Enhancing smart thermostat performance through customer data analysis and the integration of AI techniques

Enhancing the efficiency of expanding HVAC systems is a complex task. Utilizing Artificial Intelligence (AI) techniques like Machine Learning (ML) and Deep Learning offers promising solutions. ENA Solution’s smart thermostats provide the potential to employ AI methods for temperature control and energy optimization. In recent years, the company has made efforts to integrate AI for automated solutions in their smart products, but the research is still in its early stages. Moreover, considering recent advancements in AI algorithms, active R&D is essential to maintain product accuracy and performance leadership in the market.
The goal enhancing the efficiency of HVAC systems can be achieved by studying available methods, analyzing company resources, proposing practical solutions, and monitoring small-scale candidate approaches. Analysis of customer feedback and prototype data, recognizing vulnerabilities and technology bottlenecks, and applying stable, optimized methods on a large scale are also vital. The study will leverage practical knowledge of complex systems, thermodynamics, and statistical mechanics to understand influential factors and design AI models based on scale and complexity. Potential approaches may include Convolution Neural Networks (CNNs) for large-scale data or techniques like k-nearest neighbors algorithm (KNN) and decision tree algorithms for smaller scales. Alternative solutions will also be explored based on study outcomes.

Faculty Supervisor:

Raymond Spiteri;Terry Peckham

Student:

Partner:

ENA Solution Inc

Discipline:

Computer science

Sector:

Manufacturing

University:

University of Saskatchewan

Program: