• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    In order to be able to adapt the operation of a long-stator linear motor better to the requirements or the states of the individual transport units or the transport path, it is provided that the control variables (StG) of a drive coil (7, 8) of the long-stator linear motor are superimposed with a starting signal (AS) with a predetermined frequency band In this case, actual variables (IG) of the drive coil control are determined, a frequency response is determined from the manipulated variables (StGAS) superimposed with the excitation signal (AS) and the determined actual variables (IG), and from the frequency response the controller parameters (RP) for this transport unit (Tx ) are determined and the transport unit (Tx) for moving along the transport path with these determined controller parameters (RP) is controlled.
    公开号:AT518734A1
    申请号:T50494/2016
    申请日:2016-05-31
    公开日:2017-12-15
    发明作者:Ing Andreas Weber Dipl；Joachim Weissbacher Dr
    申请人:Bernecker + Rainer Industrie-Elektronik Ges M B H；
    IPC主号:
    专利说明:

    Method for operating a long-stator linear motor
    The subject invention relates to a method for operating a Langstatorlinearmotors with a transport path along which a plurality of drive coils are arranged one behind the other and with at least one transport unit which is moved along the transport path, each drive coil is controlled by a drive coil controller with associated controller parameters by the Drive coil controllers for the cooperating with the transport unit drive coils manipulated variables are specified.
    In virtually all modern production facilities, it is necessary to move components or components, even over long transport distances, with transport facilities between individual handling or production stations. For this purpose, a variety of transport or conveyors are known. Frequently, continuous conveyors in various designs are used. Conventional continuous conveyors are conveyor belts in the various embodiments, in which a rotary motion of an electric drive is converted into a linear movement of the conveyor belt. With such conventional continuous conveyors one is considerably limited in flexibility, in particular, an individual transport of individual transport units is not possible. To remedy this and to meet the requirements of modern, flexible transport equipment, so-called Long Stator Linear Motors (LLM) are increasingly used as a replacement for conventional continuous conveyors.
    In a long stator linear motor, a plurality of electric drive coils constituting the stator are arranged along a transport path. On a transport unit a number of excitation magnets, either as permanent magnets or as an electrical coil or short-circuit winding, are arranged, which cooperate with the drive coils of the stator. The long-stator linear motor can be designed as a synchronous machine, either self-excited o-the foreign-excited, or as an asynchronous machine. By driving the individual drive coils in the region of the transport unit for regulating the magnetic flux, a propulsion force is generated on the transport unit and the transport unit can thus be moved along the transport path. It is also possible to arrange along the transport path a plurality of transport units whose movements can be controlled individually and independently of each other by only the arranged in the region of the individual transport units drive coils are activated. A long-stator linear motor is characterized in particular by a better and more flexible utilization over the entire working range of the movement (speed, acceleration), individual regulation / control of the transport units along the transport route, improved energy utilization, the reduction of maintenance costs due to the smaller number Wear parts, a simple replacement of the transport units, efficient monitoring and fault detection and an optimization of the product flow along the transport route. Examples of such long-stator linear motors can be found in WO 2013/143783 A1, US Pat. No. 6,876,107 B2, US 2013/0074724 A1 or WO 2004/103792 A1.
    In US 2013/0074724 A1 and WO 2004/103792 A1, the drive coils of the stator are arranged on the upper side of the transport path. The permanent magnets are arranged on the underside of the transport units. In WO 2013/143783 A1 and US Pat. No. 6,876,107 B2, the permanent magnets are provided on both sides of the centrally arranged drive coils, with which the permanent magnets surround the stator of the long stator linear motor and the drive coils interact with the permanent magnets arranged on both sides.
    The transport units are guided along the transport path either mechanically, for example by guide rollers, as for example in WO 2013/143783 A1 or US Pat. No. 6,876,107 B2, or by magnetic guidance, as for example in WO 2004/103792 A1. Combinations of mechanical and magnetic guidance are also conceivable. In the case of magnetic guidance guide magnets may be provided on both sides of the transport units, which cooperate with arranged on the transport path guide rods. The guide rods form a magnetic yoke which close the magnetic circuit of the guide magnets. The magnetic guide circles thus formed counteract a lateral movement of the transport units, whereby the transport units are guided laterally. A similar magnetic side guide can also be found in US 6,101,952 A.
    In many transport devices transfer positions, e.g. in the form of turnouts, necessary to enable complex and intelligent rail planning or rail realizations of the transport device. These transfer positions have often been realized with the help of additional mechanical release units. An example of this is found in US 2013/0074724 A1 in the form of a mechanically triggered switch by means of movable deflection arms or a turntable. However, transport devices have also become known in which additional electrical auxiliary coils are used in order to realize a switch initiation. In US Pat. No. 6,101,952 A, the auxiliary coils are arranged, for example, on the magnetic yoke of the magnetic guide circle, while the auxiliary coils in US 2013/0074724 A1 are arranged laterally on the transport path. In DE 1 963 505 A1, WO 2015/036302 A1 and WO 2015/042409 A1, magnetically activated switches of a long-stator linear motor are described which manage without additional auxiliary coils. _O_
    A long stator linear motor places high demands on the regulation of the movement of the transport units. For this purpose, a plurality of regulators are usually arranged along the transport path, which regulate the stator currents of the drive coils in order to move the transport units as intended along the transport path. For the movement of the transport units, it is necessary that each drive coil is controlled separately to ensure a smooth, controlled and stable movement of the transport units along the transport route. On the transport route, however, move a variety of transport units, which is why various transport units are controlled by different drive coils. However, the transport units moved along the transport path can differ in their properties. For example, transport units can be loaded differently, have different states of wear, cause different executives due to manufacturing imperfections, cause different frictional forces, etc. It is also conceivable that transport units are moved with different design or different size along the transport route. All these factors influence the regulation of the transport units.
    However, since control of the drive coils must be stable and reliable for all transport units, a conservative control strategy has hitherto been pursued. With this, however, dynamics were lost in the regulation, whereby fast control interventions, for example an abrupt change in the speed of the transport unit, are limited.
