WO1998024008A1

WO1998024008A1 - Control circuit with a digital controller regulating the input current of an electronic actuator by means of pulse width modulation

Info

Publication number: WO1998024008A1
Application number: PCT/EP1997/006079
Authority: WO
Inventors: Alwin Becher; Michael Genzel; Rodolfo Möller; Ari Ojamies; Armin Tonn; Horst Wild
Original assignee: Temic Telefunken Microelectronic Gmbh
Priority date: 1996-11-22
Filing date: 1997-11-04
Publication date: 1998-06-04

Abstract

The invention relates to a control circuit to regulate the input current of an electrical actor, avoiding to a large extent switching hysteresis phenomena at set-point step change.

Description

Control loop consisting of a digital controller and a controlled system for controlling the input current of an electrical actuator using the

Pulse width modulation

In the further development of assemblies in the automotive sector, such as a clutch, regulations are increasingly being used instead of controls. Regulations have a closed action cycle consisting of a controller and a controlled system with feedback. The controller is given a setpoint of a controlled variable, which first transfers this to the controlled system, within the controlled system the actual value of the controlled variable is determined and fed back to the controller of the control loop via the feedback. Deviations between the actual value and the setpoint of the controlled variable are automatically corrected in the controller by changing a manipulated variable. The controlled variable is understood to mean the physical variable to be controlled, such as the current or the voltage. The actual value is the actual value of the controlled variable, which is determined (measured) once per control cycle. The setpoint is the value which is specified by a setpoint generator, for example a clutch pedal, and which the controlled variable is to assume exactly. The control variable is controlled by changing the manipulated variable. The sum of the changes in the controlled variable causes the control deviation of the control loop. The control deviation is the negative (* - 1) control deviation.

Influences on the control loop that cause an undesired change in the controlled variable are referred to as disturbance variables. Such disturbance variables are, for example, temperature or frequency dependencies of the components of the controlled system and / or the controller. A pilot control element can be inserted into the controller to compensate for disturbance variables. The disturbance variables are compensated on the basis of sensor data or characteristic values. In addition, variables entering a pilot control can be changed on the basis of mathematical functions, provided that this is advantageous for the control.

According to the requirements placed on the control, different control devices are used in the controller, which have a different behavior when a control deviation occurs, in the literature the proportional (P), integral (l) and differential (D) control are used, and the combinations PI, PD and PID control are known, and control loops can also be constructed by branching out from several different control devices and pilot elements. The branches of the controller are usually combined in an addition element or a multiplication element. Controllers can be implemented in analog or digital circuit technology, the controlled system is always implemented in analog circuit technology. In the case of control in digital circuit technology, if the setpoints of the controlled variable are not available as digital values, an analog-to-digital converter is arranged in front of the controller and a digital-analog converter is arranged between the controller and the controlled system An analog-digital converter is arranged between the controller and the controlled system. The digital-to-analog conversion can advantageously be implemented by pulse width modulation.

Control devices have a number of technical problems that differ in the various control devices. These are e.g. the tendency to oscillate, which can be suppressed by a suitable attenuator, or the formation of switching hysteresis.

A switching hysteresis is understood to mean the behavior of a control in the event of a jump in the setpoint between two different setpoints to cause an error in the manipulated variable, which makes it difficult to stabilize the controlled variable.

In many control loops, the temperature and frequency dependence of the resistance of the controlled system must be taken into account when controlling a controlled variable become. This usually takes place within the pilot control of a controller, as shown in FIG. 3. The temperature and frequency dependent resistance estimates are multiplied by the soli value of the controlled variable. The the digital-to-analog converter supplied control signals s consists of the portion S1 of the feedforward control is added to the portion S2 of the control device together, in the Figure 4, the behavior of the regulator at a setpoint step of l _soMιA on l _S0 _"ιB of the 3 shown controller for regulating the input current of an electrical actuator, in the diagram the control signal s is plotted against the setpoint l _ste n of the controlled variable l.

While the setpoint ι _soπιA is present, the control signal S _{A is} composed of the components S1 _A and S2 _A. In the control cycle immediately after the setpoint jump to setpoint l _S0) | _ιB , the control signal S is composed of the components S1 _B and S2 _B and a dynamic control _{value error} in the amount of | U2 _A - U2 _B I together, which leads to the formation of the switching hysteresis. The dynamic manipulated value error is caused by the component S2 of the control device, which in the first control cycle after the setpoint jump contributes the component S2 _A to the control signal s instead of the component S2 _B. Only after the end of a transient process does the control device transfer the correct portion S2 _B and the controller the correct control signal S _B to the controlled system.

