JP2003202901A - Sliding mode controlling apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a sliding mode controlling apparatus capable of satisfying a plurality of requirements without any contradiction at the time of controlling the behavior of an elastic system such as a spring mass system. <P>SOLUTION: An exhaust valve 1 is provided with a valve body 2 and an armature 34 coupled to a valve shaft 4. Then, the valve body 2 is attracted to its valve closing position by the electromagnetic force of a valve opening drive electromagnet 36 acting to the armature 34. Also, the valve body 2 is attracted to its valve closing direction by the electromagnetic force of a valve closing drive electromagnet 38 acting to the armature 34. At the time of controlling valve opening and valve closing, the displacement of the valve body 2 is detected by a displacement quantity sensor 42. Then, a sliding mode is controlled by using a switching hyperplane variably set in accordance with the detected displacement. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sliding mode controller for controlling the behavior of an elastic system such as a spring mass system.

[0002]

2. Description of the Related Art For example, a control system having a spring mass system such as an electromagnetically driven valve disclosed in Patent Document 1 below is well known.
Usually, in such a control system, feedback control such as PID control is performed so that the state quantities such as the displacement of the controlled object and the displacement speed follow the desired target state quantities. However, it has been difficult for these conventional control methods such as PID control to sufficiently secure robustness against disturbance and characteristic changes of the controlled object.

Therefore, in recent years, sliding mode control has also been applied to such a control system. In this sliding mode control, high gain control is performed so that the state quantity of the controlled object is converged on the switching hyperplane expressed by a desired designed linear function, and is controlled on the switching hyperplane. .

Therefore, according to the sliding mode control, the state quantity of the controlled object can be bound on the switching hyperplane while sufficiently ensuring the robustness against the disturbance and the characteristic change of the controlled object. Become.

[0005]

[Patent Document 1] JP-A-9-217859

[0006]

By the way, when it is attempted to determine the control performance so as to control the controlled object by using this sliding mode control, when designing the switching hyperplane based on a plurality of requirements, a plurality of those switching hyperplanes are selected. The problem may arise that requirements conflict with each other.

For example, regarding the control of the electromagnetically driven valve described in the above publication, in addition to ensuring its operational stability, the power consumption is suppressed as much as possible, and the generation of noise accompanying the opening and closing of the valve body is suppressed. However, it is an important factor in questioning its performance. For such an electromagnetically driven valve, when it is intended to suppress the generation of noise associated with the opening and closing of the valve body through the sliding mode control, the following switching hyperplane is usually designed. That is, when displacing the valve body to be controlled from one displacement end to the other displacement end side, aiming to reduce the displacement speed of the valve body immediately before the other displacement end, the vibration thereof should be suppressed. A switching hyperplane is designed to enhance damping. But,
When such a switching hyperplane is set, the displacement time required for the valve body to be controlled to be displaced from one displacement end to the other displacement end is extended in reverse.

As described above, when designing a switching hyperplane for sliding mode control, a plurality of elements are often considered in order to improve the control performance, but the required elements are the switching hyperplanes. It often happens that design conflicts with each other. And, such a contradiction of the requirement elements is a major obstacle to the improvement of the control performance.

The present invention has been made in view of these circumstances, and an object thereof is to satisfy a plurality of requirement elements without contradiction when controlling the behavior of an elastic system such as a spring mass system. An object of the present invention is to provide a sliding mode control device capable of performing the above.

[0010]

[Means for Solving the Problems] Means for achieving the above-mentioned objects and their effects will be described below. The invention according to claim 1 sets a switching hyperplane when displacing a control target to which a biasing force is applied by an elastic member from one displacement end side to the other displacement end side, and the state quantity of the control target is set. Is a sliding mode control device for controlling the controlled object so as to converge on the set switched hyperplane, and comprising a setting means for variably setting the switched hyperplane according to the displacement of the controlled object. The summary will be given.

According to the above configuration, by variably setting the switching hyperplane according to the displacement of the controlled object, it becomes possible to satisfy a plurality of required factors as the control performance of the controlled object.

According to a second aspect of the present invention, in the first aspect of the present invention, the setting means is a hyperplane that is in contact with a preset reference model for the state quantity of the control target at its corresponding displacement point. Its gist is to variably set the switching hyperplane.

According to the above configuration, by setting the switching hyperplane as a hyperplane contacting a preset reference model at its corresponding displacement point, variable control of the switching hyperplane can be easily performed. become.

According to a third aspect of the present invention, in the invention according to the first aspect, when the switching hyperplane follows the dynamic characteristic of the reference model, the state quantity and the control object indicated by the dynamic characteristic are the same. The gist is that it is defined for the deviation from the actual state quantity of.

