Master's thesis Philipp Krauss

 

Design of an automated tool to optimize the control quality of typical control loops in building energy systems

a picture Copyright: EBC

The aim of this thesis is the development and the application of an automatic and easy applicable algorithm to tune the PID controllers in typical control loops of building energy systems. For this purpose a selection of different control loops is made, which are used in the test building, the E.ON ERC Main Building at the RWTH Aachen University. Those control loops are analyzed and the algorithmis applied to them by way of example. The considered systems are first identified theoretically and described with their function charts. The proposed algorithmworks for stable control loops aswell as for unstable ones. It can be adapted to the analyzed system using tuning parameters. The user also defines a time interval for the algorithm to execute the experiments. In case of a stable control loop the algorithm starts with a step experiment to estimate the control quality. Afterwards an identification experiment is executed in closed-loopmode. If the control loop is unstable the first experiment is an identification experiment in open-loop mode. In both cases a large set of system models with different structures and orders gets identified with the measured system input and output signals. Several known PID parameter tuning methods are used to calculate the controller parameters from all identified system models. For each configuration the control quality gets estimated by simulating a closed-loop step. The results are used to evaluate the proposed PID parameter sets. For the best of them the control quality gets calculated with further closed-loop step experiments. Only with these results the quality of the identified system models and the tuning methods are rated. The application of the algorithm shows that systems with very different dynamic behavior can be analyzed. With one exception, the algorithm can enhance the closed-loop behavior in all control loops that show room for improvement. The ITAE criteria is decreased up to 95.46 %.