MONITORING AN EVENT IN A POWER CONVERTER
A method for monitoring an event in a power converter includes using record data after starting the power converter, wherein the record data do not indicate an error for a predetermined period of time after the power converter is started. Messages are assigned to error sources, and a degree of probability is calculated for an assigning process.
1 .- 11 . (canceled)
12 . A method for event monitoring in a converter, comprising:
using log data of the converter which comprise labeling containing temporal information or relating to a sequence of logged messages pertaining to a fault or a warning and which prior to a start of the event monitoring do not show a fault for a predetermined time period after the start,
using for the event monitoring an evaluation of a combination of faults or warnings of at least a first type and of a second type, wherein the first fault type or warning type depends on a type of the converter, and the second fault type or warning type comprises user-defined faults and warnings,
generating user-defined messages based on a single signal or based on a combination of signals and using the user-defined messages for event monitoring in an individual environment of the converter,
determining a most probable technical root causes of a failure or a fault by using artificial intelligence,
training a machine learning algorithm and categorizing with the trained algorithm fault events into predefined cause categories,
training the artificial intelligence to indicate one or more fault sources based on a multiplicity of status messages, warning messages or fault messages, and
calculating in each case a probability for correctness of this indication.
13 . The method of claim 12 , wherein the log data are used after a start of the converter.
14 . The method of claim 12 , wherein the status messages, warning messages or fault messages are associated with the one or more fault sources.
15 . The method of claim 14 , further comprising calculating a probability for the association of the status messages, warning messages or fault messages with the one or more fault sources.
16 . The method of claim 12 , wherein the artificial intelligence is trained to perform the event monitoring.
17 . A method for event monitoring in a converter, comprising:
recording messages of the converter, with the messages having a time stamp and an identification and the identification including a message type, a text or a source of a message;
detecting an event from a sequence of messages of a different type, wherein a message is a fault or a warning;
generating user-defined messages based on a single signal or based on a combination of signals;
using the user-defined messages for event monitoring in an individual environment of the converter,
determining a most probable technical root causes of a failure or a fault by using artificial intelligence;
training a machine learning algorithm to categorize fault events into predefined cause categories;
training the artificial intelligence so as to indicate one or more fault sources based on, wherein several fault sources can be indicated, and
calculating in each case a probability for correctness of this indication.
18 . The method of claim 17 , wherein the artificial intelligence is a cloud application, the method further comprising detecting with the artificial intelligence an output fault.
19 . The method of claim 17 , wherein the event monitoring is generated as set forth in claim 12 .
20 . Event monitoring of a converter, wherein the event monitoring comprises artificial intelligence and log file data from the converter are stored in a cloud, wherein the event monitoring is performed using a method as set forth in claim 12 .
21 . Event monitoring of a converter, wherein the event monitoring comprises artificial intelligence and log file data from the converter are stored in a cloud, wherein the event monitoring is performed using a method as set forth in claim 17 .