    The individual transport units but also subject to different wear, which makes the maintenance of the transport units or the Langstatorlinearmotors difficult. It is simple and possible to service or even exchange all transport units at predetermined times, but it is also expensive and expensive, since it may also be used to service or replace transport units in which it would not even be necessary. On the other hand, with higher wear, the traveling resistance of individual transport units may increase due to deteriorating friction between the transport units and the guide. This would also lead to higher power losses, since the drive power of the transport units would have to be increased. Last but not least, the current state of wear of the transport unit also influences its regulation,
    It is therefore an object of the present invention to better adapt the operation of a long-stator linear motor to the requirements or the states of the individual transport units or the transport route.
    This object is achieved in that the control variables of a drive coil, a start signal is superimposed with a predetermined frequency band, while actual sizes of the drive coil control are determined and from the superimposed with the start signal control variables and the determined actual variables, a frequency response is determined, wherein the frequency response the controller parameters for this transport unit are determined and the transport unit is controlled to move along the transport path with these determined controller parameters. This allows the simple determination of optimized controller parameters of the drive coil controller, which in turn enables faster control interventions than previously possible.
    In a preferred embodiment, the controller parameters are varied to set a desired, predetermined property of the frequency response. A frequency response can be easily determined, which in turn makes it possible to easily determine the control parameters.
    It may be advantageous to use the same controller parameters for different transport units or to use different controller parameters for different transport units or to determine different controller parameters for different transport sections of the controlled system or to determine different controller parameters for different loading states of a transport unit. Thus, the method according to the invention can be used very flexibly depending on the requirements of the particular application.
    The controller parameterization is repeated in a particularly preferred embodiment at predetermined intervals in order to adapt the optimal controller parameters to possibly changing properties of the transport unit or the transport path.
    Furthermore, the method according to the invention can also be used to determine characteristics of the controlled system, in particular from the frequency response, the mass of the load carried by the transport unit. In turn, the corresponding optimal controller parameters can then be selected. It is also possible to determine from the frequency response existing resonance and anti-resonance frequencies which are advantageously used to decide whether the resonance and anti-resonance frequencies are damped in the control.
    In an advantageous further development, a predefined movement profile is traversed by the transport unit, whereby at least one system parameter of a model of the controlled system is determined by means of a parameter estimation method, wherein the time profile of the value of the system parameter is detected and from the time course to a wear state of the transport unit and / or the Transport route is closed. For this purpose, the drive coil controller can first also be parameterized according to the invention. The system parameter reflects the condition of the transport route. By observing the chronological course of the system parameter, it is therefore possible to conclude on possible wear. The current state of wear of the transport unit and / or the transport route can then be used in various ways. For example, the control can be adapted to the state of wear, e.g. in which the controller parameters are changed, or it can also be carried out a maintenance of the transport unit and / or the transport route. It is an aspired goal to keep the necessary control interventions, in particular in the form of the amplitudes of the manipulated variables, as low as possible.
    The system parameter is determined in an advantageous embodiment by detecting a stator current set on a drive coil and at the same time calculating it from the model of the controlled system and minimizing an error between the detected and calculated stator current by varying the at least one system parameter of the model.
    The control behavior of the control can be improved if a feedforward control is implemented, which acts on the input of the drive coil controller. The pilot control regulates the control error to a large extent. The drive coil controller only needs to compensate for more non-linearities, unknown external influences and disturbances that are not controlled by the force precontrol.
    The subject invention will be explained in more detail below with reference to Figures 1 to 10, which show by way of example, schematically and not by way of limitation advantageous embodiments of the invention. It shows
    1 and 2 each a transport device in the form of a long stator linear motor, Figure 3 shows a cross section through a transport unit,
    4 shows the regulator scheme of the transport device,
    5 and 6 the basic concept for the identification of the controller parameters of a drive coil controller,
    7 shows a controller cascade of the drive coil controller with pilot control and guidance smoothing filter,
    8 shows the distribution of the input driving force to be regulated onto the individual acting drive coils,
    9 shows a frequency response of the controlled system and Fig.10 a drive coil controller with pilot control.
    In Fig. 1, a transport device 1 in the form of a Langstatorlinearmotors is exemplified. The transport device 1 consists of a number n of transport sections A1... A9 (generally An), which are assembled to the transport device 1. This modular design allows a very flexible design of the transport device 1, but also requires a variety of transfer positions U1 ... U9, where the on the Trans- .R. Porteinrichtung 1 moving transport units T1 ... Tx (for reasons of clarity in Fig.1 not all transport units are denoted by a reference numeral) are transferred from one transport section A1 ... A9 to another.
    The transport device 1 is designed as a long stator linear motor, in which the transport sections A1... A9 form a part of a long stator of a long stator linear motor in a manner known per se. Along the transport sections A1... A9, a multiplicity of electric drive coils are therefore arranged in the longitudinal direction in a known manner (not shown in FIG. 1 for reasons of clarity), which are connected to excitation magnets on the transport units T1... Tx (see FIG. cooperation. In a likewise known manner, by controlling the electrical stator current iA of the individual drive coils 7, 8 for each of the transport units T1... Tx independently a driving force Fv is generated, which drives the transport units T1... Tx in the longitudinal direction along the transport sections A1 , ie along the transport route, moves. Each of the transport units T1 ... Tx can be moved individually (speed, acceleration, lane) and independently (except for the avoidance of possible collisions) from the other transport units T1 ... Tx. After this basic principle of a long stator linear motor is well known, it will not be discussed in detail here.
    Along the transport path of the transport device 1 also some transfer positions U1 ... U10 are arranged. Different types of transfer positions U1... U10 are conceivable here. At the transfer positions U2 and U7, e.g. a switch is provided, while the other transfer positions U1, U3 ... U6, U8, U9 e.g. are executed as change points from one transport section A1 ... A8 to another. At the transfer position U10, e.g. a transition from a unilateral transport section A2 provided on a two-sided transport section A9. At the transfer position U2 (switch), a transport unit T6 can for example be moved further on the transport section A2 or the transport section A3. At a transfer position U1 (change position), a transport unit T5 is transferred from the unilateral transport section A1 to the unilateral transport section A2. The transfer from one transport section to another transport section can take place in any desired manner.
