System and method for monitoring a superconducting magnet

Abstract

Monitoring system (300) for a superconducting magnet (200), comprising: a computer (CMP) connectable to the superconducting magnet (200) having a cryogenic system equipped with a compressor (5) and a cooling head (6) configured to maintain a cryogenic bath (3) at an operating temperature of a superconducting coil (1). The computer (CMP) comprises a computer program (SW) configured to receive historical / statistical data relating to the magnet and calculate an unreliability index RI by means of a sum of the following terms: a first term HMinCH expressing an operating life at the current time of the cooling head (6); a second term subtractive, which expresses a minimum statistical life of cooling heads, a third term FgpCH which expresses a probability of premature failure of the cooling head (6). The monitoring system is configured to generate messages with recommendations for a user in relation to operations to be performed on the magnet.

Description

[0001] “System and method for monitoring a superconducting magnet”

[0002] DESCRIPTION

[0003]

[0001] TECHNICAL FIELD

[0004]

[0002] The present invention relates to the monitoring of superconducting magnets.

[0005]

[0003] STATE OF THE ART

[0006]

[0004] As is well known, superconducting magnets are high-tech devices used to obtain high and homogeneous magnetic fields for use in medicine, industry, and scientific research.

[0007]

[0005] They generally consist of a cryogenic container or casing, which thermally insulates the magnet from the external environment, and a cryogenic bath of liquid helium, in which the windings of the magnet itself are immersed. Superconducting magnets consisting of a series of niobium-titanium alloy windings are known. In order to keep the helium in a liquid state, the magnet is equipped with a cooling system.

[0008]

[0006] Manufacturers of superconducting magnets have implemented systems for reading at least the helium level, so that the operator can read the helium level once a day and report any levels below normal to the technical office.

[0009]

[0007] Some manufacturers have also implemented proprietary systems for automatic communication of levels and anomalies to the relevant technical office, so that the problem can be resolved by their staff.

[0010]

[0008] However, these latter systems only work if the owner has signed a maintenance contract with the system manufacturer (OEM, Original Equipment Manufacturer) and cannot be used if the customer signs a contract with a third- party service provider (known as a Multivendor).

[0009] Multivendor support companies specialize in providing technical support for machines not manufactured by them, relying on the support of technical personnel often from an OEM company, but who do not have the same technological and infrastructural support as an OEM.

[0011]

[0010] A Multivendor service company does not have access to the most advanced remote monitoring systems that the parent company has, and this can often be a source of additional costs, due, for example, to failures in the refrigeration system that can lead to serious events such as Quench or loss of liquid helium.

[0012] [Oil] Document JP-A-2005-237417 describes a magnetic resonance imaging apparatus that uses a superconducting magnet.

[0013]

[0012] Document US-A-2006-0225433 discloses a method for predicting failures in a magnetic resonance imaging machine.

[0014]

[0013] Document EP-A-2146297 describes a system and method for monitoring a medical device.

[0015]

[0014] SUMMARY OF THE INVENTION

[0016]

[0015] The purpose of the present invention is to propose a system for monitoring a superconducting magnet that provides predictions (for example, in the form of recommendations) on the possibility of malfunctions and that is alternative to those known. According to one example, this monitoring system is aimed at multivendor systems that could therefore equal or exceed the technical capabilities of proprietary remote monitoring systems.

[0017]

[0016] The subject matter of the present invention is a monitoring system for a superconducting magnet as defined in claim 1 and its particular embodiments described in dependent claims 2-11, also with reference to a magnetic field generation system. According to another subject matter, the present invention relates to a method for monitoring a superconducting magnet as defined in claim 12.

[0018]

[0017] BRIEF DESCRIPTION OF THE DRAWINGS

[0019]

[0018] The constructional and functional features of the invention will be better understood from the detailed description that follows, in which reference is made to the accompanying drawings, which represent some preferred and non-limiting embodiments, in which:

[0020]

[0019] FIG. 1 schematically shows an example of a magnetic field generation system comprising a superconducting magnet and a monitoring system;

[0021]

[0020] FIG. 2 shows an example of the architecture of said monitoring system.

