Entropy-based validation of sensor measurements

{
  "publication_number": "US2017195090A1",
  "country": "US",
  "kind": "A1",
  "title": "Entropy-based validation of sensor measurements",
  "abstract": "A method and module for detecting dependencies of at least two data streams of sensor measurements is provided, which are recorded on a technical device, in particular in an energy generation device, including: an input interface which is adapted for capturing the streams of sensor data from sensors, a data structure for analysis in form of a sliding window which is incrementally moved over the respective streams of sensor data so that there is an overlap between consecutive windows, and a processor, which is adapted to apply an entropy-based algorithm on the data within the sliding window for calculating data stream dependencies, while incrementally moving the sliding window, wherein only updated data in the sliding window is processed by the entropy based algorithm.",
  "claims": [
    "1. A method for detecting dependencies of at least two data streams of sensor measurements, which are recorded on an energy generation device, wherein the method comprises the following steps: capturing the at least two data streams of sensor data; defining a data structure for analysis in a form of a sliding window which is incrementally moved over the respective at least two data streams of sensor data so that there is an overlap between consecutive windows; and applying an entropy-based algorithm on the data within the sliding window for calculating data stream dependencies, while incrementally moving the sliding window, wherein only updated data in the sliding window is processed by the entropy based algorithm. 2. The method according to claim 1, wherein the sliding window is moved over all streams of sensor data in parallel and contains data from all streams of sensor data from a time period of specified length. 3. The method according to claim 1, wherein a breadth of the sliding window is configured in a configuration phase. 4. The method according to claim 1, wherein the method is executed in real time. 5. The method according to claim 1, wherein the method further comprises: providing a result via an output interface on an output device, representing whether or not there exists a dependency between the sensor data streams and if yes, whether the dependency is represented correctly. 6. The method according to claim 1, wherein the method is used for detecting errors and/or correlations in the data streams. 7. The method according to claim 1, wherein a minimal overlap between consecutive windows is configured. 8. The method according to claim 1, wherein the respective data streams comprise continuous and/or discrete data. 9. The method according to claim 1, wherein dependencies of the data streams are computed in a window-wise manner. 10. The method according to claim 1, wherein dependencies of the data streams are computed in an incremental manner and in particular by means of an online discretization. 11. An electronic module for detecting dependencies of at least two data streams of sensor measurements, which are recorded on an energy generation device, comprising: an input interface which is adapted for capturing the at least two data streams of sensor data from sensors; a data structure for analysis in a form of a sliding window which is incrementally moved over the respective at least two data streams of sensor data so that there is an overlap between consecutive windows; and a processor, which is adapted to apply an entropy-based algorithm on the data within the sliding window for calculating data stream dependencies, while incrementally moving the sliding window, wherein only updated data in the sliding window is processed by the entropy based algorithm. 12. A computer program product, tangibly embodying a program of machine-readable instructions executable by a digital processing apparatus to perform a method according to claim 1, if the program is executed on the digital processing apparatus."
  ],
  "cpc": [
    "G06N 7/01",
    "G06F 16/24568",
    "G06F 17/18",
    "G06F 2111/10",
    "G06F 2218/00",
    "G06F 2218/12",
    "G06F 30/20",
    "G06F 9/5066",
    "G06N 20/00",
    "G06N 5/04",
    "H04L 1/203",
    "H04L 63/1425",
    "H04W 24/10",
    "H04W 4/005",
    "H04W 4/70"
  ],
  "assignees": [
    "SIEMENS AG"
  ],
  "filing_date": "2016-12-19",
  "publication_date": "2017-07-06",
  "priority_date": "2016-01-04",
  "application_number": "US-201615383288-A",
  "family_id": "57389184",
  "citations": [
    "US2007294247A1",
    "US2011313844A1",
    "US2011320388A1",
    "US2014155043A1",
    "US2016364010A1",
    "US2017177546A1",
    "US2018307731A1",
    "US7299088B1",
    "US9245235B2",
    "US9313322B2",
    "US9961547B1"
  ]
}