Observer Design for non-stationary oscillating Disturbances in Mixing Processes

摘要

Information on the progression of the mixing process and quality are essential for mass concrete production. The power profile during the course of mixing can be used to evaluate the mixing process and its quality. Thus, a method for monitoring the average power transferred into the mixing material is presented, which enables an online mixture control based on the mechanical properties of the materials. An extensive experimental program on a 60 l laboratory twin-shaft mixer and a large-scale construction site mixer was performed to characterize mixer specific effects. For example, losses and speed dependent mixer shaft effects are examined. The portability and scalability of the results of the laboratory mixer is demonstrated by a comparison with a large construction site mixer. Based on the experimental insights, a signal model is derived which accounts for the characteristic disturbances in the power profile. Secondly, an extended Kalman filter is designed to estimate the disturbances online to separate these from the mean power signal, which can be used for evaluation of the concrete mixing progress. Finally, experimental results for different operating conditions and a comparison to classical signal filters such as a moving average filter and a low pass filter are presented, which show the increased performance of the presented approach.