|
4.3 Plant Integration with the Agent-Based System Model
In the agent-based engineering process a system model is created as an instantiated agent-system. It actively supports the engineering process and thus meets the first of the tree challenges mentioned in Section 1: Faster development and modification of automation systems. But moreover the model has further possibilities. It also absorbs the engineering knowledge about the plant like positions, tasks, associations and dependencies. And it is a standalone application, a kind of parallel high-level process control system. This high-level system with “active components” that are aware of their position and tasks contains the essential engineering knowledge to be the basis for meeting the remaining two engineering challenges discussed in sect. 1: support in plant management and engineering during operation and integration of different heterogeneous systems in the environment of the automation system. The active components (alias agents) can be used to acquire necessary information from the automation system, using the technical possibilities of existing interfaces and to coordinate all of the information flows to connected systems. And in some cases it is even possible to intervene in the running automation system. For example an agent representing a pump P1 in the model could be assigned the goal to observe the physical pump. It can address the pump via the automation system’s interfaces [3] and collect the diagnosis status signals. Moreover, by the maintenance system the agent could be given the goal to analyze the status signals and if required generate a maintenance order in the associated enterprise resource planning system. It even could guide or support the mechanic in repairing the pump and if necessary automatically re-configure the pump. Fig. 6 shows the correlation between system model, engineering tool, automation system and the environmental systems.
All necessary infrastructure for the system model to fulfill its engineering tasks during plant operation is provided by the Agent Plant Integration Framework. It provides infrastructure like access to the automation systems interfaces. External systems can be encapsulated by agent adapters and data semantics of inhomogeneous systems can be mapped within the framework. In designing the system model of the plant the framework is tailored for the real automation system. Its infrastructure combined with the agent-based system model provides the basis for all engineering tasks during operation like monitoring, supervision, diagnosis, maintenance, simulation, quality analysis as well as integration of legacy and proprietary configuration tools, enterprise & business administration systems and electronic plant documentation. 5 Conclusion The challenges on plant automation systems require improved support of engineering and integration tasks by the automation system itself. The classical fragmentation of the engineering process into separated views and view-specific tools leads to missing transferability of data, high engineering and integration effort in both plant design and plant operation phase, and to less reusable solutions. New component-based approaches integrate the specific views and data in a higher-level technological image of the system. However, they fall short due to the facts of separating engineering data from functionality and the created system model remaining a static data model. In this paper the idea of combining the component-based higher-level system model with an agent-oriented meta model of the engineering process was presented. It leads to the approach of agent-based engineering, where the engineering functionality is accomplished by the components themselves. Thereby a better and more flexible support of processes is possible. And, the essential engineering knowledge now is absorbed by the created system model as it is an agent-system. This knowledge can be used as the basis for all engineering tasks during the life cycle of a plant. The Agent Plant Integration Framework therefore provides all necessary infrastructure. The presented approach employs the merits of agent concepts like autonomous acting, flexible interactions and goal-orientation to extensively support the engineering of automation systems in plants and the integration of heterogeneous systems in the environment of the automation system. Refinement and realization of this approach is subject to work in progress. Working points are the agent-oriented engineering model, design of the framework, “agentification“ of legacy component-based engineering models and establishing technical communication to the automation system, like data access for automatic adaptation of the model to changes in the plant. Page 1, 2, 3, 4, 5, 6
Tech Materials (Free)
Industrial Automation Systems Applying Agents for Engineering of Industrial Automation Systems
Robot Team Cooperation A Descriptive Model of Robot Team and the Dynamic Evolution of Robot Team Cooperation
Kuka Robots For ONU ONU Robotics Technology Center of Excellence, powered by KUKA Robotics Corporation
Augmented reality Annotation System for Robotic Application
Modular Robots Self-Reconfiguration Planning Of Identical Modules
Autonomous robots A New Approach To Robotics
Robotic Mounting Flat Panel Displays With Robotic Mounting
Calibration of Industrial Robots A Photogrammetric Robot Calibration System Based On Off-The-Shelf Low Cost Hardware Components
Vision System For Robots Adding Vision Guidance To A Robot
Robot Assistive Tasks User-Guided Reinforcement Learning for an Intelligent Environment
More... Amazon Books
|
