the wear degree of engine components in one period of maintenance, and to determine the diagnostic parameters of wear based on the concentration of wear elements and contaminants using the non-parametric statistical method. By using the diagnostic parameters determined as a result of the above study, it can be possible to perform the diesel engine maintenance based on the actual technical condition rather than the maintenance interval recommended by the manufacturer and increase the efficiency of the diesel engine maintenance system.
The diagnostic evaluation parameters established as a result of this study will enable technicians to perform diesel engine maintenance based on the actual technical condition of the engine, rather than relying solely on the manufacturer's recommended maintenance schedule. This approach has the potential to increase the efficiency of diesel engine maintenance and reduce the cost of repairs by addressing issues before they become significant. Ultimately, the proposed methodology could help to extend the lifespan of engines and improve their overall performance, resulting in significant benefits for the transportation and heavy-machinery industries.
Keywords: spectral analysis, maintenance interval, wear elements, diagnostic parameters, diesel engine.
Сандаг Г.1, Жамъян-Осор П.2, Нацаг О.1
1Улан-Баторский железнодорожный институт (УБЖ/дИ), г. Улан-Батор, Монголия;
^Монгольский государственный университет науки и технологии (МГУНТ), г. Улан-Батор, Монголия
ОПРЕДЕЛЕНИЕ ДИАГНОСТИЧЕСКИХ ПАРАМЕТРОВ ДЛЯ ДИЗЕЛЬНЫХ ДВИГАТЕЛЕЙ ЛОКОМОТИВОВ
Аннотация. В данной работе представлены результаты исследования для определения параметров технического состояния дизеля 16ЧН26/26 тепловоза 2ТЭ116Ум, эксплуатируемого на перевозках АО «Улан-Баторская железная дорога (УБЖД)», по концентрации элементов износа (г/т) в отработанном масле. Исследования и соответствующие тесты проводились в течение одного полного периода периодичности ремонта 20 дизелей, т. е. на протяжении пяти - шести лет. При проведении исследования рассматривались три основные аспекта: определение статистических показателей износа и концентрации загрязняющих элементов в отработанном масле методом спектрального анализа; анализ износа деталей двигателя за один период межремонтного цикла; использование статистического, а не параметрического метода для определения показателей диагностики износа. Использование вновь установленных диагностических критериев позволит поддержать пересмотр графика технического обслуживания дизельных двигателей исходя из их реального технического состояния.
Параметры диагностической оценки, установленные в результате этого исследования, позволят специалистам выполнять техническое обслуживание дизельного двигателя на основе фактического технического состояния двигателя, а не полагаться исключительно на рекомендованный производителем график технического обслуживания. Такой подход сможет повысить эффективность технического обслуживания дизельных двигателей и снизить затраты на ремонт путем решения проблем до того, как они станут значительными. В конечном счете предлагаемая методология сможет помочь продлить срок службы двигателей и улучшить их общую производительность, что приведет к значительным выгодам для транспортной и тяжелой промышленности.
Ключевые слова: спектральный анализ, периодичность ремонта, критерий диагностики, дизельный двигатель, отработанное масло, элементы износа.
Introduction. The intervals /frequencies/ of routine maintenance of planned preventive maintenance system is recommended by the manufacturer, taking into account the standard parameters /values/ for the reliable operation of a diesel engine and the conditions for performing the maintenance at the lowest cost [1]. Although this maintenance system has the advantage of improving the technical inspection /checking/ and ensuring the reliability performance, it also creates wasteful costs such as mandatory disassembly to inspect components and regular planned maintenance without taking into account the real condition of machinery.
The latest application in locomotive maintenance in the world is gradually moving to the non-dismantle methods for diagnostics on the technical condition of components and the condition-based maintenance system which allows for the reduction of wasteful costs. The International Organization for Standardization (ISO) specifies guidance to machinery condition monitoring and diagnostics as follows (table 1).
