Graphic Art

6 pages
7 views

A Knowledge Based System towards Identifying Problems and Getting Instant Solutions about Our Personal Vehicles

of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
A Knowledge Based System towards Identifying Problems and Getting Instant Solutions about Our Personal Vehicles
Transcript
  Shah Murtaza Rashid Al Masud   , International Journal of Advanced Trends in Computer Science and Engineering, 2(2), March – April 2013, 15 - 20 15@ 2012, IJATCSE All Rights Reserved     ABSTRACT Identifying our personal vehicle's problems and gettingsolutions is a complicated process which demands high levelof knowledge and skills. Our goal is to develop a knowledge-based system (KBS) on vehicle conservation and troubleshooting that is capable of supporting vehicle’s owner in dealing with their vehicle problems and troubleshootingthem whenever the time is limit and the human expert, alsoknown as mechanics is not available at that very point of time.This paper provides effective planning and design issuesregarding the problems while driving the vehicle and will givea logical solution which would help in rectifying those problems. The system would contain various set of rules for detecting and distinguishing different category of failureswhich can be easily handled by the driver and will give their causes. This paper mainly deals with the starting problems of vehicle and detecting various other large scale problems. Theresearch report explains the need for a knowledge based system and some issues on developing knowledge-based systems, the vehicle failure recognition process and the problems involved in developing the system. A rule-based,web enabled expert system shell: expertise2go was used todesign the rules which involved a knowledge component,decision component, design component, graphical user interface component, and the user component. The system has been verified and tested and gave auspicious results. Key words: Knowledge-Based System, Vehicle, ExpertSystem, Rules. 1.   INTRODUCTION The terms knowledge-based system (KBS) and expert systemare normally used identically. The four main components of KBS are: a knowledge base, an inference engine, a knowledgeengineering tool, and a specific user interface. Some of KBSimportant applications include the following: medicaltreatment, engineering failure analysis, decision support,knowledge representation, climate forecasting, decisionmaking and learning, and chemical process controlling.Knowledge-based system (KBS) is a branch of artificialintelligence (AI) which was developed by the AI communityin the mid-1960s. It can be defined as "Knowledge-based System (KBS) - Typically a rule-based system for providingexpertise. Identical to expert systems, except that the sourceof expertise may include documented knowledge [1] [2]". Wecan infer from this definition that expertise can be transferred from a human to a computer and then stored in the computer in a suitable form that users can call upon the computer for specific advice as needed. Then the system can makeinferences and arrive at a specific conclusion to give advicesand explains, if necessary, the logic behind the advice. KBS provide powerful and flexible means for obtaining solutionsto a variety of problems that often cannot be dealt with byother, more traditional and orthodox methods. Normally thissystem will guide a human technician through the entireservice process, from the initial customer interview at theservice desk to the diagnosis and repair of the vehicle in thegarage.A knowledge-based system for vehicle failure diagnosis is presented in this paper. The main objectives towardsdeveloping such a system are as follows:    To help the vehicle’s owner in making decision onaction they should take in their attempt to troubleshootthe vehicle’s problem.    To save much time, cost and energy for the customer especially to those who are in urgency as they don’t haveto wait for the mechanics to repair the troubles of their vehicles.    To get the peoples even more closely the world of computerization, technology, mechanization, and automation.    