Siemens Aktiengesellschaft

Joachim Ebermann

Horst Stiehler

Kay Grammatke

Harald Eichelkamp

Dieter Schwarzbach

The invention relates to a method for process-variable-dependent identification signal emission for a closed-loop and/or open-loop control program with cyclic sampling of process variables from a technical process. A threshold value crossing time (ts, ts) is determined from at least two previous samples (AT, AT, AT-AT) of a process variable (P). At this time, an identification signal can be triggered, which can call up a single-stage or multi-stage command sequence. The threshold value crossing time (ts-ts) can likewise be determined with the aid of a mathematical approximation function and the samples (AT, AT, AT-AT). A timing mechanism can be started in the predicted sampling cycle (Ato A) preceding the threshold value crossing (SD-SD) using a time difference (ZD-ZD) remaining until the threshold value crossing (SD-SD). The threshold value crossing time (ts-ts) is determined once again using a further, subsequent sample (AT, AT, AT-AT) and the approximation function, and the remaining time difference (ZD-ZD) of the timing mechanism is adapted once again if necessary.

Method for process-variable-dependent identification signal emission

The invention concerns a method and an apparatus for time-staggered generation of identifying signals by programmable automation components (AK , AK ) with a system time recorder (SZ). A user (A) or an automation component (AK , AK ) or a user program (AP) can register a time-staggered identifying signal (K) with an event manager (EV). The event manager (EV) controls a system time counter (SZ) and signals waiting events (E -E ) back to the user (A) or the automation component (AK , AK ) or the user program (AP). The event (E -E ) is thereby enabled and an identifying signal (K) is generated.

Method and apparatus for time-staggered generation of identifying signals

Process variables (s , s ) are detected and processed cyclically in a controller (BG). Process variables (s_(x)) from the adjacent cyclically determined process variables (s_(n&minus;1), s_(n)) are determined as being event-related as a function of the time (t_(x)) of the event within the cycle. The calculation also includes derivatives of the process variable (s).

Growing community of inventors

Amtsberg, Germany

Joachim Ebermann