    Along the transport path of the transport device 1, which is given essentially by the longitudinal direction of the transport sections A1 ... A9, a number of workstations S1 ... S4 can be arranged, in which a manipulation of the transport units T1 ... Tx transported components takes place. The workstation S1 can be designed, for example, as a sluice-in and / or outfeed station, in which finished components are removed and components to be processed are transferred to a transport unit T1... Tx. In the workstations S2... S4, any processing steps can be performed on the components. In this case, the transport units T1 ... Tx can be stopped in a workstation S1 ... S4 for processing, e.g. in a filling station where empty bottles are filled or moved, e.g. be temperature-treated in a temperature control in the components, possibly also at a different speed than between the work stations S1 ... S4.
    Another example of a transport device 1 is shown in FIG. Here five self-contained transport sections A1 ... A5 are provided. The transport section A2 ... A4 serve here for the introduction of various components at the workstations S1 ... S3. In a workstation S4 of a transport section A5, these components are interconnected or otherwise processed and discharged from the transport device 1. Another transport section A1 serves to transfer the components from the transport sections A2, A3, A4 into the transport section A5. For this purpose, transfer positions U1, U2, U3 are provided in order to transfer the transport units Tx with the various components into the transport section A1. Furthermore, a transfer position U4 is provided, in which the transport units Tx are transferred with the various components in the transport section A5.
    The transport device 1 can be made almost arbitrary and can be composed of different transport section A, and if necessary, transfer positions U and work stations S can be provided.
    3 shows a cross section through an arbitrary transport section An and a transport unit Tx moved thereon. In the exemplary embodiment shown, a transport unit Tx consists of a base body 2 and a component receptacle 3 arranged thereon for receiving a component to be transported (not shown), the component receptacle 3 basically being able to be arranged at any point of the base body 2, in particular also on the underside for hanging components. On the base body 2, preferably on both sides of the transport unit Tx, the number of excitation magnets 4, 5 of the long stator linear motor is arranged. The transport path of the transport device 1, or a transport section An, is formed by a stationary guide structure 6, on which the drive coils 7, 8 of the long stator linear motor are arranged. The main body 2 with the permanent magnets arranged on both sides as excitation magnets 4, 5 is arranged in the embodiment shown between the drive coils 7, 8. Thus, in each case at least one excitation magnet 4, 5 of a drive coil 7, 8 (or a group of drive coils) arranged opposite one another and thus cooperates with at least one drive coil 7, 8 for generating a driving force Fv together. Thus, the transport unit Tx _7_ between the guide structure 6 with the drive coils 7, 8 and along the transport path movable.
    Naturally, guiding elements 9 (such as rollers, wheels, sliding surfaces, magnets, etc.) may also be provided on the base body 2 and / or on the component receptacle 3 (not shown here for reasons of clarity), in order to transport the transport unit Tx along the To lead transport route. The guide elements 9 of the transport unit Tx act together for guidance with the stationary guide structure 6, e.g. in which the guide elements 9 are supported on the guide structure 6, slide on it or unroll, etc. The leadership of the transport unit Tx but can also be done by the provision of guide magnets. Of course, other arrangements of the drive coils 7, 8 and the cooperating excitation magnets 4, 5 are conceivable. For example, it is also possible, the drive coils 7, 8 inside and the excitation magnets 4, 5 directed inward and the drive coils 7, 8 encompassing order. Likewise, excitation magnets may be provided on only one side of a transport unit Tx. In this case, drive coils on only one side of the transport unit Tx would be sufficient.
    In order to move a transport unit Tx forward, a stator current iA is impressed in the drive coils 7, 8 in the region of the transport unit Tx (FIG. 4), different stator currents iA (in magnitude and vectorial direction) being impressed in different drive coils 7, 8 can. It is also sufficient only in the drive coils 7, 8 to impress a stator current iA, which can just interact with the excitation magnets 4, 5 on the transport unit Tx. In order to generate a driving force Fv acting on the transport unit Tx, a drive coil 7, 8 is supplied with a stator current iA having a current component iAq which generates a driving force. For the movement of the transport unit Tx, however, the drive coils 7, 8 arranged on both sides do not have to be supplied with current at the same time by impressing a stator current iA. It is basically sufficient if the driving force Fv acting on the transport unit Tx for movement is generated only by means of the drive coils 7, 8 of one side. At travel sections of the transport route where a large propelling force Fv is required, e.g. in the case of a slope, a heavy load or in areas of acceleration of the transport unit Tx, the driving coils 7, 8 can be energized on both sides (e.g., transporting section A9 in Fig. 1), whereby the driving force Fv can be increased. Likewise, it is conceivable that in certain transport sections to the guide structure 6 is executed only on one side, or that in certain transport sections to the guide structure 6 is indeed carried out on two sides, but only on one side with drive coils 7, 8 is equipped. This is also indicated in FIG. 1, in which route sections with guide construction 6 on both sides and route sections with only one-sided guide construction 6 are indicated.
    It is also known to assemble a transport section An from individual transport segments TS, each carrying a number of drive coils 7, 8. A transport segment TS can be controlled in each case by an associated segment control unit 11, as described, for example, in US Pat. No. 6,876,107 B2 and shown in FIG. 4. A transport unit Tx, which is located in a transport segment TSm, is therefore regulated by the associated segment control unit 11m. In essence, this means that the segment control unit 11m controls the drive coils 7, 8 of the associated transport segment TSm such that the transport unit Tx is moved in the desired manner (speed, acceleration) along the transport segment TSm by the generated driving force Fv. If a transport unit Tx moves from a transport segment TSm to the next transport segment TSm + 1, the control of the transport unit Tx is also transferred in an orderly manner to the segment control unit 11m + 1 of the next transport segment TSm + 1. The movement of the transport unit Tx by the transport device 1 can be monitored in a higher-level system control unit 10, which is connected to the segment control units 11. The system control unit 10 controls the movement of the transport unit Tx by the transport device 1, for example, by position specifications sSOii or speed specifications vSOii. The segment control units 11 then regulate a possible error between the target variable and actual size by applying a stator current iA to the drive coils 7, 8 of the transport segment TSm. For this purpose, it is of course necessary to measure an actual variable, such as an actual position s or an actual speed v, by means of suitable sensors or to estimate it using other measured variables or otherwise known or calculated variables. Of course, it may also be provided to provide a separate segment control unit 11 for the drive coils 7, 8, the segment control units 11 also being connectable on each side via a data line and data, for example measured values of an actual variable, being interchangeable.
