Abstract:This paper presents a method of performing topology optimisation of transient heat conduction problems using the parallel-in-time method Parareal. To accommodate the adjoint analysis, the Parareal method was modified to store intermediate time steps. Preliminary tests revealed that Parareal requires many iterations to achieve accurate results and, thus, achieves no appreciable speedup. To mitigate this, a one-shot approach was used, where the time history is iteratively refined over the optimisation process. The method estimates objectives and sensitivities by introducing cumulative objectives and sensitivities and solving for these using a single iteration of Parareal, after which it updates the design using the Method of Moving Asymptotes. The resulting method was applied to a test problem where a power mean of the temperature was minimised. It achieved a peak speedup relative to a sequential reference method of $5\times$ using 16 threads. The resulting designs were similar to the one found by the reference method, both in terms of objective values and qualitative appearance. The one-shot Parareal method was compared to the Parallel Local-in-Time method of topology optimisation. This revealed that the Parallel Local-in-Time method was unstable for the considered test problem, but it achieved a peak speedup of $12\times$ using 32 threads. It was determined that the dominant bottleneck in the one-shot Parareal method was the time spent on computing coarse propagators.

What problem does this paper attempt to address?

### Problems the paper attempts to solve The paper aims to solve the problem of low computational efficiency in topological optimization in transient heat conduction problems. Specifically, the author proposes a one - shot topological optimization method based on the Parareal parallel - in - time method to accelerate the solution process of transient heat conduction problems. #### Main problem description: 1. **High computational complexity**: Traditional topological optimization methods need to solve the state (and adjoint state) of the system in each iteration, which makes the computational cost very high. Especially for transient problems, the computational cost will increase significantly with the increase of the number of time steps. 2. **Parallelization requirements**: In order to accelerate the computation, researchers have tried to use various parallelization techniques. However, for transient problems, although spatial parallelization is effective, its computational cost is still proportional to the number of time steps. Therefore, a method that can be parallelized on the time axis is needed. 3. **Limitations of the Parareal method**: Parareal is a parallel - in - time method, but it was found in the initial attempts that it requires multiple iterations to obtain accurate results, resulting in no obvious acceleration effect. Therefore, the Parareal method needs to be improved to increase its efficiency. #### Proposed solutions: - **One - Shot method**: By iteratively refining the time history in the optimization process, estimating the objective function and sensitivity, and updating the design after a single Parareal iteration. - **Accumulated objective and sensitivity**: Introduce the concepts of accumulated objective and sensitivity, and solve these quantities through one Parareal iteration, thereby reducing the number of iterations. - **Modified Parareal method**: In order to adapt to the adjoint analysis, the Parareal method is modified so that it stores the states of intermediate time steps, so that the temperature fields at all fine - grained time points can be calculated. #### Experimental results: - **Acceleration effect**: Experiments show that this method achieves a 5 - fold acceleration relative to the sequential reference method (using 16 threads). - **Result quality**: The obtained design results are similar to the reference method in terms of the target value and qualitative appearance. - **Comparison with other methods**: Compared with the Parallel Local - in - Time method, the latter is unstable in the test problem, but can achieve a higher acceleration ratio (reaching a 12 - fold acceleration when using 32 threads). In conclusion, the paper successfully improves the computational efficiency of topological optimization in transient heat conduction problems by proposing an improved Parareal method while maintaining the quality of the results.

One-shot Parareal Approach for Topology Optimisation of Transient Heat Flow

Parameterized level set method based topology optimization of transient heat conduction structures

A "poor man's" approach for high-resolution three-dimensional topology optimization of natural convection problems

A Two-Level Parallel Decomposition Approach for Transient Stability Constrained Optimal Power Flow

Multi-step variant of the parareal algorithm: convergence analysis and numerics

A Fast Methodology for Topology Optimizations of Electromagnetic Devices

Transient topology optimization for efficient design of actively cooled microvascular materials

Multi-objective topology optimization of heat transfer surface using level-set method and adaptive mesh refinement in OpenFOAM

Topology optimization of heat sinks for instantaneous chip cooling using a transient pseudo-3D thermofluid model

Parallel-in-Time Integration of Transient Phenomena in No-Insulation Superconducting Coils Using Parareal

Neural-Parareal: Dynamically Training Neural Operators as Coarse Solvers for Time-Parallelisation of Fusion MHD Simulations

One-shot procedures for efficient minimum compliance topology optimization

Parallel-in-time integration of Kinematic Dynamos

Simultaneous and Meshfree Topology Optimization with Physics-informed Gaussian Processes

A micro/macro parallel-in-time (parareal) algorithm applied to a climate model with discontinuous non-monotone coefficients and oscillatory forcing

Efficient parallel inversion of ParaOpt preconditioners

Triply periodic minimal surfaces based topology optimization for the hydrodynamic and convective heat transfer

Heuristic optimality criterion algorithm for topology optimization of conjugate heat transfer problem

Dynamic optimization analyses and algorithm design for the parallel heat exchange system in spacecraft

Two-scale topology optimization for transient heat analysis in porous material considering the size effect of microstructure

Stabilization of parareal algorithms for long time computation of a class of highly oscillatory Hamiltonian flows using data