Discrete Event Modeling Environment for Control Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current time-based modeling environments, such as Simulink, are inefficient for modeling complex systems that require event-driven behavior, as they rely on continuous time operations, leading to noticeable delays and computational costs, especially when modeling discrete events like network traffic or emergency scenarios.
Innovation Solution
An event-driven discrete event system (DES) modeling environment is introduced, where state transitions depend on asynchronous discrete events rather than continuous time, allowing for the creation of graphical representations with entities and blocks that manage and manipulate these events independently of continuous time, enabling efficient modeling of control systems with discrete event-driven components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If time-based modeling environments (e.g., Simulink) are used to model complex systems, then continuous time operations can be performed, but processor demand increases and efficiency decreases due to the need to model all time intervals including 'empty time'
Solution Approach 1:
The patent extracts and removes 'empty time' intervals from the simulation model, keeping only the discrete event times when state changes occur. This eliminates unnecessary computational operations during periods when system state remains unchanged, directly reducing processor demand while maintaining modeling efficiency.
Solution Approach 2:
The patent implements discrete event simulation where the system is updated only at specific event times rather than continuously. This periodic action approach processes the system state only when events occur, avoiding continuous time operations and reducing computational energy consumption while improving productivity.
2Reliability
If time-based modeling environments are used, then continuous time operations can be performed, but noticeable delays occur due to the continuous time stepping requirement
Solution Approach 1:
The patent removes continuous time stepping and retains only discrete event timestamps. By extracting the essential event timing information and eliminating continuous time intervals, the system achieves accurate event sequencing without the computational delays inherent in continuous time-based simulations.
Solution Approach 2:
The patent skips over 'empty time' intervals where no events occur, rushing through these periods without performing computational operations. This allows the simulation to jump directly from one event time to the next, eliminating unnecessary time delays while maintaining accurate event timing and system behavior.
3Productivity
If event-driven discrete event system modeling is used, then processor demand and computational costs are reduced, but the ability to model continuous time operations is limited
Solution Approach 1:
The patent creates a discrete event simulation framework that can model both discrete event systems and continuous time systems by treating continuous time as a special case where events occur at regular intervals. This universal approach maintains computational efficiency while providing the flexibility to model various system types.
Solution Approach 2:
The patent allows the simulation model to change parameters dynamically, switching between discrete event and continuous time modeling approaches based on the specific system being modeled. By adjusting the event time interval parameter, the same framework can accommodate both discrete and continuous time operations, maintaining both efficiency and versatility.
Data Source
AI summary
A discrete event system (DES) modeling environment models the occurrence of events independent of continuous model time. In a DES modeling environment, state transitions depend not on time, but rather asynchronous discrete incidents known as events. A discrete event modeling environment can be used to model a control system using one or more discrete event-driven components. The event-driven components can be used to model certain portions of a control system that cannot be accurately modeled using a time-based model.