    Each segment control unit 11 generates from the setpoint specifications sS0n or vson and the actual variables s and v a stator current iA, with which the required drive coils 7, 8 are acted upon. Preferably, only the drive coils 7, 8 are regulated, which interact with the transport unit Tx or their excitation magnets 4, 5. The stator current iA is a current vector (current space vector) which comprises a propulsion-force-forming q-component iAq for generating the propulsive force Fv, and optionally also a lateral force-forming d-component iAd, and which effects a magnetic flux ψ. _Q_
    To control the movement of a transport unit Tx, a drive coil controller 20 is implemented in a segment control unit 11, which regulates all drive coils 7, 8 of the transport segment TSm, as shown in FIG.
    FIG. 6 shows the basic control concept and the basic concept according to the invention for identifying the controller parameters RP of a drive coil controller 20 of a drive coil 8a, 8b as a block diagram. The controlled system 21 (essentially the technical system or the components between introducing the manipulated variable, for example in the form of the stator current iA and detecting (measuring, estimating, calculating) the actual size IG in the form of an actual position s or actual speed v of the transport unit Tx, ie in particular Drive coils 8a, 8b, transport unit Tx with excitation magnet 5 and also the interaction of the transport unit Tx with the transport section An) is controlled by the drive coil controller 20 for each drive coil 8a, 8b in a conventional manner in a closed loop. For this purpose, an actual variable IG, e.g. an actual position s or actual speed v of the transport unit Tx, detected and returned. The actual size IG can be measured, derived from other measured, calculated or known quantities or determined in a control-technical observer. The actual size IG can therefore be assumed to be known and can also be supplied to the drive coil controller 20, as indicated in Figure 6. From a control error E from the difference between setpoint SG, e.g. a setpoint position sS0n or setpoint speed vson, and actual variable IG, the drive coil controller 20 determines a manipulated variable StG, e.g. a stator current iAa, iAb for each drive coil 8a, 8b to be energized.
    The drive coil controller 20 may include a regulator cascade of a position controller RL and a speed controller RV, as shown in Fig.7. In itself but only a position controller RL or only a speed controller RV would be sufficient. As is known, the position controller RL calculates a speed vR to be adjusted from the setpoint variable SG and the actual variable IG, from which the speed controller RV again calculates a driving force FR to be adjusted, wherein the actual variable IG can also be taken into account. This input driving force FR is ultimately converted in a conversion block 25 into the stator current iA as a manipulated variable StG. For this purpose, by way of example, with the suffix K, take iAd = 0 or iAd «iAq, the known relationship FR = - ^ iAq, with the known
    Motor constants Kf, are used. If the stator current iA is calculated directly as a manipulated variable StG in the speed controller RV, the conversion block 25 can also be omitted.
    After a transport unit Tx always interacts with a plurality of drive coils 7, 8 at the same time, the input driving force FR, or the stator current i A, is also applied by all drive coils 7, 8 acting on the transport unit Tx. The input driving force FR to be adjusted is therefore still to be divided according to the current (known) position s of the transport unit Tx onto the individual acting drive coils 7, 8, as shown in FIG. The manipulated variable StG in the form of the stator current iA is in a power split unit 23 in the individual target drive coil currents iAsoii ', iAson ", iAson'" the acting drive coils 7, 8 divided. From the current position is known at any time, what proportion each acting drive coil 7, 8 contributes. From the desired drive coil currents Iasoii ', Iasoii ", Iasoii'", then in the respective drive coils 7, 8 associated with individual coil controllers 24 ', 24 ", 24'" the required coil voltages uA ', uA ", uA'" the acting drive coils 8 ', 8 ", 8'" which must be applied to the drive coils 8 ', 8 ", 8'" to set the target drive coil currents iAson ', iAson ", iAsoii'". Of course, it can also be provided for this purpose that the actual coil sizes 24 ', 24 ", 24'" are also supplied with current actual variables of the stator currents iA.
    After an individual coil controller 24 is dependent only on the concrete realization of the drive coils 7, 8, the single coil controller 24, or its parameters, can be set in advance, or assumed to be known. For this reason, the individual coil controllers 24 are preferably also assigned to the controlled system 21, as shown in FIG. Likewise, the distribution of the manipulated variable StG in sizes of the individual acting drive coils 7, 8 is preferably assigned to the controlled system 21. The coil voltages uA ', uA ", uA"' of the acting drive coils 8 ', 8 ", 8'" are then applied to the motor hardware 26 of the long stator linear motor.
    The distribution of the manipulated variable StG in sizes of the individual acting drive coils 7, 8 could of course also be performed in the drive coil controller 20. The output of the drive coil controller 20 would then be a manipulated variable StG for each acting drive coil 7, 8. In this case, of course, a plurality of excitation signals AS, just an excitation signal AS for each acting drive coil 7, 8, should be provided. Likewise, it is possible to realize the single coil controller 24 in the drive coil controller 20. In this case, the manipulated variables StG would just be voltages, which means that the starting signal AS is also a voltage. However, this does not change the inventive idea.
    In this control concept, one can think of the position controller RL and the speed controller RV belonging to the transport unit Tx. So there are just as many position controllers RL and speed controller RV as transport units Tx. For each drive coil 7, 8 there is a subordinate individual coil controllers 24 ', 24 ", 24'".
    As usual, the drive coil controller 20, or the controllers implemented therein, has a number of controller parameters RP to be set, so that a stable and sufficiently dynamic control of the movement of the transport unit Tx is possible. The controller parameters RP are usually set once, usually before or during startup of the transport device 1, for example via the system control unit 10. It should be noted that the controller parameters of the individual coil controller 24 are not parameterized in the rule, since the single coil controller 24 essentially only from the concrete, known embodiment of the drive coils 7, 8 is dependent. These controller parameters of the single coil controllers 24 are therefore normally known and need not be changed. Consequently, the controller parameters of the controller units assigned to the transport unit Tx, ie, e.g. of the position controller RL and the speed controller RV.
