A First Look at our Basic Building Blocks for Urban Energy Simulation

[prudhomm]

A First Look at Basic Building Blocks for Urban Energy Simulation

Authors: L. Berti, V. Chabannes, J. Cladellas, A. Diallo, M. Maskek, P. Pincon, C. Prud'homme
Cemosis, Université de Strasbourg

Introduction

This article delves into the foundational aspects of building modeling at Level of Detail 0 (LoD0) as part of our Ktirio platform for building energy simulation, a component designated at Cemosis for its streamlined representation of buildings as bounding boxes with flat roofs. Within the framework of Ktirio, we explore the construction and parameterization of our model, demonstrating the platform's capability to automatically generate models with varying numbers of floors. We further illustrate the practical application of our LoD0 building model through an energy simulation of Strasbourg's city center. Additionally, we provide an overview of the subsequent levels of detail, LoD1 and LoD2, which introduce more complex modeling approaches, thereby enriching our platform's versatility in addressing diverse urban energy simulation needs.

Model construction and parameterization

The structural components of our LoD0 building model include:

• 1 slab on the ground floor;
• 4 walls per floor;
• 1 slab between each floor;
• 1 window per floor;
• 1 door on ground floor;
• 1 flat roof;

The walls are composed of several layers of materials and insulation. The principal material compositions are as follows:

• Bottom slab : [styrofoam, filler, heavy concrete, mortar, floor tiles]
• Intermediate slabs between floors : [plaster board, filler, heavy concrete, mortar, floor tiles]
• Walls : [external coating, bricks, wood wool, air, plaster board]
• Roof : [mortar, heavy concrete, filler, plaster board]

Our thermal simulation model accounts for various functionalities, including dynamic thermal comfort parameters such as clothing insulation, metabolic rate, ambient humidity, level of activity, air speed, air renewal per hour, shading masks according to sun position, and ideal heating and cooling systems. These factors demonstrate the model's ability to reflect real-world conditions and their impact on a building's energy consumption.

Modelica Schematic of our Building Model with a unique floor:

Automatic model generation with multiple floors

Once a building model has been converted to fmu format, its geometric parameters (height, width, volume, thickness) and comfort parameters can be parameterized and modified, but the building structure cannot be modified. The exact number of walls, windows, thermal zones, roof shape, ... are intrinsic components to the model. They define the structure of the model. If we want to use our model but with one more window, we need to regenerate a new fmu. The latter will be a tiny variant of our building model.
For LoD0, each floor of the building contains a single thermal zone. That's why we need an automatic model generator which allows to create new variants of our model by slightly modifying its structure.

To do so, we use modelica template files of our model, parameterizing all the structures in our model that we want to vary.
We use the python-liquid library to generate files from template files.

Template files

• BuildingModel.mo : file describing the building structure;
• ParameterMap.mo : file describing all the parameters used in BuildingModel.mo which can modified bu the user when initializing a simulation;
• BuildingApplication.mo : file that instantiates the building model, parameter map, weather, dynamic parameters such as solar shading, comfort parameters and connection equations between weather and building;
• simulatordesc.json : file describing model inputs and outputs for energy simulation.

Program file : writeMoBuildingModel.py

• Parameters :
  • number of floors (number of walls and thermal zones adapted to the number of storeys);
  • if the building is insulated or not (the number of layers to match the building's thermal insulation);

Illustration of the algorithm adapting the multi-storey model:

Simulation case of Strasbourg city center (LoD0)

One of our test cases is downtown Strasbourg. A bounding box is assigned to each building. Depending on the height of the building, we assign a number of thermal zones according to the number of floors. A floor is at least 3 meters high.
So far, we've generated around ten fmus for our LoD0 model, for buildings ranging from 1 to 10 storeys. We can of course generate as many fmus with as many storeys as desired, without the acceptable limit for the number of storeys a building can have.

indoor and outdoor temperature in about 17000 buildings in Strasbourg over 2 months simulated by EuroHPC JU Discoverer (click on the image to watch the video)

Other Detail Levels

Level of Detail 1 (LoD1)

The LoD1 model defines buildings with a more detailed structure, constructing the geometry from its footprint, height and actual number of floors; where each floor corresponds to a thermal zone. The roof shape is selected from a set of predefined roof types (flat, gable, hip, conical, etc). The geometry of this model includes windows and doors, and each construction element has a thickness and a material composition. The building's information is requested from the national database Base de données Nationale des Bâtiments (BDNB). Additionally, missing information on building materials is completed using a machine learning algorithm.

