Method for detecting errors in sensor values and error detection device
Method provided for implementation in a data processing device for monitoring a permissible integrity of a measured variable of a dynamic system with the steps: Determination of at least one measurement series with respectively one number (n) of measurements (z k ) by means of at least one sensor device with respectively one predetermined measuring accuracy (Δz k ) and a predetermined actual sensor integrity ({1−p k }) related thereto, determination of a number (m) of measuring errors as a function of the actual sensor integrities ({1−p k }), as a function of the number (n) of measurements and as a function of a minimum integrity ({1−p 0 }), assignment of the determined accuracy range to an accuracy measurement, comparison of the determined accuracy measurement with a threshold value, and issuing a warning as soon as a desired value has been reached, fallen below or exceeded.
1. A method for implementation in a data processing device for monitoring a permissible integrity of a measured variable of a dynamic system comprising:
determining at least one measurement series with respectively a number (n) of measurements (z k ) by at least one sensor device with respectively one predetermined measuring accuracy (Δz k ) and a predetermined actual sensor integrity ({1−p k }) related thereto;
determining a number (m) of measuring errors as a function of the predetermined actual sensor integrities ({1−p k }), as a function of the number (n) of measurements and as a function of a predetermined minimum integrity ({1−p 0 });
determining for each measurement of the at least one measurement series of a measurement inaccuracy range (X k ) and, based on the measurement inaccuracy ranges (X 1 , . . . , X n ) and the number (m) of measuring errors, determination of an accuracy range (X 0 ) for a measured variable (x 0 ) to be determined, so that the accuracy range (X 0 ) contains the measured variable (x 0 ) with the predetermined minimum integrity {1−p 0 };
assigning the accuracy range to an accuracy measurement;
comparing the accuracy measurement with a threshold value; and
issuing a warning as soon as a desired value has been reached, fallen below or exceeded.
2. A method for implementation in a data processing device for monitoring a permissible integrity of a measured variable of a dynamic system, comprising: determining a number (m) of measuring errors by using an inequality
∑
j
=
m
-
1
n
n
!
j
!
(
n
-
j
)
!
p
j
(
1
-
p
)
n
-
j
<
p
0
,
where {1−p}={1−p 1 }, . . . , {1−p n } designates a uniform sensor integrity for a measurement series, n designates a number of measurements, p designates a sensor integrity risk, and p 0 designates a permissible integrity risk.
3. The method of claim 2 , wherein the determining of the number of measuring errors comprises inserting values m into the inequality and carrying out a comparison of the right and left side of the inequality and incrementing the value m inserted into inequality until the inequality is no longer satisfied.