Betriebsassistenz für solarthermische Kraftwerke zur Effizienzsteigerung auf Basis regelungstechnischer Methoden

• Operation assistance for concentrating solar power plants to increase efficiency based on control engineering methods

Nolteernsting, Felix; Abel, Dirk (Thesis advisor); Pitz-Paal, Robert (Thesis advisor)

Aachen : RWTH Aachen University (2020, 2021)
Dissertation / PhD Thesis

Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2020

Abstract

Concentrating Solar Power (CSP) plants concentrate sunlight for converting it into electricity. In CSP plants, thermal storage can offer an efficient way of storing renewable energy in the form of heat. The use of thermal storage increases the complexity of process control in CSP plants. The optimal operation of CSP plants is further complicated by their limited distribution and the resulting prototypical character. Hence, the experience of plant operators is of enormous importance for the successful operation of CSP plants. In this thesis a concept for an operational assistance system in CSP plants is developed. The assistance system is based on control engineering methods. It serves to increase the efficiency of power plant operation. The assistance concept consists of four successive levels with different degrees of automation. The plant operator retains most of the responsibility in prediction mode, the first level of automation. It is used to predict the future state of the power plant with constant control variables. The prediction mode is implemented in this thesis for the state of the thermal fixed-bed storage of the Jülich Solar Tower. Furthermore, a prototypical implementation of the prediction mode in the power plant is presented. For the implementation of the prediction mode, a state estimator for the thermal storage is designed and validated on measurement data. A complete automation of the power plant is carried out in the fourth automation level. The manipulated variables are determined by an optimization within an economic model predictive control. The electrical output of the power plant is maximized. Within the scope of the present work, this procedure is tested for the operation of the power plant simulatively on the scenario of a strong radiation collapse. The results of the simulation are presented and discussed. The second and third level of automation of the operational assistance system for CSP plants are introduced and motivated as well. The result of the present work is a concept, which opens up the possibility to increase efficiency in the complex prototypical process of CSP power plants. The concept is based on control engineering methods and obtains the possibility of including the experience of a plant operator.