According to the above configuration, it is possible to variably set the manner of convergence of the actual state quantity to the dynamic characteristics of the reference model according to the displacement of the valve body, and thus, a plurality of requirements as control performance of the controlled object. You will be able to satisfy the elements.

According to a fourth aspect of the invention, in the invention according to the second or third aspect, the reference model is set based on a transition of the state quantity of the controlled object by the urging force of the elastic member. And

According to the above configuration, it becomes possible to control the controlled object while making use of the natural vibration of the physical system including the elastic member and the controlled object. Consequently, the controlled object moves from one displacement end side to the other. It is possible to shorten the time required to displace the displacement end side of the.

A fifth aspect of the present invention is characterized in that, in the fourth aspect, the reference model is set as a hyperplane in the vicinity of the other displacement end of the controlled object.

According to the above configuration, it becomes possible to control the controlled object while reducing the natural vibration of the physical system including the elastic member and the controlled object in the vicinity of the other displacement side.

According to a sixth aspect of the invention, in the invention according to any one of the second to fifth aspects, the reference model is derived from different physical characteristics for each displacement region divided into a plurality of the controlled objects. The main point is that it is set as a model to be exposed.

According to the above construction, when different displacement regions have different requirements for control, it is possible to set a reference model satisfying those demands, and thus satisfy a plurality of requirement factors as control performance of the controlled object. You will be able to.

According to a seventh aspect of the invention, in the invention according to any one of the second to sixth aspects, the control of the controlled object based on the switching hyperplane variably set by the setting means is:
The gist is that it is performed based on a switching hyperplane that is variably set with respect to the displacement point on the other displacement end side by a predetermined amount from the displacement point detected each time of the controlled object.

In actual control equipment, there is usually a response delay with respect to the control of the controlled object. In this respect, according to the above configuration, by performing control of the controlled object based on the switching hyperplane variably set with respect to the displacement point on the other displacement end side by a predetermined amount from the detected displacement point, This response delay can be taken into consideration.

The invention according to claim 8 is the invention according to any one of claims 1 to 7, wherein the control of the controlled object based on the switching hyperplane variably set by the setting means is:
When the state quantity of the controlled object is in the region near the switching hyperplane that is variably set, it is essential to reduce the external force for reaching the switching hyperplane as compared with when it is in another region. And

When the sliding mode control is performed every sampling period, a kind of control delay is caused due to this, so that so-called chattering occurs in which the state quantity of the controlled object vibrates at high frequencies in the vicinity of the hyperplane. There is. In this respect, according to the above configuration, when the state quantity of the valve body is in the region near the switching hyperplane, the external force for reaching the switching hyperplane is smaller than that in the other regions. As a result, it becomes possible to suppress the occurrence of chattering.

The invention according to any one of claims 1 to 8 is, as in the invention according to claim 9, an engine valve of an internal combustion engine, wherein the control target is a valve body connected to an armature, The state quantity of the engine valve may be controlled by attracting the valve element to the one displacement end side and the other displacement end side by an electromagnetic force acting on the armature. As a result, the effects of the invention described in each of the first to eighth aspects can be suitably exhibited.

According to a tenth aspect of the invention, in the ninth aspect of the invention, the feedback gain for causing the state quantity of the valve element to reach the switching hyperplane that is variably set is set to the rotational speed of the engine, and The gist of the present invention is to further include means for variably setting in accordance with at least one of the engine load and the displacement of the valve element.

According to the above configuration, the control is performed in consideration of the fact that the external force such as the cylinder pressure applied to the valve body changes depending on the engine operating state, and the biasing force of the elastic member and the electromagnetic force acting on the armature change depending on the displacement. It can be performed.

[0029]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS (First Embodiment) A first embodiment in which a sliding mode control device according to the present invention is applied to a control device for an electromagnetically driven engine valve will be described below with reference to the drawings. . Since the intake valve and the exhaust valve as the engine valve are basically the same in configuration and drive control mode, in the present embodiment, a device for controlling the drive of the exhaust valve as the engine valve will be described as an example. .

The exhaust valve to be driven and controlled in the control apparatus of the present embodiment includes a first elastic member for urging the valve element toward one of the displacement ends and a first elastic member for urging the elastic member toward the other of the displacement ends. Two
And an elastic member. An armature is connected to the valve body, and when electromagnetic force acts on the armature, the valve body is attracted to one displacement end side and the other displacement end side, and the valve body is electromagnetically driven. It

Specifically, as shown in FIG. 1, the exhaust valve 1 includes a valve shaft 4 and a valve body 2 provided at one end of the valve shaft 4.
And springs 14 and 24 corresponding to the first and second elastic members, and an electromagnetic drive unit 30 that reciprocates the valve shaft 4 in the cylinder head 10.