[0022]

[0021] FIG. 1 schematically shows an embodiment of a magnetic field generation system 100 comprising a superconducting magnet 200 and a remote monitoring system 300.

[0023]

[0022] The superconducting magnet 200 may be part of high magnetic field equipment, operating as a magnetic field source. For example, the superconducting magnet 200 may be part of a magnetic resonance imaging (MRI) apparatus, a device for extracting single silicon crystals using a Czochralski method applied to the magnetic field, or, for example, a particle accelerator such as a cyclotron or other equipment.

[0024]

[0023] The superconducting magnet 200 comprises a superconducting coil 1 (formed by multiple conductive windings or turns), a cryogenic container 2, and a cryogenic bath 3 in which the superconducting coil 1 is immersed. Preferably, the cryogenic container 2 is enclosed in a vacuum container 4 that thermally insulates the cryogenic container and the superconducting coil 1.

[0024] In addition, the superconducting magnet 200 comprises a cryogenic system configured to cool the cryogenic bath 3. According to a particular form of implementation, the cryogenic system is of the Gifford-McMahon (GM) type and comprises a compressor 5 (external to the shield 4) and a cooling head 6.

[0025]

[0025] For example, the cryogenic bath 3 comprises liquid and / or gaseous helium, and the superconducting coil 1 has windings made of a niobium-titanium alloy. The cryogenic system is such as to keep the helium in a liquid state by maintaining the cryogenic bath at a temperature not exceeding 20° K. Alternatively, it should be noted that the superconducting magnet 200 can be made of superconducting material and / or with a cryogenic fluid other than that mentioned.

[0026]

[0026] Compressor 5 is configured to compress the refrigerant gas (e.g., helium) drawn in through an intake opening and discharge the refrigerant gas from an exhaust opening. A high-pressure pipe 13 connects the compressor 5 to the cooling head 6, and a low-pressure pipe 14 connects the cooling head 6 to the compressor 5, so that the low-pressure refrigerant gas can be recovered at the compressor 5. For example, pipes 13 and 14 may be formed from rigid pipes or flexible pipes (also called "flexlines").

[0027]

[0027] The compressor 5 includes, for example, a purification filter (also known as an absorber) for the helium gas that must reach the cooling head 6. This purification filter (not shown in the figure) removes traces of other gases and lubricating oil and has a fixed service life, depending on the model. For example, the service life of the purification filter is between 7,000 and 15,000 hours.

[0028]

[0028] The cooling head 6, in turn, is equipped with a drive unit 7, an electric motor 8, and a cooling stage 9 in thermal contact with the cryogenic bath 3. The electric motor 8 is configured to drive the cooling head 6 and, in particular, to move a dislocator (not shown) and a switching valve arranged in the cooling head

[0029] 6, so as to perform a cycle GM.

[0030]

[0029] The displacer controls the volume of a refrigerant gas expansion chamber in the cooling head 6, while the switching valve controls the pressure of the refrigerant gas in the expansion chamber, switching between supplying and recovering refrigerant gas from the compressor 5.

[0031]

[0030] In addition, the drive unit 7 is equipped with a high-pressure port 10 connected to the high-pressure pipe 13 and a low-pressure port 11 connected to the low-pressure pipe 14. The cooling head 6 receives high-pressure refrigerant gas from port 10 and discharges low-pressure refrigerant gas (which has expanded in the expansion chamber) from low-pressure port 11 through the switching valve. The compressor 5 and the drive unit 7 are electrically powered by an electrical power source, which may be the mains supply.

[0032]

[0031] The cooling stage 9, connected to the drive unit 7 via a first heat exchanger

[0033] 15, is such as to cool the cryogenic container 2 and, via a second heat exchanger

[0034] 16, the cryogenic bath 3, maintaining the superconducting coil 1 at the desired temperature. For example, the cryogenic bath 3 is maintained at temperatures of 1-5 K. More specifically, the temperature is maintained below the boiling point of helium, approximately 4.2 K.

