JPH07325019A - Vehicle speed controlling apparatus - Google Patents

Abstract

PURPOSE:To keep vehicle speed deflection zero even during acceleration and at the same time carry out excellent follow-up control without causing overshoot. CONSTITUTION:Accelerator open degree thetaA is directly computed and sent out from a vehicle speed order Vs by using a reverse function MG(S)<-1> of a transmission function of a vehicle for a feed forward transmission function, the acceleration degree of the model vehicle is computed by differentiation by feed forward control circuits 1-3 which make the vehicle speed V of the vehicle 3 be the same to the vehicle speed order Vs by operation of the accelerator 2, a vehicle control model transmission function circuit 4 which has the same transmission function MG(S)<-1>, Go(S), Mg(S) as those of the control circuits and to which the order VS is sent, and a D computation circuit 5 while using the output of the circuit 4 as speed order VS' of compensation control, the deviation between the order VS' and the vehicle speed V is computed by P computation circuit 6, the vehicle speed V is differentiated by d D computation circuit 7, the output signals of the circuits 5, 6 are added with the illustrated polarity, and the output computed from the obtained output by a PI computation circuit is sent back as an accelerator pedal stroke correcting order thetaC, to compensate the error difference due to the feed forward control.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle speed control device for driving a vehicle on a vehicle testing device such as a chassis dynamometer.

[0002]

2. Description of the Related Art Conventionally, a speed control system for driving a vehicle on a chassis dynamometer has a P control as shown in FIG.
There is an ID method or a speed control method with an acceleration minor as shown in FIG. 7 and 8, G _C (S): Transfer function of PID calculator G _O (S): Transfer function of stroke control of accelerator actuator MG (S): Transfer function of vehicle G _C1 (S): P calculation Transfer function of the calculator G _C2 (S): Transfer function of the PI calculator G _C3 (S): Transfer function of the D calculator In the PID system shown in FIG. 7, the deviation between the vehicle speed command V _S and the vehicle speed V is calculated by the PID calculator 11 to obtain the accelerator pedal stroke command θ _S , and the accelerator pedal is operated to operate the vehicle speed V
Is controlled so as to match the vehicle speed command V _S. In the system shown in FIG. 8, the deviation between the vehicle speed command V _S and the vehicle speed V is calculated by the P calculator 1
2 and the difference between the output and the output of the vehicle speed V calculated by the D calculator 13 is calculated by the PI calculator 14 to obtain the accelerator pedal stroke command θ _S to operate the accelerator pedal,
The vehicle speed V is controlled so as to coincide with the command vehicle speed V _S.

An example of the transfer function MG (S) of the vehicle is shown in FIG. In FIG. 2, G ₁ (S): Non-linear transfer function from accelerator pedal stroke to throttle valve opening G ₂ (S): Transfer function from throttle valve opening θ _B to intake pressure G ₃ which is one of engine characteristics (S): One of engine characteristics, a transfer function from intake pressure to engine output torque i _Tn : Transmission gear ratio i _D : Differential gear ratio R: Tire radius W: Vehicle weight G ₄ (S): Vehicle running Indicates resistance.

(1) The transfer function G ₁ (S) is a transfer function from the accelerator pedal stroke θ _A to the throttle valve opening θ _B , in which the pedal play and the nonlinear characteristic of θ _A -θ _B are included. included.

(2) The transfer function G ₂ (S) is a transfer function from the throttle valve opening θ _B to the intake pressure Pb, which has a non-linear characteristic with the engine speed as a parameter, and is different for each engine type. The characteristics also differ depending on the presence or absence of a turbocharger.

(3) The transfer function G ₃ (S) is the intake pressure Pb
To engine output torque τe, which also has a characteristic that is almost linear with the engine speed as a parameter. The characteristics differ depending on the type of engine.

(4) The transfer function from the engine output torque τe to the axle shaft (tire shaft) torque τa is i _Tn × i
_Represented by _D. (However, excluding mecha loss). Here, i _Tn is a transmission ratio, and n = 1 to 5 is first speed to 5
Correspond to speed. i _D is a differential gear ratio (differential ratio), which varies from vehicle to vehicle. τa = i _Tm × i _D × τe.

