Ground fault detection of UPS battery
11500033 · 2022-11-15
Assignee
Inventors
- Daniel Aggeler (Jona, CH)
- Esa-Kai Paatero (Helsinki, FI)
- Ivan Furlan (Brissago, CH)
- Nicola Notari (Gentilino, CH)
- Yannick Maret (Dättwil, CH)
Cpc classification
G01R31/389
PHYSICS
G01R31/385
PHYSICS
International classification
G01R31/52
PHYSICS
G01R31/385
PHYSICS
G01R31/389
PHYSICS
Abstract
The present invention provides a method for detecting a ground fault in a battery of a uninterrupted power supply, the battery includes at least one string with multiple battery cells, the method including the steps of defining multiple individual battery blocks of battery cells along the at least one string, performing an reference impedance measurement for the multiple individual battery blocks at a first point of time, performing a verification impedance measurement for the multiple individual battery blocks at a second point of time, evaluating a change of measured impedance between the reference impedance and the verification impedance for the multiple individual battery blocks of the at least one string, and identifying a ground fault based on a correlated change of measured impedance of the multiple individual battery blocks along the at least one string. The present invention also provides a battery management system for managing a battery of a uninterrupted power supply, which is adapted to perform the above method. The present invention further provides a UPS device and a UPS system, each of which including an above battery management system.
Claims
1. A method for detecting a ground fault in a battery of an uninterrupted power supply, the battery comprises at least one string with multiple battery cells, the method comprising the steps of: defining, by a battery management system, multiple individual battery blocks of battery cells along the at least one string, performing, by the battery management system, a reference impedance measurement of the multiple individual battery blocks at a first point of time, performing, by the battery management system, a verification impedance measurement for the multiple individual battery blocks at a second point of time, evaluating, by the battery management system, a change of measured impedance between the reference impedance and the verification impedance for the multiple individual battery blocks of the at least one string, and identifying, by the battery management system, a ground fault based on a correlated change of measured impedance of the multiple individual battery blocks along the at least one string, wherein the step of identifying a ground fault based on a correlated change of measured impedance of the individual battery blocks along the at least one string comprises a step of localizing a position of the ground fault by identifying a sequence of individual battery blocks on one side of the ground fault with a correlated change of measured impedance compared to another sequence of individual battery blocks on the other side of the ground fault along the at least one string.
2. The method according to claim 1, wherein the step of defining multiple individual battery blocks of battery cells along the at least one string comprises defining each of the multiple individual battery blocks comprising few individual battery cells.
3. The method according to claim 2, wherein the steps of performing a reference impedance measurement and performing a verification impedance measurement for the multiple individual battery blocks each comprise: generating, by the battery management system, at least one current pulse through the battery, measuring, by the battery management system, a voltage across each of the multiple individual battery blocks as response to the at least one current pulse, measuring, by the battery management system, a current across the multiple individual battery blocks as response to the at least one current pulse, and determining, by the battery management system, the impedance of each individual battery block based on the voltage and the current measured across the individual battery block as response to the at least one current pulse.
4. The method according to claim 3, wherein the step of generating at least one current pulse through the battery comprises generating at least one charge pulse and/or at least one discharge pulse.
5. The method according to claim 1, wherein the steps of performing a reference impedance measurement and performing a verification impedance measurement for the multiple individual battery blocks each comprise: generating, by the battery management system, at least one current pulse through the battery, measuring, by the battery management system, a voltage across each of the multiple individual battery blocks as response to the at least one current pulse, measuring, by the battery management system, a current across the multiple individual battery blocks as response to the at least one current pulse, and determining, by the battery management system, the impedance of each individual battery block based on the voltage and the current measured across the individual battery block as response to the at least one current pulse.
6. The method according to claim 5, wherein the step of generating at least one current pulse through the battery comprises generating at least one charge pulse and/or at least one discharge pulse.
7. The method according to claim 6, wherein the step of generating at least one current pulse through the battery comprises generating a current pulse train.
