First automated PV and storage recommendations in the SmartFarm 2 project

 in Press release

Owners of agricultural businesses or other SMEs in rural areas face insurmountable challenges

The need to increase the share of renewable energies in energy generation is hardly questioned any more. Not least due to the current events in Ukraine. But how this can be implemented in specific cases is an unsolvable question for most decision-makers. The SmartFarm 2 project is taking a closer look at farms in particular, but also at SMEs in rural areas. The owners of these companies are faced with the decision of whether to invest in renewable energies and, if so, to what extent:

  • Should it be the 10 kWp PV system or rather the 30 kWp, which is just about exempt from the EEG levy, or even the 100 kWp system?
  • Is it worth investing in an additional battery storage system to temporarily store the surplus energy?
  • Existing PV systems will no longer be eligible for EEG remuneration in the next few years. Should the systems continue to be operated? How can the electricity generated be used?

If you want to answer these and many other questions, you always have to ask yourself: "How much of the electricity I generate can I use myself?" Feeding self-generated electricity into the grid is hardly profitable. The aim must be to consume as much of the self-generated electricity as possible so that less has to be purchased from the public grid. On the one hand, this increases the financial advantage for the companies and, on the other, decentralized energy generation and use helps to relieve the burden on the public grid.

And what do mathematicians have to do with it?

In the SmartFarm 2 project, a team of experts from the field of AI and optimization, together with experts in the field of measurement sensor technology and data transmission, is developing a software tool that automatically calculates the optimal dimensioning for a PV system or battery storage system.

With the help of measurement sensors from the partner Enerserve GmbH, high-resolution measurement data of a company's consumption is recorded and collected in a database using the expertise of the technicians at the University of Bremen. Based on this data, mathematical algorithms from the partners Steinbeis Innovation Center for Optimization, Control and Regulation and the Optimization and Control working group at the University of Bremen are used to calculate the optimum system sizes. Various factors are taken into account, from acquisition and maintenance costs to electricity prices and inflation over the next 20 years. Our partner Q3 Energie GmbH contributes the relevant know-how to the software.

The development in recent years that more and more information can be obtained from data is also being utilized in this project. The data-based approach using the high-resolution consumption data of the companies offers the previously unused opportunity not only to take into account the total consumption over a year, but also to include daily and seasonal fluctuations in the calculations. This is of great value for the dimensioning of a battery storage system, for example, as this also has a daily charging and discharging cycle. In addition, the optimum battery storage size is directly related to the PV energy generated, which varies greatly over the course of the year, so this must also be taken into account.

Status report and open discussions at the second milestone meeting

The second milestone in the SmartFarm 2 project was reached at the beginning of February. In a corresponding meeting with external guests, the first results of the recommendation software were presented using three real examples from the test field in the SmartFarm 2 project in the districts of Osterholz and Verden. An initial comparison with a conventional recommendation for a PV system and a battery storage system based on empirical values is very promising.

The status report was followed by a discussion with the external guests and further questions were discussed. The online meeting was a complete success for both the project participants and the guests.

Chip shortage and corona: delays in setting up the test field

The aim of the SmartFarm 2 project is to set up a test field with numerous demonstrators. This means that at least one measuring device should be installed in these companies or municipal facilities to record total electricity consumption, which can then be used to draw conclusions about individual consumers and producers using AI. After a very successful call for participation in the project, the waiting list of interested companies is large. Unfortunately, there are delays in the installation of the measuring devices. Due to the Covid-19 pandemic, there are recurring illnesses and quarantines and the global chip shortage is causing a continuing hardware shortage in the field of sensor technology. Despite the delays in the installations, however, the previous objectives were successfully implemented and the software development was carried out as planned using the first available real data.

Plans for the next two years

In line with the good progress made, the achievement of the next goals in the project is also viewed optimistically. In addition to the further development of the recommendation software, the focus over the next two years will be on developing an energy management system. So if you already have your own generation systems installed, the existing consumers and, if applicable, storage systems should be controlled in such a way that as much of the self-generated electricity as possible is also consumed during the course of the day. With the help of mathematical expertise in the field of control and regulation, this can also be automated and optimized. Another new feature is that decisions are made dynamically and are not tied to fixed rules, as is the case in current energy management systems. The more complex the systems become, the more likely it is that these rule-based systems will reach their limits, but conversely, mathematicians are delighted with the greater potential for optimization. This is where the AI-based, mathematical approach comes into play, meaning that the SmartFarm 2 team is already working on tomorrow's solutions today.