Identification of Energy Management Configuration Concepts from a Set of Pareto-optimal Solutions

TL;DR

In this study, for a set of 20000 Pareto-optimal building energy management configurations, resulting from a many-objective evolutionary optimization, multiple concept identification iterations are conducted to provide a basis for making an informed investment decision.

Abstract

Implementing resource efficient energy management systems in facilities and buildings becomes increasingly important in the transformation to a sustainable society. However, selecting a suitable configuration based on multiple, typically conflicting objectives, such as cost, robustness with respect to uncertainty of grid operation, or renewable energy utilization, is a difficult multi-criteria decision making problem. The recently developed concept identification technique can facilitate a decision maker by sorting configuration options into semantically meaningful groups (concepts). In this process, the partitioning of the objectives and design parameters into different sets (called description spaces) is a very important step. In this study we focus on utilizing the concept identification technique for finding relevant and viable energy management configurations from a very large data set of Pareto-optimal solutions. The data set consists of 20000 realistic Pareto-optimal building energy management configurations generated by a many-objective evolutionary optimization of a high quality Digital Twin energy management simulator. We analyze how the choice of description spaces, i.e., the partitioning of the objectives and parameters, impacts the type of information that can be extracted. We show that the decision maker can introduce constraints and biases into that process to meet expectations and preferences. The iterative approach presented in this work allows for the generation of valuable insights into trade-offs between specific objectives, and constitutes a powerful and flexible tool to support the decision making process when designing large and complex energy management systems.

Figures (8)

• Figure 1: Process of evaluating the choice of description spaces: A data set, obtained from a many-objective evolutionary optimization study in previous work Liu2022, is investigated for concepts. The additional step of selecting different description spaces is added in an iterative way to the standard concept identification Lanfermann2022a procedure.
• Figure 2: Several alternative choices for selecting the description spaces for an example with three features. The selected features and their associated description space are indicated by the color of the axis labels. Black labels indicates that the corresponding feature is not associated with any description space. The colored boxes indicate possible concepts for the different choices of description spaces for a case where two concepts should be identified. (a) A single one-dimensional description space (Investment cost in blue). (b) Two separate one-dimensional description spaces (Investment cost in blue, Resilience in orange) (c) A single two-dimensional description space (Investment cost and Resilience in orange) (d) One two-dimensional and one one-dimensional description space (Investment cost and Resilience in orange, Yearly total cost in blue).
• Figure 3: Simulation model of the energy management system: The model simulates the building power and heat demand based on time and weather conditions. Energy is provided by the grid connection, a pv system, a combined heat and power plant (chp) and a stationary battery. The battery's charging and discharging behavior is controlled depending on a predetermined control strategy and internal reference values, e.g., the overall power consumption level.
• Figure 4: (a) Identified concepts from experiment 1: The process identifies three concepts that satisfy the given requirements imposed by the two description spaces. (b) The same concepts are not reasonable for a different combination of description spaces. (c) A separate concept identification process for the second distribution of description spaces also leads to reasonable concepts.
• Figure 5: Identified configuration concepts based on four description spaces: (a) $P_{PV}$ and $C_b$, (b) $C_\text{invest}$ and $P_{p}$, (c) $C_\text{annual}$ and $E_{d}$, (d) $\bar{b}$ and $E_{f}$. The colored samples (purple, green, and yellow dots) are associated with concepts 1, 2, and 3, respectively. The grey samples are not associated with any concept. The two parameters of description space 1 (top left plot) are normalized.