METHOD FOR CONTROLLING THE FEED RATE OF A FEED PUMP

Abstract

A method for controlling the feed rate of a feed pump, including a drive part having a drive motor and a hydraulic part having an intake opening, a discharge opening and a feed mechanism situated in between, a setpoint feed rate being predefined and the feed pump being triggered based on the setpoint feed rate, the temperature of the fluid and a pressure difference between the intake opening and the discharge opening of the hydraulic part of the feed pump.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

The present application claims priority to Application No. 10 2010 001 150.9, filed in the Federal Republic of Germany on Jan. 22, 2010, which is expressly incorporated herein in its entirety by reference thereto.

FIELD OF THE INVENTION

The present invention relates to a method for controlling the feed rate, i.e., the feed volume per unit of time, of a feed pump.

BACKGROUND INFORMATION

Feed pumps for fluids are widely used. In the automotive field, for example, feed pumps are used for feeding fuel to the engine. These feed pumps are usually designed as vane pumps or rotary vane pumps. In internal combustion engines in particular, it is important to accurately preselect the feed rate in order to obtain the desired injection pressure, the desired combustion performance and also low-emissions combustion. It is therefore conventional to regulate the feed rate, i.e., the setpoint feed rate is to be compared with the actual feed rate, and the feed pump is to be controlled according to a control deviation. This requires actual feed rate sensors, which makes regulation of feed rate relatively complex.

German Published Patent Application No. 10 2008 043 127 describes the regulation of the pump pressure. It is unnecessary to provide a pressure sensor if the actual pressure is ascertained by a so-called control observer. The feed pressure is determined on the basis of the motor current and the motor speed. No feed rate is determined.

It is therefore desirable to regulate the feed rate of a feed pump without measuring the actual feed rate.

SUMMARY

Example embodiments of the present invention include the provision of not measuring the actual feed rate of a feed pump but instead determining it based on the temperature of the fluid and the pressure difference of the intake opening and the discharge opening of the pump part or hydraulic part of the feed pump. Complex additional cost-intensive sensors may be omitted in this manner. The determination may be performed in practice on the basis of a characteristic map, for example, which extends over the temperature and pressure difference. The pressure difference to be taken into account includes the counter-pressure minus the inlet pressure.

For ascertaining the pressure difference, a drive torque of the drive motor, which is proportional to the pressure difference, may be used. A viscosity and temperature of the fluid are expediently also taken into account as these also have an influence on the pressure difference.

A relationship between drive torque M_ZP and pressure difference Δ_p may be written, for example, as:

$M_{\mathrm{ZP}}=\frac{V_{\mathrm{theo}}·\Delta p}{2·\pi {\eta }_{\mathrm{ZP}}}$

where:

V_theo represents the theoretical feed volume per revolution;

η_ZP represents the overall efficiency of the pump.

The drive torque may in turn be determined relatively easily based on known or easily determinable variables. The drive torque may be derived from the motor current, for example, if an engine characteristic map is known. This current measurement may be implemented inexpensively in the power electronics equipment.

A highly accurate quantitative regulation may be achieved even without performing a flow measurement by taking into account the pump geometry, for example, by performing a single measurement and storing additional measured values to correct the characteristic map.

Conventional feed pumps include a hydraulic part and a drive part flange-connected to the former. In addition, there are certain variants in which an internally or externally geared pump axially flange-connected to a motor shaft. The drive motors are arranged as DC variants as well as brushless DC variants. All these electric feed pumps are always arranged such that the feed part and the drive part are separate units. However, example embodiments of the present invention provide for the use of a pump of an integrated configuration, i.e., when the drive part and the hydraulic part form an inseparable unit. Examples of such a pump are described in U.S. Pat. No. 2,761,078 and European Published Patent Application No. 1 803 938. The use of such integrated pumps offers the advantage of a close spatial contact between the fluid and the electronics, so that a temperature sensor may be installed easily and without complex cabling, for example. If the control electronics or power electronics are connected directly to the feed medium, a temperature measurement cell may be accommodated here inexpensively and used for the regulation described herein.

