A novel multivariate linear prediction model for the marine rotary desiccant air-conditioning by adding a dynamic correction factor

Many research results suggested that a good estimation model often played a crucial role in the design, optimization and analysis for the HVAC system, especially during the preliminary design stage. Based on the multivariate linear regression analysis method, this paper presented a simple and high-accuracy prediction model by adding a dynamic correction factor. Newly developed model was not only particularly used to the marine rotary desiccant air-conditioning, but also its veracity and reliability were verified by a series of sample data and three evaluation indicators. Meanwhile, the prediction and optimization schemes of system performance are also introduced in detail. As expected, it was found that the dynamic correction factor can make the fitting value of prediction model close to the real value infinitely, and almost achieved linear fitting perfectly. As the number of correction increased, the residual and the residual standard deviation close to zero rapidly, and the relative error doubled decreased nearly. Besides, the square of multiple coefficient correlation (R2) of the prediction models reached 0.999 after the seventh corrected and the relative error much less than 1%. Furthermore, it was believed that the methodology developed here can be applied to other related fields.