DE102013207498B4 - Procedure for implementing a regulation - Google Patents

Abstract

Method for implementing a control in a control loop whose control system (72) has a PT1 behavior, using a controller (62) which outputs a manipulated variable which has a duty cycle (68) of a PWM signal (20) with a fixed switching frequency represents, whereby a state machine (80) serves as the controller (62), which assumes different states depending on a determined control deviation, which is determined by comparing an actual value (54) with a target value (12, 56), and depending on the assumed state outputs the manipulated variable.

Description

The invention relates to a method for implementing a control, in particular to a control in which a PWM signal is output as the control variable, the duty cycle of which represents the manipulated variable of the control, and an arrangement for carrying out the method presented.

State of the art

The method presented is used to implement a control in a control loop with PT1 behavior of the controlled system. In particular, the method is used to determine a manipulated variable for controlling inductors. A pulse width modulated (PWM) current signal is typically used to control inductors.

A PT1 behavior shows a PT1 element that has a proportional transfer behavior with a first-order delay.

Pulse width modulation is a type of modulation in which a quantity, for example an electrical current, changes between two values, switched on state and switched off state. Here, the duty cycle or the duty cycle of a square-wave pulse and thus the width or width of the pulses forming it are modulated at a constant frequency. The duty cycle, English duty cycle, indicates the ratio of the length of the switched-on state to the period length of a square-wave signal. If a technical unit is controlled with a PWM signal, the duty cycle can be specified as a manipulated variable by a controller to regulate the operation of this technical unit.

For current regulation of inductors, such as B. DC motors, stepper motors and electromagnetic valves, which are controlled by ICs such as H-bridges or power amplifier modules, are now in most cases used in circuits integrated 2-point controllers. For this purpose, the exponential behavior of the current over time is used to set the desired current strength on average over time. The disadvantage is that with this 2-point control the switching frequency depends on the current setpoint, the switching hysteresis and the disturbance variable.

From the DE 29 36 627 A1 A method for implementing a regulation in a control loop is already known.

Disclosure of the invention

Against this background, a method with the features of claim 1 and an arrangement according to claim 7 are presented. Explanations result from the dependent claims and the description.

The method presented is generally suitable for control loops with PT1 behavior of the controlled system. In particular, the method can be used in connection with the integration of the control device into ICs (integrated circuits) for controlling inductors. Furthermore, the method enables current control with a constant switching frequency without exceeding the complexity of a 2-point control.

In the process, a state machine is used as a controller, which enables very precise current control over five states, for example. This state machine can be integrated as a simple and cost-effective logic part into an IC, such as an integrated H-bridge or an output stage module.

The controller typically includes a comparison point between the controlled variable and the reference variable, which is referred to as a setpoint/actual value comparison. A normalization of the control deviation to the reference variable, a limitation of the standardized control deviation, the state machine with time-delayed feedback and a limitation of the manipulated variable. Further components such as the adjusting and measuring device are part of the control circuit, whereby the adjusting device can be in the form of a semiconductor switch, for example as a MOSFET. The measuring device must make the controlled variable, i.e. the current, available to the reference junction in a suitable form.

The controller, i.e. the state machine or state machine, English. state machine, processed, as in 2 The following variables are described in detail: the standardized control deviation, adjustable constants such as the duty cycle increment ΔDC, the gain and the time-delayed feedback of the manipulated variable DC _i-1 . The controller outputs the duty cycle with which the actuating device is switched. The constant switching frequency is fed directly to the control device and is independent of the states of the state machine.

The states of the state machine affect the duty cycle depending on the control deviation. In the embodiment, the states are achieved via the first to fifth state conditions. States 2, 3 and 4, 5 can only be achieved via the first state condition.

Condition 1) The control deviation is very small, i.e. H. in this example +5% of the setpoint. The duty cycle remains unchanged.

Condition 2) The control deviation is between +5% and +50%. The duty cycle becomes The switching frequency is reduced by the constant ΔDC for each new cycle until state 1 is reached.

Condition 3) The control deviation is more than 50%. The duty cycle is reduced by the factor of the two constants ΔDC x gain for each new cycle of the switching frequency until state 2) is reached.

Condition 4) The control deviation is between -5% and -50%. The duty cycle is increased by the constant ΔDC for each new cycle of the switching frequency until state 1) is reached.

Condition 5) The control deviation is less than -50%. The duty cycle is increased for each new cycle of the switching frequency by the factor of the two constants ΔDC x gain until state 4) is reached.

The sampling time of the actual value plays an important role; this will be discussed below 1 referred. This is chosen, for example, so that it records the measured value delayed by half the reciprocal of the switching frequency. The sampling frequency corresponds to the switching frequency. The measured value remains until the next switching event.

Further advantages and refinements of the invention result 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 alone, without departing from the scope of the present invention.

Brief description of the drawings

• 1 shows sampling times in a graph
• 2 shows an embodiment of the presented arrangement in a block diagram.
• 3 shows a flowchart of the state machine described.
• 4 shows a step response of a controlled system in graphs.

