 |
Description  |
|
|
CROSS REFERENCE TO RELATED APPLICATIONS
This invention refers to the subject matter of copending U.S. Patent
Application Ser. No. 186,091 by the same Inventor.
BACKGROUND OF THE INVENTION
This invention refers to a reluctance synchronous electric machine fed by
an electronic power circuit by vectorial control of the feed current.
According to the state of art preceding this invention, the system formed
by an alternate current electric machine (and therefore, in particular, by
a reluctance synchronous machine), by the pertinent electronic feed
circuit (inverter) and by its circuit for a vectorial control (in
amplitude and phase) of the feed current, cannot be designed such as to
exploit in the more favorable manner both the characteristics of the
machine and the characteristics of the inverter.
In effect it is noticed that, in order to suitably exploit the
characteristics of the inverter, the machine should be proportioned for
delivering a maximum power greater than that really allowed by the
inverter, whereas on the contrary, in order to suitably exploit the
characteristics of the machine, the inverter should be proportioned for
delivering a maximum power greater than that allowed by the machine. This
is particularly burdensome in the frequent case in which the machine is
required to deliver a constant torque (corresponding to a prefixed maximum
value) at any speed lower than an intermediate speed at which, with said
prefixed constant torque, a fixed maximum value of power is delivered,
whereas the machine should deliver a constant power, corresponding to the
above cited fixed maximum value of power, within a large range of speed
over said intermediate speed. In order that this may be possible, the
inverter should be proportioned for feeding the nominal current needed for
producing the prefixed maximum torque, as well as for feeding the maximum
voltage needed for delivering the prefixed maximum power up to the maximum
speed. The product of the nominal current multiplied by the maximum
voltage is a socalled "dimensioning power", for which the inverter should
be proportioned, and it may be much greater than the maximum mechanical
power intended to be delivered.
This may be well understood with reference to the vectorial diagram of FIG.
1. It should be noted that the distribution of sinusoidal magnetomotive
force generated by the stator winding of the machine may be construed as
resolved in two sinusoidal distributions of magnetomotive force, directed
along the direct axis (d) of the rotor and the quadrature axis (q) of the
rotor, respectively. On their turn, these distributions of magnetomotive
force may be construed as produced by two conductor distributions through
which flow electrical currents i.sub.d and i.sub.q, respectively. The
magnetic fluxes .lambda..sub.d and .lambda..sub.q, directed along the axes
d and q, respectively, which chain the two windings corresponding to said
conductor distributions, are given by the respective currents i.sub.d and
i.sub.q multiplied by the respective self-inductances L.sub.d and L.sub.q,
which are characteristic of the magnetic construction of the rotor:
.lambda..sub.d =L.sub.d .multidot.i.sub.d ; .lambda..sub.q =L.sub.q
.multidot.i.sub.q. The total vectors of current (i) and magnetic flux
(.lambda.) result from the vectorial sum of the respective components
(i.sub. d, i.sub.q and .lambda..sub.d, .lambda..sub.q) along the axes d
and q, respectively, of the rotor. The angular speed of the rotor with
respect to the stator will be indicated as .omega.. The analysis of the
system shows that, if the resistive drops are disregarded, and a
stationary distribution of the magnetic state is taken into account, the
resulting voltage vector v, which is suitable for giving rise to the
current vector i into the stator winding, is in advance and in quadrature
with respect to the flux vector .lambda.. The vector v subtends with the
vector i an angle .phi.. By denoting as T the produced mechanical torque,
and henceforth .omega..multidot.T the delivered mechanical power, the
apparent input electrical power is v.multidot.i, the active input
electrical power is v.multidot.i.multidot.cos.phi.=.omega..multidot.T, and
the amplitude of the voltage v needed for producing the magnetic fluxes
involved is v=.omega..sqroot..lambda..sub.d.sup.2 +.lambda..sub.q.sup.2.
