KIT.TC1 FlexOffice Thermal MPC
KIT.TC1 FlexOffice Thermal MPC
KIT.TC1 FlexOffice Thermal MPC Test CAse
KIT.TC1 FlexOffice Thermal MPC - overview
Test Case Overview
Author / organization: Alexander Engelmann / KIT
Test objective: characterization
Object Under Investigation: Deliverable D4.2 (Appendix E, page 48) defines the OF and KPIs that are to be minimized.
System configuration: FlexOffice
Use case: UC11 – Minimize heating, cooling and electrical energy consumption via Model Predictive Control
KIT.TC1 FlexOffice Thermal MPC- description
Short Description
The purpose of this test case is to characterize the effect of a Model Predictive Controller (MPC) on the thermal behaviour of the FlexOffice SC. As the controller is based on optimal control, optimization is used within the controlling process. In contrast to e.g. sizing of components, an operational optimization is performed, meaning that the optimal inputs to the system are computed. In other words, the result of the optimization is an optimal input sequence, not a specific parameter or location of a component. The question arises, how the system performance improves when using MPC. Therefore, KPIs and the objective function developed in D4.2 are evaluated to assess the controller performance. As it is hard to evaluate these values directly due to the absence of a physical meaning, the closed loop performance between a base case with no control, i.e. u=0, to the closed-loop trajectories when using MPC with different stage costs is compared.
Media Gallery
KIT.TC1 FlexOffice Thermal MPC - details
Test Case Details
| Test objective name | KIT.TC1 FlexOffice Thermal MPC | ||
| Author / organization | Alexander Engelmann / KIT | ||
| Short description | This test case evaluates the performance improvement of the thermal behaviour of FlexOffice in case of using Model Predictive Control. | ||
| Present use / development status | V1.0 |
| Test objective | characterization | ||
| Description incl. justification | The purpose of this test case is to characterize the effect of a Model Predictive Controller (MPC) on the thermal behaviour of the FlexOffice SC. As the controller is based on optimal control, optimization is used within the controlling process. In contrast to e.g. sizing of components, an operational optimization is performed, meaning that the optimal inputs to the system are computed. In other words, the result of the optimization is an optimal input sequence, not a specific parameter or location of a component. | ||
| System configuration | FlexOffice (see deliverable D2-3, Appendix F) | ||
| Use case | UC11 – Minimize heating, cooling and electrical energy consumption via Model Predictive Control (see deliverable D3.1, Appendix 5.10, page A-31) |
| System Under Test |
The FlexOffice system configuration (see Figure E.1) contains several thermal devices (see Figure E.2) such as radiators, a concrete core activation system and a heat buffer. Furthermore, we have a heat pump as conversion device from the electrical to the thermal domain. All components belonging to the thermal domain are tested simultaneously here.
| ||
| Object Under Investigation | The MPC for thermal building control. | ||
| Function Under Investigation | The function under investigation is the ability of the MPC to lower the fluctuation in energy demand while keeping a certain level of comfort. |
|
| |||
| Objective Function / Target Metrics | Deliverable D4.2 (Appendix E, page 48) defines the OF and KPIs that are to be minimized. More specifically, that is the level of comfort
and the fluctuation in energy demand
In the objective function these KPIs are then considered via a weighted sum as
A similar stage cost is considered within the MPC scheme. | ||
| Acceptable test result | - | ||