8. The method according to claim 7, wherein the method further comprises verifying a validity of the reference impedance measurement based on an elapsed period of time between the reference impedance measurement and the verification impedance measurement, and performing a further reference impedance measurement when the elapsed period of time exceeds a given time limit.
9. The method according to claim 5, wherein the step of generating at least one current pulse through the battery comprises generating a current pulse train.
10. The method according to claim 6, wherein the method further comprises verifying a validity of the reference impedance measurement based on an elapsed period of time between the reference impedance measurement and the verification impedance measurement, and performing a further reference impedance measurement when the elapsed period of time exceeds a given time limit.
11. The method according to claim 1, wherein the method further comprises verifying a validity of the reference impedance measurement based on an elapsed period of time between the reference impedance measurement and the verification impedance measurement, and performing a further reference impedance measurement when the elapsed period of time exceeds a given time limit.
12. The method according to claim 1, wherein the step of identifying a sequence of individual battery blocks with a correlated change of measured impedance compared to another sequence of individual battery blocks along the at least one string comprises determining a change of measured impedance individually for each individual battery block.
13. The method according to claim 12, wherein the step of identifying a sequence of individual battery blocks with a correlated change of measured impedance compared to another sequence of individual battery blocks along the at least one string comprises performing a change detection algorithm or a signal segmentation algorithm.
14. The method according to claim 1, wherein the step of identifying a sequence of individual battery blocks with a correlated change of measured impedance compared to another sequence of individual battery blocks along the at least one string comprises performing a change detection algorithm or a signal segmentation algorithm.
15. A battery management system for managing a battery of an uninterrupted power supply, the battery comprising at least one string with multiple battery cells, the battery management system is operable to: define multiple individual battery blocks of battery cells along the at least one string; perform a reference impedance measurement of the multiple individual battery blocks at a first point of time; perform a verification impedance measurement for the multiple individual battery blocks at a second point of time; evaluate a change of measured impedance between the reference impedance and the verification impedance for the multiple individual battery blocks of the at least one string; and identify a ground fault based on a correlated change of measured impedance of the multiple individual battery blocks along the at least one string; wherein the step of identifying a ground fault based on a correlated change of measured impedance of the individual battery blocks along the at least one string comprises a step of localizing a position of the ground fault by identifying a sequence of individual battery blocks on one side of the ground fault with a correlated change of measured impedance compared to another sequence of individual battery blocks on the other side of the ground fault along the at least one string.
16. A UPS device comprising a central DC link, a power supply side AC/DC converter, a power supply side DC/DC converter, and a load side output converter, whereby all converters are connected to the DC link, and the DC/DC converter is connected to a battery at its power supply side, wherein the UPS device comprises the battery management system according to claim 15.
17. A UPS system comprising multiple UPS devices, each UPS device comprising a central DC link, a power supply side AC/DC converter, a power supply side DC/DC converter, and a load side output converter, whereby all converters are connected to the DC link, and the DC/DC converters of the multiple UPS devices are connected to a battery at their power supply side, wherein the UPS system comprises the battery management system according to claim 15.
Description
BRIEF DESCRIPTION OF DRAWINGS
(1) These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter.
(2) In the drawings:
(3)
(4)
(5)
(6)
(7)
DETAILED DESCRIPTION OF THE INVENTION
(8)
(9) The UPS device 10 of the first embodiment comprises a central DC link 12, a power supply side AC/DC converter 14, a power supply side DC/DC converter 16, and a load side output converter 18, which is a DC/AC converter 18 in this embodiment. All converters 14, 16, 18 are connected to the DC link 12. Although the AC/DC converter 14 and the DC/DC converter 16 are connected to different types of power supplies, they are both here considered as connected at a power supply side of the UPS device 10. The DC link 12 further comprises storage capacitors 13, one of which is shown by way of example in
(10) The battery 22 comprises multiple strings 26, which are provided in parallel within the battery 22. Each of the strings 26 comprises multiple individual battery blocks 28, which are connected in series in each string 26. According to the first embodiment, each individual battery block 28 comprises one battery cell. Hence, each of the individual battery blocks 28 has the same number of battery cells. The parallel strings 26 have the same setup with the same number of individual battery blocks 28. The setup of a string 26 of the battery 22 can be seen e.g. in
(11) The battery 22 can be integral part of the UPS device 10, or the battery 22 can be a separate component, which is operated through the UPS device 10.
