Correlating Values of Parameters with Feed Properties

4 Correlating Values of Parameters with Feed Properties 

Once the set of optimized parameters for all correlations was determined, an attempt was made to correlate those parameters with properties of the feedstock such as API gravity, sulfur, metals, nitrogen, asphaltenes, and Conradson and Ramsbottom carbon contents. Only correlations with acceptable values of ( 0.8), determined previously, for at least four different feedstocks were selected, these correlations were 

Multiple nonlinear correlations for sulfur removal. Equations 12.84,12.87, 

As mentioned previously, the exponential equations were derived from a process different from hydrotreating/hydrocracking however, it is not surprising that metals and Conradson carbon content are well predicted when they are related to asphaltene concentration since it is well known that metals concentrate in the asphaltene fraction and that asphaltenes are directly associated with coke formation (Ancheyta and Speight, 2007). 

The form of correlations varied from simple equations with one parameter to more complex ones with four or more parameters. Firstly, individual properties of the feedstock were correlated with each optimized parameter, and then correlations with the combination of two properties were also developed. Trying to correlate more than two properties with parameters was not considered to avoid complexity of equations. Neither trigonometric functions nor more than two parameters were chosen to maintain the correlation as simply as possible. The highest value of was found for the first case, that is, correlation with one feedstock property. In the second case (correlation with two feedstock properties), three parameters were allowed to improve accuracy. 

---