In this work, we study simulation-based optimization, where the agent aims to select best configuration from design space with as few possible iterations. Inspired by success of deep reinforcement learning (DRL), formulate sampling process policy searching and give a solving method perspective iteration. Concretely, surrogate model for predicting performance each parameterized are applied, whic...

abstract:a mathematical model for the performance prediction of an industrial glass furnace with six ports on each side was developed. this model comprises of two main sub-models for the combustion chamber and glass-melting tank. the first sub-model consists of the models for the combustion and the heat transfer model including, radiation, convection and conduction. the fuel combustion in atmos...

Determination of optimum location for drilling a new well not only requires engineering judgments but also consumes excessive computational time. Additionally, availability of many physical constraints such as the well length, trajectory, and completion type and the numerous affecting parameters including, well type, well numbers, well-control variables prompt that the optimization approaches b...

WORK ACCOMPLISHED DURING THIS REPORTING PERIOD: During this reporting period, data obtained from the experiments designed and carried out during the previous reporting period was analyzed. These experiments covered a case study of MiamiDade County, Florida, using the simulation-based optimization framework developed during the previous two reporting periods. Conclusions drawn from this analysis...

Simulation optimization studies the optimization problem of simulation-based objectives in the foundation of mature modern optimization theory system. In order to improve the disadvantages of current selection of reduplicate simulation times in simulation optimization, this paper presents an adaptive selection of reduplicate simulation times in simulation optimization by using empirical knowled...