Microwave-Based Measuring Device
AssigneeBerthold Technologies GmbH & Co. KG
InventorsJoerg BAEUERLE, Rachid El Korch, Steffen MUELLER
Life of this patent
Estimating the term…
Abstract
A microwave-based measuring device includes a number n of sensors, each sensor of the number n of sensors being embodied to generate associated sensor data such that, overall, a number n of items of sensor data are generated by way of the number n of sensors. A measurement variable calculation unit is embodied to calculate a number m of measurement variable values depending on the number n of items of sensor data on the basis of values of a number d of parameters. A learning unit is embodied to calculate the values of the number d of parameters on the basis of training data.
Claims9
Microwave-based measuring device (1), comprising:
- a number n of sensors (2_1 to 2_n), with a respective sensor (2_i) of the number n of sensors (2_1 to 2_n) being designed to generate associated sensor data (xi), so that a total of n number of sensor data (x_1, . . . , x_n) is generated by means of the number n of sensors (2_1 to 2_n),
- a measured variable calculation unit (4) which is designed to calculate a number m of measured variable values (y{circumflex over ( )}_1, . . . , y{circumflex over ( )}_m) as a function of the number n of sensor data (x_1, . . . , x_n) based on values of a number d of parameters (θ_1, . . . , θ_d), and
- A learning unit (5), the learning unit (5) being designed to be based on training data ((xt_1{circumflex over ( )}((i)), . . . , xt_n{circumflex over ( )}((i)));
- (y{circumflex over ( )}S_1{circumflex over ( )}((i)), . . . , y{circumflex over ( )}s_m{circumflex over ( )}((i)))) to calculate the values of the number d of parameters (θ_1, . . . , θ_d).
Microwave-based measuring device (1) according to one of the preceding claims, characterized in that
- The learning unit (5) is designed to use the training data ((xt_1{circumflex over ( )}((i)), . . . , xt_n{circumflex over ( )}((i))));
- (y{circumflex over ( )}s_1{circumflex over ( )}((i)), . . . , y{circumflex over ( )}s_m{circumflex over ( )}((i)))) a number n of training sensor data (xt_1{circumflex over ( )}((i)), . . . , xt_n{circumflex over ( )}((i))) and a number m of associated setpoint values (y{circumflex over ( )}s_1{circumflex over ( )}((i)), . . . , y{circumflex over ( )}s_m{circumflex over ( )}((i))) of the number m of measured variable values (y{circumflex over ( )}_1, . . . , y{circumflex over ( )}_m) to extract
- The measured variable calculation unit (4) being designed to calculate a number m of training values (y{circumflex over ( )}t_1{circumflex over ( )}((i)), . . . , y{circumflex over ( )}t_m{circumflex over ( )}((i))) of the number m of measured variable values (y{circumflex over ( )}_1, . . . , y{circumflex over ( )}_m) depending on the number n of training sensor data (xt_1{circumflex over ( )}((i)), . . . , xt_n{circumflex over ( )}((i))), and
- The learning unit (5) being designed to be based on the number m of setpoint values (y{circumflex over ( )}t_1{circumflex over ( )}((i)), . . . , y{circumflex over ( )}t_m{circumflex over ( )}((i))) of the number m of measured variable values (y{circumflex over ( )}_1, . . . , y{circumflex over ( )}_m) and the number m of training values (y{circumflex over ( )}t_1{circumflex over ( )}((i)), . . . , y{circumflex over ( )}t_m{circumflex over ( )}((i))) the number m of measured variable values (y{circumflex over ( )}_1, . . . , y{circumflex over ( )}_m) to calculate the values of the number d of parameters (θ_1, . . . , θ_d).
microwave-based measuring device (1) according to any one of the preceding claims, characterized in that
- the at least one measured variable is selected from the set of measured variables consisting of:
- humidity,
- dry matter content,
- sugar content,
- Concentration and
- material composition.
Dependencies are inferred from the claim text.