a PLM 베스트 프랙티스 컨퍼런스 2012 a 인지 제조시스템 - Cognitive Manufacturing System - June 21 st , 2012 Prof. Dr. -Ing. Hong–Seok Park Laboratory for Production Engineering School of Mechanical and Automotive Engineering University of ULSAN
Contents 인지 제조시스템 - Cognitive Manufacturing System - 1. Introduction 2. Classification of disturbances through analyzing current manufacturing system 3. Elementary technology for developing self adapting manufacturing system 3.1. Cognitive agent 3.2. Biology inspired strategy 4. Development of self adapting manufacturing system (SAMS) 4.1. Concept of SAMS 4.2. Information module of SAMS 4.3. Algorithm of SAMS 5. Implementation of SAMS 5.1. Hardware architecture of SAMS 5.2. Software architecture of SAMS 5.3. Communication network of SAMS 6. Conclusion. Lab. for Production Engineering
Necessity for developing a new manufacturing concept Arriving New manufacturing Work-piece Self-adapting system Loading machine Current manufacturing Adjusting Agent #1 parameter Begin Disturbance Stop machine Tool-wear Machining Parameter Tool-wear request Machining change Tool-break Tool change Tool-break Finish Machining Intervention of human Unloading operator Agent #n machine Next Machines v Reducing productivity v Self-adapting v Decreasing the utilization v Reasoning ability in decision Intelligent& Self- Cooperation of machining shop Genetic adaptive making behaviours v Measures depending on Downtime Stop machine Experience v Self-controlling ability the experience of operators Lab. for Production Engineering
Analyzing current manufacturing in consideration of self adapting concept Machining Shop Product Disturbance { Downtime: 20-25% of total planned time Clutch housing Clutch housing v Processing operations per product: 17 v Machines: 12 Recovery method Current recovery method: Proposed method: Self-adaptive manufacturing system Stop the machining shop to repair and reset Self - Recovery Centralized control system Machine Machine Transporter Work-piece … § Rigid Control: Top-down Agent 1 Agent 2 Agent Agent problem solving § Low scalability § Low adaptability Machine 1 Machine 2 Work-piece Transporter Lab. for Production Engineering
Disturbance Classification and Management methods Disturbance Information v Data collection time: 2006.08.31-2009.08.18 v Disturbance numbers: 685 Disturbance Classification 11.4% Disturbance class Type of disturbance Related to resources Machine breakdown Non-negotiation 47.7% Maintenance of machine Negotiation Tool breakdown 40.9% Tool wear Rescheduling Operator absenteeism Related to orders Unavailability of raw material Cancellation order Rework Disturbance Disturbance Machining System Arrival of a new job order Urgent job Delay in transport using material handling system Out sourcing Related to measurement of data Process time variation Variation of set-up times Change of priority Control software Malfunction and Communication networks Malfunction of machine Tool-break MES Event Event (short recovering time) Event Event Agent Tool-wear Malfunction of machine Agent (long recovering time) Agents Rescheduling Non-negotiation Negotiation Negotiation type: Rescheduling type: Non-Negotiation type: Recovering time > 1 hour 30 mins < Recovering time < 1 hour Recovering time < 30 mins Lab. for Production Engineering
Concept of cognitive agent Sensors Reasoning Percepts Rule- based Perception Decision making Actions Environment Actions Conventional agent Effectors Cognitive technology Agent technology Synthesis of agent and cognitive technologies Update Cognitive BDI Architecture knowledge Knowledge Reasoning Beliefs Agent Learning & Experience Grasping the information of the current states of an agent’s environment (Intentions) Decision New situation Making Desires Interpretation All the possible states of tasks that agent (Desires) Familiar could carry out situation Information Plan input Intentions Action Event The states of tasks that the agent has Perception Cooperation Event Command decided to work towards (Beliefs) recognition output Communication Lab. for Production Engineering
Biology inspired strategy to adapt to disturbance Pheromones evaporate Ant travel rule: when no ant pass Each ant always try to chose the trail has higher pheromone concentration Ants lay pheromone on the trail Pheromones accumulate with when they moves food back to nest multiple ants using same path From Natural to Manufacturing Systems Ability list The machine with the shortest processing time for carrying out a specific Ability Machine tool operation will has the highest pheromone Information Node Task1 Machine 1 Ability Pheromone Task 2 Machine 2 Task 1 Pheromone Task 3 Machine 3 value 1 Task 2 Machine 1 Task 2 Pheromone value 2 Task 2 Machine 3 Machine breakdown Cognitive agent M2,T2 M3,T3 M1,T1 Product Route 2 M3,T2,T3 M1,T1 Work-piece Lab. for Production Engineering
