VALIDATION OF PHASE CURRENTS IN A MULTI-PHASE SYSTEM

Abstract

In order to ensure safe operation of a multi-phase system, even a system including a plurality of phases, a number of phase groups is provided, which comprises some of the phases, wherein phase currents of the number of phase groups are merged in a group node to form a group sum current and a group sum current measurement value of the group sum current is captured. The current measurement values belonging to the number of phase groups are summed up to form a group sum and the group sum is compared with the group sum current measurement value to validate the phase currents of the phases in order to ensure safe operation.

Description

The present invention relates to a method for safely operating a multi-phase system, in which method a phase current flows in each of a plurality of phases of the multi-phase system, and current measurement values of the phase currents are captured by phase current sensors in order to validate the phase currents. Furthermore, the present invention relates to a multi-phase system comprising a plurality of phases, a phase current flowing in each phase, current sensors being provided for capturing current measurement values of the phase currents

In multi-phase systems, for example electromagnetic transport units, such as long stator linear motors or planar motors, phase currents of the individual phases are determined. To ensure that no errors occur when determining or processing the phase currents, it is often desirable to validate the phase currents, i.e. to ensure that the determined phase currents are correct. One option for validating the determined phase currents is to use Kirchoff's rule of the sum of currents. According to Kirchhoff's first law, the sum of all currents in a node equals zero. In other words, the sum of all currents flowing into a node corresponds to the sum of all currents flowing out of the node. Accordingly, in order to validate the phase currents, the phase currents can be merged into a common node to form a phase current sum and the phase current sum can be determined. If the determined phase current sum in the node corresponds to the sum of the individually determined phase currents, the phase currents can be validated, which means that the measurement of the phase currents is to be considered as valid. EP 3 109 999 A2 discloses such a method for validating the phase currents of a multi-phase system.

When determining the individual phase currents, however, a permissible tolerance, i.e. a maximum permissible error, must be taken into account in each case. However, an error occurring in a phase, e.g. a measurement error in the phase current, which is in the tolerance range, cannot be detected in this way. Since a maximum permissible error must be taken into account in each case for all phase currents and the phase currents are added to form a phase current sum, the maximum permissible error also adds up. This means that the maximum permissible error in the phase current sum corresponds to the sum of the maximum permissible errors of the individual phase currents. The permissible error in the phase current sum increases in a proportionally linear manner together with the number of phases (and thus the number of phase currents to be determined). The high maximum permissible error in the phase current sum results in a low level of accuracy when validating the phase currents, since this level of accuracy deteriorates in a linear manner together with the number of phases.

In particular in the case of a large number of phases, the permissible error in the phase current sum can thus reach such high values that it is not possible to meaningfully validate the phase currents on the basis of the phase current sum, which means that safe operation of the multi-phase system cannot be ensured.

The problem addressed by the present invention is therefore that of providing a method for the safe operation of a multi-phase system even having a large number of phases.

According to the invention, this problem is solved by providing a number of phase groups which comprise some of the phases, the phase currents of the number of phase groups being merged in a group node to form a group sum current, and by a group sum current measurement value of the group sum current being captured by a group current sensor, the current measurement values associated with the number of phase groups being summed up by a group summation unit to form a group sum, and the group sum being compared with the group sum current measurement value by a group comparison unit to validate the phase currents of the phases in order to ensure safe operation of the multi-phase system.

Furthermore, the problem is solved by a multi-phase system in which a number of phase groups is provided which comprises some of the phases, a group node being provided, in which the phase currents of the number of phase groups are merged to form a group sum current, a group current sensor being provided for detecting a group sum current measurement value of the group sum current, a group summation unit being provided, which is designed to sum up the current measurement values of the phase currents belonging to the number of phase groups to form a group sum, a group comparison unit being provided, which is designed to compare the group sum with the group sum current measurement value to validate the phase currents of the phases in order to ensure safe operation of the multi-phase system.