    However, the determination of the controller parameters RP is difficult. On the other hand, during operation of the transport device 1, the controlled path (drive coils 7, 8, transport unit Tx with excitation magnets 4, 5) and also the interaction of the transport unit Tx with the transport section An can change. Such a change may, for example, already result from the fact that the transport unit Tx is loaded with different loads. Likewise, friction between the transport unit Tx and the guide structure 6 of the transport section An also has an effect, wherein the friction in turn may depend on the current state of wear of the transport unit Tx and the transport section A. But also operating parameters, such as the current speed of the transport unit Tx or an ambient temperature, can act on the controlled system 21, for example by speed-dependent or temperature-dependent friction, and influence the control. So that the drive coil controller 20 can control robust and stable at these very different, fluctuating conditions in a wide range, the drive coil controller 20 had previously been designed to be very conservative about the controller parameters. This, however, reduces the dynamics of the controller, in the sense of rapid control interventions, such as rapid changes in speed. To improve this problem, the procedure according to the invention is as follows, reference being made to FIGS. 5 and 6.
    A measuring cell MZ is defined, wherein the measuring cell MZ comprises at least two drive coils 8a, 8b of a side which interact with the transport unit Tx, preferably at least two adjacent drive coils 8a, 8b, as shown in FIG. In Figure 5 is shown for simplicity and without limiting the generality only one side of a single transport segment TSm with the transport unit Tx. If transport segments TSm are provided with a number of drive coils 8, then the measuring cell MZ preferably comprises all drive coils 8 of a transport segment TSm or all drive coils 8 of several transport segments TSm. _10_
    First, a coarse parameterization of the controller parameters RP is performed. This can be done on the basis of the known mass of the transport unit Tx (including the expected load) and the known design data of the long stator linear motor, the controller parameters RP are usually set so that the closed loop a greatly reduced bandwidth (low dynamics), However, a great robustness (high stability) has. Depending on the drive coil controller 20 used, e.g. a conventional PI controller, various methods for controller parameterization are known, with which a rough parameterization can be performed. The rough parameterization is merely to ensure that the transport unit Tx can be moved and positioned without placing high demands on dynamics and accuracy. With this coarse parameterization, a specific operating point can be approached with the transport unit Tx by specifying a corresponding desired value SG. In this case, a specific position s (standstill of the transport unit Tx) or a specific speed v of the transport unit Tx is understood as the operating point. Start-up means, of course, that the operating point is approached in the region of the measuring cell MZ, ie that the transport unit Tx is moved through the measuring cell MZ at a certain speed, for example, or that the transport unit Tx is moved into the area of the measuring cell MZ and stopped therein.
    In the operating point, an excitation signal AS is introduced into the closed control loop by superposing the manipulated variable StG with the excitation signal AS. The starting signal AS is impressed on all drive coils 8a, 8b of the measuring cell MZ. The start signal AS comprises a specific, predetermined frequency band. Possible excitation signals AS are, for example, a known pseudo-random binary sequence signal (PRBS) or sinusoidal sweep signal. The frequencies in the start signal AS and the amplitudes of the start signal AS are chosen so that the system responses are sufficiently informative, i. that the system responses in the frequency range of interest are sufficiently large in order to be able to be evaluated by signal technology. A frequency range of interest is, in particular, the range in which resonance or antiresonance is expected. For the objective application is often a frequency range of 10Hz to 2.500Hz, especially 500Hz to 1000Hz, interesting. The amplitudes of the starting signal AS can be based on the rated current (or rated voltage) of the long stator linear motor and are typically in the range of 1/10 of the rated current (or rated voltage). The starting signal AS should preferably be zero in the mean value, with which the controlled system (controlled system 21) itself remains as uninfluenced as possible. With the excitation signal AS, the desired movement of the transport unit Tx (given by the position specifications sS0n or speed specifications Vsoii for starting the operating point) superimposed on a Anregebewe supply, which is only possible if the measuring cell MZ comprises at least two drive coils 8a, 8b.
    The control variable StGAS superposed with the starting signal AS and the response of the controlled system 21 to this excitation, which corresponds to the actual quantity IG, are fed to an evaluation unit 22. The response of the controlled system 21 is of course the current state of motion of the transport unit Tx in the form of the actual position s or actual speed v. The response of the controlled system 21 can be measured directly, derived from other measured variables or can also be calculated by an observer or otherwise estimated. In the evaluation unit 22, the frequency response (with amplitude response and phase response) is ascertained from the control variable StGAS superimposed on the excitation signal AS and the response of the controlled system 21, typically by filtering and discrete Fourier transformation of the two signals and subsequent element-wise division of the two signals according to the scheme output divided by input. The frequency response can be determined for the open and / or closed loop.
    It should be noted that although several drive coils 8a, 8b of the measuring cell MZ the starting signal AS must be impressed, for the determination of the controller parameters RP but only the superimposed control variable StGAS one of the drive coils 8a, 8b of the measuring cell MZ must be evaluated. When the frequency response is referred to below, this is the frequency response associated with the transport unit Tx and a drive coil 8a, 8b cooperating with the transport unit Tx.
    The frequency response can serve as a basis for determining the optimal controller parameters RP. Various methods known in control engineering can be used for this purpose. In this case, the controller parameters RP are varied in order to set a specific characteristic of the frequency response in the desired manner. One known method is, for example, the Maximum Peak Criteria. The maximum peak criterion method will be explained in greater detail with reference to FIG. Therein, the frequency response in the form of the amplitude response (Figure 9a top) and the phase response (Figure 9b bottom), respectively for the open (dashed) and closed loop, shown. The open loop is known to be the consideration without feedback of the actual size IG to the target size SG. The controller parameters RP are now varied at maximum peak criteria so that the maximum value of the amplitude response of the closed loop does not exceed a certain, predetermined value MT. This value MT results e.g. from desired limits for the gain and phase margin of the open loop. This ensures that the open loop has sufficient phase reserve PM (phase cp at gain zero dB) and gain margin GM (gain G at phase -180 °). Of course, depending on the implementation of the drive coil controller 20, different controller parameters _Ί / 1_ RP have to be varied, such as, for example, a gain and a reset time for a PI controller. There are also various methods for varying the controller parameters RP. For example, an optimization problem could be formulated to minimize the distance of the maximum of the closed loop amplitude response to the value MT.
    In this way, the optimum controller parameters RP are obtained for the respective transport unit Tx. These controller parameters RP can now also be used for the same transport units Tx. Likewise, it is conceivable in the same way to determine the respective optimal controller parameters RP for each or several transport units Tx.
