A new swarm intelligence optimized multiclass multi-kernel relevant vector machine: An experimental analysis in failure diagnostics of diesel engines

2017 ◽  
Vol 17 (6) ◽  
pp. 1503-1519 ◽  
Author(s):  
Zhixiong Li ◽  
Yu Jiang ◽  
Zhihe Duan ◽  
Zhongxiao Peng
Keyword(s):  
Diesel Engine ◽  
Swarm Intelligence ◽  
Condition Monitoring ◽  
Manifold Learning ◽  
Diesel Engines ◽  
Learning Algorithms ◽  
Failure Detection ◽  
Kernel Functions ◽  
Weighted Sum ◽  
Vibration Data

This work attempts to introduce a new intelligent method for condition monitoring of diesel engines. Diesel engine is one of the most important power providers for various industrial applications, including automobiles, ships, agricultures, construction, and electrical machinery. Due to harsh working environment, diesel engines are vulnerable to failures. This article addresses a significant need to improve predictive maintenance activities in diesel engines. A new failure diagnostics approach was proposed based on the manifold learning and swarm intelligence optimized multiclass multi-kernel relevant vector machine. Three manifold learning algorithms were first respectively used to fuse the features that extracted from the original vibration data of the diesel engines into a new nonlinear space. The fused features contain the most distinct health information of the engine by discarding redundant features. Then, the swarm intelligence optimized multiclass multi-kernel relevant vector machine was proposed to identify the failures using the fused features. The contribution of this research is that the dragonfly algorithm is employed to optimize the weights of the multi-kernel functions in the multiclass relevant vector machine. It was also applied to establishing a weighted-sum model by combining the outputs of swarm intelligence optimized multiclass multi-kernel relevant vector machine models with different manifold learning algorithms. Robust failure detection of diesel engines is achieved owing to combined strengths of multiple kernel functions and weighted-sum strategy. The effectiveness of the proposed method is demonstrated by experimental vibration data collected from a commercial diesel engine. The failure detection capability of the proposed manifold learning and swarm intelligence optimized multiclass multi-kernel relevant vector machine method for diesel engines will potentially benefit the machine condition monitoring industry by improving budgeting/forecasting and/or enabling just-in-time maintenance.

Tribological failure detection and condition monitoring for diesel engines

Wear ◽  
1989 ◽  
Vol 130 (1) ◽  
pp. 189-201 ◽  
Author(s):  
M.S. Ozogan ◽  
A.I. Khalil ◽  
P.S. Katsoulakos
Keyword(s):  
Condition Monitoring ◽  
Diesel Engines ◽  
Failure Detection

Application of Oil Analysis Techniques to Diesel Engine Condition Monitoring

2012 ◽  
Vol 224 ◽  
pp. 217-220
Author(s):  
Yong Guo Zhang ◽  
Xu Feng Jiang ◽  
Xiao Wen Wu ◽  
Zong Ying
Keyword(s):  
Diesel Engine ◽  
Condition Monitoring ◽  
Diesel Engines ◽  
Lubricating Oil ◽  
Oil Contamination ◽  
Oil Analysis ◽  
Mechanical Wear ◽  
Analysis Techniques ◽  
Wear Condition ◽  
Engine Condition

In order to verify the validity of oil analysis for heavy diesel engine condition monitoring, the lubricating oil were sampled from the lubricating system of the domestic diesel engines, and then were tested by oil analysis (including contamination detection, periodic sampling test and ferrography technology). The results showed that oil analysis could monitor the lubricating oil contamination and mechanical wear condition to make diesel engines avoid early mechanical failure.

scholarly journals Inspection of artificially built mechanical failures through innovative condition monitoring techniques

2018 ◽  
Vol 22 ◽  
pp. 01025
Author(s):  
S.Seçkin EROL ◽  
Cemal MERAN
Keyword(s):  
Condition Monitoring ◽  
Research Work ◽  
Failure Detection ◽  
Monitoring Techniques ◽  
Failure Initiation ◽  
Vibration Data ◽  
Study Results ◽  
Data Acquisition Card ◽  
New Research ◽  
Electrical Consumption

