Newsbeitrag

What if all would like to become self-sufficient? EUF’s part in the carpeDIEM research project

It might be tempting for an operator of a local power system to become independent of the national power system, for instance due to the price development of electricity supply from power utilities....

It might be tempting for an operator of a local power system to become independent of the national power system, for instance due to the price development of electricity supply from power utilities. Becoming self-sufficient is a challenging and ambitious task and a couple of smaller villages and houses demonstrate that it is technically and economically feasible. A question is, however, how local power system and the overlaying power system fit together if the sub-system is still linked to the national power grid but reaches a certain level of energetic autarky. At Europa-Universität Flensburg (EUF) researchers have addressed that question, in particular for the showcase village of Dörpum in Germany.

 

The researchers analyse the impact of such largely self-sufficient sub-systems in the national and European power systems. A key question is how well the residual load pattern of the sub-system goes together with the load patterns and production patterns in the overlaying systems. Relevant is also what CO2 emissions are induced or avoided due to the “optimization” of the sub-system. And finally such impacts need to be evaluated economically.

At EUF, an energy system simulation model named “renpass” has been substantially revised and upgraded to give answers to the above questions. renpass is an application using functionalities of the Open Energy Modeling Framework (OEMoF) and it has already been utilized for several research projects on the future power systems of Germany, Europe and also Morocco, Tunisia and Jordan. In the model an energy system is abstracted by nodes and flows. By interlinking several system components with each other – e.g. one grid with a neighbouring grid or further power plants, loads or storages connected to the electrical grid – the entire power system can be mapped in great detail (Fig. 1)

 

Figure 1: Energy system as oemof-network with components (blue) and buses (red)

Sources: Own illustration (EUF). Created in Python with networkx, matplotlib and oemof.

 

The model allows a least-cost optimization of the operation of all system components with a high spatial and temporal resolution. This includes a detailed representation of the demand side as well as of the supply side and the transmission grid, including technical, economic and operational parameters of all individual elements of the system as well as hourly-resolved renewable production and load sequences.

The renpass model, the OEMoF model framework and all input data seek to be “open source” and they use “open data” wherever possible, meaning that they are openly and freely available to anyone interested to utilize them. The latest version of the model can be found online on the Github platform at github.com/znes/renpass.

EUF’s activity in the carpeDIEM research project so far has been the creation and evolvement of this simulation tool and further analyses tools and their adaption especially to the showcase of Dörpum within the overlaying German and European power systems. So far the German power system as of 2015 has been modelled with the newly developed model, accompanied by a simulation of the European power system, i.e.  Germany plus those countries Germany has electrical transmission links with. The so-derived preliminary results have been contrasted with the residual load curves from the optimized showcase system in Dörpum on an hourly basis as simulated by the project partners from the University of Applied Sciences Lübeck. For the comparison, three basic system states were identified that can be detected for every hour of the year of analysis:

Case a: The sub-system’s residual load supports the overlaying national or international power system. Example: Excess power from the sub-system meets shortage power in the overlaying system. 
Case b: The sub-system’s residual load is zero or nearly zero, i.e. it neither supports nor opposes the overlaying system.  
Case c: The sub-system’s residual load is opposing to what is required in the overlaying national or international power system. Example: a power shortage in the sub-system meets shortage power in the overlaying system.

The situation in the overlaying system can be described with its residual load available in different degrees, in every hour of the year, which is reflected in the assessment’s result. In some of the scenarios of the optimized sub-system, for instance, power surpluses can be found mainly around midday due to substantial power production from PV that cannot be directly used nor stored locally. The comparison with the first-degree residual load of the overlaying system – i.e. load minus all uncontrollable renewable production –  suggests that this production behaviour fits well to the national system in which power is required during the same period. The comparison with the second-degree residual load of the national power system in which the power production from dispatchable technologies is taken into account might differ from that assessment. In Figure 2 different sequences during the first days of the year of analysis are exemplarily depicted and related to each other. For display reasons, all values in the diagram have been normalized with their annual maximum. The diagram indicates that the first-degree residual load curve of the national power system and the spot market price for electricity practically function in parallel, which can be interpreted as the modelling approach in the renpass model being appropriate to the task and realistic. 

 

Figure 2: Exemplary sequence comparison for the first four days of the simulated year 

Sources: EEX 2018, University of Applied Sciences Lübeck, own calculations with renpass

 

The relation between different modelled sequences of the sub-system and the overlaying system can be illustrated in different ways. In Figure 3, two sequences are related to each other and displayed in the form of a heat map. In the diagram the abscissa represents the days of one year whereas the ordinate represents the hours of a day. Areas marked red are hours in which the sub-system opposes the overlaying system. Areas marked blue are moments in which the sub-system supports the overlaying system. Areas marked white represent rather neutral states of the sub-system. The diagram exemplarily shows a case in which the sub-system supports the overlaying system mainly during daytime while it opposes the overlaying system at night time. This can be explained with a net power demand at night and net power surpluses due to power generation from PV during daytime. The heat map also visualizes the change of seasons, i.e. differences of the sunshine duration in the course of the year. 

 

Figure 3: Exemplary sequence comparison (heat map) 

Coloured: relation between the optimized sub-system and the national power system (red: opposing; white: neutral; blue: supporting)  
Sources: own calculations.

 

Besides the analysis of how well or bad the sequences and systems fit together, the EUF researchers also analyse the resulting CO2 emissions and reductions of CO2 emissions, respectively, induced in the overlaying systems. By contrasting the sub-system’s residual load on the one hand and the operation of dispatchable technologies in the overlaying national and international systems on the other, the amount of induced or avoided direct CO2 emissions in the power system is calculated on an hourly basis. This approach is based on the idea that a potential positive residual load in the sub-system requires power from the overlaying system, thus power generation in the overlaying system, which again induces direct CO2 emissions. Vice versa, emissions can be avoided if the sub-system produces surplus power and power production in the overlaying system can be curtailed. As the model finds the system state with the least cost for every hour of the year, the utilization of dispatchable technologies in the overlaying system changes on an hourly basis due to different fuels, efficiencies and costs. The specific direct CO2 emissions of the operational plant mix are therefore variable by the hour. All induced or avoided CO2 emissions based on the sub-system’s residual load pattern in the overlaying system, however, can be added up and compared with other scenarios. Preliminary results show that all scenarios in which the sub-system is different from its 2015 original status might avoid direct CO2 emissions in the overlaying system.

In order to have a fair comparison of the scenarios it is necessary not only to compare the residual load sequences of the systems and to calculate induced or avoided CO2 emissions but it is also necessary to include economic parameters in the assessment. An economic evaluation therefore will show which of the sub-system’s scenarios might be most beneficial in terms of system support to the overlaying system and CO2 abatement cost. Such research outcome is expected to be completed and available by the end of 2018.

Contact: Sönke Bohm, EUF