In determining the pressure difference, a temperature-dependent leakage is expediently taken into account. This may be accomplished in particular from the following standpoints:

Based on a leakage cross section, such that positions 1 and 2 having pressures p₁ and p₂ are adjacent in the direction of the backpressure, and positions 3 and 4 having pressures p₃ and p₄ are adjacent in the intake pressure direction, it holds that:

p₁≈p₂ pump backpressure

p₄≈p₃ pump intake pressure

Since fluids are usually incompressible media, density ρ₁ is the same in positions i=1 through 4: ρ₁=ρ₂=ρ₃=ρ₄=ρ

Using a Bernoulli equation with a loss term, the influence of β_p on the leakage flow is estimated as follows:

$\begin{array}{cc}\frac{v_2^2}{2}+\frac{p_1}{\rho }=\frac{v_3^2}{2}+\frac{p_4}{\rho }+\frac{\Delta p_v}{\rho }+\rho {\int }_{s1}^{s2}\frac{\partial v}{\partial t}s\mathrm{assuming}\frac{\partial v}{\partial t}=0\mathrm{and}v_2=v_3,\mathrm{it}\mathrm{follows}\mathrm{that}& \left(1\right)\\ \frac{p_1}{\rho }=\frac{p_4}{\rho }+\frac{\Delta p_v}{\rho }\mathrm{or}& \left(2\right)\\ \frac{\Delta p_v}{\rho }=\frac{p_1}{\rho }-\frac{p_4}{\rho }\mathrm{or}\Delta p_v=p_1-p_4& \left(3\right)\end{array}$

The loss term for a constant cross section is

$\begin{array}{cc}\Delta p_v=\frac{\lambda ·l}{d}·\rho ·\frac{v^2}{2}& \left(4\right)\end{array}$

It thus follows that:

$\begin{array}{cc}v=\sqrt{\frac{2}{\rho }}·\Delta p·\frac{d}{\lambda ·l}\mathrm{where}& \left(5\right)\\ \lambda =\frac{\rho ·64}{\mathrm{Re}}=\frac{\rho ·64}{\frac{v·d}{\gamma }}& \left(6\right)\end{array}$

A friction moment estimate M_Reib of a radial friction bearing is given, for example, as:

$M_{\mathrm{Reib}}=\mu ·F_{\mathrm{Lager}}$ $\mathrm{where}$ $\mu ={\mu }_0·{}^{\left(-\frac{a·h_{\min }}{\mathrm{Rq}}\right)}$

where

a represents a constant; and

Rq represents a standard deviation of roughness Rq for contact pairing;

where:

$h_{\min }\approx \frac{B}{F_{\mathrm{Lager}}}·\frac{\eta ·D^3}{C}·\frac{\pi ·n}{60}{\left(1+\frac{\sqrt{2}·\left(1-{\gamma }^2\right)F}{B·E·h_{\min }}\right)}^{2/3}$

where

B represents a supporting width;

η represents a dynamic viscosity;

E represents a modulus of elasticity;

γ represents a transverse contraction number;

D represents a diameter;

n represents a rotational speed [1/min]

Thus a loss term which depends on rotational speed may be given.

Frictional resistance M of the rotor is formulated in a manner similar to that of a rotating disk:

$\begin{array}{c}M=2\int rF_r\\ =2\int {\mathrm{rc}}_F\frac{\rho v^2}{2}A\\ ={\int }_0^{d/2}{\mathrm{rc}}_F·\rho ·{\omega }^2r^2·2\pi rr\\ =\underset{\underset{C_M}{}}{\frac{4·\pi ·c_F}{5}}·\frac{\rho ·{\omega }^2}{2}·{\left(\frac{d}{2}\right)}^5\end{array}$

where for laminar flow and Re<3·10⁴ it holds that:

$C_M=\frac{2·\pi ·d}{s·\mathrm{Re}}$

where s represents an axial distance between the rotor and the housing;

A loss term as a function of rotational speed may in turn be given using ω=2πn.

The frictional resistance on the outer cylindrical surface is already taken into account in the bearing calculation.

Thus to determine the feed rate, a characteristic map as a function of temperature and motor current may be used. This is particularly simple because these parameters may be determined relatively accurately but nevertheless inexpensively and with little effort. A preferred relationship is obtained as follows:

$\stackrel{.}{V}=n·\left(V_{\mathrm{theo}}·K_1\right)-{\stackrel{.}{V}}_{\mathrm{Temp}}-{\stackrel{.}{V}}_{\Delta \rho }+\underset{\underset{\mathrm{drehzahlabh}\stackrel{..}{a}\mathrm{ngige}\mathrm{Verluste}}{}}{n^2·K_{10}+n^{1/2}·K_{11}+K_{12}}$

where drehzahlabhängige Verluste refers to rpm-dependent losses;

where

{dot over (V)} _Temp =T ² −K ₂ +T·K ₃ +T ^1/2 ·K ₄ +K ₅

and

{dot over (V)} _Δρ =I _Motor ² ·K ₆ +I _Motor ·K ₇ +I _Motor ^1/2 ·K ₈ +K ₉

where V_theo denotes the theoretical feed volume per revolution of the pump.