The general formula which gives the torque is T=.lambda..sub.d
.multidot.i.sub.q -.lambda..sub.q .multidot.i.sub.d.
It may be shown that the maximum torque which may be produced without
overcoming a maximum voltage V.sub.M is obtained when
.vertline..lambda..sub.d .vertline.=.vertline..lambda..sub.q
.vertline.=V.sub.M /.omega..sqroot.2, and it is T.sub.max =(1/L.sub.q
-1/L.sub.d).multidot.V.sub.M.sup.2 /2.omega..sup.2, whereby the maximum
power is P.sub.max =(1/L.sub.q -1/L.sub.d).multidot.V.sub.M.sup.2
/2.omega..
As a consequence, the maximum voltage V.sub.M which should be applied in
order to obtain the power P.sub.max increases with the maximum angular
speed .omega. at which said power is to be delivered, namely with the
extent of the range wherein the operation at constant power is required.
If I.sub.o is the nominal current foreseen for the machine, the
proportioning power (for which the inverter should be proportioned) is
V.sub.M .multidot.I.sub.o : it is much greater than the mechanical power
.omega..multidot.T which may be delivered by the system, and it increases
with the extent of the speed range wherein an operation at constant
mechanical power is required.
BRIEF SUMMARY OF THE INVENTION
In view of the above, the object of this invention is to allow a more
favorable exploitation of the characteristics of a system comprising a
reluctance electric machine and the pertinent electronic power circuit
(inverter) intended for feeding the machine, by avoiding the need for
proportioning the inverter in view of a dimensioning power much greater
than the active power required. The conception on which the invention is
based is to foresee for the rotor a particular structure which, in the
nominal conditions of operation, gives rise to a reduced value (tending to
null) of the magnetic flux along the quadrature axis, which gives a
negative contribute to the produced torque and corresponds to a voltage
component along the direct axis, which increases the amplitude of the
voltage when a mechanical power is generated.
The object of the invention is attained by means of a reluctance
synchronous electric machine having a rotor with an axial magnetic
segmentation comprising axial layers of ferromagnetic material alternated
with intercalary layers of non-ferromagnetic material, characterized in
that some permanent magnets are inserted within said intercalary
non-ferromagnetic layers, with an orientation such as to give rise to a
magnetic flux along the direction of the quadrature axis and in the sense
opposite the magnetic flux produced by the quadrature component of the
current which flows through the stator windings.
Preferably said permanent magnets are proportioned in such a way that the
magnetic flux produced by them has equal amplitude and sense opposite the
magnetic flux produced by the quadrature component of the current which
flows through the stator windings when the machine is in the nominal
conditions of operation.
Due to this characteristic, the magnetic flux component present along the
quadrature axis of the rotor is in any event smaller than that which would
be present in the absence of the features according to the invention and,
when the above stated preferred proportions are adopted, it may even be
null during the nominal operation of the machine. As a consequence, the
component of counterelectromotive force produced by said magnetic flux
within the stator winding by action of the rotor rotation is smaller or
respectively null. It is this component of counterelectromotive force
which should be opposed by a component along the direct axis of the
voltage applied to the stator in order that the current needed for the
machine operation can be caused into the stator winding. By reducing the
component along the direct axis of the voltage to be applied, the
amplitude of this voltage is reduced too, and therefore the proportions of
the inverter may be reduced. At the limit condition, namely when the above
stated preferred proportions are adopted, the inverter needs only to be
proportioned in view of the active electrical power corresponding to the
mechanical power delivered by the machine at full load.
The vectorial diagram of FIG. 2 refers to the case of the above stated
preferred proportions, and the component of the current along the direct
axis is supposed to be negligeable, as it really happens during the
operation of the machine at high speed and with a reduced magnetic flux.
The absolute value of the magnetic flux produced by the permanent magnets
is indicated as .psi..sub.M. The component of magnetic flux along the
quadrature axis is .lambda..sub.q =L.sub.q .multidot.i.sub.q -.psi..sub.M.
