Desiccant cooling systems (DCS) represent a suitable alternative to conventional systems for air-conditioning purposes. Their benefits should be correctly assessed by means of dynamic simulations, taking into account both the operating context and the available control variables. Several models are available in literature to model solid DCS based on desiccant wheels (DW). Nevertheless, physical models are rather complex to be implemented in dynamic simulation tools of building-integrated energy systems, while constant effectiveness models have low performance. Regression models can represent a suitable alternative, as they can provide high accuracy but with a low modeling effort. In this paper, experimental data are used to develop correlations to predict the dehumidification and thermal performance of a DW, as a function of inlet air temperature and humidity ratio, regeneration temperature, air flow rates, and rotational speed. Statistical tools are used to investigate the effect of those independent operating variables. Furthermore, the selection software provided by another DW manufacturer is used to generate operational data of a further desiccant rotor and to derive the related correlations. At last, the proposed model is compared with the correlations found in the relevant literature. The results show that a very good agreement is found in the comparison between measured and predicted values, with maximum relative errors not higher than 5%. Furthermore, an excellent behavior of the proposed model is also found when it is used to simulate a generic desiccant wheel, without the need of a detailed physical model. Finally, a better agreement is found with respect to other models based on correlations developed in literature, even using a higher number of coefficients to be calibrated.