New approach to staying in condition

The monitoring of differential pressure across air handling unit filters is typical of the simpler CBM applications for which systems are commonly used. When pressure breaches a pre-set limit, the BMS generates an alarm which indicates that the filter is dirty and needs to be replaced. Monitoring for a condition such as valve leakage is a similarly straightforward application.

Systems are also able to highlight progressive changes in conditions. For instance, at Seacroft Hospital in Leeds, the site’s Trend BMS uses a dynamic graphics display with a four colour ‘traffic light’ system to warn of changes in cold water tank temperatures. The latter have to be closely monitored to ensure compliance with the Legionella Code of Practice.

A BMS will often need to monitor and compare a number of different variables in order to determine plant performance. For example, for plant items like calorifiers and cooling towers, a key performance indicator is heat transfer effectiveness. To calculate this value, the system must take regular readings from various temperature sensors.

In the case of high cost plant such as chillers, CHP units and large boilers it is especially important to maintain optimum performance. Because this requires a very detailed assessment of plant efficiency it places heavy demands on the BMS. On steam boilers it would call for the monitoring of back-end, feed water and ambient temperatures plus steam pressure, gas usage and flue gas analysis. The high speed of Trend’s BMS controllers is crucial in such situations since it enables data to be continuously logged and thus allows efficiency to be calculated over a short time period.

Boiler efficiency, like certain other values, does of course vary naturally. For instance, it changes at different times in the plant’s operational cycle. Ideally, what is needed is an efficiency profile - ie, a time-based graph of expected readings which can be compared with actual measurements. This is one of the elements provided by the Trend Bureau’s newly introduced remote condition monitoring service.

The service is based on a software package (originally developed for energy monitoring and targeting) that automatically compares actual and expected profiles for a variable. When a deviation from the norm occurs, the program consults an internal ‘rulebook’ which explains the exception it has found. It then generates a report informing the end-user. Importantly, the predicted profile and rules database can be easily refined to take account of operational experience.

Uploading of plant performance data from the BMS is done completely automatically (via modem or other means).

Providing adequate data is available, the program can actually specify what maintenance work needs to be done. Were it monitoring a generator, say, and the information it was receiving included vibration analysis data, it should be able to advise when particular bearings needed to be replaced. 

The new service is likely to appeal to many system users because it means they do not have to invest time and effort setting up and implementing their own data analysis routines.