    The determination of the controller parameters RP can also be carried out for different operating points and / or different loads of the transport unit Tx. Likewise, the controller parameters RP for a transport unit Tx can also be determined for different measuring cells MZ. Thus, during operation of the long-stator linear motor for a transport unit Tx, it is also possible to switch between different controller parameter sets. For example, the controller parameter set may be selected that best matches the current load carried by a transport unit Tx or that best fits the current speed or position of the transport unit Tx. In this way, one or more controller parameter sets can be created for each transport unit Tx. In this way, consideration can also be given to the differences between the various transport units Tx. Ideally, it is already known in advance how and with what load the transport units Tx in the transport device 1 will be moved. This means that the right operating point or the appropriate measuring cell MZ can already be approached for controller parameterization.
    However, the frequency response also includes other essential characteristics of the controlled system 21. For example, the current total mass mG of the transport unit Tx can be determined during operation from the amplitude response. From this, in turn, a loading of the transport unit Tx can be concluded, since the mass mTx of the transport unit Tx is known. Any difference must therefore be attributed to the load, with which the load can be determined. For a known load, then, e.g. In turn, the appropriate controller parameter set for optimal control of the transport unit Tx be selected. To determine the total mass mG, e.g. the amplitude response | G (j ^ f) | at low frequencies f and the relation | G (j27i: f) | = ---, with the known, normalized motor constant Kf and the
    This relationship applies to sufficiently small values of the frequency f under the assumption of low viscous friction (friction force is proportional to and opposite to the magnitude of the velocity), which can be assumed in the present case of application. From this, the total mass mG can be calculated.
    Furthermore, from the frequency response (FIG. 9), it is possible to determine, as a characteristic of the controlled system 21, any resonant and antiresonant frequencies which may always occur in pairs. A resonance / anti-resonant frequency can be assumed at local or global maxima / minima of the amplitude response. By evaluating the amplitude response of the open loop such local or maximum maxima / minima can be easily found, even automated. If resonant frequencies fR and antiresonant frequencies fAR are present, depending on the location of the resonant frequencies fR and anti-resonant frequencies fAR on the frequency axis, a classification of the controlled system 21 into categories such as rigid, stiff and flexible can be performed. In this case, a controlled system 21 can be classified as rigid if the resonance / antiresonance pair with the lowest frequency values (fR, fAR) is significantly larger than the phase frequency fD. The phase transmission frequency fD is known to be the frequency at which the phase φ of the open loop intersects the value -180 ° the first time. The controlled system would be stiff if the frequency values (fR, fAR) of the resonant / antiresonant pair are in the range of the phase frequency fD and flexible if the frequency values (fR, fAR) of the resonant / antiresonant pair are significantly smaller than the phase transmission frequency fD are. Depending on the category, it is decided whether the resonance / anti-resonance frequencies (fR, fAR) are disturbing and with what measures they are eliminated or damped, for example by means of a suitable filter.
    The controller parameterization and / or the determination of the characteristics of the controlled system 21 can also be repeated during ongoing operation at certain intervals. Thus, the drive coil controller 20 can be continuously adapted to the changing state of wear of the transport unit Tx and thus to a changing controlled system 21. The controller parameterization can for example be performed every day before switching off the transport device 1 or before the commissioning of the transport device 1.
    The determined controller parameters RP could then be checked for plausibility. For example, the operating point used for the controller parameterization could be approached with the drive coil controller 20 with the determined optimum controller parameters RP, and the starting signal AS could be switched on again. The frequency response of the closed loop is again determined and based on its maximum resonance peak decided whether the behavior of the closed loop is satisfactory. Likewise, one could, additionally or alternatively, the location of the resonance frequency _1R. Check the fR or anti-resonance frequency fAR and / or the phase transmission frequency fD and thus check the plausibility of the controller parameters RP.
    Along the transport path of the transport device 1, a plurality of measuring cells MZ may be provided. Thus, different optimal controller parameters RP can be determined for different sections of the transport path. The determined controller parameters RP for a transport unit Tx preferably always apply from a first measuring cell MZ1 to the next measuring cell MZ2.
    With a parameterized drive coil controller 20, the controlled system 21 can now also be analyzed with regard to further system parameters of interest for the process. For this purpose, the controller parameters RP of the drive coil controller 20 can be used e.g. can be identified as described above, but can also be determined otherwise or may also be known. Basically, the only requirement is that a predetermined motion profile can be traversed with the drive coil controller 20. The movement profile should stimulate the controlled system 21 sufficiently to be able to identify the system parameters. For this purpose, a transport unit Tx is given a certain movement profile, e.g. in the form of a temporal course of different speeds and accelerations (also in the sense of delays), given. It is advantageous if movements are contained in both directions to detect direction-dependent system parameters. This movement profile, as desired variables of the control, is traversed by the transport unit Tx under control of the drive coil controller 20. For this purpose, the drive coil controller 20 generates control variables StG in accordance with the motion profile, which act on the controlled system 21 and effect actual variables IG of the controlled system 21, which act back on the setpoint variables SG in the closed control loop. For the controlled system 21, a model with system parameters is now assumed which describes the controlled system 21 as well as possible. For example, for the transport unit Tx the equation of motion
    with the total mass mG of the transport unit Tx, a coefficient kv for the viscous friction, a coefficient ks for the static friction, the current speed v of the transport unit Tx and the sign function sign. The driving force Fv acting on the transport unit Tx is composed, as described above, of the effects of all drive coils 7, 8 acting on the transport unit Tx, that is to say with the force Fvasi applied by a drive coil 7, 8 Long stator linear motor known in the form
    be modeled. In this case, ψρ denotes the magnetic flux produced by the excitation magnets 4, 5 and linked to the drive coil 7, 8, tp corresponds to the pole width of the excitation magnets of the transport unit Tx, and x denotes the position of the transport unit Tx. LAci and LAq denote the known inductances of the drive coil 7, 8 in d- and q-direction. With the allowable assumption iAd = 0 or iAd «iAq, this equation can be added to
    The stator current iAq of a drive coil 7, 8 is then obtained from the respective portion of the drive coil 7, 8 at the drive force Fv.
