Redundant Flow Control for Deterministic Virtualized Control Loops
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In virtualized control systems or operating systems without real-time extensions, time-critical control applications face delays or prioritization issues due to lack of direct hardware access, leading to non-deterministic execution and potential delays or disruptions.
Innovation Solution
Implementing mutually redundant sequence control components that determine manipulated variables from measured or state variables, transmit them with sequence numbers, and utilize quality of service reservations in communication devices to ensure deterministic and redundant data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If control applications are executed in virtualized control systems or operating systems without real-time extensions, then system flexibility and resource utilization are improved, but execution determinism and real-time performance deteriorate
Solution Approach 1:
The system segments control applications into multiple mutually redundant sequence control components that operate in parallel. Each component independently processes control logic and transmits manipulated variables, ensuring that if one component fails or experiences delays, others can maintain deterministic execution. This segmentation allows the system to achieve real-time performance in virtualized environments by distributing control functions across multiple independent execution paths.
Solution Approach 2:
The system implements beforehand cushioning by transmitting manipulated variables through multiple redundant paths before they are needed at the actuators. Quality of service reservations are established in advance for these transmission paths, and duplicate filters are pre-configured at network nodes. This preparatory redundancy ensures that even if some paths experience delays in virtualized environments, the control system maintains deterministic behavior through the cushioning effect of pre-established alternative paths.
2Productivity
If network resources are shared between real-time data streams and large payload data frames, then network resource utilization is improved, but quality of service for real-time communications deteriorates
Solution Approach 1:
The system applies local quality by configuring duplicate filters at specific network nodes along redundant transmission paths. These filters are selectively placed at strategic locations in the network to ensure that real-time data frames are properly identified and prioritized only where needed, while allowing non-real-time traffic to use network resources elsewhere. This localized quality enhancement maintains QoS for real-time communications without unnecessarily restricting overall network resource utilization.
Solution Approach 2:
The system performs preliminary action by establishing quality of service reservations and configuring duplicate filters in advance during resource reservation, before actual data transmission begins. This advance preparation ensures that when real-time data frames need to be transmitted, the network path is already reserved and configured to prioritize them, preventing QoS degradation even when network resources are shared with large payload traffic.
3Adaptability or versatility
If multiple control applications simultaneously request exclusive access to the same resource in virtualized environments, then application diversity is improved, but execution timing and determinism deteriorate
Solution Approach 1:
The system merges multiple control applications into a unified redundant architecture where multiple sequence control components execute the same control logic independently. Instead of allowing applications to compete for exclusive access to resources, the system combines their functions into parallel redundant paths that transmit manipulated variables through separate network paths to the same actuators. This merging eliminates resource contention while maintaining application diversity through the redundant components.
Solution Approach 2:
The system uses copying by creating duplicate sequence control components and duplicate transmission paths for each control application. Each component independently processes control logic and transmits manipulated variables with sequence numbers through separate network paths. The actuators receive multiple copies of the same control commands and use duplicate filters to identify and process them, ensuring deterministic execution timing even when multiple applications run simultaneously in the virtualized environment.
Data Source
AI summary
Time-critical control applications are provided by way of flow control components (113, 123, 133) that are each able to be loaded into a flow control environment (112, 122, 132) formed by way of a server device (101-103) and executed there. The control applications each periodically determine, from periodically acquired measured and/or state variables (12), manipulated variables (11) for a process to be controlled or regulated. A plurality of mutually redundant flow control components (113, 123, 123) are executed in parallel with one another. The mutually redundant flow control components (113, 123, 133) determine the manipulated variables (11) for a respective process cycle that follows a respective acquisition time of the measured and/or state variables (12) as soon as a majority of the mutually redundant flow control components signal error-free presence of the measured and/or state variables for the respective acquisition time. As an alternative or in addition, the determined manipulated variables are transmitted, together with a sequence number assigned to the respective process cycle, to actuators (320) and/or control units (31-32).