Table 1 - ISO standards for condition monitoring and diagnostics of machines
Machinery monitoring ISO reference
Vibration Thermography Acoustic emission and ultrasound Tribology-based monitoring and lubricant analysis ISO 13373-1:2002; 13373-2:2005; 16587; 18436-2 ISO 18434-1:2008; 18436-7 ISO 22906:2007; 29821-1:2011; 18436-6 ISO 14830-1; 18436-5
Tribology in the tribology and lubricant analysis category analyzes wear degrees, while lubricant analysis determines the machinery technical condition by analyzing the state of the lubricant. Lubrication condition monitoring (LCM) has become an important factor in decision-making on engine maintenance systems and it can be seen from the increase in the number of research papers published in the last five years [9]. This monitoring consists of analysis of physical-chemical parameters in used oil of an engine, oil additives, contaminants and wear elements [5]. In particular, it is possible to get a lot of information from the accumulation of wear particles in used oil, such as the technical condition of the engine and the normal operation of the subsystems [2; 4; 6].
The manufacturers of diesel engines recommend following the operation instructions for use, which provide the physical-chemical and other parameters of lubricating oils. They have also developed standards for diagnosing the technical condition of an engine and determining the remaining useful life of components without disassembly using the parameters for the normal, abnormal, and critical limits of concentration of wear elements and contaminants in used oil during engine operation.
The parameters in GOST20759 standard are followed for determining the qualitative and quantitative variables (concentration) of wear elements in used oil of 16ChH26/26 diesel engine by spectral analysis method, diagnosis before damages and failures and estimation of wear degree and remaining useful life of components (table 2).
Table 2 - Limits of wear elements concentration for diesel engine diagnosis
Element Symbol Concentration Components
Normal Abnormal Critical
Iron Fe <65 65-100 >100 Cylinder, Rings, Crankshafts, Gears, Rust
Lead Pb <15 15-20 >20 Bearings, babbitt
Aluminium Al <20 20-30 >30 Bearings camshaft
Copper Cu <50 50-100 >100 Bronze bushing
Chrome Cr <10 10-15 >15 Rings
Tin Sn <5 5-10 >10 Piston skirt
■-M^RFrTHq Транссиба з
The diagnostic parameters of elements concentration given in the standard takes into account many factors, such as natural and climatic conditions, adjusted power of the engine, quality of technical servicing, schedule of oil change, addition of new oil [8;10].
One. Principles on determination of diagnostic parameters. The study aims to determine the diagnostic parameters for assessing the technical state of components by the concentration of wear elements, K present in the used oil of an engine. To achieve it, it is necessary to determine the critical value KQ of the concentration of each wear element [3]. In other words, if K > K0, the engine is not allowed to be used and it needs to be repaired, but if K < K0, it is allowed to continue using it. Let's denote: D± the normal state of machinery, D2- the faulty state of machinery.
Then, if the decision rule or the diagnostic parameter is K < K0, the state D± is accepted or valid, and if K > K0, the state D2 is accepted or valid. The functions of concentration density (figure 1) corresponding to D± as a normal state and D2 as a faulty state are described as follows.
A faulty solution in diagnosis refers to a solution that does not correspond to the actual state of the diagnostic object. If, according to the diagnostic results, the state of the object is diagnosed as «faulty» (D2), but according to the results of the examination, the state is assessed as «normal», then the false alarm or its probability is a first-type error, while the solution is «normal» and in reality it is «faulty», its probability is called the second type error respectively [2].
M./O)'1
Figure 1 - The density function of diagnostic parameters in normal and faulty states
As mentioned above, Ctand C2 costs due to false alarms and errors not to detect the failure during technical diagnostics will occur. The cost estimation on an error is often not clear. While it is very difficult to determine it correctly, it is important to reduce the concerns to other errors as much as possible when any error is at a certain (acceptable) level.
According to the Neyman-Pearson approach, when the probability of a false alarm is at a given acceptable level, the probability of missing an error without detection is minimal.
<X>
Pt I f(K/D1)dK < A. (1)
Where: A - a given and acceptable level of the probability of false alarm; Pt - the probability in a state of without faults.