    The system parameters of the model of the controlled system 21, in this case the total mass mG of the transport unit Tx, the coefficients kv for the viscous friction, the coefficients ks and for the static friction, can be determined from this with the assumption of a known motor constant Kf by known parameter estimation methods , If another system parameter is known, e.g. the total mass mG as described above, the motor constant Kf can also be estimated therefrom. For the parameter estimation, the predetermined motion profile is traversed, whereby the speed v (or equivalently the position s) and the acceleration - is defined as input into the parameter estimation method. The stator current iAq set at a drive coil 7, 8 corresponds to the manipulated variable StG and is known or can also be detected otherwise, e.g. be measured. At the same time, the stator current iAq is calculated from the model of the controlled system 21 and the error (e.g., mean square error) between the calculated and measured stator currents is minimized by varying the system parameters of the model. Known parameter estimation methods are e.g. the Least Square method, the Recursive Least Square method, a Kalman or Extended Kalman filter.
    The system parameters determined thereby identify the controlled system 21, ie in particular also the transport path or a transport section An or a transport segment TSm via the coefficients kv for the viscous friction and the coefficient ks for the static friction, as well as the air gap between the excitation magnet 4, 5 By observing the time course of these system parameters on the same section of the transport path, conclusions can be drawn about the wear state of the transport unit Tx and / or the transport path, in particular the transport section An or a transport segment TSm. If the system parameters of the controlled system 21 are determined regularly, e.g. every day once, then from their temporal change from the coefficient kv for the viscous friction and the coefficient ks for the static friction, on a possible wear can be concluded. If these coefficients increase, then this is an indication of progressive wear. Similarly, from the motor constant Kf a change in the air gap can be detected, which may also point to progressive wear. In the case of impermissible changes, for example, determined by exceeding a predetermined limit, maintenance of the transport unit Tx and / or the transport section An can also be initiated.
    In order to improve the guiding behavior of the regulation of the movement of the transport units Tx by the drive coil controller 20, the drive coil controller 20 can also be supplemented by a feedforward control V. The precontrol V acts (for example, by addition) on the input of the drive coil controller 20. This is illustrated in FIG. 10 using the example of a cascaded drive coil controller 20. The precontrol V acts (eg by addition) in each case on the input of the associated controller, ie a speed precontrol vVs on the input of the speed controller RV and a force precontrol FVs on the input of the conversion block 25. The feedforward V can in a conventional manner on a model of Controlled system 21 are based, as a precontrol V usually the inverse of the model of the controlled system 21 is used. The model is preferably implemented in the form of equations of motion of the transport unit Tx, as stated above. The model is determined by the identified system parameters, which also determines the feedforward (as inverse of the model). Instead of a model of the controlled system 21, however, it is also possible to implement any other feedforward control law. For a speed feedforward vVs, for example, the following model can be used,
    with the current actual position s as actual variable IG.
    The speed controller RV therefore only controls more nonlinearities, unknown external influences and disturbance variables which are not regulated by the speed precontrol vVs. _4Q_ For a force precontrol FVs, for example, the above model can be used,
    with the coefficient kv for the viscous friction, the
    Coefficients ks for the static friction, the current speed of the transport unit Tx and the sign function sign.
    From the force demand thus determined, which is required to compensate for the current control error E, the conversion block 25 calculates the manipulated variable StG for a drive coil 7, 8, e.g. in the form of the stator current iA to be set. The current controller RS controls with a force precontrol only more nonlinearities, unknown external influences and disturbances, which are not controlled by the force precontrol.
    Furthermore, the drive coil controller 20 can be supplemented in a known manner by a guide smoothing filter FF, even without feedforward V, as shown in Fig.10. The guidance smoothing filter FF can from a control engineering point of view, e.g. be implemented as a finite impulse response (FIR) filter with a time constant T. The guide smoothing filter FF is used to filter the target quantity SG to prevent the excitation of certain unwanted frequencies. For example, the guidance smoothing filter FF could be implemented as jerk limitation (with the jerk as time derivative of the acceleration).
    The target quantity SGf filtered by the guide smoothing filter FF is then used for the feedforward control V and the control by the drive coil controller 20.
    From a specification of a movement profile in the form of a point-to-point positioning of the transport unit Tx, the following error behavior (difference between the target movement profile and the actual movement profile) can be evaluated at the end of this movement profile. From the period of the decaying oscillation of the following error (for example, as the amplitude ratio of the first two half-waves) and the period of the first oscillation, the time constant T of the guide smoothing filter FF, which corresponds to the period duration, can then be calculated in a known manner.
    The determination of the system parameters of the model of the controlled system 21 and / or the parameters of the guidance smoothing filter FF are of course dependent on the route due to the specification of the motion profile. This can also be derived properties of the transport route, such as static or dynamic friction parameters. On the basis of these properties of the transport route, in particular on the basis of the temporal change of these properties, it is therefore also possible to draw conclusions about the condition of the transport route. If the same properties are determined on the same transport path for different transport units Tx, a comparison of the properties can also be made as to the (wear) state of the transport unit Tx.
    The imprinting of a movement profile for determining the system parameters and / or parameters of the guidance smoothing filter FF preferably takes place on a transport path along which no high demands are placed on the movement of the transport unit Tx (speed specification, position specification).
    It is also conceivable to determine the system parameters and / or the parameters of the guidance smoothing filter FF at different transport sections An, for example for each transport segment TSm. In this way, by observing the time course of the system parameters of different transport sections An conclusions on the state of wear of the various transport sections to be drawn. _01_
    权利要求:
    Claims (4)
    [1]
    claims
    A method of operating a long stator linear motor with a transport path along which a plurality of drive coils (7, 8) are arranged one behind the other and with at least one transport unit (Tx), which is moved along the transport path, each drive coil (7, 8) of a Drive coil controller (20) is controlled, characterized in that the transport unit (Tx) is given a motion profile, which is traversed by the transport unit (Tx) that while at least one system parameter of a model of the controlled system (21) is determined by a parameter estimation method and that the temporal course of the value of the at least one system parameter is detected and is concluded from the time course to a state of wear of the transport unit (Tx) and / or the transport path.
    [2]
    2. Method according to claim 1, characterized in that a stator current (iA) set at a drive coil (7, 8) is detected and at the same time calculated from the model of the controlled system and an error between the detected and calculated stator current is minimized by the at least one system parameter of the model is varied.
    [3]
    3. The method according to claim 1 or 2, characterized in that a feedforward control (V) is implemented, which acts on the input of the drive coil controller (20).
