Browsing by Author "Weber, Simon Oskar"

Now showing 1 - 1 of 1

Open Access
Model predicitve approaches for building climate and seasonal energy storage control
(2020) Weber, Simon Oskar
The aim of this thesis is the integration of a seasonal energy storage system into the heat supply of a building system under consideration of weather and occupancy forecasts. A thermochemical storage system based on the material system slaked lime / burnt lime is applied as seasonal energy storage. Methodically, a model of an energy system consisting of a building, a water buffer storage tank, a heat pump and a lime storage module is developed. In addition, model-predictive control concepts are developed, which optimally operate the system over a period of one year. For an effective integration of the seasonal lime storage, weather forecasts for an entire year are required. However, public weather forecasts are only considered reliable in the time range of several days. Due to this problem the so-called base year is introduced. The base year data approximate the weather forecast beyond the public forecast period. The weather data of the base year are based on those of the typical meteorological year, which are weather data averaged over several years. On this basis, three model-predictive control concepts are developed. The hierarchy of two concepts provides for a superordinate optimal generation scheduling as well as a subordinate model-predictive control. The optimal generation scheduling uses the disturbance variable inputs of the base year and finds those system inputs through single optimisation which minimise the annual operation costs while keeping all system limits. The resulting lime storage trajectory serves as a reference for the subordinate model predictive control. These concepts now try to follow the lime storage trajectory in an optimal way on the one hand, and on the other hand to realise possible increased or decreased yields due to the public weather data of the current year. The third control concept does not require a superordinate hierarchical level. It uses the public weather forecast for the coming days and the subsequent weather data of the base year to find optimal inputs to the overall system. This thesis demonstrates that the hybrid system consisting of heat pump and lime storage module allows the lowest operating costs for the heat supply of a building. Furthermore, all control concepts presented demonstrate that operation cost savings can be achieved by using the weather forecasts of the base year and integrating the lime storage tank. Depending on the applied configuration of the heat supply, these savings lie between 10% and 30% compared to the model- predictive control, which only includes forecasts over a period of several days and not months.