A hybrid optimization algorithm for gate locations in the liquid composite molding process

Abstract

It is costly to optimize the location of multiple injection gates through a trial and error-based method in the liquid composite molding, even though there are high fidelity physics-based numerical models. A hybrid optimization method called the Simulated Annealing Genetic Algorithm is proposed in this article, which uses the genetic algorithm to provide a global search for a predetermined time and then is further improved by the simulated annealing algorithm. The optimization results of multiple injection gates show that the number of convergence iterations using the Simulated Annealing Genetic Algorithm is less than that using the genetic algorithm, and the phenomenon becomes more obvious as the number of injection gates increases. The case shows that the Simulated Annealing Genetic Algorithm can solve the multiple injection gate configuration problems of highly anisotropic laminates without extra work. The optimization results are in good agreement with the experimental results.

Review of Approaches to Minimise the Cost of Simulation-Based Optimisation for Liquid Composite Moulding Processes

“…In both academic and industrial settings, the technique of parallel computing is often employed by researchers and industry practitioners alike to hasten the simulation-based optimisation processes [7,10,19,20]. The streamlining of the simulation-based optimisation process via parallel computing for LCM process optimisation is no exception [7,20,21]. Parallel computing can be understood as the act of breaking down a larger, complex problem into numerous smaller, independent sub-tasks and computing them simultaneously across multiple processing units.…”

“…Most importantly, parallel computing allows parallelisable algorithms and applications to be computed within a shorter wall-clock time than serial computing (i.e., faster algorithm execution). While the total computational load remains unchanged, independent computing tasks can be distributed across multiple processors or computing machines, drastically compressing the computing time required from start to finish [15,17,21,22]. The computational time saving is commonly quantified by the speedup, which is defined as the proportion of the cost of solving a parallelisable problem/algorithm via a single processing unit versus that of solving it parallelly across multiple processing units.…”

“…The transient mould-filling process is often treated in the finite elements (FE) approaches as a quasi-static/quasi-steady process, assuming a steady state is achieved at each time step throughout the temporal integration [28,[30][31][32]. The numerical analysis can then be computed by either the explicit time- In summary, while the adoption of parallel computing has great potential in the application of simulation-based optimisation, its efficacy and applicability are highly dependent on the degree of achievable parallelism imposed by the algorithm's framework and its flow dependencies [8,14,17,21]. Besides that, the application of parallel computing requires the development and execution of additional auxiliary algorithms to parallelise the existing optimisation framework (e.g., for task partitioning, task scheduling, task synchronisation, etc.)…”

“…If the time-step size adopted is large enough, the overall computational cost of the implicit time-integration can be significantly less than that of the explicit scheme. Nonetheless, the implicit time-integration scheme may still be inaccurate if the time-step size chosen is unproportionately large with respect to the phenomenon evaluated [21,31,33,38].…”

“…The implicit time-integration scheme is also often preferred as many problems in practice are stiff (numerically unstable), whereas the explicit scheme will require an impractically small time-step size to bound the error in the results. For the same process duration and desired accuracy, the explicit scheme's adoption of smaller time steps will result in an enormous amount of iterations required, which could be cheaper to solve by the implicit scheme with larger, albeit more costly, time steps [7,17,21,30].…”