In this study, the factors influence on root causes of failure initiation are examined with the original test implementation and condition monitoring techniques are emphasized. In working toward this goal, the laboratory test setups and tests which have been created by international research bodies are examined and in order to practise a new research work, a unique test setup system and a test plan is built. In this target of the root cause failure detection, vibration data at radial direction and electrical consumption data are collected through the analysis by comparing two different condition monitoring techniques. In this research, fault detection in modeling fault conditions and vibration, electrical consumption measurement have been let us examination in depth. During the tests, data are collected simultaneously in vibration by four-channel Data Acquisition Card (DAC) and electrical consumption by Motor Condition Monitoring (MCM) system which are integrated with an computer system. Respect to the study results; vibration analysis in detection of defects has been judged to be more successful in comparison with electric consumption analysis under the test conditions in perspective of condition based predictive maintenanceIn the study, it is mentioned that detection of failure initiations at sensitive levels and importance of consequent results.

scholarly journals Fault Detection of Diesel Engine Air and after-Treatment Systems with High-Dimensional Data: A Novel Fault-Relevant Feature Selection Method

Processes ◽  
2021 ◽  
Vol 9 (2) ◽  
pp. 259
Author(s):  
Qilan Ran ◽  
Yedong Song ◽  
Wenli Du ◽  
Wei Du ◽  
Xin Peng
Keyword(s):  
Genetic Algorithm ◽  
Diesel Engine ◽  
Fault Detection ◽  
Correlation Analysis ◽  
Canonical Correlation Analysis ◽  
Diesel Engines ◽  
Canonical Correlation ◽  
High Dimensional Data ◽  
High Dimensional ◽  
Relevant Variables

In order to reduce pollutants of the emission from diesel vehicles, complex after-treatment technologies have been proposed, which make the fault detection of diesel engines become increasingly difficult. Thus, this paper proposes a canonical correlation analysis detection method based on fault-relevant variables selected by an elitist genetic algorithm to realize high-dimensional data-driven faults detection of diesel engines. The method proposed establishes a fault detection model by the actual operation data to overcome the limitations of the traditional methods, merely based on benchmark. Moreover, the canonical correlation analysis is used to extract the strong correlation between variables, which constructs the residual vector to realize the fault detection of the diesel engine air and after-treatment system. In particular, the elitist genetic algorithm is used to optimize the fault-relevant variables to reduce detection redundancy, eliminate additional noise interference, and improve the detection rate of the specific fault. The experiments are carried out by implementing the practical state data of a diesel engine, which show the feasibility and efficiency of the proposed approach.

Effect of altitude conditions on combustion and performance of a turbocharged direct-injection diesel engine

2021 ◽  
pp. 095440702110262
Author(s):  
Zhentao Liu ◽  
Jinlong Liu
Keyword(s):  
Diesel Engine ◽  
High Altitude ◽  
Diesel Engines ◽  
Direct Injection ◽  
Pressure Rise ◽  
Ambient Conditions ◽  
Allowable Limit ◽  
Spray Formation ◽  
Direct Injection Diesel Engine ◽  
And Performance

Market globalization necessitates the development of heavy duty diesel engines that can operate at altitudes up to 5000 m without significant performance deterioration. But the current scenario is that existing studies on high altitude effects are still not sufficient or detailed enough to take effective measures. This study applied a single cylinder direct injection diesel engine with simulated boosting pressure to investigate the performance degradation at high altitude, with the aim of adding more knowledge to the literature. Such a research engine was conducted at constant speed and injection strategy but different ambient conditions from sea level to 5000 m in altitude. The results indicated the effects of altitude on engine combustion and performance can be summarized as two aspects. First comes the extended ignition delay at high altitude, which would raise the rate of pressure rise to a point that can exceed the maximum allowable limit and therefore shorten the engine lifespan. The other disadvantage of high-altitude operation is the reduced excess air ratio and gas density inside cylinder. Worsened spray formation and mixture preparation, together with insufficient and late oxidation, would result in reduced engine efficiency, increased emissions, and power loss. The combustion and performance deteriorations were noticeable when the engine was operated above 4000 m in altitude. All these findings support the need for further fundamental investigations of in-cylinder activities of diesel engines working at plateau regions.