In accordance with the preferred proportions, it is .psi..sub.M =L.sub.q
.multidot.I.sub.o, and therefore the component .lambda..sub.q is null when
the machine operates under the nominal current I.sub.o. In these
conditions, the total magnetic flux .lambda. is equal to the component
.lambda..sub.d of the magnetic flux, and it is oriented along the direct
axis d; the component i.sub.q of the current along the quadrature axis is
equal to the nominal current I.sub.o ; and the total voltage applied is
equal to the voltage component along the quadrature axis:
v=.omega..multidot..lambda.=v.sub.q, and it is oriented along the
quadrature axis q. There is no phase shift .phi. between voltage and
current, namely the machine, in the stated conditions, is completely
corrected from phase shift; the apparent electric power is equal to the
active electric power and to the delivered mechanical power, and therefore
the inverter may be proportioned in view of this latter only.
Of course, such conditions are not verified during the operation under a
current smaller than the nominal current, and also when non-stationary
conditions of the magnetic circuit are taken into account. In addition,
theoretically the current component i.sub.d along the direct axis is to be
taken into account, but in a reluctance electric machine with axial
magnetic segmentation this current component, although not null, is much
smaller than the quadrature component i.sub.q, and therefore it modifies
only to a very little extent the above conclusions.
The phase correction of the machine is obtained thanks to the magnetic flux
produced by the permanent magnets, and this magnetic flux should be
proportioned to the magnetic flux component along the quadrature axis
produced by the current. Therefore it is of advantage, in the interest of
the proportions of the permanent magnets, that the self-inductance L.sub.q
shown by the rotor along the quadrature axis is as low as possible. In
addition, the current component along the direct axis, needed for
producing the magnetic flux, does not correspond to the ideal case of FIG.
2, and therefore it is of advantage that the self-inductance L.sub.d shown
by the rotor along the direct axis is as high as possible. Both above
considerations are abridged by the condition that the ratio L.sub.q
/L.sub.d should be as low as possible. It is therefore of advantage that a
suitable structure is adopted for the rotor, more particularly a structure
of the kind according to the copending U.S. Patent Application Ser. No.
186,091. This latter is formed by an axial magnetic segmentation
comprising layers of ferromagnetic material alternated with intercalary
layers of non-ferromagnetic material having a thickness of not less than
2/3 of the thickness of the layers of ferromagnetic material.
In view of the above, a reluctance electric machine embodying the features
of the present invention is intrinsecally phase-corrected in the operative
conditions in which it delivers the prefixed mechanical power required.
BRIEF DESCRIPTION OF THE DRAWINGS
The above and other characteristics and advantages of the subject of the
invention will appear more clearly from the following description of some
preferred embodiments, having exemplary and not limitative character and
diagrammatically shown in the annexed drawings, wherein:
FIG. 1 shows a vectorial diagram, already discussed in the preamble, which
refers to a machine according to the former state of art;
FIG. 2 shows a vectorial diagram, already discussed in the preamble, which
refers to a machine embodying the invention;
FIG. 3 diagrammatically shows a first embodiment of the structure of a
machine having a rotor with axial magnetically decreased segmentations,
embodying the features of the invention;
FIG. 4 shows on a greater scale the detail IV of FIG. 3;
FIGS. 5 and 6 show, in a way similar to that of FIG. 3, two further
embodiments of the structure of a machine embodying the features of the
invention; and
FIG. 7 shows suitable proportioning ratios between the rotor magnetic
segmentations and the stator slot pitch in a machine according to the
invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
With reference to FIGS. 3 and 4, letter S denotes an electric machine
stator, whose structure is not better detailed because it may be of any
kind of stator known for the considered type of machines or for other
types of synchronous or asynchronous electric machines. In most cases, but
not in binding manner, stator S may have a structure with a polyphase
distributed winding. It should be noted that the structure of stator S has
per se no connection with the embodiment of the invention.