The states given in (1) refers to the probability of a false alarm condition (no Pi multiplier). It can be seen from Figure 1 that when an error of false alarm increases (section, K0 has moved to the left), an error of missing a fault decreases.
<X>
Pt I f(K/D1)dK = A. (2)
Its minimum concern corresponds to the state of having the equal sign of the state given in (1) [2].
Two. Experimental result. The 16ChH26/26 diesel engine of a diesel locomotive was selected as an object of the study. The engine was produced by JSC "Kolomensky Zavod" in Russian Federation with 16 cylinders and a power of 2650 kW (figure 2).
The operation instruction recommends that M14G2TsS motor oil should be changed between
every 50-100 thousand km, and M14D2 motor oil should be changed between every 75-150 thousand km, depending on the physical and chemical characteristics. According to the study of the oil consumption of the locomotive depot, about 70 % of the total oil was changed when it reached the maximum useful life of locomotives, about 20 % was changed when the oil was mixed with fuel and coolant, and about 10 % was changed in case of failure in the physical and chemical characteristics of the oil.
2.1 Analysis on wear and wear rate in components of a diesel engine in one cycle of maintenance. When the technical state of a diesel engine is diagnosed without disassembly with its lubricating oil, it is necessary to examine carefully whether the wear and wear intensity of the same type of components are different or not from each other. The fact that some components wear out faster than others (too different intensity of wear) is the condition for making wrong decisions on diagnostics.
Therefore, in accordance with the rule of our traditional scheduled preventive maintenance, the engine was disassembled, a number of observations /measurements/ were made through the routine maintenance (RM-2, RM-3), a statistical analysis was performed on the wear and its intensity (table 3), and the distribution law was predicted (table 4).
Table 3 - The statistical parameters of wear in diesel engine components
Figure 2 - 16XH26/26 diesel engine of diesel locomotive
95 % confidence
Detal N Average Standard deviation Standard interval of the mean Minimum value Maximum value Variance of interference
error Less More
Piston Pins 128 0,025 0,013 0,0015 0,02 0,027 0,000 0,080
Bronze bushing 128 0,131 0,043 0,003 0,12 0,138 0,075 0,260
Crank Main journals 80 0,023 0,008 0,0005 0,05 0,055 0,040 0,070
Crank pin journals 64 0,019 0,009 0,0007 0,04 0,050 0,030 0,060
Cylinder liner 128 0,032 0,023 0,0015 0,02 0,035 0,000 0,105
Connecting rod bronze bushing 64 0,111 0,020 0,0014 0,10 0,114 0,070 0,145
Pin connecting rod 34 0,032 0,014 0,0012 0,03 0,035 0,005 0,080
Total 1190 0,071 0,056 0,0014 0,06 0,073 0,000 0,330
Model No random effect 0,022 0,0005 0,06 0,072
Random effect 0,0187 0,02 0,114 0,003
Table 4 - The result of Kolmogorov-Smirnov and Shapiro-Wilk tests
Components Kolmogorov-Smirnova Shapiro-Wilk
Statistics Degree of freedom Negative probability Statistics Degree of freedom Negative probability
Piston Pins 0,108 128 0,001 0,949 128 0,000
Piston bronze bushing 0,144 128 0,000 0,902 128 0,000
Crank Main journals 0,171 80 0,000 0,898 80 0,000
Crank pin journals 0,228 64 0,000 0,843 64 0,000
Cylinder liner 0,109 128 0,000 0,948 128 0,000
Connecting rod bronze bushing 0,156 64 0,000 0,936 64 0,000
Pin connecting rod 0,123 34 0,000 0,957 34 0,000
According to the statistics of the study on determining wear rate of diesel engine components, 0,05 - 0,2 mm of wear occurred in 95 % of components between every RM-2 maintenance, and 0,2 - 0,24 mm in 95 % of all types of bearings. Compared to the wear rate required to replace the components recommended in the rule for routine maintenance, there is a residual useful life of 0,7 - 0,85 percent.
