The paper shows an investigation of the accuracy and reliability of industrially available payload estimation methods in construction machinery, specifically hydraulic excavators. "Payload" refers to the weight exerted at the operating end (bucket) of the working arm of an excavator due to the amount of material in the bucket. There are several challenges which make dynamic and accurate payload estimation a difficult task - the complex kinematics of the excavator, the complicated operations and movements of the arm, as well as the requirements on the robustness of the excavator operation etc. Current payload estimation systems are based on patent EP0736752A1 'Wägeverfahren und Hubfahrzeug zur Durchführung des Verfahrens" from 1995. This empirical method has three main inputs to the system, consisting of boom angle, arm angle and the pressure difference across the boom. Using these inputs, a characteristic matrix is generated by calibrating the system using several different position and load configurations. The dynamic effects such as mass inertia or centrifugal forces are partially or completely ignored in this method. It is therefore very sensitive to changes in measuring speed, accelerations during the measurement or changes in the relative center of gravity position of the payload. This paper is an investigation and evaluation of the functionality and performance of current payload estimation systems using a multi-body-simulation model of an excavator. Using this model-based approach, this investigation can be simplified by eliminating additional influences such as hydraulic or mechanic friction or other effects of the hydraulic power train. This makes it possible to evaluate specifically the influence of static and dynamic behavior of the machine on the described method. Finally, based on the simulation results, possible improvements to the method are considered.
Payload Estimation in Excavators. Model-Based Evaluation of Current Payload Estimation Systems
2014
14 Seiten, Bilder, 9 Quellen
Conference paper
English
Payload Estimation in Excavators Using a Machine Learning Based Parameter Identification Method
Springer Verlag | 2020
|