P26
VALUE FROM DATA Despite the issues with the current meters, a lot of value is still to be harvested in the available data. Depending on the time resolution of the measured parameters, a number of analytical methods may be applied to mine the data for valuable information and to look for typical signatures of the installations and networks. The percentage of sub-stations requiring service in analyzed networks is surprisingly high. As an example, a recent study monitoring Swedish sub-stations revealed 9 faults on temperature difference appearing in a group of 140 sub-stations during only one year, translated to a fault frequency of 6.5 %. Actively using data from the smart meters can help identify such sub-stations, allowing for the initiation of corrective measures, thereby optimizing the entire distribution network. Where the billing function in the meter needs a single value once a year, the frequency of the readout of the installed smart meters is always considerably higher, ranging from daily averages to measurements every hour or even every minute. This factor must be considered in relation to the purpose of the analysis to be performed. Very detailed information at the scale of a minute allows us to e.g. inspect collective peak loads and distinguish between domestic hot water consumption and the overall heat load at the highest precision. Hourly values often contain sufficient information for customer classification and reveal for example the operation control system of installations and specific customer habits. On the other hand, daily averages are still useful in combination with outdoor temperature in order to determine if a sub-station requires technical service. In all cases, anomalies and outliers that require a physical inspection at the address can be identified, and active measures can be conducted.
ADDITIONAL SENSORS Although much value can be found in flow and temperature data from the meters, there are restrictions, especially because of the low time resolution. Getting higher frequency measurements from the meters is not necessarily a feasible approach, as some meters run on battery where no electrical outlet is available nearby. High time resolution may drain the battery faster than desired, requiring frequent maintenance to change battery packs.
Figure 1: How much data and of which time resolution? The sharp peaks from domestic hot water consumption are not visible in data of 1-hour resolution.
Because of this, it may be relevant to install additional sensors in the network at critical locations and thereby retrieve additional readings with at a higher frequency. Pressure is a relatively straightforward parameter to measure in a water-based system, such as a district heating network. Furthermore, pressure sensors can be easily installed in most homes where outlets are already available. The setup of additional sensors makes it possible to obtain information on pressure pulse propagation through the system, vibrating valves, and fast leak detection. As such, the sensors may be used both for looking at the individual installation and the overall network, especially in the remote branches of the system where low pressure may be an issue. Here, the pressure characteristics also hold key information about the dimensioning of the district heating system and will provide useful information for decisions and evaluations of renovating activities.
Figure 2: Data with a high time resolution obtained by using an external pressure sensor, measuring at 10 Hz, can show pressure fluctuations in domestic installations when valves are opened and closed.
E N E R G Y A N D E N V I R O N M E N T
Made with FlippingBook - Online Brochure Maker