To give the developer a chance to get involve directlyinto the development of knowledge based systemsynonymously expert system, and at the same time helpto build the programming and software design skill, aswell as problem explaining and solving capabilities.As the proposed system is web-based, so at anytime from any place any user can access to it and get recommendations heneeded.This paper is organized in this way, Section 2 Problemidentification. Section 3 Analytical model for the proposed KBS system. Section 4 Design of the proposed KBS systemSection 5 Implementation of the proposed KBS system, and finally Section 6 Conclusion. A Knowledge Based System towards Identifying Problems and Getting InstantSolutions about Our Personal Vehicles Shah Murtaza Rashid Al Masud Faculty of Computer Science and Information Systems Najran University, Najran, Saudi Arabiasmramasud@gmail.com   ISSN No. 2278-3091 Volume 2, No.2, March - April 2013International Journal of Advanced Trends in Computer Science and EngineeringAvailable Online at http://warse.org/pdfs/2013/ijatcse03222013.pdf     Shah Murtaza Rashid Al Masud   , International Journal of Advanced Trends in Computer Science and Engineering, 2(2), March – April 2013, 15 - 20 16@ 2012, IJATCSE All Rights Reserved  2.   PROBLEM   IDENTIFICATION   Vehicle failures can be divided into three major types: start-upstate, run-stable state, and movement state. In our proposed system we took consideration only the start-up state, problemsthat may occur when a person try to start up the vehicle, for example; engine does not work, some sounds noticed, engineworks ones and stops. These problems could be due to one or more failures; will happen, the battery needs to be recharged,the dynamo is dead, or the battery is dead. If the vehicle is inthe start-up state and doesn't start, then the motive could beone of three main reasons: a bad gasoline mix, deficiency of compression or absence of spark. In addition, thousands of minor things can create problems, but these are the main three.Bad gasoline mix: A bad gasoline mix can occur in severalways:1. The vehicle ran out of gas, so the engine is getting air but nofuel.2. The air intake might be clogged, so there is fuel but notenough air.3. The gasoline system might be supplying too plentiful or toodiminutive gasoline to the mix, meaning that ignition does notoccur properly.4. There might be an contamination in the fuel (like water inyour gas tank) that makes the fuel not burn.Deficiency of compression : If the air and gasoline cannot becompacted properly, the ignition process will not work like itshould. Lack of compression might occur for these reasons:1. Piston rings are worn2. The intake or exhaust valves are not sticking properly3. There is a hole in the cylinder.Absence of spark  : There are a number of reasons for what thespark might be weak.1. If the spark plug or the wire leading to it is damaged or worn out.2. If the wire is cut or missing3. If the system that sends a spark down the wire is notworking properly4. The fuel will not burn properly if the spark occurs indifferent cycle.Other Problems:   The following problems are also taken intoconsideration in the system.1. If the battery is dead, the engine cannot turn over.2. If the bearings that allow the crankshaft to turn freely areworn out, the crankshaft cannot turn so the engine cannot run.3. If the valves do not open and close at the right time or at all,air cannot get in and exhaust cannot get out, so the enginecannot run.4. If someone sticks a potato up your tailpipe, exhaust cannotexit the cylinder so the engine will not run.5. If you run out of oil, the piston cannot move up and downfreely in the cylinder, and the engine will seize.The proposed KBS with facts and rules would easily identifythe vehicle’s problems and give us instant solutions. 3.   ANALYTICAL   MODEL   FOR   THE   PROPOSED   KBS   SYSTEM   The knowledge-based system (KBS) developed for this work depends on the structure of expert system. The structure of expert system has five main components: user interface,working memory, inference engine, knowledge base, and explanatory facility [3][4]. The major components of KBS asshown in Figure 1 are briefly explained below.Knowledge base - declarative representation of the expertise,often in IF THEN rules;Working storage - the data which is specific to a problem being solved;Inference engine - the code at the core of the system whichderives recommendations from the knowledge base and  problem-specific data in working storage;User interface - the code that controls the dialog between theuser and the system.Explanation Facility-Illustrates to the user how and why thesystem gave a certain cause for the failure, i.e. explains thereasoning of the system to the user.In this proposed expert system we used forward chaining(reasoning) method where an interpreter uses a set of rules and a set of facts to perform an action [5]. This method