Simulating combined faults in split-type air conditioners across tropical island climates: Energy impacts and low-cost indicators for predictive maintenance

In tropical regions, split air-conditioning systems are essential in working and tertiary environments but account for significant electricity consumption. The presence of multiple simultaneous faults may progressively degrade system performance and increase energy consumption. However, the combined effects of vapor compression system faults and sensor biases remain insufficiently investigated under such climatic conditions. This study uses OpenStudio/EnergyPlus simulations coupled with an automated Python workflow to evaluate these combined effects. Results show that vapor compression system faults reduce cooling COP and increase energy consumption, whereas sensor faults alter control behavior and generate significant consumption variations. Interaction analysis of combined faults reveals nonlinear behavior at higher severity levels. A low-cost electricity consumption-based indicator is proposed to support predictive maintenance and optimized maintenance decision-making.