When determining a current measurement value of a phase current, a maximum deviation of the current measurement value from the actual phase current must be taken into account in the form of a maximum permissible error. According to the invention, some of the phases are merged in at least one phase group in a group node, which results in a group sum current for the respective phase currents in the group node. Furthermore, the current measurement values of the phases belonging to the at least one phase group are summed up to form a group sum. For the group sum, this results in a maximum permissible error as sum of the permissible errors in the phases contained in the particular phase group. Furthermore, the group sum is compared with the group sum current measurement value in order to validate the phase currents of the phase group.

If, instead, all the phases of the multi-phase system were to be merged to form a total sum current in a total node and the current measurement values of all the phases were added up to a total sum, the maximum permissible error would be the sum of the permissible errors of all the phases.

According to the invention, combining some of the phases into a phase group results in a maximum permissible error in the group sum current measurement value, which is less than it would be for the total sum current measurement value. This allows the phase currents to be validated with greater accuracy. This means that the tolerance for the deviation from the group sum current to the group sum current measurement value is lower, which means that smaller errors can also be detected during validation. By combining phases into phase groups, the accuracy of the validation is independent of the number of phases or phase currents in the multi-phase system. This validation with a high level of accuracy can ensure safe operation of the multi-phase system. It is ensured with a low error tolerance that no, or only small, errors occur during the measurement and/or processing of the phase currents. If the validation fails, at least one action can be triggered by an action unit, for example an optical and/or acoustic signal can be output. Likewise, in the event of a failed validation, the multi-phase system or part of it can be shut down as an action. As an action, for example, the multi-phase system can also be prevented from switching on. A signal can also be generated as an action, which signal is processed further, for example by a control unit of the multi-phase system.

The principle of forming phase groups to increase the accuracy of the safe determination of a phase current can be applied to any physical measuring principles for current measurement and can also be used in inhomogeneous systems involving a different current measurement. Phase groups can thus be formed independently of the measuring principle used (use of shunts, use of the Hall effect, . . . ), it being possible to use the same or at least partially different measuring principles in the method according to the invention for measuring different currents.

Of course, in addition to the method according to the invention, all the phases can also be combined in a total node to form a total sum current, and a total sum current measurement value can be determined, which is compared with the sum of the current measurement values of all the phases. An (additional) validation based on the total sum current is, however, subject to a maximum permissible error as the sum of the maximum permissible errors of all the phases.

Preferably, the phase currents of all the phases of the multi-phase system are divided into a number of phase groups, in each phase group the phase currents are merged in a group node to form a group sum current and a group sum current measurement value of the group sum current is captured, the current measurement values of the phase currents in each phase group being summed up by a group summation unit to form a group sum, and, for each phase group, the group sums being compared with the associated group sum current measurement values by a group comparison unit to validate the phase currents of all the phases within the number of phase groups.

By dividing all the phases into phase groups, the phase currents of all the phases can be validated. For each phase group, this results in a maximum permissible error as the sum of the permissible errors of the phases contained in the particular phase group. Since each phase group comprises only some of the phases of the multi-phase system, the maximum permissible error in the phases contained in the phase group is thus a lower maximum permissible error in the validation than would be the case for a total sum current. Viewed as a whole, the accuracy of the validation of the phase currents of all the phases depends on the greatest maximum permissible error in the phase groups involved.

The number of phase groups can comprise the same number of phases. This results in a symmetrical and simple structure. The phase groups can also be nested within one another. However, each phase is preferably only comprised by one phase group, in particular if all the phases are divided into phase groups.

Phase groups having a different number of phases can also be provided. This can be particularly advantageous if different maximum permissible errors are to be expected when determining the current measurement values of the respective phases. This can be the case in particular when different sensors are used.

It can be advantageous if phases of which the current measurement values have a larger maximum permissible error are captured in smaller phase groups, i.e. phase groups having fewer phases, and phases of which the current measurement values have a smaller maximum permissible error are captured in larger phase groups, i.e. phase groups having more phases.