When the wear values of the components are analyzed with the Kolmogorov-Smirnov and Shapiro-Wilk tests, it was determined that there is a normal distribution with a probability of 0,95. The normal distribution of the wear rate in the components shows that the wear and its intensity of all components are equal.
2.2 Diagnostic parameters. The experimental study was carried out in the oil of 20 engines of 10 diesel locomotives of 2TE116Um series through 319 tests during every technical maintenance TM, routine maintenance RM and every oil changes and in any other necessary cases.
The analysis on the used oil of an engine was carried according to the ASTM D6595 standard in the laboratory of Techonomics Mongolia LLC, a branch company of the Australian Techenomics International Company in Mongolia, which is accredited by the MNS/IES 17025 standard.
A model [7] was developed that can determine the amount of wear metals of components by the concentration of wear elements in engine oil. During the development of this model, it was assumed that the speed of the wear elements entering the oil and the filtration parameters are constant.
Diesel engine oil was completely changed 3-5 times or every 75,000 km on average depending on whether it can meet the physical and chemical specifications between routine maintenance of 16ChH26/26 engine (300,000 km). During the operation of the diesel locomotive, the total mileage (running) for complete change of oil was divided into periods I-IV, and the amount of the concentration of wear elements (g/ton) and its statistical parameters were determined by the spectral analysis in used oil (Table 5; Figure 3).
Table 5 - The statistical parameters of the concentration of wear elements in used oil in periods I-IV of locomotive mileage for the complete oil change
I II III IV
Sym Count Aver dis Count Aver dis Count Aver dis Count Aver Dis
Fe 33,0 20,1 11,38 36,0 30,5 17,70 25,0 34,4 15,94 16,0 32,3 13,37
Pb 33,0 1,1 1,24 36,0 1,8 2,23 25,0 2,1 1,08 16,0 2,2 0,98
Al 33,0 3,8 1,79 36,0 4,9 2,48 25,0 5,4 2,48 16,0 5,2 1,60
Cu 33,0 2,0 1,29 36,0 3,7 2,29 25,0 4,3 3,17 16,0 6,3 4,63
Cr 33,0 1,0 0,82 36,0 1,7 0,99 25,0 1,9 1,44 16,0 2,5 1,38
■ в! i
1 11 H Ph IV
.'1 =
It til N
1 II ill IV I ■ HI IV
Figure 3 - The changes in the concentration of wear elements in used oil in periods I-IV of locomotive mileage
for complete oil change; g/ton
A histogram of the concentration (g/ton) of wear elements such as iron, copper, aluminum, chromium and lead in oil was created (Figures 4 - 5). The probability distribution function K/Dt can
№ 1(53 2023
be used because it was normal or undamaged when the engine was disassembled to check the technical state through the routine maintenance.
0 20 60 80 0 5 10 15
Figure 4 - A Histogram of iron Figure 5 - A Histogram of aluminum
concentration concentration
During the observation period, based on the statistical data on the concentration of wear elements in used oil of an engine, when the acceptable limit value of the diagnostic parameters in normal state without faults is determined by model 2 with confidence between 0,9 - 0,975 using the non parametric statistical method, the following results were obtained (Table 6).
Table 6 - The diagnostic parameters of analysis
Wear elements; g/ton Probability in confidence interval
0,9 0,925 0,95 0,975
Iron 41 43 50 58
Copper 7 8 9 9
Aluminum 5 6 7 10
Chrome 3 3 3 4
Lead 3 3 3 4
The probability that the diagnostic parameters corresponding to the probability in confidence interval of 0,95 is simultaneously fulfilled by all parameters was estimated by the relative frequency approach.
p = no = 214 = (3)
n 312 v 7
where n0- number (value) obtained by all parameters; n - random number (value). The number of all parameters n0 is defined as
n0 = I?=11(klt < 50.35) • I(k2l < 7.035) • I(k3l < 8.67) • I(k4i < 3.25) • I(k5l < 3.1), (4)
(1 if К < К ■
'j. J, . j?'
