Refine
Document type
Language
- English (9)
Is part of the Bibliography
- Yes (9)
Keywords
- Data-driven modeling (2)
- External gear pump (2)
- Neural network (2)
- Physics informed machine learning (2)
- DEAP (1)
- Dielectric electroactive polymer (1)
- Small data (1)
- Tactile actuator (1)
- Tactile display (1)
Semi-rigid ring-shaped electrode dielectric electroactive polymer membrane as buckling actuator
(2019)
Using Neural Networks with Linear Regression as a scalable model to predict the behaviour of pumps
(2024)
This article presents a method for predicting the
behavior of external gear pumps using a neural network, to
optimize a higher-level dosing process. Sparse neural networks
learn to map the characteristic curves of fluid machinery. The
presented method achieves an error flow rate of approximately
3.28 milliliters per minute. The neural networks make it possible
to directly substitute the volume flow sensor in order to design
dosing applications without cost-intensive volume flow sensors.
However, due to the existing manufacturing tolerances of the
used pumps, the overall pump efficiency differs significantly.
For this reason, it is not directly possible to apply the networks
universally to an entire pump series, which strongly limits the
industrially scalable solution. For this reason, the algorithms are
extended by a linear regression model, which enables users to
calibrate the neural networks to the individual pumps based on
very few reference measurements.
This study presents a method for predicting the volume flow output of external gear
pumps using neural networks. Based on operational measurements across the entire energy chain, the neural network learns to map the internal leakage of the pumps in use and consequently to predict the output volume flow over the entire operating range of the underlying dosing process. As a consequence, the previously used volumetric flow sensors become obsolete within the application itself. The model approach optimizes the higher-level dosing system in order to meet the constantly growing demands of industrial applications. We first describe the mode of operation of the pumps in use and focus on the internal leakage of external gear pumps, as these primarily determine the losses of the system. The structure of the test bench and the data processing for the neural network are discussed, as well as the architecture of the neural network. An error flow rate of approximately 1% can be achieved with the presented approach considering the entire operating range of the pumps, which until now could only be realized with multiple computationally intensive CFD simulations. The results are put into perspective by a hyperparameter study of possible neural architectures. The biggest obstacle considering the industrial scaling of this solution is the data generation process itself for various operating points. To date, an individual dataset is required for each pump because the neural architectures used are difficult to transfer, due to the tolerances of the manufactured pumps.