The current measurement values of the phase currents are preferably reliably captured in accordance with a predefined safety requirement, a single-fault security measure being preferably required as a safety requirement for determining the current measurement values.

If the current measurement values are determined by means of a sigma-delta conversion, this safety requirement can relate to an implementation in an FPGA; in the case of LEM converters, it can relate to an ADC converter.

The method according to the invention can be used in all multi-phase systems having a plurality of n>2 phases, in particular in long stator linear motors or planar motors. If all the phases of the multi-phase system are to be combined in phase groups, a plurality of n>3 phases is required.

In the case of a number of four phases, for example, two phase groups each having two phases can be provided.

In the following, the present invention shall be described in more detail with reference to FIGS. 1 and 2 which, by way of example, show schematic and non-limiting advantageous embodiments of the invention. In the drawings:

FIG. 1 shows a validation of the phase currents according to the prior art;

FIG. 2 shows a validation of the phase currents in respective phase groups.

FIG. 1 shows a validation of n phase currents i₁, . . . , i_n of a multi-phase system 1 according to the prior art. The multi-phase system 1 can be, for example, an electromagnetic transport system, e.g. a long stator linear motor or planar motor, and comprises a plurality n of phases L₁, . . . , L_n. A phase current i₁, . . . , i_n flows in each phase L₁, . . . , L_n, a phase current sensor S₁, . . . , S_n being provided in each phase L₁, . . . , L_n to determine a current measurement value i_M1, . . . , i_Mn of the phase current i₁, . . . , i_n. The individual current measurement values i_M1, . . . , i_Mn each have a maximum error ε₁, . . . , ε_n in relation to the actual phase current i₁, . . . , i_n, e.g. due to system-related measurement errors. This maximum error ε₁, . . . , ε_n must be taken into account during validation. The first current measurement value i_M1 thus consists of the first phase current i₁ and a first maximum error ε₁: i_M1=i₁+ε₁; the second current measurement value i_M2 consists of the second phase current i₂ and a second maximum error ε₂: i_M2=i₂ ε₂, etc.; and the n-th current measurement value i_Mn consists of the n-th phase current i_n and an n-th maximum error ε_n: i_Mn=i_n+ε_n.

The maximum errors ε₁, . . . , ε_n themselves can each be assumed to be positive or negative, it being possible for an error tolerance band to also be assumed as the maximum possible error ε₁, . . . , ε_n. A positive maximum error ε₁, . . . , ε_n means a maximum possible excess of the particular phase current i₁, . . . , i_n by the associated current measurement value i_M1, . . . , i_Mn; a negative maximum error ε₁, . . . , ε_n means a maximum possible shortfall of the particular phase current i₁, . . . , i_n by the associated current measurement value i_M1, . . . , i_Mn. A maximum error ε₁, . . . , ε_n as the tolerance band means that the maximum possible excess and/or shortfall of the particular phase current i₁, . . . , i_n is assumed by the associated current measurement value i_M1, . . . , i_Mn. In a summation unit E0, the current measurement values i_M1, . . . , i_Mn are summed up to form a measurement value total sum

$i_M=\sum _{j=1}^ni_{\mathrm{Mj}}.$

The measurement value total sum i_M thus comprises the sum of then phase currents i₁, . . . , i_n, plus the sum of the maximum errors ε₁, . . . , ε_n:

$i_M=\sum _{j=1}^ni_{\mathrm{Mj}}=\sum _{j=1}^ni_j+\sum _{j=1}^n{\varepsilon }_j.$

Furthermore, the phase currents i₁, . . . , i_n are merged into a total node K_g to form a sum current i_g, which results in the sum current i_g=i₁+i₂++i_n. A sum current measurement value i_Mg of the sum current i_g is determined, in this case using a sum current sensor S_g, the sum current measurement value i_Mg also having a maximum error ε_g, and thus the sum current i_g and the maximum error ε_g result in the following: i_Mg=i_g+ε_g.

