SimulationUsingDistributedModels - Simulation Using...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Simulation Using Distributed Models In the pervious section it was shown how the simulation / control tool can be use to control an automated system. We then used the same model to simulate the system by simply running the tool in simulation mode. Next we saw how to control a system using a distributed collection of independent controllers. These controllers communicate with each other. Now the question is: how can we simulate the system using the same model if the controller consists of a collection of models and the simulation tool studied thus far needs one model? In the following section a methodology will be presented that used to create a supervisory function that, along with some modifications to the tool, can simulate a system using a collection of models rather that only one model. The Distributed Modeling Paradigm The simulation of distributed systems discussed here is based on the Hierarchical Object Oriented Programmable Logic Simulator (HOOPLS) methodology. HOOPLS is a simulation methodology that is founded upon the belief that in order to accurately model a system with a distributed control architecture, the interactions among the controllers must be considered within the model. The single most important characteristic of the HOOPLS methodology, and what separates it from other object-oriented research efforts in simulation, is its focus upon modeling the flow of messages among the included controllers. The question may be asked, why is modeling the interactions among the controllers important to the accuracy of the simulation? To answer this question imagine a team of people that control a company, say a car manufacturing company. The managers, starting with the CEO all the way to the floor supervisor, control the company. This management team provides an example of a distributed discrete-event control architecture. It can easily be seen that the team spends most of its time communicating with each other. As such, if you were to model this team, ignoring the communications among the team members, your model would suffer from a significant loss in accuracy. Under the distributed modeling paradigm, a controller, in conjunction with its controlled subsystem, is referred to as a coordinated object. The modeling of the controllers must take into account the manner in which each element of the coordinated object can modify the state of a given entity. In addition, the coordinated object, itself, can reassign the ownership of an entity to any subordinate subsystem for the implementation of a specified set of tasks. The material handling systems (which may also be coordinated objects) that are included within the coordinated object can change only the location of the entity. Finally, it is the primary unit processes which are contained within the coordinated object that execute the assigned processing tasks that physically modify the state of an entity. The distributed modeling explicitly considers these constraints when it defines the state-transition mechanisms for the evolution of the discrete-event
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 7

SimulationUsingDistributedModels - Simulation Using...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online