(5) The transfer function from the axle shaft torque τa to the driving force F _V at the tire contact surface is expressed by 1 / R. However, R is the tire radius (conditions without tire slip). F _V = (1 / R) τa.

(6) The transfer function G ₄ (S) of the running resistance is expressed as a function of the vehicle speed V and changes from vehicle to vehicle.

Running resistance F _RL = G ₄ (S) V.

(7) The acceleration force F _α (= F _V −F _RL ) is obtained by subtracting the running resistance F _RL from the driving force F _V.

(8) The transfer function from the acceleration force F _α to the vehicle speed V is represented by K / SW. However, W is the vehicle weight, K is a constant, and S is a Laplace operator. Vehicle speed signal V = (K / SW) F
It becomes _α .

Since the above-mentioned conventional speed control is feedback control, it has the following drawbacks peculiar to feedback control.

(1) In order to move the feedback control loop stably, the computing units G _C (S) and G _C1 (S) to G _C
It is necessary to adjust the gain of _C3 (S).

(2) The transfer function MG (S) of the vehicle has a large number of items which vary depending on the vehicle, and when speed control with the same response is required in the feedback control loop, these changing factors change during driving. Control calculator G every time
Gain adjustment such as _C (S) must be performed.

(3) The feedback control loop is expressed as a closed loop transfer function, and there is a delay in the control response determined by the gain of the control calculator G _C (S) and the like. This delay is much larger than the delay (about 0.1 seconds) of the forward transfer function of G _O (S) × MG (S), and is usually 1 to 2 seconds, which is slow.

Therefore, the present applicant has previously proposed a feedforward control system using an inverse function MG (S) ^-1 of the transfer function of the vehicle as shown in FIG. (Japanese Patent Application No. 6-75
532).

This control method is based on the transfer function MG of the vehicle.
The accelerator pedal stroke θ required to match the vehicle speed V with the vehicle speed command V _S using the inverse function MG (S) ⁻¹ of (S)
_A is calculated by direct calculation, and this θ _A is output as the accelerator actuator operation signal θ _S. As a result, the accelerator actuator is actuated to obtain the accelerator actuator stroke θ, θ = θ _A , the accelerator pedal of the vehicle is moved, and as a result, the vehicle speed V equal to the vehicle speed V _S can be obtained.

FIG. 3 shows an example of the inverse function MG (S) ^-1 of the transfer function MG (S) of the vehicle. In FIG. 3, (1) The transfer function from the speed command V _S to the acceleration force F _α is
It is represented by F _α = (SW / K) · V _S , and from the speed V _S to the running resistance F _RL is represented by F _RL = G ₄ (S) V _S. (W is vehicle weight).

(2) The driving force F _V is F _V = F _α + F _R , which is the sum of the acceleration force F _α and the running resistance F _RL .

(3) Axle shaft torque τ _a from the driving force F _V
The transfer function up to is τ _a = R × F _V. (R is the tire radius).

(4) The transfer function from the axle shaft torque τ _a to the engine output shaft torque τe is τe = (1 / (i
_Tn × i _D )) × τ _a . (I _Tn is transmission ratio, i _D is differential ratio).

(5) The transfer function from the engine output shaft torque Te to the intake pressure Pb is represented by G ₃ ′ (S). (The engine speed Ne is a parameter).

(6) The transfer function from the intake pressure Pb to the throttle valve opening θ _B is represented by G ₂ ′ (S). (The engine speed Ne is a parameter).

(7) The transfer function of the non-linear characteristic including the pedal play from the throttle valve opening θ _B to the accelerator pedal stroke θ _A is represented by G ₁ ′ (S).

(8) The engine speed Ne is the transmission speed in the case where the transmission is in one of the first speed to the fifth speed and the clutch is engaged
Ne = (1000 / 2πR60) × i _Tn × i _D × V _S

All the transfer functions of this vehicle are obtained by calculation, and the vehicle weight W, the tire radius R, the transmission ratio i _Tn , and the differential ratio i _D can be obtained from the vehicle specifications. Further, the transfer function G ₄ (S) of the running resistance calculation can also be obtained from the road running resistance of the actual vehicle.