It is diagrammatically shown in FIG. 3 that stator S is connected to a feed
circuit C inserted between stator S and an electric branch network R. The
feed circuit C should be construed as including a power inverter, which
feeds stator S, and an electronic circuit for vectorial control (in
amplitude and phase) of the feed current delivered by the inverter to the
stator. The structure of circuit C is not better detailed because such
control circuits are per se well known in the application to other types
of electric machines and, generally speaking, any circuit of this kind may
be used for feeding the electric machine according to the invention.
In FIGS. 3 and 4, reference T denotes a magnetic gap, of a thickness t,
provided between the inner surface of stator S and the outer surface of
the rotor. The rotor shown in these Figures has four poles and it is
mounted onto a shaft A, which may be of normal steel and extends to form a
spider with four arms A' oriented along the two direct axes of the rotor.
In the four quadrants delimited by the four arms A' of shaft A there are
inserted four curved axial laminations, whose general structure is well
known for this type of machines and includes a number of superimposed
ferromagnetic sheets L alternated with intercalary non-ferromagnetic
layers I. According to the teaching of the copending U.S. Patent
Application Ser. No. 186,091 by the same Inventor, said intercalary
non-ferromagnetic layers I have a substantial thickness, at least not less
than 2/3 of the thickness of the ferromagnetic sheets L.
According to the main feature of this invention, a number of permanent
magnets M (FIG. 4) are inserted within the intercalary non-ferromagnetic
layers I alternated with the ferromagnetic sheets L. Each permanent magnet
M is magnetized along a direction perpendicular to the plane tangent to
the intercalary layer I in the point where the considered permanent magnet
M is located. The sense of the magnetization of the permanent magnets M is
such that all permanent magnets of each quadrant produce, in their whole,
a magnetic flux oriented along the direction of the quadrature axis q
which passes through the considered quadrant and opposite the magnetic
flux generated, in the considered quadrant, by the quadrature component of
the electric current which flows through the stator winding.
Taking into account what has been explained with reference to the vectorial
diagram of FIG. 2, due to the described arrangement of components of
magnetic flux present in the rotor along its quadrature axes are smaller
that the components of magnetic flux which would be present in the absence
of the permanent magnets M. As a consequence, is also smaller the
component along the direct axis of the voltage which the feed circuit
should apply to the stator winding; the amplitude of said voltage is
reduced too, and the inverter which is part of the feed circuit C may be
more strictly dimensioned.
Preferably the characteristics of the permanent magnets M are chosen in
such a way that the magnetic flux produced by them is at least
approximately equal in amplitude and opposite in sense to the magnetic
flux produced by the quadrature component of the current which flows
through the stator winding when the machine operates in its nominal
operating conditions. In this preferred case, as already explained in the
foregoing with reference to the vectorial diagram of FIG. 2, it is
sufficient that the inverter be dimensioned in view of an electric power
value near the active electric power corresponding to the mechanical power
delivered by the machine at full load.
In the middle portion of each curved lamination L,I there is a member B
which, for practical reasons of structure and of manufacture, is not
segmented, and which defines one of the interpolar spaces of the rotor.
When this invention is embodied, it is of advantage that the members B are
of ferromagnetic material in order to support the magnetic induction
distribution at the magnetic gap, which is produced by the permanent
magnets M.
From the constructional point of view, the distribution of permanent
magnets M inserted within the intercalary layers I, which in this case are
curved, may be embodied by applying a number of permanent magnets M of a
little size onto a flexible supporting sheet F (FIG. 4). The pliability of
the supporting sheet F allows easily inserting a layer of permanent
magnets M between two ferromagnetic sheets L, notwithstanding that these
latter are curved.
