Failure prediction apparatus for superconductive magnet and magnetic resonance imaging system

• US 6,774,632 B2
• Filed: 09/13/2002
• Issued: 08/10/2004
• Est. Priority Date: 09/14/2001
• Status: Expired due to Fees

• Current Assignee: GE Medical Systems Global Technology Company LLC (GE Healthcare Technologies, Inc.)
• Original Assignee: GE Medical Systems Global Technology Company LLC (GE Healthcare Technologies, Inc.)
• Inventors: Kono, Kazuhiro
• Primary Examiner(s): Gutierrez, Diego
• Assistant Examiner(s): Vargas, Dixomara
• Application Number:
  US10/243,785
• Publication Number:
  US 20030052681A1
• Time in Patent Office: 697 Days
• Field of Search: 324/307-309, 324/318
• US Class Current: 324/318
• CPC Class Codes: G01R 33/3815 with superconducting coils, e.g. power supply therefor

Abstract

In order to allow easy estimation of the operating condition of a magnet by a technician of any skill level, and to allow accurate prediction of a potential failure beforehand, a data processing section 52 is provided for resolving pressure data for a cooling helium gas, level (remaining amount) data for a liquid helium, first temperature data for a first stage of a refrigerator, second temperature data for a second stage of the refrigerator, and third temperature (room temperature) data for an equipment room 13 in which a compressor 40 is placed, as prespecified parameters, into a plurality of elements, using the elements as magnet data, calculating the Mahalanobis distance D² of the normalized magnet data, and comparing the calculated Mahalanobis distance D² with data stored in a database 53 obtained by developing magnet data in a normal condition into a Mahalanobis reference space to estimate the operating condition.

Claims

1. A failure prediction apparatus for predicting a failure of a superconductive magnet for forming a static magnetic field space employed in a magnetic resonance imaging apparatus, comprising:

a pressure sensor for detecting a pressure of a cooling medium in said superconductive magnet;

a level sensor for detecting a level of said cooling medium;

a first temperature sensor for detecting a temperature of a predefined portion of a refrigerator;

a second temperature sensor for detecting a temperature of a room in which a compressor is placed for cooling the cooling medium from said refrigerator and supplying the cooling medium to said refrigerator; and

a calculating device for calculating a Mahalanobis distance of magnet data including, as prespecified parameters, the pressure of the cooling medium detected by said pressure sensor, the level of the cooling medium detected by said level sensor, the temperature of the refrigerator detected by said first temperature sensor, and the room temperature detected by said second temperature sensor for determining whether the magnet is normal or not.

2. The failure prediction apparatus for a superconductive magnet of claim 1, wherein

said calculating device resolves the pressure of the cooling medium detected by said pressure sensor, the level of the cooling medium detected by said level sensor, the temperature of the refrigerator detected by said first temperature sensor, and the room temperature detected by said second temperature sensor, as the prespecified parameters, into a plurality of elements, normalizes the elements to form magnet data, and calculates the Mahalanobis distance of the normalized magnet data for determining whether the magnet is normal or not.

3. The failure prediction apparatus for a superconductive magnet of claim 2, comprising;

a database storing data obtained by resolving magnet data previously sampled from a normally operating magnet including, as prespecified parameters, a pressure of the cooling medium detected by said pressure sensor, a level of the cooling medium detected by said level sensor, a temperature of the refrigerator detected by said first temperature sensor, and a room temperature detected by said second temperature sensor, into a plurality of elements, and developing the resolved magnet data comprised of the plurality of elements into a Mahalanobis reference space; and

a device for determining whether the operating condition of the superconductive magnet is normal or not by comparing the Mahalanobis distance obtained by said calculating device with the stored data in said database.

4. The failure prediction apparatus for a superconductive magnet of claim 1, comprising:

a database storing data obtained by developing magnet data previously sampled from a normally operating magnet including, as prespecified parameters, a pressure of the cooling medium detected by said pressure sensor, a level of the cooling medium detected by said level sensor, a temperature of the refrigerator detected by said first temperature sensor, and a room temperature detected by said second temperature sensor, into a Mahalanobis reference space; and

a device for determining whether the operating condition of the superconductive magnet is normal or not by comparing the Mahalanobis distance obtained by said calculating device with the stored data in said database.

5. A magnetic resonance imaging system comprising:

a superconducting magnet for forming a static magnetic field space;

a data processing section for forming an image of a region within a subject placed within the static magnetic field space;

a failure prediction apparatus for the superconductive magnet, including;

a pressure sensor for detecting a pressure of a cooling medium in said superconductive magnet;

a level sensor for detecting a level of said cooling medium;

a first temperature sensor for detecting a temperature of a predefined portion of a refrigerator;

a second temperature sensor for detecting a temperature of a room in which a compressor is placed for cooling the cooling medium from said refrigerator and supplying the cooling medium to said refrigerator; and

a calculating device for calculating a Mahalanobis distance of magnet data including, as prespecified parameters, the pressure of the cooling medium detected by said pressure sensor, the level of the cooling medium detected by said level sensor, the temperature of the refrigerator detected by said first temperature sensor, and the room temperature detected by said second temperature sensor for determining whether the magnet is normal or not.

