EP4299904B1 - Method for controlling variable-speed fluid pumps - Google Patents

Description

The present invention relates to a method for controlling variable speed fluid pumps, as well as a computer program and a computing unit for carrying out the method.

Background of the invention

In hydraulics, fluid pumps are often used with a pumping mechanism with a fixed delivery volume, which is driven by a drive with variable speed (so-called variable speed constant pump). When operating such pumps, the volume flow and/or the delivery pressure are usually regulated by adjusting the speed accordingly. Such pumps are used, for example, for hydraulic units, pump drives or servo-hydraulic actuators, for example in machine tools or in plastics processing, such as in injection molding machines, blow molding machines, presses, etc., where high delivery pressures are required.

The control of such variable speed pumps can be implemented, for example, by a cascaded controller structure in which a target pressure is specified and includes a pressure control circuit, a speed control circuit and a current control circuit for the drive motor. However, the speed control circuit is disturbed by the load torque M _L , which is set depending on the current pressure via the pump displacement volume. The pressure control circuit is also disturbed by the leakage volume flow of the pump. These disturbances impair the control quality of the drive system.

WO 2018/0207157 A2 shows a method for controlling pressure and flow of hydraulic fluid through a hydraulic control unit, which includes receiving a pressure command from an electronic control unit. The received pressure command is processed to generate a pressure-based torque command and a shaft angle-based torque command or a volume-based torque command. The generated pressure-based torque command and the generated shaft angle-based torque command or the volume-based torque command are processed to further generate a duty cycle for a field-oriented controlled motor.

The DE 10 2019 220 322 A1 relates to a method for operating a variable speed variable displacement pump, in which a conveyor mechanism adjustable in a displacement volume per working cycle is driven by means of a variable speed drive, wherein within the framework of a control at least one variable is controlled to a setpoint by specifying a speed setpoint for a speed of the drive and a setpoint for a parameter determining the displacement volume per working cycle, wherein the speed setpoint is adapted by means of a feedforward control in the event of a change in the setpoint for the parameter determining the displacement volume per working cycle.

It is therefore desirable to improve the control of variable speed fluid pumps.

disclosure of the invention

According to the invention, a method for controlling a pump with a variable speed drive as well as a computer program and a computing unit for carrying out the method with the features of the independent patent claims are proposed. Advantageous embodiments are the subject of the subclaims and the following description.

In particular, a method is proposed in which a target pressure value is first entered into a cascaded controller structure, wherein the controller structure comprises a pressure control loop and wherein the control system of the pressure control loop comprises a speed control loop into which a target speed is entered as a control variable of a pressure controller of the pressure control loop. The control system of the speed control loop also comprises a drive control loop into which a target drive torque is entered as a control variable of a speed controller of the speed control loop. In addition, a pre-control value for the drive torque is as a result of a time differential equation of the target pressure, and this pilot value is added to the target drive torque, which is input into the control system of the speed control loop.

According to the invention, the temporal differential equation of the target pressure comprises an inverse system model of the controlled system of the cascaded controller structure.

This inverse system model can, for example, be formed depending on a pressure dynamics in a load volume in which the pressure is controlled and a torque dynamics of the drive.

The pilot control value for the drive torque can be determined in exemplary embodiments by the following equation: $$M_{\mathit{cmd}}^{\mathit{ff}}=\frac{J_{\mathit{eq}}2{\mathit{\pi C}}_h}{V_g}{\ddot{p}}_L^d+\frac{2\pi }{V_g}\left(J_{\mathit{eq}}{C}_l+d_M{C}_h\right){\dot{p}}_L^d+\left(\frac{d_{M}2{\mathit{\pi C}}_l}{V_g}+\frac{V_g}{2\pi }\right)p_L^d$$

Where J _eq is an equivalent moment of inertia, C _h is a hydraulic capacity, V _g is a displacement volume of the pump, p _L ^d is a target load pressure, C _l is a pressure-dependent leakage coefficient and d _M is a speed-related friction torque of the pump.

It is also possible to create a pre-control value for the speed, whereby the pre-control value for the speed is determined by: $${\mathrm{n}}_{\mathit{cmd}}^{\mathit{ff}}=\frac{C_h}{V_g}{\dot{p}}_L^d+\frac{C_l}{V_g}\cdot p_L^d$$

Where C _h is the hydraulic capacity, V _g is the displacement volume of the pump, p _L ^d is the target load pressure and C _l is the pressure-dependent leakage coefficient.