[0035]

[0032] An electrical cable 17 connects the superconducting coil 1 to a power supply device (not shown), located outside the vacuum container 4. The electrical cable 17 comprises a metallic current conductor 18 connected to the power supply device through a feedthrough portion installed in the vacuum container 4 and a superconducting current conductor 19 connected to the metallic current conductor 18. The superconducting current conductor 19 is connected to the superconducting coil 1.

[0036]

[0033] The superconducting magnet 200 is also equipped with at least one temperature sensor 20 configured to measure the temperature of the cryogenic bath 3 and at least one level sensor 21 that measures the level of the cryogenic bath fluid 3.

[0037]

[0034] The superconducting magnet 200 also includes a water cooling system designed to cool the helium gas exiting the compressor 5, which has undergone a temperature increase due to the compression itself.

[0038]

[0035] The superconducting magnet 200 includes a controller 25 (connected to a power source) for the cryogenic system, which can be located near the compressor 5 or remotely from it. More specifically, the controller 25 is configured to control the operation of the cryogenic system and, in particular, that of the compressor 5 and the cooling head 6, also based on the output of the temperature sensor 20. In addition, the controller 25 can also be configured to control electrical aspects of the superconducting magnet 200.

[0039]

[0036] The controller 25 comprises a CPU, at least one mass memory and working memory, and may be, for example, a microprocessor or a computer, connected, for example, to other computers via a telematic network. The controller 25, or the telematic network in which it is included, contains software for controlling / monitoring the operation of the cryogenic system and the superconducting magnet as a whole. The controller 25 is connected to the cooling head 6 by means of a first connection line 26 and to the sensors included in the cryogenic container 2 by means of a second connection line 27, so as to receive / send data, messages, and / or commands, schematically represented in FIG. 1.

[0040]

[0037] Although a particular form of superconducting magnet 200 has been described above, it can be of various types. For example, magnets manufactured by General Electric, Siemens, and Philips are suitable. Magnets manufactured by General Electric are equipped with a monitoring system (integrated into the controller 25 or a computer connected to it) called MM3 or MM4, depending on the version. Magnets manufactured by Philips and Siemens are equipped with an OEM (Original Equipment Manufacturer) remote monitoring system that is not accessible from the outside.

[0041]

[0038] The monitoring system (MON) 300 can be advantageously added to the proprietary monitoring systems managed by the manufacturers of the superconducting magnet 200. The monitoring system 300 comprises one or more computers (for example, microprocessors, PCs, server computers) connected to the superconducting magnet 200to receive historical and / or statistical data relating to the magnet itself.

[0042]

[0039] The monitoring system 300 may be connected to a dedicated interface of the controller 25 or the network in which it is included (for example, in the case of a magnet manufactured by General Electric that has such an interface) or may be connected to the first communication line 26 and / or the second communication line 27 if an accessible interface is not available. Typically, the monitoring system 300 is managed by a supplier other than the company that manufactures the superconducting magnet 200, such as a multivendor.

[0043]

[0040] According to one example (FIG. 2), the monitoring system 300 comprises at least one computer CMP (e.g., a server), a database DB (which stores data useful for monitoring the superconducting magnet 200), and software SW for calculating the unreliability of the cryogenic system of the superconducting magnet 200.

[0044]

[0041] According to a particular embodiment, the monitoring system 300 also comprises a plurality of connection interfaces INTF to the superconducting magnet 200. These interfaces are specific to different brands and models of the magnet. In addition, the monitoring system may be equipped with a broker BRK (in hardware or software) that translates data from the proprietary monitoring system into the language appropriate for the monitoring system 300 and HTML consultation pages.

[0045]

[0042] In addition, the monitoring system 300 hosts, for example in the computer CMP, software SW for calculating an unreliability index RI of the cryogenic system of the superconducting magnet 200. This software SW corresponds to an algorithm that takes into account historical or statistical data relating to certain components of the superconducting magnet 200 and also statistical data relating to other superconducting magnets of a type similar to that of the magnet 100 actually being monitored. This data is stored in the database DB and is obtained by monitoring the superconducting magnet 200 or has been collected as a result of monitoring other superconducting magnets.