In principle, positive maximum errors ε₁, . . . , ε_n and/or negative maximum errors ε₁, . . . , ε_n could be added separately, in particular if different values are assumed for the positive and negative maximum errors ε₁, . . . , ε_n. If an error tolerance band is assumed as the maximum possible error ε₁, . . . , ε_n, then a summation of the maximum errors ε₁, . . . , ε_n means expanding the error tolerance band. For ease of illustration, an addition of the maximum error ε₁, . . . , ε_n is only shown in a general manner. A summation of the maximum errors ε₁, . . . , ε_n can fundamentally take place for negative maximum errors ε₁, . . . , ε_n and/or for maximum errors ε₁, . . . , ε_n and/or for maximum errors ε₁, . . . , ε_n in the form of a tolerance band.

A comparison unit V0 now compares the measurement value total sum i_M with the sum current measurement value i_Mg, the phase currents i₁, . . . , i_n being validated if there is a sufficiently precise match. If the validation fails, an action A is triggered, for example.

During validation, however, the maximum errors ε₁, . . . , ε_n of the current measurement values i_M1, . . . , i_Mn and the maximum error ε_g of the sum current measurement value i_Mg must be taken into account, as is shown in FIG. 1 in a general manner. The maximum errors ε₁, . . . , ε_n in the current measurement values i_M1, . . . , i_Mn with respect to the phase currents i₁ . . . , i_n are therefore included in the measurement value total sum i_M. When comparing the measurement value total sum i_M with the sum current measurement value i_Mg (shown as i_M⇔i_Mg in FIG. 1), not only are the maximum error ε₁, . . . , ε_n of the current measurement values i_M1, . . . . , i_Mn also taken into account, but also the maximum error ε_g of the sum current measurement value i_Mg with respect to the sum current i_g.

This results in a high maximum permissible total error

${\varepsilon }_V={\varepsilon }_1+{\varepsilon }_2+\dots {\varepsilon }_n+{\varepsilon }_g=\sum _{j=1}^n{\varepsilon }_j+{\varepsilon }_g$

for the validation of the phase currents i₁, . . . , i_n, in particular in the case of a large number n of phases L₁, . . . , L_n. Thus, an occurring error which causes the sum current measurement value i_Mg to deviate from the total measurement value i_M by less than the maximum permissible total error

${\varepsilon }_V=\sum _{j=1}^n{\varepsilon }_j+{\varepsilon }_g$

cannot be detected. This occurring error can relate to a phase current i₁, . . . , i_n or can accumulate from a plurality of occurring errors related to a plurality of phase currents i₁ . . . , i_n.

Assuming that the maximum error ε₁, . . . , ε_n of the current measurement values i_M1, . . . , i_Mn and the maximum error ε_g of the sum current measurement value i_Mg each correspond to the identical maximum error ε₁=ε₂= . . . =ε_n=ε_x (not shown), the result for the total measurement value sum i_M is the sum of the phase currents i₁, . . . , i_n plus the n-th maximum error ε_x: i_M=i₁+i₂+, . . . , +i_n+n ε_x. In contrast, the sum current measurement value i_Mg corresponds to the sum of the phase currents i₁, . . . , i_n plus the maximum error ε_x: i_Mg=i_g+ε_x=+i₂+, . . . , +i_n+ε_x. In order to be able to compare the total measurement value sum i_M with the sum current measurement value i_Mg

$\left(i_M\stackrel{{\varepsilon }_V}{}i_{\mathrm{Mg}}\right),$

a maximum permissible total error in the size of ε_V=(n+1) ε_x must be taken into account as the error limit, the factor n originating from the determination of the current measurement values i_M1, . . . , i_Mn and corresponding to the number of phases L₁, . . . , L_n while the additional factor 1 is derived from the measurement of the sum current measurement value i_Mg.

This means that, during validation, an error that occurs which causes the sum current measurement value i_Mg to deviate from the total measurement value sum i_M by less than the maximum permissible total error ε_V=(n+1) ε_x cannot be detected.