However, the transfer functions G ₃ '(S), G ₂ '
(S) and G ₁ ′ (S) cannot be easily obtained.
Therefore, in the above-mentioned prior application, the vehicle is driven on a chassis dynamo, and the accelerator pedal stroke θ _A , the intake pressure Pb, and the engine output torque τe of the calculation result are used as the engine rotation speed Ne parameter and 10 rotations per rotation. The point data was measured, and these data were shown in Fig. 4 (see CMAC).
C) The learning table is used to create and store the weight table.

FIG. 4 shows an example in which a vehicle is driven on the chassis dynamo and a load table is created by the seamak learning function. The seamak learning circuit 41 uses the engine output torque τe, the engine speed Ne and the intake pressure Pb. The seamak Te / Pb load table 42 is created by using the seamak learning circuit 45, and the seamak Pb / θ _A load table 46 is created by the intake pressure Pb, the engine speed Ne and the accelerator pedal stroke θ _A , and stored.

In the speed control of FIG. 6, the inverse function MG (S) ⁻¹ of the vehicle transfer function shown in FIG.
The load tables 42 and 46 are operated as ₃ ′ (S), G ₂ ′ (S) and G ₁ ′ (S) as shown in FIG. 5, and Pb is calculated from Te and Ne by the table 42. Then, the table 46 calculates θ _A from Pb and Ne and outputs it. As a result, the output of the entire region is obtained from the data of 10 points × about 10 parameters.

[0031]

In the feedforward control using the inverse function of the transfer function of the vehicle, theoretically, all the drawbacks of the feedback control can be solved. When calculating, there is always an error. Therefore, there is a drawback that an accurate accelerator pedal stroke command cannot be obtained and an error occurs between the vehicle speed command and the detected vehicle speed.

The present invention has been made to solve the problem of such feed-forward control, and its object is to make the vehicle speed deviation zero during acceleration and to provide a follow-up performance without overshoot. An object of the present invention is to provide a vehicle speed control device capable of good control.

[0033]

In order to achieve the above object, a vehicle speed control device according to the present invention uses an inverse function of a transfer function of a vehicle as a feedforward transfer function to directly open an accelerator opening from a vehicle speed command. And a feedforward control circuit for matching the vehicle speed to the vehicle speed command by the accelerator operation, a vehicle control model transfer function circuit having the same transfer function as the control circuit and receiving the vehicle speed command, and the output of this circuit Is used as a speed command for compensation control, and a feedback control circuit for compensating for an error amount due to the feedforward control is provided.

[0034]

When the inverse function of the vehicle transfer function of the feedforward control circuit is found, there is always an error, which causes an error in the vehicle speed command and the vehicle speed. However, the transfer function of the vehicle control model transfer function circuit is the feedforward control circuit. The feedback control circuit uses the output signal of this vehicle control model transfer function circuit as the speed command for compensation control to detect and compensate for the error component, so that there is no vehicle speed error even during acceleration and Good speed control is possible.

[0035]

Embodiments of the present invention will be described with reference to FIGS. In FIG. 1, A is a vehicle model reference feedforward control circuit configured in the same manner as shown in FIG. 6 except for the adder A ₃ portion, and this circuit is the main part of this system. B is (detected) vehicle speed V obtained by the feedforward control newly provided in the present invention.
When there is a deviation between the vehicle speed command and the vehicle speed command, the accelerator pedal stroke (main) from the inverse function circuit 1 of the transfer function of the vehicle
It is a vehicle model compensation control circuit that outputs an accelerator pedal stroke correction command that uses a command as a correction.