scholarly journals Nonlocal Estimation of Manifold Structure

2006 ◽  
Vol 18 (10) ◽  
pp. 2509-2528 ◽  
Author(s):  
Yoshua Bengio ◽  
Martin Monperrus ◽  
Hugo Larochelle
Keyword(s):  
Large Class ◽  
Manifold Learning ◽  
Learning Algorithms ◽  
Tangent Plane ◽  
Training Data ◽  
Kernel Learning ◽  
Manifold Structure ◽  
Prediction Function ◽  
Underlying Manifold ◽  
Shared Structure

We claim and present arguments to the effect that a large class of manifold learning algorithms that are essentially local and can be framed as kernel learning algorithms will suffer from the curse of dimensionality, at the dimension of the true underlying manifold. This observation invites an exploration of nonlocal manifold learning algorithms that attempt to discover shared structure in the tangent planes at different positions. A training criterion for such an algorithm is proposed, and experiments estimating a tangent plane prediction function are presented, showing its advantages with respect to local manifold learning algorithms: it is able to generalize very far from training data (on learning handwritten character image rotations), where local nonparametric methods fail.

scholarly journals MODIFYING MATHEMATICAL MODELS FOR CALCULATING OPERATIONAL CHARACTERISTICS OF DIESEL ENGINES BURNING RME BIOFUELS / MATEMATINIŲ MODELIŲ PRITAIKYMAS DYZELIŲ, DIRBANČIŲ RME BIODEGALAIS, DARBO PROCESUI IR EKSPLOATACINIAMS RODIKLIAMS SKAIČIUOTI / ДОРАБОТКА МАТЕМАТИЧЕСКИХ МОДЕЛЕЙ РАСЧЕТА ПОКАЗАТЕЛЕЙ РАБОТЫ ДИЗЕЛЯ НА БИОДИЗЕЛИНАХ РРМЭ

Transport ◽  
2011 ◽  
Vol 26 (1) ◽  
pp. 50-60 ◽  
Author(s):  
Sergejus Lebedevas ◽  
Galina Lebedeva ◽  
Kristina Bereišienė
Keyword(s):  
Diesel Engine ◽  
Mathematical Modelling ◽  
Mathematical Models ◽  
Diesel Engines ◽  
Fuel Injection ◽  
Internal Heat ◽  
Computer Programs ◽  
Toxic Substances ◽  
Load Range ◽  
Operational Characteristics

