UNIK-OPT/NN Neural network based adaptive optimal controller on optimization models

摘要

When the future information for an optimization model is not complete, the model tends to incorporate such uncertainties as some assumptions on the coefficients. As time passes and more precise information is accumulated, the initial optimal solution may no longer be optimal, or even feasible. At this point, model builders want to modify the assumed and controllable coefficients to obtain the desired values of designated decision variables. To aid this process, a neural network could effectively be applied. So we develop a tool UNIK-OPT/NN which can support the construction and recall of the neural network model on top of the knowledge assisted optimization model formulator UNIK-OPT and the semantic neural network building aid UNIK-NEURO. By adopting a commonly interpretable semantic representation of optimization and neural network models, UNIK-OPT/NN can effectively automate most of the neural network construction and recall procedure for optimal control.