Anomaly detection and correlation system
AssigneeSiemens Healthcare GmbH
InventorPaul William Edgley
Life of this patent
Estimating the term…
Abstract
An anomaly detection system comprising a diagnostic interface and a data acquisition system capable of recording data indicating variation of voltages at tapping points in a superconducting magnet. Data representing those variations in voltages may be stored as a data log and a processor may analyze the data in order to identify a mechanical impact.
Claims11
A method for detecting anomalies in voltages within a magnet system comprising series-connected current-carrying superconducting coils, comprising:
- receiving voltages from tapping points located electrically between superconducting coils within the magnet system;
- converting the voltages into digital representations of the voltages;
- storing the digital representations of the voltages;
- receiving signals from mechanical vibration sensors;
- and
- processing the digital representations of the voltages and the signals received from the mechanical vibration sensors to categorize a detected variation in the voltages at the tapping points,
- wherein the detected variation in the voltages at the tapping points is categorized to identify a cause of the variations in voltage as one of (i) a first type of mechanical impact, (ii) a second type of mechanical impact, or (iii) interference in a magnetic field due to a nearby object not resulting in a mechanical impact,
- wherein the storing the digital representations of the voltages and processing the digital representations of the voltages and the data received from the mechanical vibration sensors to categorize the detected variation in the voltages at the tapping points is performed in response to the detection of a voltage at the tapping points in excess of a voltage threshold which lasts for a duration of time in excess of a time threshold.
A method for detecting anomalies in voltages within a magnet system comprising series-connected current-carrying superconducting coils, comprising:
- receiving voltages from tapping points located electrically between superconducting coils within the magnet system;
- converting the voltages into digital representations of the voltages;
- storing the digital representations of the voltages;
- and
- processing the digital representations of the voltages to calculate a periodicity of time identified with successive instances in which a coil voltage disturbance value is in excess of a voltage threshold, and to categorize a detected variation in the voltages at the tapping points based upon the periodicity of time to identify the cause of the variations in voltages as one of (i) a first type of mechanical impact, (ii) a second type of mechanical impact, or (iii) interference in a magnetic field due to a nearby object not resulting in a mechanical impact.
A method for characterising mechanical impacts to a magnet system that is used as part of a magnetic resonance imaging (MRI) system, comprising:
- receiving power from an independent power source that is separate from the MRI system to power a data acquisition system;
- via the data acquisition system:
- receiving signals from one or more vibration sensors attached to the magnet system;
- converting the signals into digital representations of the signals;
- storing the digital representations of the signals;
- and
- processing the digital representations of the signals to categorize a detected vibration,
- wherein the detected vibration is categorized to identify a cause of the vibrations while the magnet system is not in use in the MRI system as one of (i) a first type of mechanical impact, (ii) a second type of mechanical impact, or (iii) interference in a magnetic field due to a nearby object not resulting in a mechanical impact.
Dependencies are inferred from the claim text.