This pre-control value can then be added to the target speed, which is output by the pressure controller as a control variable. This can prevent the speed control loop from working against the torque pre-control.

Optionally, the target pressure value can be entered into a target value filter which is configured to output a target trajectory for the target pressure value and the derivatives of the target pressure value, and this target trajectory can then be entered into the differential equation for forming the pilot control value.

Such a setpoint filter may, for example, comprise a second-order state variable binomial filter.

Overall, the flatness-based feedforward control proposed here dynamically compensates for load torques and disturbance volume flows. This improves the control behavior and the stability reserve of the system. The drive controller can be designed for interference immunity, since the follow-up behavior is significantly improved by the feedforward control. This relieves the load on the speed and pressure controllers. At the same time, the feedforward control is insensitive to sensor noise.

A computing unit according to the invention, e.g. a control unit of a hydraulic system, is set up, in particular in terms of programming, to carry out a method according to the invention.

The implementation of a method according to the invention in the form of a computer program or computer program product with program code for carrying out all method steps is also advantageous, since this causes particularly low costs, especially if an executing control unit is also used for other tasks and is therefore already present. Suitable data carriers for providing the computer program are in particular magnetic, optical and electrical storage devices, such as hard disks, flash memories, EEPROMs, DVDs, etc. It is also possible to download a program via computer networks (Internet, intranet, etc.).

Further advantages and embodiments of the invention emerge from the description and the accompanying drawings.

It is understood that the features mentioned above and those to be explained below can be used not only in the combination specified in each case, but also in other combinations or on their own, without departing from the scope of the present invention.

The invention is illustrated schematically in the drawing using embodiments and is described in detail below with reference to the drawing.

character description

• Figure 1 shows schematically a fluid pump with a variable speed drive; and
• Figure 2 shows an exemplary controller structure according to a possible embodiment.

Detailed description of the drawing

Figure 1 shows schematically components of a variable speed fluid pump for which the methods described here can be applied. In this example, the pump 10 has a conveyor 20 which can convey in two directions, a variable speed drive 30 and a computing unit 40 for operating the pump.

The conveyor 20 can be designed, for example, as a constant pump that achieves a certain delivery volume per stroke or per revolution. The drive 30 is designed here as a standard motor that has an asynchronous motor 31 and a frequency converter 32. Alternatively, synchronous servo motors or other drive units can also be used. The speed n of the asynchronous motor 31 is variable, with the speed being set, for example, by the current applied to the drive motor 31.

The computing unit 40 can be used to control the pump 10, to which at least one target delivery pressure and optionally a target volume flow are supplied. The computing unit 40 can comprise a drive controller implemented in hardware or software, which determines the required control variables and transmits them to the motor 30. Optionally, the actual pressure can be determined using a sensor. The computing unit 40 can further determine, e.g. by monitoring the electrical current and the speed, which drive torque is currently being delivered by the drive, and can then, for example, transmit a target speed to the drive unit 30 as a control signal. Alternatively, it is also possible for the drive unit 30 itself to comprise certain control or regulation elements, and thus for the computing unit 40 to, for example, only specify a target torque for controlling the drive and thus the pump and transmit this to the drive 30 as a control signal.

It is understood that this pump drive system is only described as an example and the controller variants described below are not limited to this system. In particular, the present invention can also be used with any other pumps and is not limited to fixed displacement pumps. It is also possible that certain parts of the cascaded controller structure are implemented in several separate computing units, that certain sensors are not present in other systems or additional parameters are measured via sensors, or that the drive unit is designed differently than described here.

Figure 2 shows a controller structure for controlling a variable speed fluid pump according to a possible embodiment.

A cascaded controller structure is provided which nests a pressure control loop, a speed control loop and a drive control loop such as a current control loop. The current control loop 230 for the pump drive is shown here only in a simplified and exemplary manner and can be implemented in any way, e.g. as a PI controller; with a different drive variant, a different control loop can also be used at this point which controls a drive torque of a pump drive. The current control loop 230 receives a target drive torque M ^R _cmd as a reference variable and feeds back the actual value 232 of the drive torque M _act . A controlled drive torque M _M is thus obtained.