A computation unit, for example, a control unit of a motor vehicle, is equipped, in particular as far as programming is concerned, to perform a method described herein.

It should be understood that the features mentioned above and those yet to be explained below may be used not only in the particular combination given but also in other combinations or alone.

Example embodiments of the present invention are illustrated schematically in the Figures and are described below in more detail with reference to the Figures.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 schematically shows a feed pump, which is suitable in particular for performing a method according to an example embodiment of the present invention.

FIG. 2 shows in a diagram the relationship between feed rate and rotational speed as a function of the pressure difference at a constant fluid temperature.

FIG. 3 shows in a diagram the relationship between feed rate and rotational speed as a function of the inlet pressure at a constant pressure difference and a constant fluid temperature.

FIG. 4 shows in a diagram the relationship between feed rate and rotational speed as a function of the fluid temperature at a constant pressure difference.

DETAILED DESCRIPTION

FIG. 1 shows an electric feed pump of an integrated configuration, in which the drive part and the hydraulic part or feed part form an inseparable unit 120, which is diagramed schematically and labeled as 100 as a whole. In the present example, the integrated configuration is achievable by the fact that the rotor of the drive motor at the same time also forms the moving pump element of the hydraulic part, as described in European Published Patent Application No. 1 803 938, for example, which is expressly incorporated herein in its entirety by reference thereto. Hydraulic part 120 thus includes drive motor 121, which also acts as feed mechanism 121, drawing in a fluid, fuel in particular, through an intake opening 122 and discharging it through a discharge opening 123. There is therefore a pressure difference Δ_p between intake opening 122 and discharge opening 123.

The pump also includes an electronic part 110. A regulating module 111 and a power module 112 are provided in electronic part 110. Regulating module 111 receives a setpoint feed rate {dot over (V)}_setpoint from a motor control unit 150 and determines therefrom a setpoint rotational speed n_setpoint for the drive motor, which is transmitted to power module 112. Power module 112 may have, for example, an inverter for operation of the drive motor. Motor current I_motor is determined in power module 112 and transmitted to regulating module 111.

Based on the integrated configuration of pump 100, there is a close spatial contact between electronic part 110 and drive and hydraulic part 120, so that fluid temperature T_actual-fluid is easily measurable by a measurement performed by a sensor 113 provided within electronic part 110.

The feed rate of feed pump 110 may be controlled on the basis of measured motor current I_motor and measured fluid temperature T_actual-fluid. A characteristic map as a function of temperature T_actual-fluid and motor current I_motor is used in regulating module 111 according to the equation:

${\stackrel{.}{V}}_{\mathrm{Soll}}=n_{\mathrm{Soll}}·\left(V_{\mathrm{theo}}·K_1\right)--\left(T_{1\mathrm{st}-\mathrm{Fluid}}^2·K_2+T_{1\mathrm{st}-\mathrm{Fluid}}·K_3+T_{1\mathrm{st}-\mathrm{Fluid}}^{\frac{1}{2}}·K_4+K_5\right)--\left(I_{\mathrm{Motor}}^2·K_6+I_{\mathrm{Motor}}·K_7+I_{\mathrm{Motor}}^{\frac{1}{2}}·K_8+K_9\right)++n_{\mathrm{Soll}}^2·K_{10}+n_{\mathrm{Soll}}^{\frac{1}{2}}·K_{11}+K_{12}$

where

Soll denotes a setpoint, Ist denotes actual; and

V_theo denotes the theoretical feed volume per revolution of the pump and is usually given on the data sheet. Characteristic map constants K₁-K₁₂ are ascertained empirically. To do so, a sufficient number of measured points [{dot over (V)}, n, T, I] is preferably measured and evaluated using known fitting methods (e.g., least squares fitting).

Based on the characteristic map, setpoint rotational speed n_setpoint is determined and transmitted to power module 112. To regulate the feed rate, actual rotational speed n_actual of drive motor 121 is regulated at setpoint rotational speed n_setpoint. A known rotational speed regulation may be used to do so.

Alternatively it is possible to use actual rotational speed n_actual together with measured motor current I_motor and measured fluid temperature T_actual-fluid to determine the actual feed rate via the characteristic map and to regulate the actual feed rate at the setpoint feed rate, again with the setpoint rotational speed being regulated.

Various relationships are explained purely qualitatively below with reference to FIGS. 2 to 4 merely for the purpose of illustration.

FIG. 2 shows a diagram 200, illustrating the relationship between feed rate {dot over (V)} on the ordinate as a function of rotational speed n on the abscissa at a constant temperature. Three feed rate curves 210, 220 and 230 are shown in diagram 200, each curve being characterized by a different pressure difference Δp between the intake opening and the discharge opening. Thus a first pressure difference Δp₁ is assigned to feed rate curve 210, a second pressure difference Δp₂ is assigned to feed rate curve 200, and a third pressure difference Δp₃ is assigned to feed rate curve 230, the pressure difference increasing, so that it holds that: Δp₁<Δp₂<Δp₃. The feed volume/rotational speed characteristic curve is shifted to the right with an increase in pressure difference Δp because internal leakage increases. In other words, a higher rotational speed is also necessary to supply a certain feed rate at a higher pressure difference.

Each of the three feed rate curves includes a first essentially linearly increasing range A and a following curved range B. The slope in range A is constant and depends essentially only on the geometric displacement volume of the pump. The feed volume curve flattens out in range B due in particular to partial cavitation phenomena on the intake end, caused in particular by high local flow velocities.

FIG. 3 shows in a diagram 300 the influence of pressure at the intake opening, i.e., inlet pressure p_inlet on the feed volume/rotational speed characteristic curve. Diagram 300 shows three characteristic curves 310, 320 and 330 at a constant pressure difference Δp, these characteristics differing in their inlet pressure. Characteristic curve 310 is defined by inlet pressure p_inlet1 characteristic curve 320 is defined by inlet pressure p_inlet2 and characteristic curve 330 is defined by p_inlet3 where the following holds: p_inlet1>p_inlet2>p_inlet3.

A variation in the inlet pressure produces a shift in ranges A and B such that the stable, i.e., linear operating range A becomes smaller with a drop in inlet pressure. In other words, the stable range is smaller the higher the inlet pressure p_inlet. It is thus advisable to provide a limit in the pump specification to avoid operating in range B.

FIG. 4 shows the influence of the fluid temperature on the feed volume/rotational speed characteristic curve in a diagram 400. Three characteristic curves 410, 420 and 430 are shown in diagram 400, a different fluid temperature T₁, T₂ and T₃ being assigned to each diagram, where it holds that T₁<T₂<T₃. The characteristic curves are shifted to the right with an increase in fluid temperature because the temperature influences the viscosity of the fluid and thus affects the leakage. Furthermore, the pump components expand, so that different materials are usually used for different components and thus there is different thermal expansion. For example, the housing is often made of aluminum, whereas the feed mechanism often has steel elements, which thus have a lower thermal expansion than the housing. As a result, the leakage increases with an increase in temperature. On the whole, it is apparent that a higher rotational speed is also needed at a higher fluid temperature to supply a certain feed rate.

Claims

1. A method for controlling a feed rate of a feed pump, including a drive part having a drive motor and a hydraulic part having an intake opening, a discharge opening and a feed mechanism situated in between, comprising: specifying a setpoint feed rate; andtriggering the feed pump based on the setpoint feed rate, a temperature of the fluid, and a pressure difference between the intake opening and the discharge opening of the hydraulic part of the feed pump.

2. The method according to claim 1, further comprising determining the pressure difference based on a drive torque of the drive motor.

3. The method according to claim 2, further comprising determining the drive torque based on a motor current flowing through the drive motor.

4. The method according to claim 3, further comprising determining at least one of (a) an actual feed rate and (b) a setpoint rotational speed using characteristic map based on the temperature and the motor current.

5. The method according to claim 1, further comprising determining the pressure difference based on a viscosity and the temperature of the fluid.

6. The method according to claim 1, further comprising determining the pressure difference taking into account a temperature-dependent leakage.

7. The method according to claim 1, further comprising triggering a setpoint rotational speed of the drive motor.

8. The method according to claim 1, wherein the feed pump is arranged as a pump of an integrated configuration, in which the drive part and the hydraulic part form an inseparable unit.

9. The method according to claim 8, further comprising determining the temperature by measuring in an electronic part of the feed pump.

10. The method according to claim 9, further comprising determining a motor current in a power module of the electronic part.

11. A computation unit configured to perform a method for controlling a feed rate of a feed pump, including a drive part having a drive motor and a hydraulic part having an intake opening, a discharge opening and a feed mechanism situated in between, the method including: specifying a setpoint feed rate; andtriggering the feed pump based on the setpoint feed rate, a temperature of the fluid, and a pressure difference between the intake opening and the discharge opening of the hydraulic part of the feed pump.