    [4]
    4. The method according to claim 3, characterized in that the drive coil controller (20) comprises a speed controller (RV) with a speed input and / or a conversion block (25) with a force input and the feedforward control (V) Ge-speed precontrol (vvs) and or a force precontrol (Fvs) is calculated, wherein the speed precontrol (vvs) acts on the speed input and / or the force precontrol (Fvs) on the force input. _oo_
    类似技术:
    公开号 | 公开日 | 专利标题
    EP3251986B1|2021-01-27|Method for operating a long stator linear motor
    EP3379719B1|2021-04-14|Method for transferring a transport unit to a transferring position
    EP3341803B1|2019-06-19|Method for the open-loop and/or closed-loop control of a linear drive, a control device, a linear drive and a system
    EP3385110A1|2018-10-10|Method for controlling the normal force of a transport unit of a linear motor with guideway stator
    EP3385803B1|2020-11-18|Method for operating a long stator linear motor
    DE10334736A1|2005-02-17|Linear motor with advancement or motion regulation for use in industrial automated processes, with bearing unit for guiding a secondary component along the desired path
    DE102012003479A1|2013-08-22|Method and device for carrying out a manipulator process
    EP3547530B1|2020-08-05|Method for operating a transport device in the form of a linear motor with guideway stator
    EP0452375B1|1992-12-30|Automatic goods transport device with transport elements driven by a linear motor
    WO2015140155A1|2015-09-24|Transportation apparatus for moving and/or positioning objects
    EP3251985A1|2017-12-06|Method for operating a long stator linear motor
    EP1864370A1|2007-12-12|Linear motor and method for operation of a linear motor
    DE3411190C2|1986-11-06|Magnetic regulator for long-stator magnetic levitation vehicles
    DE102016222806B3|2018-04-05|Transport goods carrier, transport system and method for transporting a cargo
    EP3363751A2|2018-08-22|Method for transfering a transport unit of a linear motor conveyor to a transfer position
    EP3575250A1|2019-12-04|Method for operating a conveying device in the form of a long stator linear motor
    DE2446936A1|1976-04-08|Hover vehicle dynamic uncoupling control - with magnetic flux or temporary change introduced as additional quantitive dimension
    DE102011011810A1|2012-08-23|Electromagnetic abeyance concept for contactless generation of magnetic supporting force, guiding force, and drive force, comprises track, on whose both sides rails are arranged to generate field effects lower side in air gap of magnets
    DE2446851A1|1976-04-08|Magnet guided rail hover vehicle - has magnetic current of carrier or guide magnets introduced as special quantitative dimension
    EP3887911A1|2021-10-06|Method for controlling an automation process in real time
    WO2021185910A1|2021-09-23|Energy transmission in a linear transport system
    同族专利:
    公开号 | 公开日
    AT518734B1|2018-05-15|
    EP3251985A1|2017-12-06|
    US20170346380A1|2017-11-30|
    CN107453679A|2017-12-08|
    US10554111B2|2020-02-04|
    CA2968931A1|2017-11-30|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    DE102011075174A1|2011-05-03|2012-11-08|Robert Bosch Gmbh|Transfer device for objects|
    US20140097783A1|2011-06-15|2014-04-10|Kabushiki Kaisha Yaskawa Denki|Conveying system|
    DE102012025326A1|2012-12-22|2014-06-26|Festo Ag & Co. Kg|Method for operating electro-magnetic transportation system of car, involves generating control commands to control units to operate movement profile for transport device based on activations of magnetic coils and magnetic coil groups|
    US20150303841A1|2014-04-18|2015-10-22|Canon Kabushiki Kaisha|Linear motor control apparatus and linear motor control system|
    FR1596424A|1968-12-26|1970-06-15|
    US6208497B1|1997-06-26|2001-03-27|Venture Scientifics, Llc|System and method for servo control of nonlinear electromagnetic actuators|
    US6101952A|1997-12-24|2000-08-15|Magnemotion, Inc.|Vehicle guidance and switching via magnetic forces|
    US6876107B2|2002-06-05|2005-04-05|Jacobs Automation|Controlled motion system|
    US20080115372A1|2003-05-20|2008-05-22|Hanspeter Vogel|Rail Assembly, Rail Switch And A Transport Device Provided With A Magnetostrictive Sensor|
    US9032880B2|2009-01-23|2015-05-19|Magnemotion, Inc.|Transport system powered by short block linear synchronous motors and switching mechanism|
    DE102011003682A1|2011-02-07|2012-08-09|Robert Bosch Gmbh|Transport device with recognition function|
    DE102012204919A1|2012-03-27|2013-10-02|Beckhoff Automation Gmbh|STATOR DEVICE FOR A LINEAR MOTOR AND LINEAR TRANSPORT SYSTEM|
    DE102013218389A1|2013-09-13|2015-03-19|Krones Ag|Device and method for switching a passive switch for transport systems with linear motors|
    DE102013218394A1|2013-09-13|2015-03-19|Krones Ag|Device and method for the maintenance of transport elements in a container treatment plant|
    KR102331404B1|2013-09-21|2021-11-25|마그네모션, 인코포레이티드|Linear motor transport for packaging and other uses|US11165372B2|2017-09-13|2021-11-02|Rockwell Automation Technologies, Inc.|Method and apparatus to characterize loads in a linear synchronous motor system|
    DE102018209723A1|2018-06-15|2019-12-19|Krones Ag|Method and device for monitoring the wear of a long-stator linear motor system|
    法律状态:
    优先权:
    申请号 | 申请日 | 专利标题
    ATA50494/2016A|AT518734B1|2016-05-31|2016-05-31|Method for operating a long-stator linear motor|ATA50494/2016A| AT518734B1|2016-05-31|2016-05-31|Method for operating a long-stator linear motor|
    EP17172343.0A| EP3251985A1|2016-05-31|2017-05-23|Method for operating a long stator linear motor|
    US15/608,095| US10554111B2|2016-05-31|2017-05-30|Method for operating a long stator linear motor|
    CN201710405059.2A| CN107453679A|2016-05-31|2017-05-31|Method for running long-stator linear motor|
    CA2968931A| CA2968931A1|2016-05-31|2017-05-31|Method for operating a long stator linear motor|
    [返回顶部]