Establishing precise data correlations is a fundamental necessity for effective design, simulation, and optimization of chemical processes. While several correlation-generating methods for data fitting exist, each brings a unique degree of complexity and accuracy. This study introduces a distinctive method, appropriately named, Proportional Nodes Method, for correlating data on surfaces with two independent variables, focusing on precise estimations of correlations for boundary curves and nodal variation at selected points. The application of the proportional node method results in a lower average percentage deviation than the Chebyshev polynomial method (±0.978 for the method utilizing Chebyshev polynomials and ±0.5378 for the Proportional Nodes Method). This novel method exhibits lower complexity and generates correlations with superior simplicity and accuracy. Furthermore, the method effectively develops correlations for three properties—specific volume, enthalpy, and entropy of superheated Opteon, a widely-used refrigerant in automotive air-conditioners. The resulting data shows a satisfactory percentage deviation from the actual data (<+0.22% for a specific volume, <+0.124 for enthalpy and <+0.125 for entropy). In addition, correlations have been generated for the saturated vapour. This method's potential extends to developing correlations for non-overlapping data with two independent variables, making it a powerful tool for chemical process optimization.
Read full abstract