Pacific Northwest National Laboratory
Bulding Systems Program
Collage photo of a boiler, chiller, and cooling tower.

Automated Diagnostic Algorithms for Chillers, Boilers, Cooling Towers, and Chilled Water Distribution

Laboratory researchers in collaboration with staff of Architectural Energy Corporation (AEC) developed logic for diagnosis of problems with chillers, cooling towers, and chilled water distribution systems in commercial buildings. This work was done in a project funded jointly by the California Energy Commission Public Interest Energy Research Program (PIER) and the U.S. Department of Energy Building Technologies Program. The logic was derived from diagnostics based on visual interpretation of data plots developed and used over many years by AEC staff. The logic is documented and requirements specified for software for automating the diagnostics in a PNNL report (Automated Diagnostics Software Requirements Specification Version 1.1) as well as a final report to the California Energy Commission (Final Report Compilation for Pattern Recognition-Based Fault Detection and Diagnostics).

The methods rely on fixed data that characterizes the equipment (e.g., rated current or power of electrical motors, temperature set points for water flows) and on metered data (e.g., outdoor ambient conditions, electrical current to fan and pump motors, and temperatures of water at various points in systems) sampled over time. Fixed data are input during set up of the software, and metered data are collected from sensors during equipment operation.

The outputs of the diagnostic processes are indicators of the condition of the various equipment. Improper operating conditions or faults lead to energy waste and accelerated wear on equipment. The major faults detectable with the diagnostic logic follow. Some of these faults have several subcategories, which isolate or specify the fault to higher resolution.

Cooling tower faults: Chiller faults: Boiler faults: For more information, see: