Facility Deployment Decisions Through Warp Optimization of Regressed Gaussian Processes

A method for quickly determining deployment schedules that meet any given fuel cycle demands is presented here. This algorithm is fast enough to perform in situ within low-fidelity fuel cycle simulators. It uses Gaussian process regression models to predict the production curve as a function of time and the number of deployed facilities. Each of these predictions is measured against the demand curve using the dynamic time warping distance. The minimum-distance deployment schedule is evaluated in a full fuel cycle simulation, and the generated production curve then informs the model on the next optimization iteration. The method converges within five to ten iterations to a distance that is less than 1% of the total deployable production. This speed of convergence makes it suitable for use even when fuel cycle realizations are expensive, as in higher-fidelity or agent-based simulators. A representative once-through fuel cycle is used to demonstrate the methodology for reactor deployment. However, the algorithm itself is multivariate and may be used to determine the deployment schedules of many facility types that meet a number of independent criteria simultaneously. The once-through, electricity production example was chosen for the simplicity of illustrating the method.