(12) The UPS device 10 of the first embodiment further comprises a bypass 30 with a bypass switch 32, which is provided in this embodiment as silicon controlled rectifier, also referred to as scr. The bypass 30 provides a connection between the AC power supply 20 and the load 24, which is provided in parallel to the AC/DC converter 14, the DC link 12, and the output converter 18.
(13) The UPS device 10 of the first embodiment also comprises a controller 34, which controls the operation of all controllable components of the UPS device 10, i.e. the AC/DC converter 14, the DC/DC converter 16, the output converter 18 bypass switch 32. Furthermore, the controller 34 receives measurement results from current and/or voltage measurements as performed by the DC/DC converter 16.
(14) The UPS device 10 of the first embodiment further comprises a battery management system (BMS) 36. The BMS 36 is part of the UPS device 10. However, in an alternative embodiment, the BMS 36 is a separate device. Since the BMS 36 is part of the UPS device 10, it can interact with the UPS device 10, e.g. when doing the impedance measurements of the individual battery blocks 28. The BMS 36 is provided to monitor different parameters of the battery 22, which is done by measurements of voltage and current of the individual battery blocks 28, as discussed later in more detail.
(15) Further components of the UPS device 10, which are not essential for understanding the present invention, are not shown in
(16) A UPS system 40 according to a second embodiment can be seen in
(17) As can be further seen in
(18) Subsequently, a method for detecting a ground fault 60 in a battery 22 of the uninterrupted power supply device 10 or the uninterrupted power supply system 40 will be discussed with respect to
(19) The method starts with step S100, which refers to defining multiple individual battery blocks 28 of battery cells along each string 26 of the battery 22. In this embodiment, each individual battery block 28 comprises one battery cell.
(20) According to step S110, a reference impedance measurement is performed for each of the multiple individual battery blocks 28 at a first point of time. Hence, the BMS 36 uses the UPS device 10 to create a current pulse train through each string 26 of the battery 22. The current pulse train comprises a sequence of multiple pulses, which a sequence of different or identical discharge and charge pulses, so that a sum of energy transfer between the battery 22 and the DC/DC converter 16 is almost zero. The current pulse train comprises a break or gap between two subsequent pulses, where no current flows through the battery 22. The sequence of pulses is generated in order to chemically excite the battery 22 and to clear effects such as “coup de fouet”. “Coup de fouet” refers to a phenomenon associated with voltage drop at the beginning of discharge of the battery 22, in particular a lead acid battery.
(21) Measurements of a voltage and a current across each of the multiple individual battery blocks 28 are performed as response to the sequence of current pulses after clearing this effect. The impedances can be seen in
(22)
(23) The current pulses have any suitable shape, as long as their frequency content is rich, e.g. square, pure sinusoidal signals, square wave, PRBS, white noise, or others. A pulse current of the current pulse is sufficiently large in order to produce a well measurable voltage drop. Additionally, the pulse current and the duration of the pulse are small enough to not increase the temperature of the battery 22 by more than a fraction of Kelvin. Typically, a pulse current of 5-10 A is used for monitoring the impedance of the battery 22, i.e. the individual battery blocks 28, when the battery 22 has an impedance in the order of tens of milliohms. A pulse duration of the current pulse is in the order of hundreds of milliseconds. A rising edge and a falling edge of each current pulse is sharp enough in order to have a rich-enough frequency content of e.g. up to a few hundred hertz. In general, the bandwidth of a step signal is given by 0.35/RT, where RT is the rise time of the edge. Accordingly, a bandwidth of e.g. 300 Hz requires a rise time of less than 1 ms. Together with the measurement of the reference impedance, environmental parameters including e.g. a temperature and an humidity are measured and recorded. Furthermore, also other parameters including e.g. a state of charge of the battery 22 are performed and recorded.
