The research focus "Intelligent Energy Monitoring" is at home in the context of Smart Metering with Smart Grids. Past efforts to reduce energy consumption were hindered by the fact that the general public has little awareness for their actual energy consumption. This is mainly due to the fact that the user until now could only see the total consumption and the change of this over years or months. The user has to interpret the data himself with great difficulty in order to work out measures to reduce the amount of energy used. Willingness to take this approach is understandably minimal.

The research focus aims to develop a system which will automatically separate the energy consumption of individual devices from the total consumption, to show the user how much energy is used by each device. In doing so, the idea is not to equip every energy consuming device with a measuring device which would cause more work and drive up the installation costs. Instead measurements will be made at a central point, at the building entry point.

This would lead to a major change in the visualization and the association of electrical energy consumption for the user. One example for this change is shown in Picture 1 below. The electricity bill of the future shows the consumption of each individual piece of equipment. Thus they can see in detail what costs were generated by which device. With this information private and industrial customers can for the first time become aware of the costs connected to their equipment and how long each device is used daily.

Picture 1: The electricity bill of the future




The research focus is at the core of the "Non-intrusive Load Monitoring" (NILM) research area. Also spoken of, normally in connection with the environment of private homes, is "Non-intrusive Appliance Load Monitoring" (NALM). NILM algorithms are at the basis of intelligent energy monitoring. They recognize and classify devices in the measurement of the total electricity consumption. The various methods are grouped at this stage by the type of characteristics used.

The first approach for a NILM system was developed by George Hart at the Massachusetts Institute of Technology (MIT), who worked on the analysis of the load in private homes at the end of the 80s and beginning of the 90s. With his measurements, active and reactive power P and Q, was recorded in second intervals. This resulted in a rectangular progression in the ideal case of the recorded power depending on the switching status of the device. The switched on user is self-contained by using the size of the power jump. In order to recognize several devices, for every activation operation of a device the corresponding flank when switching off must be registered.

Modern approaches are based on a linear process chain consisting of Event, Desertion, Extraction, Classification and Load Tracking blocks (see Picture 2).

Picture 2: NILM Processing Chain


The ISS Research Institute has been involved for several years in the development of algorithms and processing structures for intelligent monitoring. In the ZAFH-AAL research project, the algorithms which are developed are used for the recognition of dangerous situations.