If a long stator linear motor which has, for example, a plurality of n=42 phases L₁, . . . , L_n=42, is operated as a multi-phase system 1, then summing up the current measurement values i_M1, . . . , i_Mn=42 to form a total measurement value sum i_M and comparing with the sum current measurement i_Mg of the sum current i_g results in a maximum permissible total error of ε_V=ε₁+ε₁+ . . . ε_n=42+ε_g.

Assuming that the maximum error ε₁, . . . , ε_n=42 in the current measurement values i_M1, . . . , i_Mn=42 and the maximum error ε_g of the sum current value i_Mg each correspond to a maximum error ε_x of 100 mA, the maximum permissible total error in ε_V=(n+1) ε_x=43*100 mA=4.3 A. This means that during validation, individual errors that occur in current measurement values i_M1, . . . , i_Mn, or even accumulative errors that add up to a total error that is less than the maximum permissible total error ε_V=4.3 A are not detected.

In contrast, FIG. 2 shows a preferred embodiment of the method according to the invention and the device according to the invention. A multi-phase system 1 having a plurality n of phases L₁, . . . , L_n is also shown, phase current sensors S₁, . . . , S_n being provided in each phase L₁, . . . , L_n for measuring the respective phase currents i₁, . . . , i_n. A current measurement value i_M1, . . . , i_Mn is thus determined for each phase current i₁, . . . , i_n. The individual current measurement values i_M1, . . . , i_Mn, in turn, each have a maximum error ε₁, . . . , ε_n. As described above with reference to FIG. 1, this results in the first phase current i₁ together with a first maximum error ε₁ for the first current measurement value i_M1: i_M1=i₁+ε₁, and the second phase current i₂ together with a second maximum error ε₂ for the second current measurement value i_M2: i_M2=i₂+ε₂, up to the n-th current measurement value i_Mn, which consists of the n-th phase current i_n and an n-th maximum error ε_n: i_Mn=i_n+ε_n.

In contrast to FIG. 1, however, a plurality of m>1 phase groups G₁, . . . , G_m an are provided, each of which comprises some of the phases L₁, . . . , L_n. According to the invention, each phase group G₁, . . . , G_m comprises some of the plurality n of phases L₁, . . . , L_n. The number of phases L₁, . . . , L_n comprised by a phase group G₁, . . . , G_m is referred to in the following as the group phase number p₁, . . . , p_m. The group phase number p₁, . . . , p_n, of the phase groups G₁, . . . , G_m can be identical, or at least partially vary between the phase groups G₁, . . . , G_m. In FIG. 2, each phase group G₁, . . . , G_m comprises three phases, so that the group phase number p₁, . . . p_n, of each phase group G₁, . . . , G_m is three.

The number m of phase groups G₁, . . . , G_m in total can also comprise only some of the phases L₁, . . . , L_n of the multi-phase system 1. However, only this portion of the phases L₁, . . . , L_n, of the particular phase current i₁, . . . , i_n are thus validated according to the invention.

In each phase group G₁, . . . , G_m, the current measurement values i_M1, . . . , i_Mn of the associated phases L₁, . . . , L_n are each summed up by a group summation unit E₁, E₂, . . . , E_m to form a group sum i_G1, . . . , i_Gm.

The group comparison units V₁, . . . , V_m and/or group summation units E₁, E₂, . . . , E_m can be an integral component of the multi-phase system, for example integrated on a control unit of the multi-phase system 1, e.g. of an electromagnetic transport system. The same applies to the action unit, if provided. The group comparison units V₁, . . . , V_m and/or group summation units E₁, E₂, . . . , E_m and/or action unit can each be designed as hardware or software.

