Abstract
Simulation aided building design optimization process can support architects and engineers in selecting high-performing building materials, glazing size, type, and precise sizing of air conditioners. To this end, this paper presents the application of two stochastic algorithms, i.e. Particle Swarm Optimization (PSO) and Genetic Algorithm (GA), to minimize annual cooling energy consumption for a residential building prototype situated in a hot and dry (Köppen climate classification: BSh) region of India. Total search space of 1890 unique design configurations is explored via variations across seven wall construction materials, i.e. (Aerated Concrete, Compressed Stabilized Earth and Calcium Silicate) blocks and (Concrete, Fly Ash, Hand and Machine Moulded Burnt Clay) bricks, single and double glazed windows, wall to window sizes of 10, 15, 20, 25 and 30 percent of the wall and three air conditioning capacities of 1, 1.5 and 2 tons. Both GA and PSO are performed over a population of 50 solutions for 25 generations using a flexible modelling and data exchange framework between Matlab, EPlauncher and Energy Plus tools. The same process employing GA and PSO repeated along four building orientations of 0°, 90°, 190° and 270° produce 53.78% and 52.39%, 52.52% and 52.29% energy reductions compared to the base case. The final optimal design recommends aerated concrete blocks 10% WWR, double glazed window (U ∼ 3.094 W/m2K), SHGC = 0.45 and 1 ton AC in both bedrooms for all four building orientations. Although both algorithms identify the same optimal solutions, GA requires double time (∼24 h, ∼12 generations) needed by PSO (∼12 h, ∼6 generations). Therefore, it may be more economical to adopt PSO over GA for simulation aided building optimization process.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call