Method for controlling multidimensional manipulated variable of sensor data, involves subjecting coarse actuator to sum coarse adjustment function obtained by processing coarse actuator function with coarse adjustment function

摘要

The sensor data received through neural networks is processed to recognize a deviation function and to issue several detected gross deviation functions. The database is queried with a record number of the detected gross deviation function to obtain the coarse adjustment function. The coarse actuator function is processed with coarse adjustment function to obtain a sum coarse adjustment function. The coarse actuator is subjected to sum coarse adjustment function, while changing the variable. The sensor data obtained from neural networks, is processed in parallel. An independent claim is included for device for controlling multidimensional manipulated variable of sensor data.