Fault Diagnosis Expert System For Marine Power Plant Based On Network

Mechanical faults usually occurre onboard ocean-going ships, which challenge marine engineers onboard the ship. In the early years, marine engineers had to analyses the symptoms and solve these problems by themselves.Nowadays, with the rapid development of mechanical fault diagnosis technology, marine fault diagnosis expert system has been a very useful and very effective tool for the technical management of marine power plant because it can help marine engineers solve the mechanical faults very quickly and very effectively. The research work and development process for a fault diagnosis expert system for marine power plant based on network is introduced in this paper.Firstly, the fundamental knowledge of fault diagnosis expert system for marine power plant is introduced in chapter one. Diagnosis is a process of locating the exact causes of a failure or a fault. Once a failure has occurred, the maintenance engineer will identify the symptoms, analyze the symptomatic information, interpret the various error messages and indications and come up with the right diagnosis of the situation in terms of which components may have caused the fault and the reasons for the failure of the components. Since a machine may have many components and may be highly complex, diagnosis of a machine fault usually requires technical skill and experience. It also requires extensive understanding of the machine's structure and operation, and some general concepts of diagnosis. Fault diagnosis is important to safe and reliable operation of a machine. Modern marine power plant system is no exception. Moreover, the ability to correctly select the proper maintenance for a given application is also important. To obtain this ability, a certain amount of knowledge is necessary along with experience.Marine power plant is composed of propulsion system, marine auxiliary machinery system, piping system, deck machinery system, automatic equipment, pollution prevent equipment and so on. Modern marine power plants use increasingly complex machines, some with extremely demanding for performance criteria. Attempting to diagnose the faults in these systems is often a very difficult and daunting task for operators and maintainers. Failed machines can lead to economic loss and safety problems due to unexpected and sudden stoppages. These machines need to be monitored during working process to improve machine operation reliability and reduce unavailability. Therefore, effective condition monitoring brings significant benefits to industry. However, condition monitoring requires further fault diagnosis, which is a labor-oriented exercise for marine engineers. Without efficient diagnosis, one is unable to make reliable prediction of failure. A natural progression is the automation of this labor-oriented process of diagnosis by implementing intelligent diagnosis techniques so that marine engineers or technicians can be relieved of this relatively expensive task.Faults of marine power plant can be classified by faults consequence, faults characteristic, faults rule, faults reason and the severity of faults. Fault diagnosis methods can be categorized such as visual inspection method, mathematical model method, fault tree analytical method, neural network method and fuzzy logic method. Artificial intelligence can be taken as intellectualized marine fault diagnosis and signal processing method can be used in marine fault diagnosis together with many kinds of technical fusion comprehensive technologies of marine fault diagnosis. Fault diagnosis expert system along with domestic and international research work will be introduced in this chapter. By comparing to the traditional fault diagnosis method, advantages of using the fault diagnosis expert system is described with the improvement of the shipping service and complexity of regulation enhancement, equipment faults have had a remarkable influence on ships. Further development and its forecast of the fault diagnosis expert system will be discussed. The research work and schedule of this dissertation will be finally concluded in chapter one.Secondly, the principle of expert system will be introduced in chapter two. An expert system is an intelligent computer software program that uses knowledge and inference procedures to solve problems that are difficult enough to require significant human expertise for their solutions, that is, an expert system which imitates human expert's decision-making way to trace fault from imputed symptoms with in-store knowledge. The architecture of an expert system will be presented, which includes knowledge base, inference engine, global database, explanation mechanism, knowledge acquisition module and man-machine interfaces. The most important parts in an expert system are knowledge base and inference engine, which determines the capacity and efficiency of the expert system to solve the diagnosis problems. An expert system can usually be designed to have general characteristics such as high performance, adequate response time, good reliability, understandable, flexibility. It can also list all the reasons for and against a particular hypothesis, list all the hypotheses that may explain the observed evidence, explain all the consequences of a hypothesis, give a prognosis or prediction of what will occur if the hypothesis is true, justify the knowledge of the program, justify the questions that the program asks user for further information and so on. Expert systems have their power from knowledge. The most important property of any expert system is the knowledge it contains and it is the effective use of this knowledge that makes its reasoning successful. So, there is necessity for knowledge representation. To represent knowledge, production rules are commonly used as a knowledge base in expert system because their advantages greatly outweigh their disadvantages.The knowledge base is a symbolic representation of power plant common trouble shooting facts in form of production rules based on Access Database. Relationships between these facts are also included in the knowledge base. The knowledge base is composed of hypothesis rule tables, conclusion rule tables and rule dictionary tables containing codified hypothetical results of rules. Rule dictionary tables contain rules according to different hypothesis code and conclusion code. All the tables are connected together by keywords in Access Database to form the knowledge base. In an expert system, knowledge acquisition process is one of the most difficult phases of expert system building. The process of building an expert system is called knowledge engineering and is done by knowledge engineering. Knowledge engineering refers to the acquisition of knowledge from a human expert or other source and its coding in the expert system. The knowledge acquisition is an optional feature in many systems. The inference engine in an expert system interprets the knowledge stored in the knowledge base, performs logical deduction process. Reasoning method, reasoning direction and searching strategy should be taken into account in the design of inference engine. Advices and recommendations are produced by the inference engine, which operates on the knowledge encoded in the knowledge base. There are various methods of reasoning or inference, which is very important for expert systems because reasoning is the standard technique for expert systems to solve problems. The expert system introduced in this research paper utilized an uncertain reasoning method and a data-driven forward inference engine.Thirdly, expert system tools and design methods are presented in chapter three. The expert system classification model and expert system definition are introduced. The classification model has proved to be an excellent representation for diagnosis problems or interpretation. Several expert system tools have been developed using the classification model. This approach with well understood and classification systems will have many similarities in design. It may be possible to program the expert system as a conventional program. The task of designing a specific application 'model' involves extensive experimentation and modifications. A general approach is necessary to the design of expert system. The general systems for developing expert models can be viewed as experimental tools for rapidly building of a running system.How to build an expert system for a specific problem depends on how to decide to structure of the problem, and how to manage human expert's knowledge into the structure. Firstly one must try to see how general problem is broken down into sub-problems, and how these interact. Then one must decide on a particular choice of representation for expert's knowledge which is suitable for more or less powerful methods of reasoning. The expert knowledge expression is also briefly introduced such as expressing conclusions, expressing observations and expressing reasoning rules. The selection of conclusions and questions is presented for the use of the expert system. The explanation of decisions and the expert system design is finally concluded in this chapter.Fourthly, development of the fault diagnosis expert system with Microsoft Access database and Macromedia Dreamweaver MX programming language is described in chapter four. A database is simply an organized collection of related information stored in a file. Many software packages have the function of creating databases. To create a database in Dreamweaver, the database structure, database relationships, database connections should be taken into account. To create a working database, it's important to understand the database structure. There are different kinds of relationship such as one-to-one relationships, one-to-many relationships and many-to-many relationships. In Dreamweaver MX, no matter what kind of database is used, Dreamweaver should be connected to the database. There are different types of database drivers depending on different kinds of database such as OLE DB connections, ODBC connections, ColdFusion and JDBC Connections. Microsoft Access Database is suitable for Dreamweaver website application. But ODBC connection is necessary with ODBC driver and interaction between database and Dreamweaver should be taken into account.Finally, fault diagnosis expert system software with Microsoft Access database for marine power plant program was debugged. Some programming codes are expressed in chapter five with some demonstrations. Some conclusions for further development of the marine fault diagnosis system are drawn out, together with future works in the last chapter six.