Developing surrogate ANN for selecting near-optimal building energy renovation methods considering energy consumption, LCC and LCA

Buildings are responsible for more than 30% of the total energy consumption and an equally large amount of related greenhouse gas emissions. Improving the energy performance of buildings is a critical element of building energy conservation. Furthermore, renovating existing buildings' envelopes and systems offers significant opportunities for reducing Life Cycle Cost (LCC) and minimizing negative environmental impacts. This approach can be considered as one of the key strategies for achieving sustainable development goals at a relatively low cost, especially when compared with the demolition and reconstruction of new buildings. One of the main methodological and technical issues of this approach is selecting a desirable renovation strategy among a wide range of available options. The main idea and motivation behind this study relies on trying to bridge the gap between Simulation-Based Multi-Objective Optimization (SBMO) and Artificial Neural Network (ANN). For a whole building simulation and optimization, current SBMOs often need thousands of simulation evaluations. Therefore, the optimization becomes unfeasible because of the computation time and complexity of the dependent parameters. To this end, one feasible technique to solve this problem is to implement surrogate models to computationally imitate expensive real building simulation models. The objective of the research focuses on developing a robust ANN to explore vast and complex data generated from the SBMO model. More specifically, this research aims to propose an accurate ANN to predict energy consumption using data from the SBMO model. The proposed model will potentially offer new venues to predict Total Energy Consumption (TEC), LCC, and Life Cycle Assessment (LCA) for different renovation scenarios, and select the optimum scenario. To illustrate the applicability of the model, a case study was developed and the accuracy of the proposed model was evaluated. Results show that models constructed using ANNs are considerably less time-consuming than the conventional Building Energy Model (BEM) while achieving acceptable accuracy.