Our energy system tool develops cost-optimised, reliable and sustainable energy systems. For the valid optimisation of highly complex energy systems, the planning tool takes into account high-resolution temporal and spatial data and integrates them using modern aggregation methods. In this way, we combine different generation, storage, sector coupling, transmission technologies and the temporal fluctuation of energy availability and energy demand over several years in our optimisation. Furthermore, the optimisation takes into account the change from the existing to the future energy system for intelligent and valid investment decisions.
We have identified five megatrends that are disruptively changing the energy sector and that are often missed by conventional energy system planning:
In the last 9 years, the share of photovoltaics and wind in Germany’s gross electricity generation has increased by 400%. In practice, however, conventional static methods (e.g. merit order) are often used, but they do not adequately design expansion capacities, locations and times for either non-dispatchable generations or storage systems.
The number of decentralised power generators in Germany already exceeds 1.5 million solar, 27,000 wind and 9,000 biogas power plants. Nevertheless, the usual single-node representation of the energy system does not provide a high geographical resolution. No high geographical representation lacks representation of decentralisation and network bottlenecks.
In private households, we are observing a massive expansion of 182,000 photovoltaic home storage units in 2019 alone. Here too, the use of conventional methods leads to avoidable misjudgements and investments.
Interconnecting the electricity, heat and transport sectors is a core EU strategy for the cost-effective decarbonisation of the energy system. Nevertheless, conventional energy system planning often looks exclusively at the electrical energy system and thus misses the potentials of sector coupling.
While we have seen a 90% reduction in photovoltaic prices over the last nine years, the CO2 price has risen by 500% in the last three years alone. However, it is common practice to look at individual investment years and thus fail to take long-term effects of future technology and emission prices into account.
Our progressive optimisation focuses on these megatrends.
Emission reduction targets and falling prices for renewable generation and storage technologies are leading to structural changes in energy systems. This increases their dependence on weather conditions. The planning of future energy systems is supported by mathematical simulation and optimisation models, which are, however, limited by available storage and the time required for solving them. This leads to a balancing act between modelling of temporal complexity, spatial complexity and mapping of physical properties and ends in compromises.
To enable more detailed physical and spatial representation, it is common practice to reduce the temporal complexity of models by aggregation. For this aggregation, representative periods are usually modelled instead of the complete time series. We have shown that the aggregated representation of wind time series to representative days reduces the reliability of the energy system, leads to considerable distortions and underestimates the total costs. This phenomenon is particularly relevant for simulations and planning of energy systems with low emission limits. We find that the required capital costs are underestimated by more than 35% and the emissions using three weeks on representative days are more than twice as high as the planned emission limit.
Neither an increase in the number of representative periods, nor any other representation of the clusters, nor any different attribute weighting can solve the three problems mentioned. We recommend not to use time series aggregation to representative days to design energy systems with low emissions, especially if wind resources are important. Planning should consider cross-sectoral different technologies as well as high-resolution weather, consumer and grid data in time and space.
The solutions offered by the market so far are no longer up to date because they use less adaptable and outdated algorithms, depend on expensive local computing power and require time-consuming data preparation.
We look forward to explaining our value for your business in a personal conversation and present you a demo!