HOT|COOL NO. 3/2019 - "Big Markets: China and Poland

P27

INTELLIGENT DATA ANALYSIS To help the efficient analysis of smart meter data, intelligent algorithms and so-called machine learning are used. A computer program learns the data patterns and signatures of well-functioning installations and networks and - based on this - points out anomalies or deviations from predicted values. Although the true physical relationships between weather, consumption, mass flow, water temperatures, etc. remain unknown, intelligent analysis methods have proven to be successful in grasping very detailed trends, such as in careful predictions of the heat load. By wider use of these intelligent tools, we are facing a general change from physical models towards a data-driven understanding of the district heating network. PERSPECTIVES A complete exploitation of smart meters, smart meter data and additional measurements will lead to a number of challenges that need to be overcome in the years to come. Supported roll-out of smart meters internationally is still needed, and with regard to the heat meter interfaces, documentary standards are not available. Energy suppliers are forced to build their own data transfer infrastructure for meter reading. In line with this, requirements on data time resolution, completeness, i.e. occurrence of missing values, presence of incorrect meter readings and sensor breaks, as well as quantification of the associated uncertainties, should be established.

To promote the interest and attention on these issues, there is an urge for demonstrations and proven examples of business cases, including development of automated fault detection supervision systems to detect anomalies in the networks and in the buildings’ sub-stations. Over the next few years, efforts and technological investments will reveal some of the large potential of data-driven heating and energy systems. Exactly how much value that can be retrieved held up against the expenses of smart meter installations, as well as establishment and operation of robust data collection systems, is still a question that awaits answers and pioneering work.

For further information please contact: Jakob Fester, jafe@teknologisk.dk

This article was supported by a grant from Danish Technological Institute’s performance contract 2018-2020, entered with the Danish Agency for Institutions and Educational Grants under the Ministry of Higher Education and Science.

NIRAS Energy Private as well as public investors prefer NIRAS’ international experience within district heating and biogas plants. This applies to both renovation of existing plants and establishment of new plants. We provide expert consulting through- out the entire process – from business plan and authority approvals to design and tender documents. We supervise the construction phase and during commissioning.

Expert consulting – all the way from idea to commissioning PURE ENERGY

www.niras.com

Bioenergimagasinet_nr_3_05_2018_190x130_ASM_NBA_VIG_UK.indd 1

09-05-2018 08:11:14

www.dbdh.dk

J O U R N A L N 0 . 3 / 2 0 1 9

Made with FlippingBook - Online Brochure Maker