The article considers and solves the problems of adapting the mathematical models, used in calculating operational characteristics of diesel engines burning mineral diesel oil, to engines converted to RME biofuels. The analysis of mathematical models of calculating the main technical and economic characteristics of diesel engines as well as the parameters of the in-cylinder process and the concentration of toxic substances in the exhaust gases is performed. The need for adjusting the calculation algorithms is also demonstrated. The computer programs based on single-zone thermodynamic models are used in the research. The programs of mathematical modelling are modified, i.e. supplemented with the algorithm for calculating energy characteristics of the combustion products (e.g. specific heat capacity, internal heat, the lower calorific value, etc.). Based on the computer programs, modified for examining diesel engines burning biofuels, the computer-aided mathematical modelling experiment is carried out. The results of modelling are compared with the data obtained in testing the diesel engine 1A41. The mathematical modelling performed demonstrates the accuracy acceptable for solving practical problems: the difference between the obtained calculation results and diesel engine testing data for the load range of (1.0÷0.5) Pi nom does not exceed ± 5÷7%. Higher accuracy of modelling the characteristics of diesel engines, operating in the low- and medium-load modes, may be accounted for by the adjustment of the algorithm for calculating the induction period and the on-set phase of fuel injection. Santrauka Publikacijoje pateikti dyzelio darbo proceso ir eksploatacinių rodiklių skaičiavimo matematinių modelių tobulinimo rezultatai, siekiant matematinius modelius pritaikyti dyzeliui dirbant riebiųjų rūgščių metilesterių, taip pat RME, biodegalais. Atlikta skaičiavimo algoritmų analizė ir pagrįsti pagrindiniai jų tobulinimo aspektai. Sukurtas kompiuterinis programinis modelis, skirtas darbinio kūno dyzelio cilindre energiniams parametrams skaičiuoti (specifinei šilumai, vidinei energijai), dyzeliui dirbant plačiąja elementinės cheminės sudėties biodegalų gama. Atliktas 1A41 dyzelio techninių ir ekonominių bei ekologinių rodiklių modeliavimas kompiuteriu, naudojant patobulintus matematinius modelius. Parodytas gautų rezultatų, pakankamų praktiniams uždaviniams spręsti, tikslumas: dyzeliui dirbant artimos nominaliosios apkrovos režimais (1,0÷0,5) Pi nom skirtumas tarp skaičiavimo ir dyzelio motorinių tyrimų rezultatų neviršija ±5÷7%. Dyzeliui dirbant mažos apkrovos režimu, matematinio modeliavimo rezultatų adekvatumo didinimas susijęs su kuro įpurškimo fazės bei savaiminio užsiliepsnojimo indukcijos periodo skaičiavimo algoritmo tobulinimu. Резюме Рассмотрены и решены задачи адаптации математических моделей расчета показателей дизеля при переводе их работы с минерального дизелина на биодизелины метилового эфира рапсового масла (МЭРМ). Выполнен анализ и дано обоснование направлений доработки расчетных алгоритмов ряда однозонных термодинамических математических моделей, используемых в практике исследования эксплуатационных показателей, параметров индикаторного процесса и эмиссии токсичных компонентов в выхлопных газах дизелей. В форме программного модуля составлен алгоритм расчета энергетических параметров рабочего тела в цилиндре (удельной теплоемкости, теплоты сгорания, внутренней энергии) для топлив с широким элементарным химическим составом. С использованием доработанных компьютерных программ выполнен широкий расчетный эксперимент и сопоставлены его результаты с данными моторных стендовых испытаний одноцилиндрового отсека дизеля 1А41. Показана приемлемая для решения практических задач точность математического моделирования: для нагрузочных режимов (1÷0,5) Pmi расхождения расчетных и экспериментальных значений параметров работы дизеля не превышают ±5÷7%. Повышение качества моделирования показателей дизеля на режимах средних и малых нагрузок связано с уточнением алгоритма расчета периода индукции и фазы начала видимого горения топлива в цилиндре.

scholarly journals Fault severity detection of a worm gearbox based on several feature extraction methods through a developed condition monitoring system

2021 ◽  
Vol 3 (1) ◽  
Author(s):  
Berkan Hızarcı ◽  
Rafet Can Ümütlü ◽  
Zeki Kıral ◽  
Hasan Öztürk
Keyword(s):  
Fault Detection ◽  
Condition Monitoring ◽  
Extraction Methods ◽  
Poincaré Plot ◽  
Vibration Data ◽  
Sliding Motion ◽  
Fault Severity ◽  
Fault Features ◽  
Gearbox Vibration ◽  
Condition Monitoring System

AbstractThis study presents the severity detection of pitting faults on worm gearbox through the assessment of fault features extracted from the gearbox vibration data. Fault severity assessment on worm gearbox is conducted by the developed condition monitoring instrument with observing not only traditional but also multidisciplinary features. It is well known that the sliding motion between the worm gear and wheel gear causes difficulties about fault detection on worm gearboxes. Therefore, continuous monitoring and observation of different types of fault features are very important, especially for worm gearboxes. Therefore, in this study, time-domain statistics, the features of evaluated vibration analysis method and Poincaré plot are examined for fault severity detection on worm gearbox. The most reliable features for fault detection on worm gearbox are determined via the parallel coordinate plot. The abnormality detection during worm gearbox operation with the developed system is performed successfully by means of a decision tree.

