.

Keywords- Expert system (ES), Artificial Intelligence (AI), Car Failure Diagnosis System (CNDS), Knowledge-Based System (KBS), Inference engine.

Expert system components

Structure of the car noise diagnosis system

A. User Interface

User interface goes about as the primary correspondence unit for

the master framework, from which were client can ask questions and

framework can create solutions. To assemble the user interface a PC

programming dialect or an instrument is needed. The master

arrangement of auto commotion determination utilizes the C-

programming dialect to fabricate the framework. The user interface is

represented as a menu which displays the questions to the user and the

user answers with Yes or No.

B. Explanation Facility

Illustrates to the user how and why the system gave a certain

cause for the failure, i.e. explains the reasoning of the system to the

user.

C. Knowledge Base

The knowledge of the system is collected from mechanic

experts, specialized books, and from different car websites. The

knowledge base contains about 150 production rules for different types

of car failures and causes. Knowledge base in the expert system acts

as a storage unit which stores the data given as input.

Advance Car Noise Diagnosis System (CNDS) ProcessThe CNDS is a system with multi-tasking capabilities that is

directive for car fault detection. It Provides information sources,

knowledge interaction and evaluation regarding the technical

process to be tackled. Such tasks are failure tracing, failure

recognition, and customer advising that requires it to adopt various

types of modules. Each module’s goal is achieved by applying a

number of processes. Figure shows the system components in a

process level. One of the main characteristics of the CNDS is the

addition of the database to be in parallel with the system’s

knowledge base. An interaction is held between the knowledge base

and the database including information transfer about the processes

under execution. The database works as a temporary storage due to

its continuous change in state. The knowledge base of the system

encompasses both the derived and the heuristic knowledge about

the required car failure diagnostic process. The inference engine is

built to do the reasoning based on several algorithmic procedures

that enable it to reach the required conclusion.

Fig. Advance CFDS processes

Conclusion

In this Poster, Expert system (ES), Artificial Intelligence

(AI), Car Noise Diagnosis System (CNDS), Knowledge-Based

System (KBS), Inference engine are presented. The proposed

CNDS is utilized to assist inexperienced mechanics or

individuals who face sudden failure in the car and provide

decision support system. In addition, CNDS is considered as an

interactive training tool that can provide expert guidance in car

fault detection. Using this system, Fault diagnosis in technical

systems has done a lot of theoretical and practical attention.

Diagnosis is a complex reasoning activity, which is currently one of

the domains where Artificial Intelligence techniques have been

successfully applied as these techniques use association rules,

reasoning and decision making processes as would the human brain

in solving diagnostic problems.

Reference• http://www.rebuilt-auto-engines.com

• http://www.speeding.co.uk/acatalog/engine.html

• Youtube; https://www.youtube.com/watch?v=RKjdnQ0sHQc

• http://www.writing.engr.psu.edu/posters.html

Sanjay Kumar, Sumanth, Surya kiran, Tulasiram, Venkata Surya Kumar, Viral kumar, (Group-6) Presents Poster on

‘An Expert System for Car Noise Diagnosis’Instructor

Ileana Costea, Ph.D. MSE MSE614 SP2015

D. Inference Engine

When abnormal situation arises in the car, the KBS makes

inferences by deciding which rules are satisfied by facts stored in the

working memory and executes the rule with highest priority and

propose proper correcting solution. The rules whose patterns are

satisfied by facts in the working memory are stored in the agenda part

of the inference engine. The inference engine traces the applicable

facts that represent the corresponding symptoms (i.e. the observed

Variables or behavior) and apply them to the tagged rules. In the

proposed Car Noise Diagnosis System (CNDS), production-rules is the

chosen form for knowledge representation. Additionally, the inference

engine is the forward-chaining type (data driven mode) as shown in

Figure. Forward-chaining inference engine has the advantages of

producing solutions for the unformulated problems. In this type of

engine, the satisfactory facts that meet the rules conditions are

promoted to be triggered. All the applicable rules resulting from the

previous process make up the conflict set of rules to be implemented

according to specific order that enables solution generation.

Introduction

An Expert System is an artificial intelligence program that has

expert level knowledge about a particular domain and knows how

to use its knowledge to respond properly.

The Expert System (ES) is one of the leading

Artificial Intelligence (AI) techniques that have been adopted to

handle car failure detection. This paper presents the imperatives

for an ES in developing car failure detection model and the

requirements of constructing successful Knowledge-Based

Systems (KBS) for such model. Expert system on car noise

diagnosis system has the list of deficiencies in its database which

develops the relationship, standardized mean square slip, and

formant frequencies values between knowledge base and the user

quires to draw the inferences which will give the best possible

solution. Thus, systems can be highly useful in individuals and

mechanics for failure detection and training purposes. Moreover,

capturing and retaining valuable knowledge on such domain yield

more accurate and less time consuming models

Car Noise Diagnosis Expert System

Make the selection based on the noise heard

1. Abnormal Noise.

2.Chattering.

3.Clunk.

X. No idea of the noise.

Enter your selection according to the car state;

Working Memory

The noise is from engine

compartment

While starting engine

The problem is faulty starter drive

Production Rules

Rule 1

Rule 2

Rule 3

Rule 4

Fig- The production system after Rule 1 is fired

Rule 1: IF the noise come when starting engine from the engine compartment,

THEN the problem is Faulty starter drive and inspect starter drive.

Rule 2: IF the noise come from under the vehicle while changing gears,

THEN the problem is damaged U-Joint and inspect U-Joint.

Rule 3: IF the noise come from tire or wheel or wheel area,

THEN the problem is Brake Pad/Shoe contaminated with grease or brake

fluid and inspect brake pad/Shoe front.

Rule 4: IF the noise come from front of vehicle while accelerating,

THEN the problem is worn or damaged CV half shaft and inspect CV half-

shaft.

Fig-Four assumed rules in agenda