Aircraft Propulsion Distributed Control Using Blockchain Coordination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Centralized coordination of distributed electrical sub-systems in aircraft propulsion systems becomes intractable due to high communication and computation demands, especially as the number of sub-systems increases, making efficient decentralized control structures necessary.
Innovation Solution
A secure, distributed transaction ledger using blockchain technology coordinates electrical sub-systems by calculating global qualities such as Lagrange multipliers and step sizes, allowing nodes to perform iterative optimization calculations and converge to an optimal control solution, thereby enabling efficient decentralized power control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If centralized coordination is used among distributed electrical sub-systems, then optimal control can be achieved, but communication and computation demands become intractable as the number of sub-systems increases
Solution Approach 1:
The patent divides the centralized coordination problem into distributed sub-problems by implementing a distributed optimization framework where each electrical sub-system independently solves local optimization problems. The global coordination is segmented into iterative exchanges of dual variables (Lagrange multipliers) between sub-systems and a coordinator, replacing the intractable centralized computation with manageable distributed calculations.
Solution Approach 2:
The patent introduces a distributed coordinator that acts as an intermediary, facilitating communication between electrical sub-systems without requiring direct peer-to-peer communication among all sub-systems. This intermediary computes and distributes dual variables that guide local optimization, reducing overall communication complexity while maintaining coordinated control.
2Adaptability or versatility
If the number of distributed electrical sub-systems increases, then system capability and flexibility improve, but centralized coordination becomes intractable
Solution Approach 1:
The patent implements a dynamic distributed optimization framework where the coordination mechanism adapts to the number of sub-systems. The iterative exchange of dual variables and primal solutions automatically scales with system size, maintaining coordination efficiency regardless of the number of electrical sub-systems through dynamic adjustment of optimization iterations.
Solution Approach 2:
The patent changes the coordination parameters from centralized state variables to distributed dual variables (Lagrange multipliers). This parameter transformation enables the system to handle increasing numbers of sub-systems efficiently, as each sub-system only needs to exchange compact dual variable information rather than full state information, reducing communication burden while maintaining adaptability.
3Ease of operation
If distributed control is implemented without optimization, then decentralization is achieved, but convergence to optimal power tracking is not guaranteed
Solution Approach 1:
The patent implements a feedback mechanism through iterative dual variable exchanges between the distributed coordinator and electrical sub-systems. Each iteration provides feedback on power tracking errors via updated Lagrange multipliers, guiding sub-systems to adjust their power output until convergence to the optimal power tracking profile is achieved, while maintaining decentralized control architecture.
Data Source
AI summary
Some embodiments provide a distributed optimization framework and technology for the control of aggregated propulsion system components that iteratively operates until a commanded propulsion-related profile is produced by aggregated components of the propulsion system. Some embodiments use a distributed iterative solution in which each component solves a local optimization problem with local constraints and states, while using global variables that are based upon information from each other component. The global variables may be determined via a distributed transaction system using component-specific information obtained from each component at each iteration. The global variables may be broadcast to the components for each new iteration.