[0046]

[0043] The calculation software SW calculates the unreliability index RI of the cryogenic system by calculating a sum T1 that has a first term HCH, a second term HMinCH and a third term F^CH ■ This sum T1 relates to the cooling head 6.

[0047]

[0044] More specifically, the first term HCH expresses the operating life at the present time (i.e., at the time the calculation is performed) of the cooling head 6. The second term HMinCH, which is subtractive, expresses a minimum statistical life found for cooling heads (of a type similar to head 6). For example, the minimum statistical life of a cooling head is 50,000 h if new and 25,000 h if it is a reconditioned head.

[0048]

[0045] The third term Ixc, expresses the probability of premature failure of cooling head 6 and depends on a sum of pre-recorded durations Hchnof malfunctions of superconducting magnet components 200 other than cooling head 6. For example, the malfunctions considered are those related to failures of components of the refrigeration chain, such as the compressor 5 or the purification filter (i.e., the absorber) or pipes 13 and 14. These malfunctions may involve problems directly influenced by the cooling head 6, such as a quench event, or a failure related to coil 1.

[0049]

[0046] From a mathematical point of view, the sum T1 can be used in the following formula to calculate the unreliability index RI (or its relevant contribution Cl): max (0, T1*KCH ) =max (0, (HCH -HMUICH +FgpcH ) *KCH ) (1)

[0050]

[0047] In formula (1), the terms defined above can be recognized, while the function f»ax(0,f(x)) is the Rectified Linear Unit (ReLU) function, which returns the positive part of its argument. The factor I<CH that multiplies the sum T1 represents a weight coefficient that can be determined based on experience.

[0051]

[0048] The third term F^CH can be calculated as expressed by the following formula (2):

[0052]

[0049] Where Bchnis a binary number that represents (when it takes the value 1) the presence of an nthrecorded malfunction, Hchnrepresents its duration, and N is the number of recorded malfunctions. As expressed by formula (2), the third term F^CH is calculated by raising the sum of the malfunction durations to an exponential factor Qch, which is chosen based on experience.

[0053]

[0050] According to a particular form of implementation, the value expressed by formula (1) is only a first contribution Cl to the calculation of the unreliability index RI, which can also be calculated by considering a second contribution C2. Preferably, this second contribution C2 depends on another sum T2 that refers to compressor 5. The calculation of the sum T2 involves the calculation of a fourth term H(Cmp), a fifth term ffcmpMin and a sixth term F^pcmp-

[0054]

[0051] More specifically, the fourth term Hemp expresses the current operating life of compressor 5. The fifth term HMincmp, which is subtractive, expresses the minimum statistical life found for compressors of a type similar to compressor 5. For example, the minimum statistical operating time of a helium compressor is 40,000 hours if new or 20,000 hours if reconditioned.

[0055]

[0052] The sixth term F^p(Cmp) expresses the probability of premature failures of compressor 5 and depends on a sum of durations Hchf»pnof pre-recorded malfunctions for components other than compressor 5. For example, such failures may include: lack of three-phase power supply or cooling water supply problems (absent or too high temperature), or cooling water with inadequate acidity.

[0056]

[0053] More specifically, the second contribution C2 to the unreliability index RI can be expressed as:

[0057] C2= max (0, T2* / fcmp)=max[0, (Hemp ■ HMinCmp + Fgpcmp) * / fcmp) (3)

[0058]

[0054] In formula (3), the terms defined above can be recognized, while the function f»ax(0,f(x)) is, also in this case, the Rectified Linear Unit function. The factor Kemp that multiplies the sum T2 represents another weight coefficient that can be determined based on experience.

[0055] The sixth term F^cmp can be calculated as expressed by the following formula (4):

[0059]

[0056] Where Bcmpnis a binary number that represents (when it takes the value 1) the presence of a recorded malfunction, Hcmpnrepresents its duration, and M is the total number of malfunctions taken into consideration (with n ranging from 1 to M). As expressed by formula (4), the sixth term F cmp is calculated by raising the sum of the durations of the malfunctions to an exponential factor (y(;zvy. which is chosen based on experience.