(24) According to step S120, a validity of the reference impedance measurement is verified based on an elapsed period of time between the reference impedance measurement and the verification impedance measurement. In case the elapsed period of time exceeds a given time limit, which is three months in this embodiment, the method returns to step S110 and a further reference impedance measurement is performed. Otherwise, the method continues with step S130.
(25) According to step S130, a verification impedance measurement is performed for the multiple individual battery blocks 28 at a second point of time. The measurement of the verification impedances is performed as discussed above for the reference impedances. Together with the measurement of the verification impedance, environmental parameters including e.g. a temperature and a humidity are measured and recorded. Furthermore, also other parameters including e.g. a state of charge of the battery 22 are performed and recorded.
(26) According to step S140, a change of measured impedance between the reference impedance and the verification impedance for the multiple individual battery blocks 28 of each string 26 is evaluated. The change of measured impedance is a difference between the reference impedance and the verification impedance for each individual battery block 28. Since battery parameters can change e.g. depending on a temperature, a state-of-charge, or other, it is verified if the measurements of the verification impedance and the reference impedance were performed under the same or at least similar conditions as defined e.g. by the above parameters. In case of a significant variation of the parameters, a correction of the measured verification impedance and/or the reference impedance is performed based on these parameters.
(27) According to step S150, ground fault 60 is identified based on a correlated change of measured impedance of the multiple individual battery blocks 28 along the string 26. The ground fault 60 refers to a connection of any point of the string 26 of battery cells to ground 62, e.g. based on a battery leakage shorting to ground 62. Ground faults 60 are detected for each string 26 individually. Already a change of an overall measured impedance of the battery 22 or of one string 26 of the battery 22 indicates presence of a ground fault 60. If a statistically meaningful change in the measured impedance of the individual battery blocks 28 cannot be found, no ground fault 60 is present. If a meaningful change, or a segmentation, can be found, this is an indication of a ground fault 60, and the location of the change in the measured impedance through the string 26 of individual battery blocks 28 indicates a location of the ground fault 60, as discussed below in more detail.
(28) The identification of a ground fault 60 is based on a correlated change of measured impedance of the individual battery blocks 28 along each string 26 of the battery 22. When considering a ground fault 60 along the string 26 of battery cells, a number of N individual battery blocks 28 is located at one side of the ground fault 60, and a number of M individual battery blocks 28 is located at the other side of the ground fault 60. The current through N individual battery blocks 28 is essentially identical, and the current through M individual battery blocks 28 is essentially identical. The current through the respective groups of individual battery blocks 28 differs in a ground current in case of the ground fault 60. When the ground fault 60 occurs, the measured impedance of the M or the N individual battery blocks 28 changes, which indicates the occurrence of the ground fault 60, as can be seen in
(29) According to a first approach, a change of measured impedance can be calculated based on
(30)
(31) With an impedance Z.sub.F of the ground fault 60 being much larger than an impedance Z.sub.G to ground 62, and the impedance Z.sub.B being the impedance of an individual battery block 28, the above can be approached by
(32)
(33) In the embodiment shown in
(34) If a ground fault 60 actually developed between the first and the second point of time, the measured impedance of a whole group of individual battery blocks 28 changed. Since all individual battery blocks 28 in this group experience the same changed current, the estimated impedances for all individual battery blocks 28 in this group exhibit the same percent error. Hence, the increase of measured impedance of each potential location of ground fault 60 is checked, i.e. between each pair of individual battery blocks 28. In other words, the BMS 36 assumes a ground fault 60 at one individual battery block 28 and checks for a consistent change of measured impedance in the group consisting of the subsequent individual battery blocks 28 in the string 26. The BMS 36 steps through the entire string 26 and evaluated the change of measured impedance for each possible group of individual battery blocks 28. This enables localizing a position of the ground fault 60 by identifying a sequence of individual battery blocks 28 with a correlated change of measured impedance compared to another sequence of individual battery blocks 28 along each string 26.
