In this research paper, I am seeking to analyze more about advanced Programmable logic accountants. Today it is difficult to conceive of a industry without PLC and other automatic accountants. As the production becomes more and more efficient, the accountant work faster and the system go more complex. Artificial intelligence ( AI ) techniques cut down the complexness and they are used through PLC-based procedure control system. The working of unreal intelligence consist of -diagnostic, cognition, expert and the construction of an AI system. Function such as AI mistake nosologies in procedure aid in commanding and successfully foretelling the results based on occupant cognition.
Here I will be researching more about the applications of PLCs such as usage of PLC s with fuzzed logic. Basic fuzzed logic and besides its cardinal constructs will be analysed. Use of the fuzzed logic accountant in practical applications include supplying existent clip logical control systems. In the terminal I will be reasoning on how advanced PLC are more efficient than the conventional PLCs.
Keywords: Programmable Logic Controller, Artificial Intelligence, Fuzzy Logic Controller
I. Introduction
In industry usage of automatic accountants is increased, the usage of PLC ‘s. The programmable logic accountants are based upon the on/off logic, in PLC ‘s we use usually closed or usually unfastened switch, and these exchange can turn on or turn off the devices. PLC ‘s contain a little processing unit, memory, input and end product interfaces. But the PLC ‘s are non able to stand for the all informations of the procedure and they are non able to take action to take the mistakes. But with the aid of the Artificial Intelligence we can do the system to do the determination on mistakes. Artificial Intelligence is the subdivision of computing machine scientific discipline. In AI we use the information from the procedure to work out the many mistakes in the industry. All the information used by the AI is gathered from the individual working in the plant/machine and log book besides provide information about mistake. The information about the fault-what sort of mistake that was and how that was solved. Once all the information we gather from the procedure, all information is stored in the memory of the PLC or elsewhere will be used by the PLC to work out the complex jobs of the procedure.
II. ARTIFICIAL INTELLIGENCE IN A PLC
A. Three type of AI system
The categorization of AI system is really hard because they are used in the many applications but nevertheless we can sort in three types: 1.diagnostic 2.knowledge 3.expert.All three type of AI system have similar features. The system go more and more sophisticated as the size of the information base additions and the extent how the procedure information is used.
1 ) Diagnostic AI system.
This type of system is the mistake observing systems. They detect the mistake in the application and they do non work out the job. For illustration if the temperature of a armored combat vehicle is decreased the diagnostic system can name the mistake by reading the thermocouple values. These system use the cognition to make on a mistake decision, these type of system are used in the applications that use a little information base and cognition.
2 ) Knowledge AI system
An cognition AI system is the enhanced diagnostic system. These type of systems are able to observe the mistake and procedure behaviors based on the cognition and they are besides able to take the determinations refering the procedure and/or the possible cause of a mistake.
3 ) Expert AI system
This type of system comes on the first place in AI applications. Adept systems are more capablenesss than the cognition system. The adept system provides a farther capableness for analyzing procedure informations with the aid of statistical analysis and the system predict results of the procedure that are based on present procedure appraisals. The result computation may be a determination and with the aid of that determination procedure maintain the end product in malice of a mistake sensing. The cognition used in the expert AI systems are more complex than in the
other AI systems ; these type of system generate more feedback information. The adept systems besides require more refined package programming to do determination, since their determination trees involve more options and properties.
The execution of the adept systems is merely done by with the aid of excess scheduling and they besides need more hardware. The system use the transducers to do the determination in the procedure and the entire figure of transducer used in this system is more than the other system. Programmable logic accountant use the AI system, it will necessitate two or more than two processer to do the all scheduling for the system. PLC system necessitate more velocity to run in existent clip, the system should be fast. The system has big informations to run in the existent clip due to big informations system besides need big memory to hive away that informations.
B. Artificial Intelligence System Architecture
The block diagram ( Figure 1 ) shows the basic architecture of an AI system. It has three primary elements: 1 Global database, 2 Knowledge database, 3 Inference engine. The block diagram show that expert block foremost, that block provide the cognition to the AI system and the cognition is received from a individual who know about the plant/process, how the machine execute their operation. The adept sends the all information ( about system care, mistakes ) to the cognition applied scientist. The procedure of conveying the cognition and assemblage informations is known as cognition acquisition.
1 ) Global Database
Global database incorporate the information about the procedure and the system, how to command them. The information contained by the planetary database is about the input and end product informations flow from the procedure. The planetary database is the storage country, the information about the procedure stored. The information stored in planetary database can be used any clip to do the AI determination to command the procedure. PLC have memories to hive away the informations and the Global database resides in the memory of the system that makes the system to take the AI determinations. We can besides utilize the AI system with computing machine and the Global database will be in the difficult disc of the computing machine.
