Back to results
US 8,731,724 B2Granted
Automated fault detection and diagnostics in a building management system
AssigneeJohnson Controls Technology Company
InventorsKirk H. Drees, James P. Kummer
Life of this patent
Estimating the term…
Abstract
Systems and methods for building automation system management are shown and described. The systems and methods relate to fault detection via abnormal energy monitoring and detection. The systems and methods also relate to control and fault detection methods for chillers. The systems and methods further relate to graphical user interfaces for use with fault detection features of a building automation system.
Claims21
A controller for presenting an estimate of fault cost in a building management system comprising:
- a processing circuit configured to maintain a history of energy usage values for the building management system and a threshold parameter establishing an energy usage value limit relative to the history of energy usage values;
- and
- wherein the processing circuit is further configured to identify energy outlier days by comparing energy use for a day to the history of energy usage values and the threshold parameter;
- wherein the processing circuit is further configured to subtract an estimated average daily cost, calculated with the outlier days, from an estimated average daily cost calculated excluding the outlier days to estimate an additional cost of the energy outlier days;
- wherein the processing circuit is configured to assign the estimated additional cost of the energy outlier days to the estimate of the fault cost in the building management system;
- and
- wherein the processing circuit is configured to output the estimated fault cost to at least one of a remote device and a local electronic display.
4 more elements
A computerized method for detecting chiller faults, comprising:
- comparing an actual performance of a chiller to an expected performance of the chiller, wherein the comparison comprises calculating an error between the actual performance and the expected performance;
- applying results of a statistical analysis to the comparison to identify a deviation from the expected performance, wherein identifying a deviation from the expected performance comprises:
- (1) maintaining an exponentially weighting moving average of the error, and (2) comparing the exponentially weighted moving average to a threshold;
- determining whether to identify the error as associated with a chiller fault based on whether the exponentially weighted moving average exceeds the threshold;
- and
- generating an output signal indicative of a chiller fault in response to the exponentially weighted moving average exceeding the threshold.
3 more elements
Dependencies are inferred from the claim text.