Regarding the compensation control circuit B, 4 is the control circuit A
Same as the transfer function of the transfer function MG (S) ^-1 , G
A vehicle control model transfer function circuit adopting _O (S) and MG (S), 5 is a D calculator composed of a transfer function G _C3 (S) for differentiating the speed command V _S ′ from the circuit 4, and A ₁ is a circuit A matching device for detecting a deviation between the vehicle speed command V _S ′ from 4 and the vehicle speed V, 6 is a P calculator for a transfer function G _C1 (S) that outputs a signal proportional to this deviation, and 7 is a vehicle speed V D operator of transfer function G _C3 (S) that differentiates A, A ₂ is an adder that adds the signals from the operators 5 to 7 with the polarities shown, and 8 is an adder A ₂
Is a PI calculator that proportionally integrates the signal from and outputs an accelerator pedal stroke correction command to the adder of the control circuit A.

Inverse function MG of the vehicle transfer function shown in FIG.
(S) ⁻¹ transfer functions G ₃ ′ (S), G ₂ ′ (S), G ₁ ′
(S) uses the load table by the seamak learning function as in the previous application (FIGS. 4 and 5).

The transfer gain of the vehicle control model transfer function circuit 4 is steady, and the response delay is the response time from the vehicle speed command V _S until the vehicle speed V is output. The purpose of this circuit is to prevent overshoot. If this circuit is not, the command vehicle speed V _S
When the vehicle is accelerated by, the stroke command of the accelerator pedal for compensating for the delay of the vehicle speed with respect to the command is the PI calculator 8
Therefore, when the vehicle speed command V _S shifts from acceleration to a steady state, the value accumulated in this integral term is released at the PI time constant, and an extra accelerator command is issued.
As a result, vehicle speed overshoot occurs. The circuit 4 prevents the vehicle speed overshoot from being accumulated in the integral term due to the vehicle speed delay. The output of this circuit is the vehicle speed detection of the vehicle model and becomes the speed command V _S ′ of the vehicle model compensation control using the feedback control.

The feedback circuit composed of the calculators 5 to 9 is a speed control circuit with an acceleration minor having an acceleration command forcing, and an acceleration / deceleration command signal from the D calculator 5 and a vehicle speed from the P calculator 6 are provided. The sum of the deviation signal is used as a new acceleration / deceleration command, and the acceleration / deceleration deviation signal with the acceleration / deceleration detection signal from the D calculator 7 is output from the adder A2.
The operator 8 proportionally integrates and outputs an accelerator pedal stroke correction command θ _C to the adder A ₃ of the control circuit A to correct the accelerator pedal stroke command θ _A to θ _S.

As described above, the vehicle model compensating control circuit B is added to the vehicle model reference feedforward control circuit A whose feedforward transfer function is the reciprocal of the vehicle transfer function.
As a result, the acceleration command forcing by the differentiation of the vehicle speed command V _S ′ from the D calculator 5 acts very effectively.

That is, the differential operation outputs of the vehicle speed command V _S ′ and the vehicle speed V are outputs of the D operation circuits 5 and 7, respectively, and they are canceled out when the acceleration of the speed command V _S and the vehicle speed V become the same. Further, the speed command V _S ′ and the vehicle speed V _S in consideration of the response delay
If the two match, the output of the P calculator 6 becomes 0, so P
The input of the I calculator 8 becomes 0, and the output of the PI calculator 8 becomes constant.

That is, the inverse function MG (S) of the transfer function of the vehicle
If there is no error in ^-1 , the output of the PI calculator 8 is constant at 0, so that it is not a disturbance of the feedforward control of the control circuit A. If the inverse function MG (S) ⁻¹ has an error, only the accelerator stroke command corresponding to the error is accumulated in the PI calculator 8 and compensated.

Further, due to the canceling effect of the acceleration of the speed V of the speed command V _S ′ of the D calculators 5 and 7, there is no deviation between the speed command V _S ′ and the speed V even during acceleration, and the vehicle speed command V is steady after acceleration. _Since overshoot does not occur even when _S changes, ideal control with good followability with respect to the vehicle speed command V _S is possible.

This is the inverse function MG (S) of the transfer function of the vehicle.
^This means the same response as that obtained by the feedforward control circuit of FIG. 6 when there is no error in ^{−1 and} it is as in theory. In the actual control, a response time for correcting the error of the inverse function MG (S) ⁻¹ of the transfer function of the vehicle is required, so that it is slightly different from the case of only the feedforward control of FIG.

