Abstract:A large-scale complex system comprising many, often spatially distributed, dynamical subsystems with partial autonomy and complex interactions are called system of systems. This paper describes an efficient algorithm for model predictive control of a class of system of systems for which the overall objective function is the sum of convex quadratic cost functions of (locally) constrained linear subsystems that are coupled through a set of (global) linear constraints on the subsystems coordination parameters.
The proposed control algorithm is based on parametrization and splitting of the underlying optimization problem into one global coordination problem and a set of local optimization problems pertaining to individual subsystems. The local optimization problems are solved off-line, via parametric optimization, while the coordination problem is solved on-line. The properties of the local parametric solutions are utilized to solve the coordination problem very efficiently. In particular, it is shown that, for a fixed number of coupling constraints, the coordination problem can be solved with a linear-time algorithm in a finite number of iterations if all subsystems have one-dimensional coordination parameters.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to carry out effective coordinated control among multiple subsystems in the system, especially when these subsystems have partial autonomy and are interconnected through linear coupling constraints. Specifically, the paper focuses on how to minimize the cumulative cost function of all subsystems under the premise of satisfying local constraints and global coordination parameter constraints. This problem usually occurs in large - scale complex systems, such as power networks, building complexes, petrochemical and chemical production facilities, water distribution systems and natural gas networks, etc.
### Main Objectives of the Paper:
1. **Simulate the Advantages of Distributed Optimization Methods**: Make the proposed control method applicable to the System - of - Systems (SoS) framework while protecting the data privacy of each subsystem.
2. **Implement an Efficient Online Computation Algorithm**: This algorithm can be easily applied to large - scale SoS containing a large number of subsystems and has high computational efficiency.
### Key Points of the Solution:
- **Problem Decomposition**: Decompose the overall control problem into smaller, decoupled local problems related to each subsystem and a coordination problem that describes the coupling of subsystems.
- **Offline Computation**: The contribution of each local subsystem to the overall cost function and its optimal control action are calculated offline as a solution to multi - parameter quadratic programming (mp - QP).
- **Online Coordination**: Based on the offline solution, determine the global optimal coordination parameters by solving the coordination problem. This process does not require iterative information exchange between the central coordinator and local subsystems, thus improving computational efficiency.
### Main Contributions:
- **Re - formulate the Global Coordination Problem as a Continuous Quadratic Knapsack Problem (CQKP)**: When there are multiple coupling constraints, this problem is further defined as a Multiple - Constrained Continuous Quadratic Knapsack Problem (MCQKP).
- **Propose an Efficient Linear - Time Algorithm**: That is, the Hyperplane Search (HPS) algorithm, which can find the solution to the global coordination problem in a finite number of iterations, so that this method can be well - scaled to large - scale SoS.
- **Verify the Effectiveness of the Method through Numerical Case Studies**: In the numerical case study of the micro - grid system, this method not only theoretically confirms its linear - time complexity, but also shows the advantage of significantly reducing online computation time in practical applications. Compared with the traditional centralized MPC method, the computation time can be reduced by up to two orders of magnitude.
In conclusion, this paper proposes a centralized Model Predictive Control (MPC) method that combines offline and online computation to solve the coordinated control problem in the System - of - Systems, especially suitable for those scenarios where the data privacy of subsystems needs to be protected.