HTW’s “Energy Storage Inspection 2022” (ESI22) assessed a relatively large number of BESSs, including established vendors such as BYD and Sonnen, but also new products from Fenecon, Solax, Varta and others. The program focuses on residential battery systems.
All BESS manufacturers were invited to participate in ESI22. 14 manufacturers participated in the comparison of their systems with measurement data from 22 systems. Laboratory tests have been carried out by independent test institutes in accordance with the “Efficiency guideline for PV storage systems”.
Along with product quality and warranty, system efficiency is one of the top five criteria for purchasing a certain storage system. Unfortunately, comparable specifications in data sheets are rarely seen. The objective of the annual storage classification is to allow a better comparison between the different products.
Since the first HTW inspection in 2018, more than 60 different storage systems have been compared with measurement data from independent test institutes. The results show that many manufacturers have identified and used key mechanisms to improve the effectiveness of their products.
Thanks to these improvements, increasingly efficient PV energy storage systems are now available on the market. Homeowners who are eager for more independence from rising electricity prices are benefiting from this development. The lower the losses of the system, the higher the degree of self-sufficiency of the owner for his electricity supply.
Around 93,000 PV systems were registered in the German Energy Market Data Register in 2019, increasing to 162,000 in 2020. The market for battery systems with a capacity of up to 20 kWh is showing growth significant similar. The number of registered storage systems doubled between 2019 and 2020, reaching 82,000 installations.
Going forward, in 2021, more than 200,000 PV systems and 130,000 battery storage systems have been installed. In 2019, only 37% of new photovoltaic systems were installed alongside a battery storage system. This number has increased to 56% in 2021. In addition to this, an increasing number of private photovoltaic system owners are deciding to add storage to their existing photovoltaic system. The Energy Market Master Data Register shows that in 2021, more than 16,000 existing PV systems received a storage upgrade. In 2019, a combined 724 MW was installed. This number nearly doubled in 2020 to 1,285 MW and increased further to 1,697 MW in 2021.
Residential storage systems were rated using the System Performance Index (SPI) metric. This is done by simulating the operational behavior of the battery system over a period of one year. The results of a real PV storage system and a theoretical lossless system are compared to calculate the SPI.
Thus, comparable to the performance ratio (PR) of a PV system, the SPI evaluates the general efficiency of a solar plus storage system. Two independent baseline scenarios were used to compare different system sizes. These two scenarios naturally differ in the size of the PV system (5 kW and 10 kW), household electricity demand and its composition (5010 kWh/a and 9362 kWh/a). The size of the PV system is responsible for the two chosen names SPI (5 kW) and SPI (10 kW).
12 systems were evaluated by SPI (10 kW) as part of the ESI22. An abbreviation (eg A1) has been assigned to each system analyzed in the report. Two manufacturers chose to participate anonymously. The G2 system managed to achieve the highest SPI (10kW) with 95.1%. The J2 system only achieved an SPI (10kW) of 89.1%, mainly due to its high conversion and standby losses.
Every conversion of energy with power electronics and battery storage leads to losses. The battery efficiency of the analyzed systems ranges from 93.3% to 97.8%. The average efficiency of the lithium-ion batteries in the study was 95.8%. While the most efficient systems manage to achieve lane efficiencies of up to 97%, less efficient systems often have lane efficiencies below 88%.
With the model-based approach, various loss mechanisms of a battery system can be isolated. The graph (right) shows a detailed overview of the different losses with the D2 system. As the system with the second highest SPI result, the D2 has only 2.9% higher conversion losses – 0.5% less than the conversion losses of the best placed G2 system.
The J2 system has the highest conversion losses at 7.2%. This system has higher conversion losses than the absolute losses of the top six ranked systems, G2 through E2. These losses are mainly the result of low conversion efficiency during discharge. Even though D2 has outstanding overall performance, system control losses are comparatively high at 0.9%. In comparison, G2 has control losses of slightly more than 0.2%. Even greater losses due to high control deviations can be found in the J2 (1.1%) and B1 (1.4%) systems. The D2 system has relatively low standby losses at only 0.4%. J2 also has the highest standby losses at 1.7%. D2’s below-average losses in almost all loss categories result in an extremely high SPI (10kW) of 94.7%.
Loss of control and standby
The common point of the various sources of loss is that they affect the energy flows between the building and the network. Dynamic and/or stationary control deviations lead to an undesirable increase in grid injection. Preventing this is the main source of loss of control of a PV energy storage system.
Independent test institutes have analyzed the behavior of the system in response to fluctuating power demand with a step response test. Manufacturers focus on optimizing system control. Average settling time was 8.1 seconds in 2018 and dropped to 4.2 seconds in 2022. Exceptionally well-behaved systems can control power fluctuations within hundredths of a millisecond. But there are still products that offer a settling time as long as 12 seconds with a three second dead time.
Another cause of energy loss is the standby power consumption of a system with a charged or discharged battery. The magnitude of this loss depends on the power demand and the time the system typically spends in a fully charged or discharged state. A PV battery spends between 2,000 and 4,000 hours annually discharged but only 1,000 to 2,000 hours when fully charged. Other components such as power meters must be supplied with electricity throughout the year.
Efficient systems can achieve less than 2 W of losses resulting from standby power consumption, but one of the analyzed systems has a standby power consumption of more than 70 W. The average standby power consumption of all systems was 20 W in 2021. Over the past five years, the standby power consumption of the analyzed systems has increased. This trend is the result of the installation of higher capacity inverters and batteries. The results of the storage comparison show that many systems have great potential for optimization in this regard.
The degree of self-sufficiency that can be achieved is determined by the share of a household’s electricity demand that can be covered by the rooftop photovoltaic system. The graph (left) illustrates the self-sufficiency that can be achieved by the assessed systems. In addition, the degree of self-sufficiency for an ideal system without losses is calculated, which is indicated by the green line. The line drawn illustrates how the degree of self-sufficiency is related to battery size. Efficiency losses reduce the degree of self-sufficiency. The greater the distance between the theoretical maximum and the actual system at the same capacity, the greater the losses.
This demonstrates that increased battery capacity does not necessarily lead to higher self-sufficiency scores. For example, if a 10 kWh battery system is coupled with a 10 kW rooftop PV system, the maximum degree of self-sufficiency that can be achieved is 57%.
Looking at the battery systems tested, F1 and D2 have only a minor difference of 2.5% between the theoretically achievable and actual degree of self-sufficiency that can be achieved. The high losses that occur on the B2 system result in a degree of self-sufficiency 5.1% lower than the theoretical maximum. The G2 system achieved a higher degree of self-sufficiency than the larger and less efficient J2 system by almost 3 kWh. This shows that system efficiency can have much more influence on self-sufficiency than can be achieved with respect to the usable battery capacity metric.
About the authors
Lucas Meissner is currently completing his bachelor’s degree in Renewable Energy Engineering at HTW Berlin. As a working student, he is part of the research group on solar storage systems.
Johannes Weniger started the annual energy storage inspection in 2018. Since 2013 he has been part of the solar storage systems research group at HTW Berlin. Johannes Weniger holds a doctorate from the Technical University of Berlin.
Nico Orth is a research associate in the solar storage systems research group led by Prof. Volker Quaschning at the University of Applied Sciences (HTW) Berlin. As part of the Perform research project, he analyzes and compares the performance of PV storage systems.
This content is copyrighted and may not be reused. If you wish to cooperate with us and wish to reuse some of our content, please contact: firstname.lastname@example.org.