Embodiments of the invention

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

1 shows progressions of sizes and sampling times in a graph. The course of a current 10 and a setpoint 12 are plotted over time. At a first sampling time 14, a second sampling time 16 and a third sampling time 18, the value of the current is read in and converted into a duty cycle of a PWM signal 20. This PWM signal 20 has a fixed switching frequency and therefore a constant period 22. The duty cycle is represented by the time periods DC ₁ 24, DC ₂ 26 and DC ₃ 28. The duty cycle therefore increases with the current 10.

In 2 shows an embodiment of the arrangement for carrying out the method, which is designated overall by reference number 50. The illustration shows a comparison unit 52, in which a comparison of the actual value 54 with the setpoint value 56 of the current takes place. This results in a control deviation 57. The illustration also shows a normalization unit 58, which normalizes the control deviation 57 to the reference variable, the actual value 54. This results in the standardized control deviation l_norm 59. Furthermore, a limiting unit 60, a controller 62, a measuring device 64, a further limiting unit 66, which outputs the actual duty cycle 68, an adjusting device 70, a controlled system 72, in this case an inductance, and a disturbance variable 74, which acts on the controlled system 72, is reproduced. Disturbance variable 74 can be, for example, a change in resistance of the inductance over temperature and time.

The controller 62 includes a state machine 80 and a delay element 82 for a time-delayed feedback.

3 shows a possible flowchart of the state machine. The illustration shows state 1 DC = DC _i-1 (reference number 100), state 2 DC = DC _i-1 - ΔDC (reference number 102), state 3 DC = DC _i-1 - gain x ΔDC (reference number 104), state 4 DC = DC _i-1 + ΔDC (reference number 106) and state 5 DC = DC _i-1 + gain x ΔDC (reference number 108).

I_norm corresponds to the standardized control deviation. At the beginning 110 it is determined: $$\text{I}\_\text{standard}=\left(\text{I}\_\text{is}-\text{I}\_\text{should}\right)/\text{I}\_\text{should}$$

In a first step 112, the first condition is checked: $$-0.05>=\text{I}\_\text{standard}<0.05$$

If this is fulfilled, state 1 100 is assumed. Starting from state 1 100, the first state condition 112 is checked again. If this is not fulfilled, the second condition 114 is checked, which is: $$0.05<\text{I}\_\text{standard}<0.05$$

If this is fulfilled, state 2 102 is assumed; if this is not fulfilled, a third state condition 116 is checked, which is: $$\text{l}\_\text{standard}>=0.5$$

If this is fulfilled, state 3 104 is assumed. If this is not fulfilled, the second condition 112 is checked again. If this is not met either, a fourth condition 118 is checked. This is: $$-0.05>\text{I}\_\text{standard}>-0.5$$

If this is fulfilled, state 4 106 is assumed. If this is not fulfilled, a fifth condition 120 is checked. This is: $$\text{I}\_\text{standard}<=-0.5$$

If this is fulfilled, state 5 108 is assumed. If this is not fulfilled, the fourth condition 118 is checked again. If this is not fulfilled, a jump back is made to check the first condition 112.

I_norm as a standardized control deviation represents a measure of this control deviation. Depending on the control deviation, different states are assumed in which the duty cycle has different values for DC, see here 1 , be determined. The controller therefore assumes different, discrete states depending on the control deviation, in which the duty cycle is set differently depending on the control deviation or is set as a manipulated variable.

In 4 different courses are shown. The illustration shows a step response 150 without a controller with disturbance variable input, the course of a disturbance variable 152, the course of a step response with a controller and with disturbance variable input, namely a setpoint 154 and an actual value 155, as well as the course of the manipulated variable 156, the duty cycle.

It can be seen how the influence of the disturbance variable 152 can be compensated for with the controller.

The presented method can be integrated into the new development of ICs in which actuators and measuring devices are provided for controlling inductors. These are, for example, H-bridges for operating DC and stepper motors, highly integrated power stage components and multifunctional ICs, which have logic and sensor as well as power elements.

Claims (9)

Method for implementing a control in a control loop whose control system (72) has a PT1 behavior, using a controller (62) which outputs a manipulated variable which has a duty cycle (68) of a PWM signal (20) with a fixed switching frequency represents, whereby a state machine (80) serves as the controller (62), which assumes different states depending on a determined control deviation, which is determined by comparing an actual value (54) with a target value (12, 56), and depending on the assumed state outputs the manipulated variable. Procedure according to Claim 1 , in which the state machine (80) can assume five different states (100, 102, 104, 106, 108). Procedure according to Claim 1 or 2 , which is used to regulate the current of inductors. Procedure according to one of the Claims 1 until 3 , in which a standardized control deviation is taken into account. Procedure according to one of the Claims 1 until 4 , in which the duty cycle (68) is output as a manipulated variable with a time delay. Procedure according to one of the Claims 1 until 5 , in which the sampling point (14, 16, 18) of the actual value (54) is delayed by half the reciprocal of the switching frequency. Arrangement for implementing a regulation in a control loop, in particular for carrying out a method according to one of the Claims 1 until 6 , with a controller (62), a comparison unit (52), an adjusting device (70) and a measuring device (64), the controller (62) comprising a state machine (80). Arrangement according to Claim 7 , which is integrated into an integrated circuit. Arrangement according to Claim 7 or 8th , in which the controller (62) additionally includes a delay element (82).

Images