2 ) Knowledge Database
The cognition database store the information as the planetary database shop about the procedure and the all information is supplied from the expert. It besides contains information about the mistakes, procedure, causes of the jobs and their solutions as good. Furthermore, all the regulations that help to do the determination are besides stored in the cognition database. The diagnostic system has knowledge database and that is less complex than the cognition system. The cognition system is less complicated than the adept system. It stored in the system memory.
3 ) Inference Engine
All the AI system has inference engine. All the determinations are made in the illation engine. Inference engine use the cognition database to do the determination about the procedure and after that illation engine execute the regulations in the procedure. It besides uses the historical information of the procedure to do the existent clip determination. PLC system contain the cardinal processing unit, CPU perform all the operation for the system and the illation engine may be inside the CPU or it may non be indoors informations that depends upon the diagnostic, cognition, expert.
C. Knowledge representation
In the cognition representation all the AI schemes are organised and the cognition applied scientist represent the input of the expert. The cognition database is used for the storage of the representation. The cognition from the expert is changed in the signifier of regulations ( IF and THEN/ELSE ) and we call it rule-based cognition representation. It make the system capable to take action and determination.
A PLC system is used with AI, all the control schemes are executed by package plans. Whenever a mistake is detected by the system and at that clip system makes a determination, illation engine besides use the cognition representation. The determination will be in the signifier of package.
D. Rule-based cognition representation
It uses the cognition from the expert and do the determination with the aid of that cognition. The regulations contain two parts, first portion ancestor ( IF something happens ) and the 2nd portion consequent ( THEN take this action ) . All the regulations are made for the procedure and they can be complex.
A simple rule-based System may do a simple diagnostic regulation, such as: IF the temperature of a armored combat vehicle is less than the set point, THEN turn on the warmer. A more complex diagnostic expression may incorporate regulations that farther depend on a more complex diagnostic expression and they involve the regulations that are depend on Parent regulations:
IF instance 1, THEN & A ; frac34 ; ELSE nil
& A ; frac12 ;
IF instance 2, THEN & A ; frac34 ; ELSE something
& A ; frac12 ;
IF instance 3 & A ; frac34 ; THEN nil
The determination tree makes the system capable to take the determinations. The figure 2 shows how the determination tree works to acquire a determination on the given procedure informations.
E. Knowledge illation
This is the method used to pull decision by garnering the information. When the system execute the chief control scheme at that clip the illation of cognition takes topographic point in the illation engine. The cognition illation is besides takes topographic point in cognition database when calculation of regulation is traveling on. In the little control system the cognition illation takes topographic point on local footing. But in instance of big systems cognition illation takes topographic point in the hierarchal system. To plan a AI based PLC we need hardware, the demand of hardware depends on the engagement of the AI. In all AI systems some common methods of regulations are used for the execution of cognition illation. Methods are: 1. Forward Chaining, 2.Backward Chaining. Forward chaining-This method is used to happen out the results of a given informations and receives the information from the planetary database. Forward Chaining is done by two methods: deepness foremost, comprehensiveness foremost hunt. Backward chaining-This is besides similar to the forward chaining. Basically it is used to happen the ancestors.
F. Basic Architecture of an AI based PLC
Large and complex distributed control systems are made by the combination of little systems. They can pass on with each other either straight or with the aid of local country web. The AI is added in the big systems, planetary database, cognition database and cognition illation from these is distributed all over the system. These big systems are made by the combination of little system and all the local system has their ain local informations base and cognition database. The PLCs in the diagram shows they perform illation engine calculations. In the big systems the supervisory PLC usage the all subsystem and their local database to do a complex determination. The chief computing machine we call it blackboard keep the all information from the little units. The chief computing machine applies all the complex AI solution.
III. FUZZY LOGIC IN A PLC
In industrial mechanization Programmable Logic Controller combine the simpleness, and dependability. Fuzzy logic is a portion of AI which deals with concluding used to copy human determination devising and thought in machines. The logical thinking is transformed in algorithms. These algorithms are used when the informations can non be converted in the binary signifier. The end product of the procedure is the input for the fuzzed accountant. Fuzzy logic performs three chief actions. First is fuzzification, in this action the information received at the input is converted in the fuzzed signifier. Second is fuzzed processing, which involves the transmutation of the input informations harmonizing to IF & A ; hellip ; THEN regulations formed by the user at the clip of design and scheduling of the fuzzed control systems. After completing the fuzzy processing ( rule-processing phase ) the fuzzed accountant reach an result. Third is defuzzification procedure, this is concluding measure of the fuzzed accountant. In this measure the concluding end product informations is converted into the existent end product informations and after that the information is sent to the procedure with the aid of end product interface. The fuzzed logic accountant is placed in the PLC rack in this instance the accountant does non hold a direct contact with the procedure, the fuzzed logic accountant will direct the defuzzification informations in the PLC memory location and PLC send that informations to the procedure by the interface faculty. In the most of the fuzzed logic accountant have their independent interface ports and they are besides connected to the PLC with the aid of the stopper. The fuzzed accountant can pass on with the procedure through the PLC & A ; lsquo ; s input/out ports. PLC can be interfaced with the intelligent fuzzy accountants.