0, IJ A s A0,
where klt - iron concentration in the measurement; k2i - copper concentration in the measurement; x3i -aluminum concentration in the measurement; x4i - chrome concentration in the measurement; k5i - lead concentration in the measurement;
ki, k2, k3, kA, k5 - the parameters (values) of the concentration of wear elements are transformed by a linear transformation of
Y-^ — tt^^fc^ + + ^13^3 + ^14^4 + ^15^5»
— a2lkl + ^22^2 + a23^3 + a24^4 + ^25^5»
— 0-3lkl + ^32^2 + a33^3 + a34^4 + a35^5; (5)
— + 0^42^2 + ^43^3 + ^44^4 + ^45^5»
— a5lki + ^52^2 + a53^3 + a54^4 +
I
№202533) И ИЗВЕСТИЯ Транссиба
and as a result of the analysis on the principal component using SPSS program, the first main component with the largest dispersion is Yt = 0,881 k1 + 0,875k2 + 0,751 k3 + 0,783x4 + 0,410x5, and the dispersion is 57,7 percent of the total dispersion and the second largest component in the remaining dispersion is Y2 = -0,210^ - 0,158fc2 + 0,424fc3 - 0,435fc4 + 0,843fc5, and the dispersion accounts for 22,96 percent of the total dispersion. The first and second components account for 85,323 % of the total dispersion, which means that they can be considered representative of the entire sample.
Therefore, with the help of the first and second components, the analytical range with a probability of 0,05 of the first type error is can be created as Y1 < 60,02; Y2 < 1,004.
2.3 The result of the diagnostic parameters into the maintenance systems. The attempts to regulate the maintenance interval recommended by the manufacturer of a diesel locomotive according to the mileage of a locomotive using the increase in the concentration of elements in used oil and the diagnostic parameters determined in the study were successful. For example, according to the operation instruction of the manufacturer, the routine maintenance-2, which involves partial disassembly of a diesel engine, is recommended every 300,000 km, and the routine maintenance-3, which is completely disassembled after taking out from the diesel locomotive, every 600,000 km. It was possible to increase the mileage of a locomotive for routine maintenance-2, if the wear element concentration is not detected in the used oil up to the maximum value of the diagnostic parameters that we determined. This case occurred in 30 percent of all locomotives and reduced maintenance costs by 20 percent per million kilometers of mileage.
Conclusions.
1. The quantitative and qualitative amounts of concentration in wear elements in used oil of a diesel engine were determined by the spectral analysis through the every technical maintenance (every 15000 - 20000 km) and the statistical analysis was performed.
2. It was determined that the wear intensity of diesel engine components and the wear of the same type of components has a normal distribution through the routine maintenance-2 or every 300000 km.
3. In accordance with the operation condition in Mongolia, when the engine is operated at a reduced power of 10 percent, the parameters for diagnostics without disassembly of components were determined by the spectral analysis method. The quantitative method has been analyzed by principal component method.
References
1. Gorsky A.V., Vorobyov A.A. Optimization of the locomotive repair system. Moscow, Transport Publ., 1994 (In Russian).
2. Chetvergov V.A., Ovcharenko S.M., Bukhteev V.F. Technical diagnostics of locomotives. Moscow, Educational and methodological center for education in railway transport Publ., 2014, 371 p. (In Russian).
3. Birger I.A. Technical diagnostics. Moscow, Mechanical engineering Publ., 1978, 240 p. (In Russian).
4. Yuegang Zhao. Oil Analysis Handbook. Spectro Scientific, Inc., 2017, 111 p. (In Chinese).
5. National Standard ASTM D5185 Standard Test Method for Determination of Additive Elements, Wear Metals, and Contaminants in Used Lubricating Oils and Determination of Selected Elements in Base Oils by Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-AES). USA.
6. Fernandez-Feal M.C., Fernandez-Feal M.L. Study of Metal Concentration in Lubricating Oil with Predictive Purposes. British Journal of Applied Science & Technology, Past, 2018.
