Design Optimization of a Passive Building with Green Roof through Machine Learning and Group Intelligent Algorithm

This paper proposed an optimization method to minimize the building energy consumption and visual discomfort for a passive in Shanghai, China. A total of 35 design parameters relating form, envelope properties, thermostat settings, green roof configurations were considered. First, Latin hypercube sampling (LHSM) was used generate set samples, samples obtained through computer simulation calculation. Second, four machine learning prediction models, including stepwise linear regression (SLR), back-propagation neural networks (BPNN), support vector (SVM), random forest (RF) developed. It found that BPNN model performed best, with average absolute relative errors 3.27% 1.25% comfort, respectively. Third, six algorithms selected couple models find optimal solutions. The multi-objective ant lion (MOALO) algorithm be best algorithm. Finally, different groups variables conducted by using MOALO associated outcomes being analyzed. Compared reference building, solutions helped reduce up 34.8% improved 100%.