Characteristics of injection of diesel fuel, rapeseed oil and their mixtures in diesel engines of various types

2021 ◽  
pp. 167-178
Author(s):  
Keyword(s):  
Diesel Engine ◽  
Diesel Fuel ◽  
Diesel Engines ◽  
Rapeseed Oil ◽  
High Speed ◽  
Fuel Injection ◽  
Experimental Studies ◽  
Fuel Injector ◽  
High Speed Diesel ◽  
Engine Type

The necessity of adapting diesel engines to work on vegetable oils is justified. The possibility of using rapeseed oil and its mixtures with petroleum diesel fuel as motor fuels is considered. Experimental studies of fuel injection of small high-speed diesel engine type MD-6 (1 Ch 8,0/7,5)when using diesel oil and rapeseed oil and computational studies of auto-tractor diesel engine type D-245.12 (1 ChN 11/12,5), working on blends of petroleum diesel fuel and rapeseed oil. When switching autotractor diesel engine from diesel fuel to rapeseed oil in the full-fuel mode, the mass cycle fuel supply increased by 12 %, and in the small-size high-speed diesel engine – by about 27 %. From the point of view of the flow of the working process of these diesel engines, changes in other parameters of the fuel injection process are less significant. Keywords diesel engine; petroleum diesel fuel; vegetable oil; rapeseed oil; high pressure fuel pump; fuel injector; sprayer

STUDYING OF THE EFFECT OF WATER-FUEL MIXTURE ON OPERATIONAL PARAMETERS OF A DIESEL

2021 ◽  
Vol 2 (143) ◽  
pp. 54-61
Author(s):  
Sergey V. Borisov ◽  
◽  
Aleksandr E. Lomovskikh ◽  
Oleg E. Prilepin ◽  
Timur R. Mamatkazin ◽  
...  
Keyword(s):  
Diesel Engine ◽  
Diesel Fuel ◽  
Diesel Engines ◽  
Aqueous Dispersion ◽  
Fuel Mixture ◽  
Research Purpose ◽  
Regression Equations ◽  
Operational Parameters ◽  
Rough Water ◽  
Minimum Number

Improving the parameters of diesel engines is an urgent task. Work has been carried out to significantly reduce the consumption of their fuel with the introduction of water dispersions into the fuel. Currently, water-fuel emulsions with exotic emulsifiers are mainly tested. (Research purpose) The research purpose is in creation of a water-fuel emulsion without an emulsifier with a simple installation and identifying the influence of the composition and quality of this WFE on the performance of the YaMZ-236 diesel engine. (Materials and methods) The article presents a plant for the preparation of a "rough" water- fuel mixture from diesel fuel according to GOST 32511-2013 and distilled water according to GOST 6709. Authors conducted standard bench tests at the KI-5540- GOSNITI stand with a YaMZ-236 diesel engine with an upgraded fuel system and performed the control of the smoke content of the exhaust gases with the gas analyzer "AUTOTEST". The dependence of diesel performance indicators on the composition and dispersion of water-fuel emulsions without an emulsifier was studied experimentally with a minimum number of tests, but with the maximum possible combination of the values of three variable factors. (Results and discussion) The influence of various water-fuel emulsions on the performance of the diesel engine was evaluated according to the plan of a full factor experiment, including 20 tests. The second-order regression equations were obtained by mathematical processing of the test results. The feasibility of using water-fuel emulsions for diesel engines was confirmed. By modeling a water-fuel mixture without emulsifiers, there was created an aqueous dispersion with drops up to two micrometers. In the load tests of the diesel engine with it, there was noticed an improvement in its performance. (Conclusions) The introduction of 17-20 percent water dispersion with drops of up to two micrometers into diesel fuel reduced the specific fuel consumption by 18 percent, the smokiness in the K indicator by 20- 22, and in the N indicator by 30-35 percent.

Sign in / Sign up

Export Citation Format

Share Document