[0060]

[0057] Advantageously, the unreliability index RI can be calculated by considering a third contribution Fgs that depends on a sum of the duration of previous failure events En, classified as serious and relating to the entire superconducting magnet 200. Examples of failures that can be defined as serious are: compressor failure with mechanical component failure resulting in contamination of the helium gas with oil / water / air; spontaneous quench resulting from the presence of ice in the magnet; quench resulting from low liquid helium level in the container.

[0061]

[0058] The third contribution Fgs can be expressed by the following formula

[0062]

[0059] where X is the maximum number of events of interest (with n ranging from 1 to X), p is an exponent, and Ke is a weighting factor, which are chosen based on experience.

[0063]

[0060] Preferably, the unreliability index RI can be calculated only by considering the term T1 and, in particular, as expressed by formula (1) showing the calculation of the first contribution Cl. Alternatively, the unreliability index RI can be calculated by also considering the term T2 and, in particular, as the sum of the first contribution Cl and the second contribution C2, expressed by formula (3).

[0064]

[0061] According to another embodiment, the unreliability index RI is also based on the third contribution Fgs and can be calculated as the sum of the first contribution Cl, the second contribution C2, and the third contribution as expressed by the particular formula (5). According to this embodiment, the unreliability index RI can be expressed by the formula (6) shown below:

[0065] According to the example described, the higher the value of the unreliability index RI, the greater the unreliability.

[0066]

[0062] For example, if the unreliability index RI, calculated according to one of the methods described above, exceeds a predetermined threshold Th, it is possible to intervene on the superconducting magnet 200 with maintenance, replacement, or repair of components, avoiding or reducing the possibility of failures. This is very useful in avoiding serious malfunctions such as those related to quench.

[0067]

[0063] It is possible to predict that when the threshold Th is exceeded, the monitoring system 300 will generate a warning message MSG (containing, for example, recommendations) that can be received by an operator of the system itself. This MSG message with a recommendation can be transmitted in various possible forms: SMS, voice call, app notification, email, etc.

[0068]

[0064] The method of calculating unreliability as described with reference to for the monitoring system 300can be additional to fault detection methods based on the reading of physical parameters provided by sensors. As is clear from the above description, the computer program SW is based primarily on statistical and / or historical data describing the history of the superconducting magnet itself or of magnets or components similar to those of the monitored device. In fact, the Applicant has noted that, by taking into account previous faults, particularly reliable predictions can be obtained, as each fault has consequences for the monitored device. This approach allows for highly reliable long-term predictions (e.g., in the form of recommendations).

[0069]

[0065] The monitoring system 300is suitable for management, for example, by a so-called Multivendor company and allows that company to offer an extremely useful service to users of the superconducting magnet 200.

[0070] Key to the symbols shown in the figures generation system 100 superconducting magnet 200 monitoring system 300 superconducting coil 1 cryogenic container 2 cryogenic bath 3 vacuum container 4 compressor 5 cooling head 6 drive unit 7 electric motor 8 cooling stage 9 high-pressure port 10 low pressure port 11 high pressure pipe 13 low pressure pipe 14 first heat exchanger 15, second heat exchanger 16 electrical cable 17 metallic current conductor superconductive current conductor 19 connected to the metallic current conductor 18. The superconducting current conductor 19 is connected to the superconducting coil 1 temperature sensor 20 level sensor 21 controller 25 first connection line 26 second connection line 27

[0071] CMP computer database DB unreliability index calculation software SW unreliability index RI

[0072] INTF interfaces

[0073] HTML reference pages broker BRIC

Claims

CLAIMS1. Monitoring system (300) of a superconducting magnet (200), comprising: a computer (CMP) that can be connected to the superconducting magnet (200) comprising: a container (2) in which a cryogenic bath (3) and a superconducting coil (1) are arranged; a cryogenic system equipped with a compressor (5) and a cooling head (6) configured to maintain the cryogenic bath (3) at an operating temperature of the superconducting coil (1); wherein said computer (CMP) includes a computer program (SW) configured to receive historical / statistical data relating to the superconducting magnet and calculate an unreliability index RI of the cryogenic system by a sum of the following terms: o a first term HCH which expresses an operating time at the current time of the cooling head (6); o a second term HMinCH, subtractive, expressing a statistical minimum duration of cooling heads, o a third term F^CH which expresses a probability of premature malfunction of the cooling head (6) and which depends on a first summation of pre-recorded durations Hch„ of malfunctions of components of the superconducting magnet (200) other than said cooling head (6).

