Evaluating different machine learning techniques as surrogate for low voltage grids

Abstract

The transition of the power grid requires new technologies and methodologies, which can only be developed and tested in simulations. Especially larger simulation setups with many levels of detail can become quite slow. Therefore, the number of possible simulation evaluations decreases. One solution to overcome this issue is to use surrogate models, i. e., data-driven approximations of (sub)systems. In a recent work, we built a surrogate model for a low voltage grid using artificial neural networks, which achieved satisfying results. However, there were still open questions regarding the assumptions and simplifications made. In this paper, we present the results of our ongoing research, which answer some of these questions. We compare different machine learning algorithms as surrogate models and exchange the grid topology and size. In a set of experiments, we show that algorithms based on linear regression and artificial neural networks yield the best results independent of the grid topology. Furthermore, adding volatile energy generation and a variable phase angle does not decrease the quality of the surrogate models.

Co-Simulation of a Cellular Energy System

“…Communication protocols are generally simulated with event-based simulators, while models of the physics of energy systems are usually based on differential and algebraic equations solved with continuous simulation [34]. Models for markets and market aggregators also require different paradigms for complex calculations and optimization, including machine learning [35]. Combining all these so-called domainspecific models with their different paradigms to simulate the entire cellular energy system necessitates co-simulation.…”

“…The mapping between these variables across different simulators has to be automated. To reduce the simulation time, parts of the energy network can be replaced by simplified models that describe their behavior with sufficient accuracy, e.g., using machine learning [35].…”

Accelerating Energy-Economic Simulation Models via Machine Learning-Based Emulation and Time Series Aggregation

“…It is based on correlations and interdependencies of the simulation models and aims to increase performance by enabling larger simulation setups. In a subsequent publication, Balduin et al highlighted the usage of surrogate models e.g., in the field of calculation and optimization of energy savings, the substitution of simulation models, uncertainty assessment, as well as micro-grids [20].…”

“…Reviewed papers make use of this approach, e.g., in chemistry and medicine [4,18,22,25], the automotive industry [23,24], geoscience [4,21], astrophysics [4], fluid dynamics [26], or various engineering challenges [16,18]. Use cases in the energy sector include fusion simulation [14], the heat demand of buildings [15], an urban energy simulator [17], vehicle energy consumption [28], and smart grids [18][19][20]. A simulation and emulation of large quantities of P2P communities, as performed in Section 6, has not been performed so far.…”