| Marine engineering maintenance is an ancient learning mechanism, which was established by the thermal dynamic machines used for the propulsion of ship. It was first used as a labor workmanship and then technical workmanship. In the last several decades, it developed into a curriculum called marine engineering curriculum. According to this dissertation, marine engineering maintenance work is getting a change over from experience to technical science to be able to cope with the advanced equipments' requirements. We can therefore foresee a farewell to periodical maintenance in the 21 century[1].Ship's engine faults are not caused by a short period but by a series latent reasons —the trend of performance parameters decrease[2]. This paper focus on a severe engine cylinder liners crack problem for the MAN B&W S50MC single action, uni-flow scavenging, direct driving low speed diesel engine installed on container vessel after 5 years service. The studied vessel was built and completed maiden voyage in March 1994. The shipping company supplied 8 years information of the subject vessel from launch till now (Table5-1). The information included main engine service data, engine department abstract log, engine maintenance report, bunkering report, lubrication oil and cooling fresh water analysis report etc. We deeply appreciated their full cooperation and kind help.This paper will mainly use fuzzy parameter method to record the important machine parameters. First, we can solve the degree of membership functions of fault fuzzy parameters. Then, the weight of the fuzzy membership function can be obtained by the Analysis Hierarchy Process (AHP) [3] with Matlab software. Finally, the result of synthetic fuzzy membership function of each cylinder can be used to find out the engine real faults and time relationship. In addition, such model analysis is suitable for the real ship and others sister ships maintenance programs. All the analysis results can be applied to the modern ship design, operation and maintenance procedures. Furthermore, this approach will provide a fundamental for the shipping reliability maintenance this century.Through the application of high technical advancement in the shipping industry, engine troubles and maintenance frequency are decreased sharply, but both the downtime and maintenance costs soars rapidly. The dilemma of conventional measures of maintenance and new equipment operating are more outstanding. In such a situation, both engine factories and marine engineers will pay more attention not only to enhance research and development in engine quality and performance, but also to increase the maneuvering and operational ability to further reduce labor and downtime costs[4].The requirement of the technical state of low speed marine diesel engine in new century, not only sets on engine parts and systems' original quality, but also depends on the working course of the engine[5]. During the process of maneuvering, through the changes in the engine performance condition, such as the temperature and pressure, the engineer can be able to predict engine troubles before engine breakdown. Then he can prolong the engine parts and safety service interval, finally improve the reliability of the engine. There are final goals of my dissertation.Chapter 1 express a brief introduction of marine maintenance progress in the history, the study and results of the research..Chapter 2 explains the low speed diesel engine improvement and new century maintenance technique further progress tendency.Chapter 3 shows integrate domestic and foreign marine diesel engine maintenance works in the present and future development, and collection of several different kinds of periodical engine performance and navigation report to choose useful data for computer analysis.Chapter 4 illustrates marine diesel engine major parts including cylinder cover, piston and cylinder liner born by thermal and mechanical stress in the design requirement, and then mentions the low speed marine diesel engine performance parameters for the effectiveness of engine.Chapter 5 is the main part of the dissertation, through the performance parameters using statistic measures for faults diagnosis analysis.Statistic analysis deals with the collection, analysis, interpretation and presentation of masses of numerical data. Several important statistical parameters, such as mean and variance, measured from observation data, are used to describe the center location and the variation of sample data.The estimations of mean and the coefficient of variation are described in the following:Assume a set of data{xt,;t=l,2,....N} consist of a part of observations, called samples. The mean value of the set of data {xt}, could be calculated by using the sample mean:Where Y represents the parameters' deviation, in some situations, we need a measure of relative variation, such as the coefficient of variance, which expresses the standard deviation as a percentage of what being measured. The coefficient of variation is defined as:) xMlMean and the coefficient of variation may represent the center point, variation and the deviation from peak value of probability density function of the system response time[6].The propose method compares the deviation degrees of really running parameters with normal parameters by using an experience judge approach. We compare the percentage of deviation of running parameters of two sister ships with same type. One is in good working condition without any engine fault and the other encounters engine serious fault caused cylinder liner continue cracked in only 2 years after 4 years operation.Chapter 6 also the main part of the dissertation, through the performance parameters using fuzzy measures for faults diagnosis analysis. Fuzzy faults diagnosis analysis:(l)Fuzzy membership function: Collecting expertise experience method. Expertise method is collecting several actually experienced engineer knowledge, according to the really working experience for the engine fault and parameters relationship, directly important parameter degree of fuzzy membership^], setting 1 for normal operating parameter, and zero are for the fault fuzzy membership, the range of scope are [0,1] region. The typical formula and fuzzy membership degree are as formula and drawing as follows:0,-........x<297,x>396*-297,297<*<33033 >3396-X,363<*<39633 1,.........330 |
PDF Full Text Request