For example, for the first phase group G₁, the first current measurement value i_M1, the second current measurement value i_M2 and the third current measurement value i_M3 are added up to form a first group sum i_G1, so that the first group sum i_G1 comprises the sum of the first, second and third phase currents i₁, i₂, i₃ plus the sum of the first, second and third maximum errors ε₁, ε₂, ε₃: i_G1=i₁+i₂+i₃+ε₁+ε₂+ε₃; the second group sum i_G2 comprises the sum of the fourth, fifth and sixth phase currents i₄, i₅, i₆ plus the sum of the fourth, fifth and sixth maximum error ε₄, ε₅, ε₆: i_G2=i₄+i₅+i₆+ε₄+ε₅+ε₆, etc.; and the m-th group sum i_Gm comprises the sum of the n−2, n−1 and n-th phase currents i_n-2, i_n-1, i_n plus the sum of the n−2, n−1 and n-th maximum errors ε_n-2, ε_n-1, ε_n: i_Gm=i_n-2+i_n-1+i_n+ε_n-2+ε_n-1+ε_n.

Furthermore, the phase currents i₁, . . . , i_n of each phase group G₁, . . . , G_m are merged in a group node K₁, . . . , K_m to form a group sum current i_g1, . . . , i_gm, which results in the sum of the phase currents i₁, . . . , i_n contained in the group for the particular group sum current i_g1, . . . , i_gm. Using a group current sensor S_g1, . . . , S_gm, a group sum current measurement value i_Mg1, . . . , i_Mgm of the group sum currents i_g1, . . . , i_gm is determined, the group sum current measurement values i_Mg1, . . . , i_Mgm each also having a maximum error ε_g1, . . . , ε_gm with respect to the associated group sum currents i_g1, . . . , i_gm.

For example, the first group sum current measurement value i_Mg1 of the first phase group G₁ consists of the first, second and third phase currents i₁, i₂, i₃ and the maximum error ε_g1, etc.

According to the invention, a group comparison unit V₁, . . . , V_m compares the group sum current measurement values i_Mg1, . . . , i_Mgm with the associated group sums i_G1, . . . , i_Gm (shown as

$i_{\mathrm{Mg}1}\stackrel{{\varepsilon }_{\mathrm{VG}1}}{}i_{G1},i_{\mathrm{Mg}2}\stackrel{{\varepsilon }_{\mathrm{VG}2}}{}i_{G2},\dots ,i_{\mathrm{Mgm}}\stackrel{{\varepsilon }_{\mathrm{VGm}}}{}i_{\mathrm{Gm}}$

in FIG. 2) to validate the phase currents i₁, . . . , i_n.

If the validation fails, i.e. if the particular group sum current measurement value i_Mg1, . . . , i_Mgm differs from the associated group sum i_G1, . . . , i_Gm, an action A can be triggered by an action unit (not shown). As an action, for example, an optical and/or acoustic signal can be output and/or at least part of the multi-phase system 1 can be shut down.

In turn, in each group comparison unit V₁, . . . , V_m, the maximum error ε₁, . . . , ε_n in the current measurement values i_M1 . . . , i_Mn (over the group sum i_G1, . . . , i_Gm) and the maximum error ε_g1, . . . , ε_gm in the particular group sum current measurement value i_Mg1, . . . , i_Mgm must be taken into account. The maximum permissible total group error ε_V1, ε_V2, . . . , ε_Vm in each phase group G₁, . . . , G_m thus comprises only the sum of the maximum errors ε₁, . . . , ε_n of the current measurement values i_M1, . . . , i_Mn of the phases L₁, . . . , L_n which are contained in the particular phase group G₁, . . . , G_m, and the maximum error ε_g1, . . . , ε_gm in the associated group sum current measurement value i_Mg1, . . . , i_Mgm.