6. The magnetic resonance imaging system of claim 5, wherein

said calculating device resolves the pressure of the cooling medium detected by said pressure sensor, the level of the cooling medium detected by said level sensor, the temperature of the refrigerator detected by said first temperature sensor, and the room temperature detected by said second temperature sensor, as the prespecified parameters, into a plurality of elements, normalizes the elements to form magnet data, and calculates the Mahalanobis distance of the normalized magnet data for determining whether the magnet is normal or not.

7. The magnetic resonance imaging system of claim 6, wherein said failure prediction apparatus comprises:

a database storing data obtained by resolving magnet data previously sampled from a normally operating magnet including, as prespecified parameters, a pressure of the cooling medium detected by said pressure sensor, a level of the cooling medium detected by said level sensor, a temperature of the predefined portion of the refrigerator detected by said first temperature sensor, and a room temperature detected by said second temperature sensor, into a plurality of elements, and developing the resolved magnet data comprised of the plurality of elements into a Mahalanobis reference space; and

a device for determining whether the operating condition of the superconductive magnet is normal or not by comparing the Mahalanobis distance obtained by said calculating device with the stored data in said database.

8. The magnet resonance imaging system of claim 5, wherein said failure prediction apparatus comprises:

a database storing data obtained by developing magnet data previously sampled from a normally operating magnet including, as prespecified parameters, a pressure of the cooling medium detected by said pressure sensor, a level of the cooling medium detected by said level sensor, a temperature of the predefined portion of the refrigerator detected by said first temperature sensor, and the room temperature detected by said second temperature sensor, into a Mahalanobis reference space; and

a device for determining whether the operating condition of the superconductive magnet is normal or not by comparing the Mahalanobis distance obtained by said calculating device with the stored data in said database.

9. A magnetic resonance imaging system employing a superconductive magnet for forming a static magnetic field space, receiving a subject into said static magnetic field, and imaging a region to be examined in the subject using magnetic resonance, comprising:

a failure prediction apparatus for the superconductive magnet, including;

a pressure sensor for detecting a pressure of a cooling medium in said superconductive magnet;

a level sensor for detecting a level of said cooling medium;

first and second temperature sensors for detecting temperatures of a plurality of portions of a refrigerator;

a third temperature sensor for detecting a temperature of a room in which a compressor is placed for cooling the cooling medium from said refrigerator and supplying the cooling medium to said refrigerator; and

a calculating device for calculating a Mahalanobis distance of magnet data including, as prespecified parameters, the pressure of the cooling medium detected by said pressure sensor, the level of the cooling medium detected by said level sensor, the temperatures of the plurality of portions of the refrigerator detected by said first and second temperature sensors, and the room temperature detected by said third temperature sensor for determining whether the magnet is normal or not.

10. The magnetic resonance imaging system of claim 9, wherein

said calculating device resolves the pressure of the cooling medium detected by said pressure sensor, the level of the cooling medium detected by said level sensor, the temperatures of the plurality of portions of the refrigerator detected by said first and second temperature sensors, and the room temperature detected by said third temperature sensor, as the prespecified parameters, into a plurality of elements, normalizes the elements to form magnet data, and calculates the Mahalanobis distance of the normalized magnet data for determining whether the magnet is normal or not.

11. The magnetic resonance imaging system of claim 10, wherein said failure prediction apparatus comprises:

a database storing data obtained by resolving magnet data previously sampled from a normally operating magnet including, as prespecified parameters, a pressure of the cooling medium detected by said pressure sensor, a level of the cooling medium detected by said level sensor, temperatures of the plurality of portions of the refrigerator detected by said first and second temperature sensors, and a room temperature detected by said third temperature sensor, into a plurality of elements, and developing the resolved magnet data comprised of the plurality of elements into a Mahalanobis reference space, and

a device for determining whether the operating condition of the superconductive magnet is normal or not by comparing the Mahalanobis distance obtained by said calculating device with the stored data in said database.

12. The magnetic resonance imaging system of claim 9, wherein failure prediction apparatus comprises:

a database storing data obtained by developing magnet data previously sampled from a normally operating magnet including, as prespecified parameters, a pressure of the cooling medium detected by said pressure sensor, a level of the cooling medium detected by said level sensor, temperatures of the plurality of portions of the refrigerator detected by said first and second temperature sensors, and a room temperature detected by said third temperature sensor, into a Mahalanobis reference space; and

a device for determining whether the operating condition of the superconductive magnet is normal or not by comparing the Mahalanobis distance obtained by said calculating device with the stored data in said database.