(35) A detailed approach is specified in mathematical terms as follows. With ΔZ.sub.k=1 . . . N being a relative change of measured impedance between the first and the second point of time for an individual battery block 28 identified by index k within the string 26 comprising N individual battery blocks 28, it is assumed that all individual battery blocks 28 share the same current under ideal conditions. Here, ΔZ.sub.k=Z.sub.k′/Z.sub.k, where Z.sub.k is the reference impedance prior to the occurrence of the ground fault 60, and Z.sub.k′ is the verification impedance after the ground fault 60 occurred. If the k-th individual battery block 28 is in a group m of individual battery blocks 28, that sees the full test current, as can be seen in
(36) If the individual battery block 28 is in the group n of individual battery blocks 28, that sees the only part of the test current, the change of measured impedance becomes ΔZ.sub.k=Z.sub.Σn/Z.sub.F, where Z.sub.Σn is the sum of the measured impedances of all individual battery blocks 28 experiencing a reduced current. Using this approach, ΔZ.sub.k has theoretically the same value for each individual battery block 28 within the group of m individual battery blocks 28 experiencing the full test current (i.e. ΔZ.sub.k=0), and ΔZ.sub.k has another value, which is constant, for each individual battery block 28 within the group of n individual battery blocks 28 experiencing only part of the test current, which is approximately Z.sub.Σn/Z.sub.F. In practice, there will be slight differences because of measurement noise and errors.
(37) According to
(38) Under further consideration of k.sub.−f being sample average of the ΔZ.sub.k for k=1 . . . (f−1) and X.sub.+f being sample average of the ΔZ.sub.k for k=(f−1) . . . N. If X.sub.+f is different from zero in a statistically significant manner, it implies that there is high likelihood of a ground fault 60 close to individual battery block 28 with index f. This can, for example, be assessed via a statistical test such as the Student's t-test.
(39) For the particular case of the Student's t-test, when considering a group of ten individual battery blocks 28 whose impedances are measured with a statistical accuracy equivalent to a standard deviation of one percent point. This means that a 95% of the measured values are within ±2% of the mean. It can then be shown that if the sample average X.sub.+f is larger than 0.7%, the difference is statistically significant with a confidence interval of 95%. Equivalently, if the impedance of each of the individual battery blocks 28 is measured with a statistical accuracy equivalent to a standard deviation of one tenth of a percent point, then a variation of a group impedance by 0.07% can be detected—which is the typical value estimated previously. This can be achieved with a measurement, which does not have to have an increased accuracy. If the measurement system is consistent, it is sufficient to average the measured impedance a sufficient number of times in order to obtain the required statistical accuracy. Alternatively, a known SNR (signal to noise ration) improvement method can be used to obtain a good impedance estimate of the individual battery blocks 28 to start with. Additionally, there is the presence of a test group, i.e. those batteries that do not experience the changed current. The group test can allow to further improve the reliability of the method.
(40) Furthermore, a change detection algorithm or a signal segmentation algorithm can be performed. A set of subsequent individual battery blocks 28 prior to a location of a ground fault 60 and a set of subsequent individual battery blocks 28 further on from the location of the ground fault 60 are determined.
(41) While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments. Other variations to be disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word “comprising” does not exclude other elements or steps, and the indefinite article “a” or “an” does not exclude a plurality. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be construed as limiting scope.
REFERENCE SIGNS LIST
(42) 10 uninterruptible power supply device, UPS device, uninterruptible power supply 12 DC link 13 storage capacitor 14 AC/DC converter 16 DC/DC converter 18 output converter, DC/AC converter 20 AC power supply 22 battery 24 load 26 string 28 battery block 29 system impedance to ground 30 bypass 32 bypass switch 34 controller 36 battery management system, BMS 40 uninterruptible power supply system, UPS system, uninterruptible power supply 42 communication bus 44 user interface 46 AC power supply bus 48 DC battery supply bus 50 load bus 60 ground fault 62 ground