Search

This paper presents a novel, simple method for reducing external operating condition datasets to be used in multi-generation system optimization models. The method, called the Characteristic Operating Pattern (CHOP) method, is a visually-based aggregation method that clusters reference data based on parameter values rather than time of occurrence, thereby preserving important information on short-term relations between the relevant operating parameters. This is opposed to commonly used methods where data are averaged over chronological periods (months or years), and extreme conditions are hidden in the averaged values. The CHOP method is tested in a case study where the operation of a fictive Danish combined heat and power plant is optimized over a historical 5-year period. The optimization model is solved using the full external operating condition dataset, a reduced dataset obtained using the CHOP method, a monthly-averaged dataset, a yearly-averaged dataset, and a seasonal peak/off-peak averaged dataset. The economic result obtained using the CHOP-reduced dataset is significantly more accurate than that obtained using any of the other reduced datasets, while the calculation time is similar to those obtained using the monthly averaged and seasonal peak/off-peak averaged datasets. The outcomes of the study suggest that the CHOP method is advantageous compared to chronology-averaging methods in reducing external operating condition datasets to be used in the design optimization models of flexible multi-generation systems. [ABSTRACT FROM AUTHOR]

Average surface roughness value ( R a ) is an important measure of the quality of a machined work piece. Lower the R a value, the higher is the work piece quality and vice versa. It is therefore desirable to develop mathematical models that can predict the minimal R a value and the associated machining conditions that can lead to this value. In this paper, real experimental data from an end milling process is used to develop models for predicating minimum R a value. Two techniques, model tree and sequential minimal optimization based support vector machine, which have not been used before to model surface roughness, were applied to the training data to build prediction models. The developed models were then applied to the test data to determine minimum R a value. Results indicate that both techniques reduced the minimum R a value of experimental data by 4.2% and 2.1% respectively. Model trees are found to be better than other approaches in predicting minimum R a value. [ABSTRACT FROM AUTHOR]