The target drive torque M ^R _cmd is in turn the manipulated variable of a speed control loop. The speed control loop receives a target speed n _cmd of the pump as a reference variable for a speed controller 220 and feeds back an actual speed n _act 222, whereby the current control loop 230 together with the moment of inertia element 224, which outputs the controlled speed n from the drive torque M _M with the moment of inertia of the pump and motor, forms the controlled system of the speed control loop.

The target speed n _cmd forms the manipulated variable of a pressure control loop, the pressure controller 210 of which receives a target load pressure p _L ^d as a reference variable and feeds back 212 an actual value p _L of the load pressure. The control system of the pressure control loop thus comprises the speed control loop 220, 222, 224, the current control loop 230, 232 and further control elements. The further control elements of the system comprise a proportional element 214 in order to convert the controlled speed n obtained from the speed control loop into a volume flow Q _p by taking into account the predetermined displacement volume V _g of the pump per revolution. The controlled load pressure p _L is then obtained by means of a PI element 216 from the volume flow Q _p by taking into account the equivalent compression module E' _oil and the load volume V _L in which the pressure is regulated. The actual pressure p _act is in turn fed back via the pressure control circuit 212 to the input of the pressure regulator 210.

The speed control loop is disturbed by a load torque M _L , which can be represented as a proportional element 240 from the existing load pressure p _L and the displacement volume V _g of the pump. The resulting load torque M _L counteracts the drive torque M _M .

In addition, the pressure control loop is disturbed by the leakage volume flow Q _L of the pump, which results from the load pressure p _L and the leakage coefficient C _l and can be modeled as a proportional element 242. The leakage volume flow Q _L counteracts the volume flow Q _p of the pump.

In order to compensate for these disturbances, a pilot control 250 can now be provided according to the invention, which corrects the commanded drive torque M ^R _cmd for the pump drive, which is output by the speed controller 220 as a manipulated variable and enters the current control loop 230 as a reference variable, by means of a pilot control value M ^ff _cmd . The actual value that is thus entered into the current control loop or the controlled system of the speed control loop is therefore the sum of the drive torque M ^R _cmd and the pilot control value M ^ff _cmd .

wobei C_h= V_L / E'_Öl die hydraulische Kapazität mit dem Ersatzkompressionsmodul E' _Öl und dem Lastvolumen V_L ist, V_g das Schluckvolumen der Pumpe ist, ω_M das Drehmoment des Antriebs ist, und C_l der druckabhängige Leckagebeiwert ist.The pre-control value can be formed in particular via an inverse system model of the controlled system. This system model can be configured according to the Figure 2 shown controller structure can be derived as follows:
The pressure dynamics in the load volume, ie the change p _L of the load pressure results in $${\dot{p}}_L=\frac{1}{C_h}\cdot \left(\frac{V_g}{2\pi }\cdot {\omega }_M-C_l\cdot p_L\right)$$

where C _h = V _L / E' _oil is the hydraulic capacity with the equivalent compression modulus E ' _oil and the load volume V _L , V _g is the displacement of the pump, ω _M is the torque of the drive, and C _l is the pressure dependent leakage coefficient.

By differentiating with respect to time we obtain $${\ddot{p}}_L=\frac{1}{C_h}\cdot \left(\frac{V_g}{2\pi }\cdot {\dot{\omega }}_M-C_l\cdot {\dot{p}}_L\right)$$

• wobei J_eq das äquivalente Trägheitsmoment ist, das sich aus den Trägheitsmomenten J_Pumpe der Pumpe und J_Motor des Antriebsmotors ergibt, J_eq= J_Pumpe + J_Motor ;
• wobei M_M das Antriebsmoment ist;
• wobei V_g das Schluckvolumen der Pumpe ist, und
• wobei d_M ein drehzahlbezogenes Reibmoment der Pumpe ist.

The torque dynamics on the motor-pump shaft are: $${\dot{\mathrm{\omega }}}_M=\frac{1}{J_{\mathit{eq}}}\left(M_M-\frac{V_g}{2\pi }{p}_L-d_M{\mathrm{\omega }}_M\right)$$

• where J _eq is the equivalent moment of inertia resulting from the moments of inertia J _pump of the pump and J _motor of the drive motor, J _eq = J _pump + J _motor ;
• where M _M is the drive torque;
• where V _g is the displacement of the pump, and
• where d _M is a speed-related friction torque of the pump.