FIG. 9 shows a response waveform in the simulation when the shift time of the transmission is 0 second in the circuit of FIG. According to this waveform, the vehicle speed control delay is about 0.4.
It can be seen that the vehicle speed deviation due to the control delay is about 0.3 Km / h.

FIGS. 10, 11, and 12 show response waveforms in the simulation when the shift time of the transmission is 1 second in the circuits of FIGS. 6, 1, and 8. In the simulation result embodiment (FIG. 1), the deviation between the vehicle speed command V _S and the (detected) vehicle speed V is smaller than that in FIGS. 6 and 8.

[0047]

Since the present invention is configured as described above, it has the following effects.

(1) MG (S) ^-1 , G _O (S), MG
Since the transfer function of (S) has only one output value for one input value, the feedforward system is always stable.

(2) The feedforward system provides the fastest control system.

(3) Inverse function MG (S) of the transfer function of the vehicle
^{By making -1} accurate, the feedback compensation loop has a smaller operation amount and becomes θ _C → 0, so that overshoot or fluctuation does not occur in control.

(4) Since the inverse function MG (S) ^-1 is obtained by calculation, there is no place for manual control gain adjustment, and control gain adjustment is unnecessary.

(5) Due to the reason (4) above, the control gain adjustment is unnecessary even if the vehicle is changed.

(6) The vehicle model compensation control circuit can reduce the vehicle speed deviation to zero even during acceleration. And no overshoot occurs.

(7) In the vehicle model compensation control, the transmission ratio, differential ratio, tire radius,
If the control gain is made variable by calculation from data such as vehicle weight, manual adjustment becomes unnecessary even if the vehicle changes.

(8) Therefore, stable and highly accurate tracking performance can always be easily obtained.

[Brief description of drawings]

FIG. 1 is a block diagram showing a vehicle speed control system according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a transfer function of the vehicle according to the embodiment and the prior application example.

FIG. 3 is a block diagram showing an inverse function of a transfer function of the vehicle according to the embodiment and the prior application example.

FIG. 4 is a block diagram showing a learning system using seamac according to the embodiment and the prior application.

FIG. 5 is a block diagram showing a dynamic driving system using seamac according to the embodiment and the prior application.

FIG. 6 is a block diagram showing a vehicle speed control system according to a prior application example.

FIG. 7 is a block diagram showing a PID speed control method according to a conventional example.

FIG. 8 is a block diagram showing a speed control system with an acceleration minor according to a conventional example.

9 is a waveform diagram showing a simulation response waveform (shift time 0 seconds) of the circuit of FIG.

10 is a waveform diagram showing a simulation response waveform (shift time 1 second) of the circuit of FIG.

FIG. 11 is a waveform diagram showing a simulation response waveform of the example.

12 is a waveform diagram showing a simulation response waveform of the circuit of FIG.

[Explanation of symbols]

A ... Vehicle model reference feedforward control circuit B ... Vehicle model compensation control circuit G _O (S) ... Accelerator actuator stroke control transfer function MG (S) ... Vehicle transfer function MG (S) ^-1 ... Vehicle transfer function Inverse function 4 ... Vehicle control model transfer function circuit 5, 7, 13 ... D calculator 6, 12 ... I calculator 8, 14 ... PI calculator 11 ... PID calculator 41, 45 ... Seamac learning circuit 42, 46 ... Load table

Claims (1)

[Claims] 1. A feedforward control circuit for calculating and outputting an accelerator opening directly from a vehicle speed command by using an inverse function of a vehicle transfer function as a feedforward transfer function, and for matching the vehicle speed with the vehicle speed command by the accelerator operation. A vehicle control model transfer function circuit having the same transfer function as that of the control circuit and to which the vehicle speed command is input; and a feedback control circuit for compensating an error amount due to the feedforward control by using an output of this circuit as a speed command for compensation control. A vehicle control device comprising: