Method and device for determining the layer thickness distribution in a paint layer
First Claim
1. A method for determining a layer thickness distribution to be expected in a paint layer produced during paint spraying after inputting specific spraying parameters into an electrostatically based paint spraying device, which comprises the steps of:
- providing a data processing device performing the steps of;
setting up and using a phenomenological mathematical model of a quasi-stationary three-dimensional spray pattern;
inputting an angle of rotation of electrodes of the spraying device and a rate of movement of the spraying device directly into the phenomenological mathematical model as fixed in put parameters;
feeding in real physical input parameters including, a paint volume, directing air data, and a voltage value, whose influence on a spraying result is not accurately known, to an artificial neural network which has previously been trained using real input data including a configuration of the spraying device used, a paint type, operating parameters, and measured values of a test layer thickness distribution, and the artificial neural network carries out a conversion of the real physical input parameters into model input parameters;
feeding the model input parameters to the phenomenological mathematical model;
integrating spray patterns formed by the phenomenological mathematical model in a function unit in dependence on movement data of the spraying device which are contained in input parameters to form an overall paint layer; and
outputting the layer thickness distribution of the overall paint layer.
1 Assignment
0 Petitions
Accused Products
Abstract
A method for determining a layer thickness distribution in a paint layer produced during paint spraying after inputting specific spraying parameters into an electrostatically based paint spraying device. A data processing device sets up and uses a phenomenological mathematical model of a quasi-stationary three-dimensional spray pattern. Specific parameters, such as an angle of rotation of electrodes and a rate of movement of the spraying device are input into the phenomenological model as fixed input parameters. In addition, real physical input parameters such as paint volume, directing air data and a voltage value, whose influence on the spraying result is not accurately known, are fed to an artificial neural network. The neural network having been previously trained using real input data such as a configuration of the spraying device, a paint type, operating parameters, and measured values of the layer thickness distribution. The neural network carries out a conversion of the input parameters into model input parameters which are fed to the phenomenological model. Spray patterns formed by the phenomenological model are integrated in a further functional unit as a function of movement data of the spraying device which are contained in the input parameters to form the overall paint layer which is output.
-
Citations
4 Claims
-
1. A method for determining a layer thickness distribution to be expected in a paint layer produced during paint spraying after inputting specific spraying parameters into an electrostatically based paint spraying device, which comprises the steps of:
-
providing a data processing device performing the steps of;
setting up and using a phenomenological mathematical model of a quasi-stationary three-dimensional spray pattern;
inputting an angle of rotation of electrodes of the spraying device and a rate of movement of the spraying device directly into the phenomenological mathematical model as fixed in put parameters;
feeding in real physical input parameters including, a paint volume, directing air data, and a voltage value, whose influence on a spraying result is not accurately known, to an artificial neural network which has previously been trained using real input data including a configuration of the spraying device used, a paint type, operating parameters, and measured values of a test layer thickness distribution, and the artificial neural network carries out a conversion of the real physical input parameters into model input parameters;
feeding the model input parameters to the phenomenological mathematical model;
integrating spray patterns formed by the phenomenological mathematical model in a function unit in dependence on movement data of the spraying device which are contained in input parameters to form an overall paint layer; and
outputting the layer thickness distribution of the overall paint layer.
-
-
2. The method according to claim 1, which comprises training a separate neural network for each desired model parameter, the separate neural network has only a single output and a number of input neurons which correspond to a portion of a totality of available input variables.
-
3. The method according to claim 2, which comprises eliminating a parameter that is acknowledged as irrelevant when a learning process of two neural networks which differ formally only by an input parameter lead to equivalent learning results in conjunction with an otherwise identical learning data record.
-
4. A data processing configuration for determining a layer thickness distribution to be expected in a paint layer produced during paint spraying after inputting specific spraying parameters into an electrostatically based paint spraying device, the data processing configuration comprising:
-
a data processing device having means for producing a phenomenological mathematical model, an artificial neural network connected to said means for producing the phenomenological mathematical model, and a functional unit connected to said means for producing the phenomenological mathematical model, said data processing device programmed to;
set up and use the phenomenological mathematical model of a quasi-stationary three-dimensional spray pattern;
input an angle of rotation of electrodes of the spraying device and a rate of movement of the spraying device directly into the phenomenological mathematical model as fixed input parameters;
feed in real physical input parameters including, a paint volume, directing air data, and a voltage value, whose influence on a spraying result is not accurately known, to said artificial neural network which has previously been trained using real input data including a configuration of the spraying device used, a paint type, operating parameters, and measured values of a test layer thickness distribution, and said artificial neural network carries out a conversion of the real physical input parameters into model input parameters;
feed the model input parameters to the phenomenological mathematical model;
integrate spray patterns formed by the phenomenological mathematical model in said functional unit in dependence on movement data of the spraying device which are contained in input parameters to form an overall paint layer; and
out put the layer thickness distribution of the overall paint layer.
-
Specification