Inserting equation (3) into equation (2) gives: $${\ddot{p}}_L=\frac{1}{C_h}\cdot \left(\frac{V_g}{2\pi }\cdot \frac{1}{J_{\mathit{eq}}}\left(M_M-\frac{V_g}{2\pi }{p}_L-d_M{\mathrm{\omega }}_M\right)-C_l\cdot {\dot{p}}_L\right)$$

verwendet werden. Dazu kann Gleichung (4) mit Hilfe der Gleichung (1) nach dem Antriebsmoment umgestellt werden, so dass sich das Vorsteuergesetz als inverses Streckenmodell ergibt: $$M_{\mathit{cmd}}^{\mathit{ff}}=\frac{J_{\mathit{eq}}2{\mathit{\pi C}}_h}{V_g}{\ddot{p}}_L^d+\frac{2\pi }{V_g}\left(J_{\mathit{eq}}{C}_l+d_M{C}_h\right){\dot{p}}_L^d+\left(\frac{d_{M}2{\mathit{\pi C}}_l}{V_g}+\frac{V_g}{2\pi }\right)p_L^d$$

This gives a second order differential equation that represents the system model. The inverse system model can now be used as a flatness-based feedforward control 250 of an additive torque for the trajectory sequence of a required pressure dynamics $\left[p_L,{\dot{p}}_L,{\ddot{p}}_L\right]$

can be used. For this purpose, equation (4) can be rearranged using equation (1) for the drive torque, so that the pre-control law results as an inverse system model: $$M_{\mathit{cmd}}^{\mathit{ff}}=\frac{J_{\mathit{eq}}2{\mathit{\pi C}}_h}{V_g}{\ddot{p}}_L^d+\frac{2\pi }{V_g}\left(J_{\mathit{eq}}{C}_l+d_M{C}_h\right){\dot{p}}_L^d+\left(\frac{d_{M}2{\mathit{\pi C}}_l}{V_g}+\frac{V_g}{2\pi }\right)p_L^d$$

bzw. nach der Drehzahl aufgelöst wird: $${\mathrm{n}}_{\mathit{cmd}}^{\mathit{ff}}=\frac{C_h}{V_g}{\dot{p}}_L^d+\frac{C_l}{V_g}\cdot p_L^d$$

Since this is a cascaded control loop, the superimposed speed control loop can optionally also be supplied with the pilot control signal from which the torque pilot control is derived. This can prevent the speed control loop from working against the torque pilot control. The corresponding pilot control signal can be calculated using equation (1) by dividing the equation according to the rotational frequency. ${\mathrm{\omega }}_{\mathit{cmd}}^{\mathit{ff}}$

or is resolved according to the speed: $${\mathrm{n}}_{\mathit{cmd}}^{\mathit{ff}}=\frac{C_h}{V_g}{\dot{p}}_L^d+\frac{C_l}{V_g}\cdot p_L^d$$

This feedforward value can then be added to the input of the speed control loop.

durch einen Sollwertfilter 260 bestimmt werden. Dazu kann beispielsweise ein Zustandsvariablen-Binomial-Filter zweiter Ordnung verwendet werden. Als Beispiel kann die Solltrajektorie nach der folgenden Differentialgleichung bestimmt werden: $${\ddot{p}}_L^d=\frac{1}{{\tau }^2}{p}_{\mathit{cmd}}-\frac{2}{\tau }{\dot{p}}_L^d-\frac{1}{{\tau }^2}{\ddot{p}}_L$$

wobei τ die Zeitkonstante des Sollwertfilters ist. Grundsätzlich können die Solldruckwerte aber auch anders bestimmt werden. Dabei ist p_cmd der kommandierte Solldruckwert.To implement the pilot control 250, the time derivatives of the target pressure $\left[p_L,{\dot{p}}_L,{\ddot{p}}_L\right]$

by a setpoint filter 260. For example, a second-order state variable binomial filter can be used for this purpose. As an example, the setpoint trajectory can be determined according to the following differential equation: $${\ddot{p}}_L^d=\frac{1}{{\tau }^2}{p}_{\mathit{cmd}}-\frac{2}{\tau }{\dot{p}}_L^d-\frac{1}{{\tau }^2}{\ddot{p}}_L$$

where τ is the time constant of the setpoint filter. In principle, the setpoint pressure values can also be determined differently. Here, p _cmd is the commanded setpoint pressure value.

The target trajectory obtained from equation (7) can now be inserted into the pre-control law, equation (5), and the pre-control torque M ^ff _cmd calculated with it can be added to the input of the current control loop 230. The target trajectories can also be inserted into the pre-control law (6) for the speed.

A simpler alternative to the pre-control for the drive controller is to compensate only for the load torque M _L , 240 , so that the pre-control value for the drive torque in this variant is M ^ff _cmd = ( V _g / 2π ) p _cmd from the commanded setpoint pressure value. A setpoint filter is not required in this case. The pre-control value is then added to the output value of the speed controller 220 or the input value of the current control loop 230 as in the previous embodiment.

By applying the model-based feedforward control according to the invention, the control behavior of the control system is improved, since the actuator (i.e. the drive motor of the pump) reacts immediately in a highly dynamic manner, even if no large control deviation occurs.

The follow-up behavior of the drive control (i.e. the current control loop) is also significantly improved by the feedforward control. This reduces the load on the speed and pressure controllers. At the same time, the feedforward control is insensitive to sensor noise.

A control system as described here can be used for all variable speed pumps, especially for pumps for controlling hydraulic units or hydraulic actuators. The described method with pilot control can also be used for the force control of servo-hydraulic actuators; in this case, a pressure is also controlled via a motor-pump unit, which can then be converted into a force value via the cylinder surfaces.

Claims (9)

1. Method for controlling a pump with a variable-speed drive, comprising: inputting a target pressure value (p_cmd ) into a cascaded controller structure, wherein the controller structure comprises a pressure control loop (210, 212, 214, 216),

   wherein the controlled system of the pressure control loop comprises a speed control loop (220, 222, 224), into which a target speed (n_cmd) is input as the manipulated variable of a pressure controller (210) of the pressure control loop;

   wherein the controlled system of the speed control loop comprises a drive control loop (230, 232), into which a target drive torque (M^R _cmd) is input as the manipulated variable of a speed controller (220) of the speed control loop;

   forming (250) a pilot control value (M ^{f f} _cmd ) for the drive torque as the result of a temporal differential equation of the target pressure; and

   adding the pilot control value to the target drive torque that is input into the controlled system of the speed control loop,

   wherein the temporal differential equation of the target pressure comprises an inverse system model of the controlled system of the cascaded controller structure.
2. Method according to Claim 1, wherein the inverse system model is formed on the basis of pressure dynamics in a load volume in which the pressure is controlled, and torque dynamics of the drive.
3. Method according to one of the preceding claims, wherein the pilot control value for the drive torque is determined by: $$M_{\mathit{cmd}}^{\mathit{ff}}=\frac{J_{\mathit{eq}}2{\mathit{\pi C}}_h}{V_g}{\ddot{p}}_L^d+\frac{2\pi }{V_g}\left(J_{\mathit{eq}}{C}_l+d_M{C}_h\right){\dot{p}}_L^d+\left(\frac{d_{M}2{\mathit{\pi C}}_l}{V_g}+\frac{V_g}{2\pi }\right)p_L^d$$

   

   where C_h is the hydraulic capacity, V_g is the displacement volume of the pump, C_l is the pressure-dependent leakage coefficient, J_eq is the equivalent moment of inertia,d_M is a speed-related friction torque of the pump, and p_L ^d is the target load pressure.
4. Method according to one of the preceding claims, furthermore comprising: forming a pilot control value (n^ff _cmd) for the speed, wherein the pilot control value for the speed is determined by: $${\mathrm{n}}_{\mathit{cmd}}^{\mathit{ff}}=\frac{C_h}{V_g}{\dot{p}}_L^d+\frac{C_l}{V_g}\cdot p_L^d$$

   

   where C_h is the hydraulic capacity, V_g is the displacement volume of the pump, C_l is the pressure-dependent leakage coefficient and p_L ^d is the target load pressure,

   and adding the pilot control value to the target speed output by the pressure controller (210) as the manipulated variable.
5. Method according to one of the preceding claims, furthermore comprising:

   inputting the target pressure value into a target value filter (260) which is configured to output a target trajectory for the target pressure value and the derivatives of the target pressure value, and

   inputting the target trajectory into the differential equation in order to form the pilot control value.
6. Method according to Claim 5, wherein the target value filter comprises a second-order state variable binomial filter.
7. Computing unit (40) comprising a processor configured to perform the method according to one of the preceding claims.
8. Computer program comprising instructions which, when the program is executed by a computer, cause the latter to perform the method according to Claims 1 to 6.
9. Computer-readable data carrier on which the computer program according to Claim 8 is stored.

Images