involveschecking the condition part of a rule to determine whether it istrue or false. If the condition is true, then the action part of therule is also true. This procedure continues until a solution isfound or a dead end is reached. Forward chaining iscommonly referred to as data-driven reasoning. This method composed of three components. These are: The rule set; aworking storage area which contains the current state of thesystem; an inference engine which knows how to apply therules.The rules are of the form: left hand side (LHS) ==> right hand side (RHS), in short: IF---THEN. The LHS is a collection of conditions which must be matched in working storage for therule to be executed. The RHS contains the actions to be takenif the LHS conditions are met.The execution cycle of the system is:Select a rule whose left hand side conditions match the currentstate as stored in the working storage.Execute the right hand side of that rule, thus somehowchanging the current state.The syntax of the rules is:rule <rule id>:[<N> : <condition>,.......]==>[<action>, ....].Where: rule id - a unique identifier for the rule; N – optionalidentification for the condition; condition - a pattern to matchagainst working storage; action - an action to take. The Figure3 mentioned below represents the concept of the forward chaining algorithm.  Shah Murtaza Rashid Al Masud   , International Journal of Advanced Trends in Computer Science and Engineering, 2(2), March – April 2013, 15 - 20 17@ 2012, IJATCSE All Rights Reserved  Figure 1: Main components of proposed KBS Figure 3: The forward chaining algorithm model for the proposed system.   4.   DESIGN   OF   THE   PROPOSED   KBS   SYSTEM   The Knowledge Engineer acquires knowledge from experts inspecific field, through standardization process, and sorts outthe collected knowledge from experts to formed rules. Invehicles’ problems identification system, the way of knowledge representation is production rule. The productionrule uses the presentation way of “IF P THEN Q”, P is precondition, and Q is conclusion [6]. Some of the rules of the proposed KBS are:Rule 1: IF the vehicle does not start,THEN check for the fuel.Rule 2: IF the vehicle does not start,AND the fuel is not empty,THEN check the battery.Rule 3: IF the vehicle does not start,AND the battery does not have a problem,THEN check whether the engine is overheated or is very cold.The proposed knowledge-based system is developed bydecision table or knowledge base table. Where the condition part of the decision table is called the condition stub and theaction part the action stub shown in Figure 3 as mentioned  below. Each column represents a rule, and when all of theconditions in a column are satisfied, the action or actionsdesignated in that column represent the recommended decision [6]. Rule1 Rule2 Rule3Condition1Condition2Action1Action2Figure 3: A decision table.   The connection between decision Tables and rules in arule-based expert system is easily seen in the followingdecision table. "Y" and "N" in the condition stub represent"Yes" and "No" ("T" and "F" could be used representing"True" and "False") and an "X" in the action stub means theaction is recommended. "-" in a cell means the condition or action is irrelevant, shown in Table 1 [7]. Using the conceptsgetting from the Table 1 we developed a new table ‘Table 2’,which shows the combination between decision table rules for our proposed system. A decision table towards vehicles’ problems identification is shown in Table 3. Table 1: Combination between decision table rules in rule based KBS Rule1 Rule2Condition1 Y Y Condition2 - - Condition3 Y Y Action1 X Action2 X Action3 Table 2: Combination between decision table rules in KBSfor our proposed system Rule1 Rule2 Rule3 Does the starter run? Y Y NDo you smell gas when thestarter cranks?Y N -replace the dead battery - - Xrefuel the vehicle - X -wait 10 minutes, restartflooded vehicleX - - Table 3: A decision table towards vehicles’ problems identification Rule1 Rule2 Rule3 the result of trying thestarter the vehiclecranksnormallythe vehiclecranksnormallynothinghappensthe smell of present while not present -  Shah Murtaza Rashid Al Masud   , International Journal of Advanced Trends in Computer Science and Engineering, 2(2), March – April 2013, 15 - 20 18@ 2012, IJATCSE All Rights Reserved gasoline trying thestarter while tryingthe starter therecommended actionwait 10minutes, restartflooded vehiclerefuel thevehiclerechargeor replacethe batteryAlthough either decision table form is easily mapped