Illustration of a building from the LoD1 model, viewed using the Solibri software:

Level of Detail 2 (LoD2)

The LoD2 model defines buildings with a maximum of detail, whether for roofs, façades, interiors, staircases. This will enable us to observe heat loss in detail, identify the key points of a building's insulation, etc.

Illustration of a building from the LoD2 model, viewed using the Solibri software:

Conclusion

In conclusion, this article has shown some of the foundational aspects of urban energy simulation through the lens of building modeling at various levels of detail, starting with the most basic Level of Detail 0 (LoD0). By presenting a structured approach to model construction, parameterization, and automatic generation, we have laid the groundwork for understanding how such simulations contribute to energy efficiency and thermal comfort in urban environments, particularly in the context of Strasbourg city center.

The transition from LoD0 to more intricate models like LoD1 and LoD2 signifies not only an increase in geometric and material detail but also a deeper insight into the thermal dynamics and energy consumption patterns of buildings. The methodology outlined for generating building models automatically allows for scalability and adaptability, catering to buildings of varying sizes and complexities with ease.

The case study of Strasbourg city center under LoD0 highlights the practical implications of our approach, demonstrating the potential for widespread application in urban planning and energy management. Future developments, as hinted with LoD1 and LoD2, promise even more precise simulations, incorporating detailed architectural features and advanced thermal properties.

As we move forward, the integration of machine learning and national databases like the Base de données Nationale des Bâtiments (BDNB) for refining building information will further enhance the accuracy and reliability of energy simulations. This journey from basic building blocks to detailed urban landscapes encapsulates a significant leap towards sustainable urban development, where energy efficiency and thermal comfort are not just ideals but achievable realities.

In essence, our exploration of building modeling for energy simulation stands as a testament to the evolving capabilities in urban planning and energy management. It underscores the importance of detailed simulations in designing energy-efficient cities, offering a roadmap for future research and application in this vital field.

Funding Support

Ktirio Urban Building is supported by the following organizations:

• Europe (EuroHPC JU)
  Funded by the European Union. This work has received funding from the European High Performance Computing Joint Undertaking (JU) and Poland, Germany, Spain, Hungary, France, Greece under grant agreement number: 101093457. This publication expresses the opinions of the authors and not necessarily those of the EuroHPC JU and Associated Countries which are not responsible for any use of the information contained in this publication.
• ANR and PEPR NumPEx
  Ktirio Urban Building has received funding from the Agence Nationale de la Recherche through the PEPR NumPEx programme.
• AMIES
  Ktirio Urban Building has received funding from the Agence des Mathématiques en Interaction avec les Entreprises et la Société through the PEPS programmes.
• CNRS
  Ktirio Urban Building has received funding from the Centre National de la Recherche Scientifique through the prematuration programme.

Partners

Ktirio Urban Building is developed by the plateform Cemosis at the University of Strasbourg.

The following partners are involved in the project:

• Synapse Concept
  Synapse Concept is a technology company specializing in creating innovative solutions that facilitate smart building management and urban development. Their focus is on integrating advanced IoT technologies and data analytics to enhance the efficiency, sustainability, and comfort of urban buildings. Synapse Concept’s expertise in smart sensors, data processing, and AI-driven insights plays a crucial role in modern urban building projects.
• Cisco Meraki
  Cisco Meraki offers cloud-managed IT solutions, known for their simplicity and efficiency. In the context of urban building, Meraki provides robust networking solutions, including wireless, switching, security, and smart camera systems. Their technology is pivotal in ensuring reliable and secure connectivity, essential for managing modern urban buildings. Meraki’s solutions are characterized by their ease of deployment, management, and scalability, making them a fitting choice for smart city projects.

The Ktirio Urban Building project is part of the CoE Hidalgo2 and benefits from the expertise of the following partners: PNSC, HLRS, ATOS, SZE, ICCS, MTG, FN.