The invention has for its object to provide a method for control according to the preamble of claim 1, in which the formation of a switching hysteresis is avoided. This object is achieved by the features in the characterizing part of patent claim 1. In the control circuit according to the invention, consisting of a digital controller and a controlled system, a digital controller with the following structure is used to control the controlled variable “input current” of an electric actuator by means of the manipulated variable “control voltage” using pulse width modulation: fed to an analog-digital converter. The output signals of the analog-digital converter are fed to a pilot control and an integrator control. In the pilot control, the nominal value of the input current is divided by a measured supply voltage. Before the integrator control is a first adder, with which the feedback of the controlled system is connected, and on which the negated actual values of the controlled variable are fed back to the digital controller. The output signals of the integrator control and the pilot control are combined in a multiplier. A second adder is arranged between this multiplier and the integrator control, which is connected to a map control in which temperature and frequency-dependent resistance values of the controlled system are stored. These resistance values were determined experimentally and should correspond as closely as possible to the resistance of the controlled system.

The output signal of the multiplier, which represents a dimensionless control signal standardized to 1, is fed to a digital-to-analog converter which works according to the technique of pulse width modulation. The mean value of the control voltage of the electrical actuator results from the control signal multiplied by the supply voltage.

The controlled system of the control loop has the following components: the electrical actuator, the input current of which is regulated, and a measuring device with which the actual value of the input current is measured. The result of the measurement is fed to a second analog-digital converter, the output signal of which is fed to the first adder. In a further development of the invention, it is provided that the electric actuator is a proportional valve for controlling a hydraulic system which causes the opening and closing of a motor vehicle clutch. The method according to the invention is described below using the exemplary embodiment of a control for setting the clutch torque of a motor vehicle clutch in connection with drawings (FIGS. 1 and 2). Show it:

1 shows the signal flow diagram of a control loop according to the inventive method.

Figure 2 is a diagram showing the basic courses of the manipulated variable in a setpoint step. The setting of the clutch torque of a hydraulic multi-plate wet clutch of a motor vehicle must be done both dynamically and statically with minimal errors. For this purpose, the input current I of a proportional valve PV is regulated within a control loop, which adjusts the level of the hydraulic pressure acting on a clutch. The setpoint generator of the control is the clutch pedal of the motor vehicle. The manipulated variable of the controlled variable input current l is the manipulated voltage U _ste n. The digital controller RE of the control loop is implemented in digital circuit technology, while the controlled system RS of the control loop with the proportional _valve PV and the device R _meas for determining the actual _value l _{ist of} the input current l is designed in analog circuit technology. Within the digital controller RE, the control voltage u _ste n is referred to as the control signal s. The digital-to-analog conversion is implemented using PWM technology. FIG. 1 shows a signal flow diagram of such a control consisting of a digital controller RE and the controlled system RS. The setpoint I _{JOII of} the input current I predetermined by the clutch pedal is fed to a first analog-digital _{converter AD-} in the digital controller RE. The output signals of the first analog-digital converter AD ₁ are fed to a pre-control vs and to the control device.

The control device is designed as an integrator control IR, which has no permanent control differences. The rate of change of the control signal s is proportional to the control difference, within the feedforward control vs the setpoint l _so n of the input current l is divided by the measured value UB _measuring the on-board voltage UB of the motor vehicle, immediately before the integrator control IR there is a first adder in the digital controller RE A _1f to which the feedback of the controlled system S is connected and to which the negated actual values i | _{St of} the input current I is fed back from the controlled system RS to the digital controller RE. The output signals of the integrator control IR and the pilot control vs are combined in a multiplier M.

Arranged between this multiplier M and the integrator control IR is a second adder A ₂ , which is connected to a map control KS, in which resistance values R (T, f) of the controlled system RS are stored as a function of temperature and frequency. For this purpose, the map control KS is connected to a temperature sensor TS, which measures the temperature T of the controlled system Determine RS.

