S.A.N. Nouwens,B. de Jager,M.M. Paulides,W.P.M.H. Heemels
Abstract:Model predictive control (MPC) is an optimization-based control strategy with broad industrial adoption. Unfortunately, the required computation time to solve the receding-horizon MPC optimization problem can become prohibitively large for many applications with a large number of state constraints. This large number of state constraints can, for instance, originate from spatially discretizing a partial differential equation of which the solution has to satisfy constraints over the full spatial domain. This is particularly the case in MPC for RF-based hyperthermia cancer treatments, which forms a strong motivation for this study. To address this problem, we propose a novel constraint-adaptive MPC framework for linear discrete-time systems. In this framework, we select at each time-step a subset of the state constraints that are included in the optimization problem, thereby reducing the online computational burden. Critically, our framework guarantees the same closed-loop performance, recursive feasibility, and constraint satisfaction properties as the original (non-reduced) MPC scheme. We achieve this result by efficiently exploiting reachable set computations and the MPC cost function. We will demonstrate our novel method using a hyperthermia cancer treatment case study showing a two-orders of magnitude improvement in computation time, with identical closed-loop performance as the original (non-reduced) MPC scheme.
What problem does this paper attempt to address?
The paper aims to solve the problem of excessive computation time faced by model predictive control (MPC) when dealing with systems with a large number of state constraints. Specifically, when applying MPC to scenarios where partial differential equations on continuous spatial domains need to be constrained, such as radio - frequency (RF) - based hyperthermia cancer treatment, the computational burden caused by these constraints makes it difficult to apply MPC in real - time. The paper proposes a new online constraint - adaptive MPC framework, which reduces the online computational burden by selecting a subset of state constraints at each time step, while ensuring the same closed - loop performance, recursive feasibility, and constraint - satisfaction properties as the original (unreduced) MPC scheme.
### Main contributions of the paper
1. **Propose a new online constraint - adaptive MPC framework**: This framework reduces the online computational burden by selecting a subset of state constraints at each time step.
2. **Utilize reachable sets and cost function level sets**: Constraints are systematically removed from the optimization problem through fast reachable set analysis and the optimal control sequence of the previous time step.
3. **Demonstrate the effectiveness of the method**: Through a case study of radio - frequency - based hyperthermia cancer treatment, it is proved that this method can reduce the computation time by two orders of magnitude while maintaining the same closed - loop performance as the original MPC scheme.
### Paper structure
- **Introduction**: Introduces the basic concepts of MPC and its wide application in industry, and points out the computational challenges when dealing with a large number of state constraints.
- **Preliminaries**: Defines linear discrete - time systems, typical MPC settings, forward and backward reachable sets, and cost function level sets.
- **Constraint removal method**: Describes in detail how to use reachable sets and cost function level sets to remove redundant constraints.
- **Numerically efficient constraint removal**: Provides methods for computing outer approximations of reachable sets and cost function level sets, and proposes strategies for efficiently identifying and removing redundant constraints.
- **Case study**: Through a case of radio - frequency - based hyperthermia cancer treatment, shows the effectiveness of the proposed method in practical applications.
- **Conclusion and outlook**: Summarizes the main contributions of the paper and discusses future research directions.
### Key formulas
- **System dynamics**:
\[
x_{k + 1}=Ax_k + Bu_k
\]
- **State and input constraints**:
\[
x_k\in\{x\in\mathbb{R}^n\mid Cx\leq b\}=:X
\]
\[
u_k\in\{u\in\mathbb{R}^m\mid Cuu\leq bu\}=:U
\]
- **MPC optimization problem**:
\[
\min_{U_k}J(x_k, U_k)
\]
\[
\text{subject to }X_k = \Phi x_k+\Gamma U_k
\]
\[
X_k\in X_N,\quad U_k\in U_N
\]
- **Cost function**:
\[
J(x_k, U_k)=\ell_T(x_N|k)+\sum_{i = 0}^{N - 1}\ell(x_i|k, u_i|k)
\]
- **Reachable set**:
\[
\rightarrow R_{k + 1}(X_0, U)=\{x^+\in\mathbb{R}^n\mid x^+=Ax + Bu,\text{ for some }x\in\rightarrow R_k(X_0, U), u\in U\}
\]
- **Cost function level set**:
\[
J_u(x_k,\tilde{U}_k)=\{U\in\mathbb{R}^{Nm}\mid J(x_k, U)\leq J(x_k,\tilde{U}_k)\}
\]
### Case study
The paper shows the effectiveness of the proposed method in practical applications through a case of radio - frequency - based hyperthermia cancer treatment. The results show that this method can reduce the computation time.