intoknowledge-based system rules, the extended entry form is probably most effective for this purpose as the following rulesextracted from the three columns of the table illustrated inTable 4 as shown bellow. Table 4: Extracted format of a decision table towards vehicles’ problems identification Rule1 Rule2 Rule3 If the result of trying the starter =the vehicle cranksnormally and the smell of gasoline = presentwhile trying thestarter Then therecommended action = wait 10minutes, restartflooded vehicleIf the result of tryingthe starter = thevehicle cranksnormally and the smell of gasoline =not present whiletrying the starter Then therecommended action= refuel the vehicleIf the result of trying thestarter =nothinghappensThen therecommended action =recharge or replace the batteryTherefore, the three rules are equivalent to the followingsingle rule that could be generated for a knowledge-based system:IF Condition1=true OR Condition2=true OR Condition3=trueTHEN Action1=true 5.   IMPLEMENTATION   OF   THE   PROPOSED   KBS   SYSTEM   To implement this KBS, web-enabled, knowledge-based e2gRuleWriter decision table software and Java applet-based e2gRuleEngine Knowledge-based shell that provides aninteractive user interface are used. All of these abovementioned software are free for private, educational, and commercial use [6].The proposed web-enabled KBS utilizes a combination of qualitative and quantitative procedures for our vehicles’ problems identification. The mixture of both providesreliability to the technical process and facilitates knowledgeinteraction and analysis. The proposed KBS for vehicles’ problems identification has been successfully implemented and validated. The results show that basic decision making procedures are very satisfactory. KBS starts by displaying itsmain window as shown in Figure 4. The user is expected toclick the button to choose the option for getting the solution of his vehicle’s problems. Figure 4: Index page for our proposed KBS Click the button labeled  "Start the consultation" as shownin Figure 4 the knowledge base will be loaded. We should seethe first PROMPT as shown in Figure 5. Figure 5: The first PROMPT appeared after clicking the "Start theconsultation"   Click the ‘ Why ask?’ button to see the inference engine'sexplanation as shown in Figure 6 of why this question has been asked. Figure 6: The inference engine's explanation when we click on the button ‘ Why ask? ’ The following figures: Figure 7, Figure 8, Figure 9, and Figure 10 will show how we can input values to get instantsolutions about our vehicles problems. Figure 7: Process of input (1) value in our KBS Figure 8: Process of input (2) value in our KBS  Shah Murtaza Rashid Al Masud   , International Journal of Advanced Trends in Computer Science and Engineering, 2(2), March – April 2013, 15 - 20 19@ 2012, IJATCSE All Rights Reserved  Figure 9: Process of input (3) value in our KBS Figure 10: Process of input (4) value in our KBS   The result or action taken by the system is shown in the followingFigure 11. Figure 11: The result we obtained after giving the input values to our KBS If we click the Explain button with all conclusions showingin the drop down list of resolved attribute values we will seethe following detailed explanation of the recommendation for the rule 5 from our research as shown in Figure 12. Figure 12: The explanation procedure of the proposed KBS The following Figure 13 shows how the knowledge base is produced and kept place on knowledge base table by using theknowledge base software e2gRuleWriter.jar. The rule writer software is compiled and managed by another softwarenamely e2gRuleApplet.jar [7].Each GOAL is defined on a single line that begins withGOAL followed by the name of the goal attribute enclosed insquare brackets. The following figures (Figure14-16) showthe development process of knowledge based system.  Condition mode: The appearance of the condition mode dataentry screen after entering data describing all the conditions asshown on the following Figure 14. Action mode: Inputs that are accepted when definingconditions but not used by actions is shown in Figure 15. Defining rules: The final knowledge base construction stepinvolves entering condition and action values to form the rulesas shown in Figure 16, then generating the e2gRuleEngineknowledge base. Figure 13: Production of knowledge base for the proposed KBS byusing e2gRuleWriter shell software. Figure 14: The conditions for the proposed system  
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x