TA-23-C090 – Real-time Deployment of Building Automation System-integrated Inverse Model-based Fault Detection and Diagnostic algorithms

The increasing complexity of HVAC systems in large commercial buildings warrants recurring implementation of novel analytical approaches to operations and maintenance, and subsequent responses to improve and maintain optimal energy performance. Albeit the established capabilities of data-driven fault detection and diagnostics (FDD) to identify suboptimal controls policies and mechanical faults resulting in poor energy performance, few attempts have been made to disseminate and deploy scalable solutions around these approaches. Furthermore, real-time FDD approaches traditionally applied within a building automation system (BAS) are predominantly rules-based, offering limited insights to operating deficiencies with gradual negative impacts on energy performance. This paper presents a scalable prototype of a webbased application employing established data-driven, inverse model-based FDD methodologies in a real-time, BAS-integrated environment; the generated insights and resulting actions taken are discussed. This novel implementation facilitates automatic and recurring analyses of HVAC controls network data and can provide on-demand actionable visualizations and key performance indicators (KPI) to building operators; the application is currently deployed in an office building in Ottawa, Canada.