In the embodiment according to FIG. 5, the ferromagnetic sheets L which
form a part of the axial laminations are not curved, but they are bent
according to a polygonal shape. In this case the intercalary spaces I are
formed by plane segments, and permanent magnets M shaped as plane plates
may be easily inserted into the intercalary spaces I. As it may be noted
from this Figure, the number of permanent magnets forming each layer may
be different from the number of permanent magnets forming the other
layers. This allows to generate in each layer the more suitable
magnetomotive force. However, instead of changing the number of permanent
magnets of the different layers, it is also possible to insert in the
different layers permanent magnets M having different magnetic
characteristics.
From the magnetic point of view, the rotor segmentations should extend
axially in order to define axially extending intercalary layers, suitable
for receiving the permanent magnet distributions which are characteristic
of the invention. To this end, in the embodiments of FIGS. 3 to 5 the
rotor laminations include ferromagnetic sheets which materially extend in
the axial direction. However it is also possible to obtain magnetically
axial segmentations by using ferromagnetic sheets which materially form a
transverse lamination, provided that said ferromagnetic sheets show
suitable flux guides formed by adequately shaped cavities. An example of
such a structure is given by FIG. 6. In this case the rotor comprises a
stack of round transverse ferromagnetic sheets, superimposed along the
axis of a shaft A onto which they are keyed. In these ferromagnetic sheets
there are cut elongated cavities which define the intercalary
non-ferromagnetic layers I. These cavities are separated by ferromagnetic
portions L which, in the whole of the stacked sheets, define axial layers
of ferromagnetic material. In this way, by using a materially transverse
lamination, which is preferable from the constructional point of view, a
behaviour magnetically equivalent to that of the above described axial
laminations may be attained. The ferromagnetic portions which, in the
formerly described structures, were formed by the arms A' of shaft A and
by the members B of the interpolar spaces, are formed in this structure by
corresponding portions A" and B' of the transverse sheets forming the
ferromagnetic layers L.
As it may be noted from the same FIG. 6, in the embodiment shown the
cavities or flux guides cut through the sheets and defining the
intercalary layers I do not extend up to the periphery of the rotor, but
they end at a short distance therefrom. In this way, along the periphery
of the rotor there is formed a succession of thin ferromagnetic bridges
closing the cavities which form the flux guides. In other possible
embodiments such bridges could also be provided in some positions
different from the peripheral position shown in FIG. 6, namely in internal
positions such as B".
Such structures are known per se in the reluctance machines, and their
purpose is to ensure to the rotor a mechanical strength suitable for
withstand the centrifugal forces which arise during the rotation, without
the need for hooping the rotor by means of a material having a high
maximum tensile stress, such as carbon fibers. Apart the economical
reasons, a hooping is undesirable because it increases the thickness of
the magnetic air gap with respect to the geometrical gap.
But on the other hand, a structure including ferromagnetic bridges, either
in peripheral or internal positions, in the known machines has the
disadvantage of considerably increasing the magnetic permeance along the
quadrature axes, and moreover this increase follows a non-linear function
of the delivered power, due to the different levels of magnetic saturation
in said bridges. In the known machines, the magnetic flux which flows
through such bridges, and eventually saturates them, should be produced by
the electric feed current, and therefore it has an unfavorable effect onto
the behavior of the feed circuit.
On the contrary, with the structure according to the invention, some
ferromagnetic bridges, either peripheral or internal, may be provided for
in the rotor without incurring any inconvenience, because they may be
magnetically saturated by action of the permanent magnets inserted within
the intercalary layers, provided, of course, that the cross section of
said bridges is suitably limited. In this case, the presence of said
ferromagnetic bridges should be taken into account in proportioning the
permanent magnets inserted within the intercalary layers: in effect, apart
from producing the magnetic flux needed for reducing or suppressing the
direct axis component of the voltage to be applied, the magnets should
also produce a magnetic flux intended to saturate the peripheral
ferromagnetic bridges. When thus magnetically saturated by the permanent
magnets, said bridges behave with respect to any other magnetic flux like
a non-ferromagnetic material and, if they are located at the periphery,
they constitute an ideal extension of the intercalary non-ferromagnetic
layers up to the magnetic gap. In these circumstances the power electronic
circuit and the control circuit do not suffer any effect from the presence
of said bridges.
