• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.34 no.7
• /
• pp.703-714
• /
• 2010

This paper describes the design optimization of a rotating rectangular channel with staggered arrays of pin-fins by Kriging metamodeling technique. Two non-dimensional variables, the ratio of the height to the diameter of the pin-fins and the ratio of the spacing between the pin-fins to the diameter of the pin-fins are chosen as the design variables. The objective function that is a linear combination of heat transfer and friction loss related terms with a weighting factor is selected for the optimization. To construct the Kriging model, objective function values at 20 training points generated by Latin hypercube sampling are evaluated by a three-dimensional Reynolds-averaged Navier-Stokes (RANS) analysis method with the SST turbulence model. The Kriging model predicts the objective function value that agrees well with the value calculated by the RANS analysis at the optimum point. The objective function is reduced by 11% by the optimization of the channel.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.35 no.8
• /
• pp.763-771
• /
• 2011

An optimization procedure for the design of a ventilation axial-flow fan is presented in this paper. Flow analyses of the preliminary fan are performed by solving three-dimensional Reynolds-averaged Navier-Stokes equations via a finite-volume solver with the shear-stress transport turbulence model as a turbulence closure. Three variables, the hub-to-tip ratio and the stagger angles at the mid and tip spans, are selected for the optimization. The Latin-hypercube sampling method as a design-of-experiments technique is used to generate twenty-five design points within the design space. and the weighted average surrogate models, WTA1, WTA2, and WTA3, are applied for find optimal designs. The results show that the efficiency is considerably enhanced.

• Journal of the Korean Society of Systems Engineering
• /
• v.18 no.2
• /
• pp.75-93
• /
• 2022

Machine learning (ML) data-driven meta-model is proposed as a surrogate model to reduce the excessive computational cost of the physics-based model and facilitate the real-time prediction of a nuclear power plant's transient response. To forecast the transient response three machine learning (ML) meta-models based on recurrent neural networks (RNNs); specifically, Long Short Term Memory (LSTM), Gated Recurrent Unit (GRU), and a sequence combination of Convolutional Neural Network (CNN) and LSTM are developed. The chosen accident scenario is a control element assembly withdrawal at power concurrent with the Loss Of Offsite Power (LOOP). The transient response was obtained using the best estimate thermal hydraulics code, MARS-KS, and cross-validated against the Design and control document (DCD). DAKOTA software is loosely coupled with MARS-KS code via a python interface to perform the Best Estimate Plus Uncertainty Quantification (BEPU) analysis and generate a time series database of the system response to train, test and validate the ML meta-models. Key uncertain parameters identified as required by the CASU methodology were propagated using the non-parametric Monte-Carlo (MC) random propagation and Latin Hypercube Sampling technique until a statistically significant database (181 samples) as required by Wilk's fifth order is achieved with 95% probability and 95% confidence level. The three ML RNN models were built and optimized with the help of the Talos tool and demonstrated excellent performance in forecasting the most probable NPP transient response. This research was guided by the Systems Engineering (SE) approach for the systematic and efficient planning and execution of the research.