The output signal of the multiplier M is the digital actuating signal s, which represents a dimensionless quantity standardized to 1. The subsequent digital-to-analog converter DA, which works according to the technology of pulse width modulation PWM, converts the digital control signal S into the analog control voltage u _ste ιι. The mean value of the analog control voltage u _ste ιι results from the control signals s (duty cycle) multiplied by the on-board voltage ÜB of the motor vehicle.

In the case of a pulse width modulator with a variable modulator frequency f, the resistance values R (T, f) are stored as a two-dimensional map in the map control. When using a pulse width modulator with a constant modulator frequency f, a characteristic curve R (T) is sufficient.

The control path RS of the control loop has as components the proportional valve PV, the input current I of which is regulated, and a measuring device, which is shown in FIG. 1 only by the measuring resistor R _meSs , and with which the actual value l, _{st of} the input current I is determined . The result of the measurement is fed to a second analog-digital converter AD ₂ , which is connected to the first adder A-. The on-board voltage UB of the motor vehicle is also determined by means of the measuring device and the measured value UB _{measured is fed to} the pilot control vs the digital regulator RE.

The controlled system consisting of proportional valve PV and measuring resistor R _{meas /} has a resistance _behavior which is dependent on the temperature T and on the frequency f of the pulse width modulator.

With this construction of the digital controller RE, the dimensionless manipulated variable s is composed of the portion of the pilot control s (i _S0 "/ UB _me β) multiplied by the portion of the integrator control IR and by the portion of the map control KS. The output signal of the map control KS is added to the regulated part of the resistor R (T, f) of the controlled system RS which has been corrected by the integrator control IR. 2 shows the behavior of the control according to the invention at a setpoint step change of the input current l from the target value ι _S0, _ιA ι to the target value l _should, s shown the reproduced in the Figure 1 control loop in the diagram is the control signal s to the desired value! _« ,,, of the input current I plotted. While the setpoint ι _soHιA is present, the control signal S _A sets in accordance with:

R, _{τ f} + portion ,. fg atorregelemπchtung

¹ .soll * - (T, f) 'Inle .A ÜB m. eat together. In the control cycle immediately after the setpoint jump from the setpoint l _soMιA to the setpoint l _{soM B} , the control signal S _B sets in accordance with:

ς, _ (T) ⁺ share iniegnuorregelemnc iung

^Ss _me ß together.

The required manipulated variable S _B is reached directly in the first control cycle after the setpoint step, a dynamic manipulated variable error does not occur. A switching hysteresis, which would prevent the input current I from settling out in a stable manner, cannot develop in the control according to the invention. This results in smooth transitions of the clutch torque, which are a basic prerequisite for achieving good driving comfort with a regulated, hydraulic multi-plate wet clutch.

Claims

claims

1. Control loop consisting of digital controller (RE) and controlled system (RS) for controlling the input current (l) of an electrical actuator (PV) using the

Pulse width modulation (PWM), the digital controller (RE) having the following components

The parallel connection of a pilot control (VS) and an integrator control (IR), the inputs of which are connected to the output of a first analog-digital converter (AD.,), _To which the setpoint (l _soM ) of the input current (l) is fed,

• A first adder (A.,) arranged in series between the first analog-digital converter (AD.,) And the integrator control (IR), to which the negated actual value (I _ist ) of the input current (I) from the controlled system (RS) is fed,

A multiplier (M), to which the output signals of the pilot control (VS) and the integrator control (IR) are fed, and which forms a digital control signal (S) as the output signal,

• A second adder (A ₂ ) arranged between the multiplier (M) and the integrator control (IR), to which the output signals of a map control (KS) are fed, in the temperature and frequency-dependent resistance values (R (T, f)) of the controlled system (RS) are filed,

• and one arranged between the multiplier (M) and the controlled system (RS), which uses pulse width modulation (PWM)

Digital-to-analog converter (DA), the output signal of which is an analog control voltage U _continuous , and the mean value of which is the analog control signal (S) multiplied by the supply voltage (ÜB) of the controlled system (RS), and the controlled system (RS) has the following components

• the electrical actuator (PV),

A measuring device (R _meas ) for determining the actual _value (l, _st ) of the input current (l) of the electrical actuator (PV),

• and a second analog-to-digital converter (AD ₂ ), which is connected to the first adder (A.,).

2. Control circuit claim 1, characterized in that the electrical actuator (PV) is an electromagnetic proportional valve.