A similar possibility may be exploited, in the case of a materially axial
lamination, by providing a hooping of ferromagnetic material (as
diagrammatically shown at H, FIG. 6), and magnetically saturating, by
action of the permanent magnets inserted within the intercalary spaces,
the hooping portions which are located in front of the ends of said
intercalary spaces and constitute ferromagnetic bridges. In this way one
can provide more economical hoopings which do not increase the magnetic
gap with respect to the geometrical gap.
In order to attain the optimum behaviour of the ferromagnetic material
forming the rotor, it is suitable that the magnetic circuit reluctance
involved in each magnetic segmentation of the rotor remains substantially
unmodified during the changes in the position of the rotor with respect to
the stator, and this notwithstanding that the stator is anisotropic due to
the presence therein of the slots for the windings. In this case, the
magnetomotive force of the magnets inserted within the intercalary layers
of the rotor does not induce during the displacements of the rotor any
substantial change of magnetic induction in the ferromagnetic layers of
the rotor. This condition may be attained, according to a further feature
of this invention, when the length of the ferromagnetic layer of each
magnetic segmentation of the rotor, measured along the magnetic gap, is
equal to the pitch of the stator slots, measured along the magnetic gap.
In effect, in this case each ferromagnetic layer of the rotor, in any
position, always faces an identical extension of the ferromagnetic
material of the stator.
As shown by the diagrammatic representation of FIG. 7, to this generally
expressed condition may correspond two different proportions of the parts,
depending upon the intercalary layers separating the magnetic
segmentations of the rotor being open towards the magnetic gap (as
according to FIG. 5) or closed towards the magnetic gap (as according to
FIG. 6).
When the intercalary non-ferromagnetic layers I of the magnetic
segmentations of the rotor are substantially open towards the magnetic gap
(top portion of the diagram of FIG. 7), whatever may be the pitch PR of
the rotor segmentation, the desired condition is attained when the width
LL (measured at the magnetic gap) of the ferromagnetic layers L is equal
to the pitch PS of the stator slots (also measured at the magnetic gap),
or to a multiple thereof. In this case, the width of the intercalary
non-ferromagnetic layers I is immaterial.
When the intercalary non-ferromagnetic layers I of the magnetic
segmentations of the rotor are substantially closed towards the magnetic
gap (bottom portion of the diagram of FIG. 7), whatever may be the width
of the ferromagnetic layers L which separate the intercalary
non-ferromagnetic layers I, the desired condition is attained when the
pitch PR of the magnetic segmentations of the rotor (measured at the
magnetic gap) is equal to the pitch PS of the stator slots (also measured
at the magnetic gap), or to a multiple thereof. As it may be understood,
this is a particular case of the former condition, obtained when the width
of the intercalary layers I at the magnetic gap is reduced to null,
whereby (only at the magnetic gap) the width LL of the ferromagnetic
layers L occupies the entire pitch PR of the magnetic segmentations of the
rotor.
In the practice, this latter condition is also valable when the intercalary
layers I of the magnetic segmentations of the rotor, instead of being
completely closed towards the magnetic gap, are shaped in such a way as to
show at the magnetic gap only a little width, of the same order of the
thickness of the magnetic gap itself.
In this way, two different possibilities are offered to the designer, who
may suitably chose according to the different requirements of the
construction. Moreover it should be born in mind that all the stated
conditions require only to be observed within a certain approximation, in
view of the lines of magnetic induction being capable of a certain
divergence in traversing the magnetic gap.
Thanks to the provision of the described means for intrinsic phase
correction, the reluctance synchronous electric machine according to the
invention allows providing systems, comprising the machine itself and the
pertinent electronic feed circuit with vectorial current control, wherein
both the characteristics of the machine and the characteristics of the
inverter included in the feed circuit are exploited in the more favorable
manner.
* * * * *
|
|
 |
|
 |
Description |
|