2. System (300) according to claim 1, in which said computer program is configured to calculate the third term by raising said first summation of durations Hch„ by an exponential factor Qch.

3. System (300) according to claim 2, in which said computer program is configured to multiply the sum of the first term, the second term and the thirdterm by a first weight factor KCH4. System (100) according to claim 1, wherein the computer program is configured to calculate the unreliability index RI by adding to said first, second and third terms: o a fourth term Hemp expressing an operating time at the current time of said compressor (5); o a fifth term / fcmpMin, subtractive, expressing a statistical minimum duration of compressors, o a sixth term Fgpcmp which expresses the probability of premature compressor malfunction (5) and which depends on a second summation of pre-recorded durations Hcm[)nof malfunctions of superconducting magnet components (200) other than said compressor (5).

5. System (300) according to claim 4, in which said computer program is configured to calculate the sixth term Fgpcmp by raising said second summation of durations Hcm[)nby an additional exponential factor Qcmp.

6. System (300) according to claim 5, in which said computer program is configured to multiply the sum of the fourth term, the fifth term and the sixth term by a second weight factor Kcmp7. System (300) according to claim 1, wherein said computer program is configured to calculate the unreliability index RI of the cryogenic system by additionally adding a seventh termrepresenting past malfunctions classified as severe.

8. System (300) according to claim 1, in which: said malfunctions of components of the superconducting magnet(200) other than said cooling head (6) include at least one of the following malfunctions: compressor malfunction (5), compressor purification filter malfunction, compressor inlet and outlet piping malfunction (13, 14), quench, coil malfunction (1): said malfunctions of components of the superconducting magnet (200) other than said compressor (5) include at least one of the following malfunctions: no three-phase power supply, no cooling water supply, cooling water temperature too high, cooling water of unsuitable acidity; such past malfunctions classified as serious include at least one of the following malfunctions: compressor malfunction with malfunction of a mechanical component with contamination of refrigerant with oil and / or water and / or air; spontaneous quench resulting from the presence of ice in the magnet; quench resulting from a low level of cryogenic bath in the container (3).

9. Magnetic field generation system (100) comprising: a superconducting magnet (200) and a remote monitoring system (300) as defined by at least one of the preceding claims.

10. Generation system (100) according to claim 9, wherein said superconducting magnet (200) is configured to operate in one of the following apparatuses: magnetic resonance apparatus, silicon single crystal extraction device using a Czochralski particle accelerator method, cyclotron.

11. Magnetic field generation system (100) according to claim 9, in which: said computer (CMP) stores a threshold value and the computer program(SW) is such that it compares the unreliability index RI and with the threshold value and generates a message (MSG) following a comparisoncontaining recommendations for at least one of the following: maintenance work, replacement of components, repair of components12. Method of monitoring a superconducting magnet, comprising: connect a computer (CMP) to the superconducting magnet (200) comprising: a container (2) in which a cryogenic bath (3) and a superconducting coil (1) are arranged; a cryogenic system equipped with a compressor (5) and a cooling head (6) configured to maintain the cryogenic bath (3) at an operating temperature of the superconducting coil (1); receive computer (CMP) historical / statistical data on the superconducting magnet; (200) calculate by computer (CMP) an unreliability index RI of the cryogenic system by a sum of the following terms: o a first term HCH which expresses an operating time at the current time of the cooling head (6); o a second term HMinCH, subtractive, expressing a statistical minimum duration of cooling heads, o a third term F^CH which expresses a probability of premature malfunction of the cooling head (6) and which depends on a first summation of pre-recorded durations Hch„ of malfunctions of components of the superconducting magnet (200) other than said cooling head (6).

Images