A. Interface of Fuzzy logic with PLC
A Fuzzy logic accountant ‘s input interface can read the informations from the 8 devices and it can convey the informations to 4 end product devices with the aid of the end product interface. This interface is able to execute 128 regulations, each regulation can hold maximal IF conditions and the action will be in the signifier of two THEN. The fuzzed logic accountant has capableness to execute all its calculations in merely in 6 millisecond if fuzzed logic unit works individually of the processer, as a consequence it supplying fast operation of fuzzed logic control.
B. Fuzzy Logic and I/O Communication
In below given Table 1, the Fuzzy Logic Unit ( FLU ) uses the programmable accountant ‘s memory to hive away the control parametric quantities and fuzzed logic accountant uses 10 words or registries. The place of the FLU faculty in the rack tells about the registries ‘ references. Assuming that the place of the FLU faculty takes the references 110 through 119, the usage of the references by the FLC faculty as follows:
- The first four spots ( 0-3 ) the first word is ( word 110 ) and its first four spots ( 0-3 ) enclose, in BCD, the FLU faculty uses as the figure of inputs. 15 figure spot turns on the fuzzed processing of this word.
- The 2nd word ( word 111 ) specifies that the location of the input informations stored in the PLC ‘s memory. It tells the get downing registry reference.
Table I
inputs: spots 0-3 of word 110 specify the figure of inputs to be read
( 8 soap ) ( e.g. , I = 8 )
Word 111: starting reference where input informations is located ( length of I )
( e.g. , reference = 120 )
End products: spots 0-3 of word 112 specify the figure of end products to be
written ( 4 soap ) ( e.g. , O = 4 )
Word 113: starting reference where end product informations is located ( length of O )
( e.g. , reference = 130 )
Word 114: used for flags and scenes
Wordss 115-119: available as working word references
- As the first word, the 3rd word is ( word112 ) and the first four spots enclose the end products in BCD.
- The 4th word ( word 113 ) store the reference where the end product informations is stored, the end product informations is obtained by the fuzzed logic calculations.
Because fuzzed logic accountant work with the other I/O interfaces, their input/output informations must be send to the I/O faculties working with them. Figure 4 shows how the memory references ( words ) used by the Fuzzy Logic Controller and it besides shows the location of the input and end product informations harmonizing to the input/output devices.
We can besides utilize the block transportation direction to reassign the informations between FLU and input/output interfaces ( Figure 5 ) .
IV CONCLUSION
When we apply unreal techniques to a system, we need to add hardware every bit good as package to in the system. The plan that system needed is depending upon the mistake in the system, the mistake sensing is complex so the plan will be more complex. We design a system that besides has intelligence ; this is possible by adding the information from the procedure. The information should be about the procedure sing the last clip mistake and what type of mistake that was, how that was solved and when was the last care performed. The add-on of unreal intelligence and fuzzed logic accountant in a PLC make the system faster and the system will be able to take determination about the procedure. The system will be better than the conventional PLC ‘s.
V REFRENCES
- Bikash Pal, Balarko Chaudhuri, Robust control in power systems, Spinger, 2005.
- Fuzzy Logic Toolbox User ‘s Guide, The MathWorks, Inc.,2008.
- PLC-5 Programmable Controlers, Rockwell Automation, USA, 2007.
- C. P. Chuang, X. Lan, J. C. Chen, A systematic process for planing province combination circuits in PLCs, Journal of Industrial Technology, 1999 ; 15 ( 3 ) :2-5.
- S. Manesis, K. Akantziotis, Automated synthesis of ladder mechanization circuits based on state-diagrams, Advances in Engineering Software, 2005 ; 36:225-233.
- A. Rullan, Programmable logic accountants versus Personal computing machines for procedure control, Computers and Industrial Engineering, 1997 ; 33:421-424.
- J. Jang, P. H. Koo, S. Y. Nof, Application of design and control tools in a multirobot cell, Computers and Industrial Engineering, 1997 ; 32:89-100.
- P. Klingstam, P. Gullander, Overview of simulation Tools for computer-aided production technology, Computers in Industry, 1999 ; 38:173-186.
- L. A. Bryan, E. A. Bryan, Programmable Controllers, Theory and Implementation, An industrial text company publication, U.S.A. 1997, p. 785.