Comparing the first group sum current measurement value i_Mg1 with the associated first group SUM i_G1

$\left(i_{\mathrm{Mg}1}\stackrel{{\varepsilon }_{\mathrm{VG}1}}{}i_{G1}\right)$

results in, for example, only a maximum permissible total group error ε_VG1=ε₁+ε₂+ε₃+ε_g1 for the first group G1; comparing the second group sum current measurement value i_Mg2 with the associated second group sum

${i_{G2}}^{i_{\mathrm{Mg}2}\stackrel{{\varepsilon }_{\mathrm{VG}2}}{}i_{G2}})$

results in a maximum permissible total group error ε_VG2=ε₄+ε₅+ε₆+ε_g2, . . . ; comparing the m-th group sum current measurement value i_Mgm with the associated m-th group sum

${i_{Gm}}^{\left(i_{\mathrm{Mg}m}\stackrel{{\varepsilon }_{\mathrm{VG}m}}{}i_{Gm}\right)}$

results in a maximum permissible total group error ε_VGm=ε_n-2+ε_n-1+ε_n+ε_gm.

For the general case of a phase group Gy, this means that the maximum permissible total group error ε_Vgy consists of the maximum errors of the phase group Gy:

$\sum ^{\mathrm{py}}{\varepsilon }_j,$

a group phase number p_y of maximum errors being summed up, i.e. one per phase contained in the phase group Gy (p_y is the group phase number of the phase group Gy), and the maximum error ε_g1, . . . , ε_gm of the group sum current measurement value i_Mgy that belongs to the phase group G_y being taken into account. The maximum permissible total group error ε_Vgy for the phase group G_y thus corresponds to

${\varepsilon }_{\mathrm{VGy}}=\sum ^{\mathrm{py}}{\varepsilon }_j+{\varepsilon }_{\mathrm{gy}}.$

The maximum permissible total group error ε_Vgy is taken into account during validation, i.e. when comparing the group sum current measurement value i_Mg1, . . . , i_Mgm with the associated group sum i_G1, . . . , i_Gm. The validation fails if the group sum current measurement value i_Mg1, . . . , i_Mgm deviates from the associated group sum i_G1, . . . , i_Gm by more than the maximum permissible total group error ε_Vgy.

It should be assumed again that the maximum errors ε₁, . . . , ε_n in the current measurement values i_M1, . . . , i_Mn and the maximum errors ε_g1, . . . , ε_gm in the group sum current measurement values i_Mg1, . . . , i_Mgm each correspond to the maximum error ε_x (not shown). For the group sum i_G1, . . . , i_Gm of each phase group G₁, . . . , G_m, the sum of the associated phase currents i₁, . . . , i_n plus a three-fold maximum error ε_x (three-fold in this case, since the group phase number p₁, p₂, . . . , p_m is three) results in the following: i_G1=i₁+i₂+i₃+3 ε_x, i_G2=i₄+i₅+i₆+3 ε_x, i_Gm=i_n-2+i_n-1+i_n+3 ε_x.

Furthermore, the sum of the phase currents i₁, . . . , i_n contained in the group G₁, . . . , G_m plus the maximum error ε results in the following for the group sum current measurement values i_MG1, . . . , i_Mgm at the group node K₁, . . . , K_m: i_Mg1=i_g1+ε_x=i₁+i₂+i₃+ε_x, i_Mg2=i_g2+ε_x=i₄+i₅+i₆+ε_x, . . . , i_Mgm=i_gm+ε_x=i_n-2+i_n-1+i_n+ε_x. In order to be able to compare the group sums jam with the associated group sum current measurement values i_Mg1, . . . , i_Mgm for the purpose of validating the phase currents i₁, . . . , i_n in the phase groups G₁, . . . , G_m, only one maximum permissible total group error ε_VG1, . . . , ε_VG1 in the particular group G₁, . . . , G_m in the amount of ε_VG1=(3+1) ε_x, ε_VG2=(3+1) ε_x, . . . , ε_VGm=(3+1) ε_x must be taken into account for each group G₁, . . . , G_m, the particular factor 3 corresponding to the group phases number p₁, . . . , p_n, and originating from the determination of the current measurement value_s i_M1, . . . , i_Mn of the phases contained in the particular group G₁, . . . , G_m, the additional factor 1 originating from the determination of the group sum current measurement values i_Mg1, . . . , i_Mgm.