Systematic procedure for developing a Self-Adaptive Manufacturing System (SAMS) Product and Machining Shop Disturbance Classification Finding measures against the Disturbance Analysis corresponding disturbances Elementary technologies Inspired Biology : Ant Colony Algorithm to develop a SAMS Cognitive Agent Model of the machining shop based on functional agents Developing Architecture Information Model of SAMS Of the machining shop Mechanism of SAMS for adapting to based on Cognitive agents disturbance Architecture of the Test-bed Algorithm for making decision of SAMS Implementing Mechanism of the implemented SAMS the Test-bed of SAMS for adapting to disturbances Model of SAMS Model of SAMS Test-bed Test-bed Strategies for overcoming disturbances Strategies for overcoming disturbances Non-negotiation Negotiation Diagnosis Disturbance Cognitive Agent Controller Lab. for Production Engineering
Concept of a self adapting manufacturing system MES Knowledge v Behaviours policies Machine § Rule-based Machine Agent #i Work-piece Transporter § Reasoning mechanism Machine Agent #3 Agent Agent v Pheromone value Agent #2 Reasoning Machine Agent #1 Decision Making Interpretation Plan Perception Control Communication Signals Tasks Work-piece Machine #i Transporter Machine #1 Machine #3 Machine #2 Disturbance Machining system Machine #1 Work-piece Machine #2 Machine #2 Transporter Machining Shop Lab. for Production Engineering
Cognitive agent based disturbance handling Previous state I, B Deliberation reasoning Update state b:=see Deciding on B:=update( B, b) what state to achieve Comparison MES D:=process(task) t:= compare (B, D) Negotiation Disturbance Inform normal state (t:=0) Deciding on what agent to be selected without (agent) Type C: Cooperative behavior Type A: rescheduling Disturbance type Negotiation + MES c:=Diagnosis (type) Select (agents) rescheduling Type B: Reactive behavior Mean-ends without (p) reasoning I:=filter(B,D,I) Agent (selected) Rule-base p:=plan(B, I, c) Deciding on Execute (p) How to achieve this state Execute (p) End End Lab. for Production Engineering
Implementing the Test-bed of SAMS Working sequence for implementing the SAMS Hardware architecture of SAMS Software architecture of SAMS Developing the database of SAMS Non-negotiation mechanism of SAMS Negotiation mechanism of SAMS Demonstration of SAMS Evaluation of SAMS Lab. for Production Engineering
Hardware architecture of SAMS on the test-bed Lab. for Production Engineering
The communication of wire network between the devices in SAMS Flowchart SEND/RECEIVE PLC S7-300 between PLC and RFID DI D0 CP343-1 CP341 module module Internet card IP: 192.168.0.30 Internet cable RS232 cable Disturbance light inputs indicator Message for SEND/RECEIVE data Light Work- piece RFID Tag RFID Reader Read/Write Lab. for Production Engineering
The communication of wireless network between agents in SAMS The communication mechanism between IP: 192.168.1.2 Access Access agents in the AMS point point Agent #2 IP: 192.168.1.3 Agent #3 Access Access IP: 192.168.1.1 point point IP: 192.168.1.n Agent #n Agent #1 Wireless Bridge in wireless network The communication bridge has function of the signal amplifier, and is a middleware node in the communication wireless network. Lab. for Production Engineering
Software architecture of SAMS Machine1 Machine2 Machine3 RFID PLC1 RFID PLC2 PLC3 RFID Work-piece Disturbance information information KepserverEX KepserverEX KepserverEX SQL SQL SQL Task OPC OPC OPC Process Disturbance Information Agent#1 Agent#2 Agent#3 Knowledge Message Message MES Wire Disturbance SQL Process Wireless Information Resource Information Lab. for Production Engineering
Database design and analysis Cooperation information Machine agent state Perception Task Information Information collection Plan OPC items A Disturbance classifying Suggested A Decision method A B B Machine Information Unknown Disturbances Disturbance information Process Information A A disturbance known B disturbance unknown Lab. for Production Engineering
Algorithm for making decision of SAMS Non-Negotiation Negotiation Lab. for Production Engineering
Recommend
More recommend
