| 摘要 |
<P>PROBLEM TO BE SOLVED: To suppress deterioration of accuracy in determining abnormality in the case where zero-pint learning is not normally performed in performing the zero-point learning for a sensor. <P>SOLUTION: In the case where zero-point learning is not normally performed, the absolute value of a current Ib from a current detection sensor is compared with a threshold Iref2 as a magnitude (¾β¾) of the tolerance (e.g., ±β) of the current detection sensor (S140), and it is determined that no abnormality has occurred if the absolute value of the current Ib is equal to or smaller than the threshold Iref2, and that an abnormality has occurred if the absolute value of the current Ib is larger than the threshold Iref2. With this, deterioration of accuracy in determining abnormality can be suppressed, which is based on the magnitude β of the tolerance functioning superimposingly on an offset current α that is to be an error caused by using a current Ib that has been subjected to correction based on zero-point learning regardless of zero-point learning not being normally performed. <P>COPYRIGHT: (C)2013,JPO&INPIT |