7. Ovcharenko S.M., Minakov V.A. Modeling accumulation of products depreciation in engine
oil of diesel D49. Izvestiia Transsiba - Journal of Transsib Railway Studies, 2014, no. 3 (19), pp. 31-36 (In Russian).
8. Organization for Cooperation of Railways (OSJD). Recommendations for the introduction of a diagnostic system for controlling the state of diesel locomotives and diesel-trains by analyzing oil results. Infrastructure and rolling stock committee OSJD, Warsaw, P647-1.2009 (In Russian).
9. James M., Wakiru, Liliane Pintelon. A review on lubricant condition monitoring information analysis for maintenance decision support. Mechanical systems and signal processing, vol. 118, 2018.
10. Macian V., Tormos B., Olmeda P., Montoro L. Analytical approach to wear rate determination for internal combustion engine condition monitoring based on oil analysis. Tribology International, 36, 2003, pp. 771-776.
ИНФОРМАЦИЯ ОБ АВТОРАХ
Сандаг Галбадрах
Улан-Баторский Железнодорожный институт (УБЖ/дИ).
Мира пр., д. 44, район Баянгол, г. Улан-Батор, 210635, Монголия.
Магистр технических наук, старший преподаватель кафедры «Подвижной состав», УБЖ/дИ.
Тел.: (976)-91-009193.
E-mail: [email protected]
Жамъян-Осор Пурэвсурэн
Монгольский государственный университет науки и технологии (МГУНТ).
Бага Тойруу ул., д. 34, Сухэ-Баторский район, г. Улан-Батор, 14191, Монголия.
Кандидат физико-математических наук, профессор математической статистики кафедры «Отделение бизнес-администрировния», МГУНТ.
Тел.: (976)-99-281475.
E-mail: [email protected]
INFORMATION ABOUT THE AUTHORS
Sandag Galbadrakh
Ulaanbaatar Institute of Railways.
Peace av., 44, Bayangol District, Ulaanbaatar, 210635, Mongolia.
Master of Science in Engineering, senior lecturer of the department «Rolling Stock», Ulaanbaatar Institute of Railways.
Phone: (976)-91-009193.
E-mail: [email protected]
Jamyan-Osor Purevsuren
Mongolian University of science and technology (MUST).
34, Baga Roundabout, Sukhbaatar District, Ulaanbaatar, 14191, Mongolia.
Ph. D. in Mathematical Statistics, professor of the department «Branch of Business Administration», MUST.
Phone: (976)-99-281475.
E-mail: [email protected]
Нацаг Одгэрэл
Улан-Баторский железнодорожный институт (УБЖ/дИ).
Мира пр., д. 44, район Баянгол, г. Улан-Батор, 210635, Монголия.
Магистр лингвистики, старший преподаватель кафедры «Социально-гуманитарные науки», УБЖ/дИ.
Тел.: (976)-88-177187. E-mail: [email protected]
Natsag Odgerel
Ulaanbaatar Institute of Railways.
Peace av., 44, Bayangol District, Ulaanbaatar, 210635, Mongolia.
Master of Linguistics, senior lecturer of the department «Social sciences and Humanities», Ulaanbaatar Institute of Railways. Phone: (976)-88-177187. E-mail: [email protected]
БИБЛИОГРАФИЧЕСКОЕ ОПИСАНИЕ СТАТЬИ BIBLIOGRAPHIC DESCRIPTION
Сандаг, Г. Определение диагностических параметров для дизельных двигателей локомотивов / Г. Сандаг, П. Жамъян-Осор, О. Нацаг. - Текст : непосредственный // Известия Транссиба. - 2023. -№ 1 (53). - С. 2 - 9 (на англ. языке).
Sandag G., Jamyan-Osor P., Natsag O. Definition of diagnostic parameters for locomotive diesel engines. Journal of Transsib Railway Studies, 2023, no. 1 (53), pp. 2-9 (In English).