If, as a comparison to the prior art, a long stator linear motor which has a plurality of n=42 phases L₁, . . . , L_n is operated in turn as a multi-phase system 1, then dividing the phases L₁, . . . , L_n into a number m=14 of phase groups G₁, . . . , G_m=14 each having a group phase number p₁, . . . , p_m=3 and a maximum error ε_x of, for example, 100 mA per phase group G₁, . . . , G_m results in a maximum permissible total group error ε_VG1, . . . , ε_VG14 of ε_VG1, . . . , ε_VG14=(p+1)=4*100 mA=0.4 A.

This means that only individual errors smaller than the maximum permissible total group error ε_VG=(p+1) ε_x=0.4 A are not detected during validation. Thus, the validation according to the invention has a much higher accuracy than a validation via a sum current measurement value i_Mg according to FIG. 1; in the example described it is higher by a factor of 10.75. The validation according to the invention thus has a high level of accuracy, and therefore safe operation of the multi-phase system 1 can be ensured.

Claims

1. Method for safely operating a multi-phase system, wherein a phase current flows in each of a plurality of phases of the multi-phase system respectively, and wherein current measurement values of the phase currents are captured by phase current sensors in order to validate the phase currents, wherein a number of phase groups is provided, which comprises some of the phases, wherein the phase currents of the number of phase groups are merged in a group node to form a group sum current, and a group sum current measurement value of the group sum current is captured by a group current sensor, wherein the current measurement values belonging to the number of phase groups are summed up by a group summation unit to form a group sum, and wherein the group sum is compared with the group sum current measurement value by a group comparison unit to validate the phase currents of the phases in order to ensure safe operation of the multi-phase system.

2. Method according to claim 1, wherein the phase currents of all the phases of the multi-phase system are divided into a number of phase groups, and in each phase group the phase currents of the phase group are merged in a group node to form a group sum current and a group sum current measurement value of the group sum current is captured, wherein, in each phase group, the current measurement values of the phase currents will each be summed up by a group summation unit to form a group sum, and wherein, for each phase group, the group sums are each compared with the associated group sum current measurement values by a group comparison unit in order to validate the phase currents of all the phases.

3. Method according to claim 1, wherein the number of phase groups each comprises the same number of phases.

4. Method according to claim 1, wherein the current measurement values of the phase currents are safely captured in accordance with a specified safety requirement.

5. Method according to claim 1, wherein an action is triggered by an action unit in the event of a failed validation.

6. Method according to claim 5, wherein an optical and/or acoustic signal is output as the action and/or at least part of the multi-phase system is shut down.

7. Multi-phase system comprising a plurality of phases, wherein a phase current flows in each of the phases, wherein phase current sensors are provided for capturing current measurement values of the phase currents, wherein a number of phase groups is provided, which comprises some of the phases, wherein a group node is provided, within which the phase currents of the number of phase groups are merged to form a group sum current, wherein a group current sensor is provided for capturing a group sum current measurement value of the group sum current, in that a group summation unit is provided, which is designed to sum up the current measurement values of the phase currents belonging to the number of phase groups to form a group sum, and wherein a group comparison unit is provided, which is designed to compare the group sum with the group sum current measurement value to validate the phase currents of the phases in order to ensure safe operation of the multi-phase system.

8. Multi-phase system according to claim 7, wherein the number of phase groups comprises all the phases of the multi-phase system, the phase currents of the phase group are merged in a group node to form a group sum current in each phase group, in that a group summation unit is to proceed for each phase group, which is designed to sum up the current measurement values of the phase currents to form a group sum respectively, and wherein a group comparison unit is provided for each phase group, which is designed to compare the group sums with the associated group sum current measurement values respectively in order to validate the phase currents of all the phases.

9. Multi-phase system according to either claim 7, wherein an action unit is provided, which is connected to the group comparison unit and is designed to trigger an action in the event of a failed validation.

10. Multi-phase system according to claim 9, wherein the action unit is designed to output an optical and/or acoustic signal and/or to shut down at least part of the multi-phase system.