Here, an exhaust port 18 communicating with the combustion chamber 17 is formed in the cylinder head 10, and a valve seat 16 on which the valve element 2 is seated is formed at the peripheral edge of the opening of the exhaust port 18. ing. That is, as the valve shaft 4 reciprocates, the valve body 2 is seated on and away from the valve seat 16, so that the exhaust port 18 is opened and closed.

Further, a spring for urging the valve body 2 in the opening direction and the closing direction is provided. That is, the lower retainer 12 is provided in a portion of the valve shaft 4 on the side opposite to the combustion chamber 17 with respect to the cylinder head 10. Between the lower retainer 12 and the cylinder head 10, the lower retainer 12 corresponds to the first biasing member. The lower spring 14 is arranged in a compressed state. The valve body 2 is biased in the valve closing direction by the elastic force (biasing force) of the lower spring 14.

On the other hand, an upper retainer 22 is provided at the end of the valve shaft 4 on the side opposite to the valve body 2, and an upper retainer 22 and an upper portion provided in a casing (not shown) of the electromagnetic drive unit 30. An upper spring 24 corresponding to the second biasing member is arranged in a compressed state between the cap 20 and the cap 20. The valve body 2 is biased in the valve opening direction by the elastic force (biasing force) of the upper spring 24.

The electromagnetic drive unit 30 includes an armature 34 fixed to the valve shaft 4, and a lower core 36C and an upper core 38C arranged so as to sandwich the armature 34.
It has and. Here, the armature 34 is made of a disc-shaped high magnetic permeability material. Also, the lower core 36
C and the upper core 38C both have a high magnetic permeability material formed in an annular shape, and the valve shaft 4
Is reciprocally inserted.

An annular groove 36h centering on the axis of the valve shaft 4 is formed on the surface of the lower core 36C facing the armature 34, and an annular lower coil 36c is provided in the groove 36h. ing. This lower coil 3
An electromagnet (opening drive electromagnet) 36 that drives the valve body 2 in the valve opening direction is configured by 6c and the lower core 36C.

On the other hand, an annular groove 38h centering on the axial center of the valve shaft 4 is formed on the surface of the upper core 38C facing the armature 34, and an annular upper coil 38c is provided in the groove 38h. ing. The upper coil 38c and the upper core 38C constitute an electromagnet (close driving electromagnet) 38 for driving the valve body 2 in the valve closing direction.

FIG. 1 shows the state of the valve body 2 when no electromagnetic force is generated in any of the electromagnets 36 and 38. In this state, the armature 34 is attached to each electromagnet 3
The springs 14 and 24 are not attracted by the electromagnetic force of the solenoids 6 and 38, and stand still at a position where the elastic forces of the springs 14 and 24 are balanced, that is, at a position approximately in the middle between the lower core 36C and the upper core 38C.

When the electromagnetic force of the electromagnets 36 and 38 is applied to the armature 34, the armature 34 is attracted to the lower core 36C side and the upper core 38C side.
This electromagnetic force is applied to each coil 36 of these electromagnets 36 and 38.
It is generated by energizing c and 38c.

In this embodiment, the electromagnets 36, 3
8 to control the energization of the coils 36c and 38c of the exhaust valve 1
This is performed based on the displacement of (valve element 2, valve shaft 4). Specifically, the engine valve control device according to the present embodiment includes a displacement amount sensor 42 in the upper cap 20. This displacement amount sensor 42 outputs a voltage (detection signal) that changes according to the distance between the sensor 42 and the upper retainer 22, and the displacement amount of the upper retainer 22, in other words, based on this voltage. For example, the displacement amount of the valve body 2 is detected. Then, the energization control is performed based on the displacement of the exhaust valve 1 by using the detection result of the displacement amount sensor 42.

This energization control is performed by the electronic control unit 40 that collectively controls various controls of the internal combustion engine. The electronic control unit 40 includes a CPU, a memory, electromagnets 36 and 38.
In addition to a drive circuit that supplies an exciting current to each of the coils 36c and 38c, an input circuit that receives the detection signal of the displacement sensor 42, an A / D converter that performs A / D conversion of this detection signal (both not shown) And so on.

Here, the operation mode of the exhaust valve 1 which is driven to open and close by controlling the energization of the exhaust valve 1 by the electronic control unit 40 will be described. In the closed state of the exhaust valve 1, the holding current is closed in order to hold the exhaust valve 1 at the fully closed position, in other words, to hold the valve body 2 at the position where it is seated on the valve seat 16. It is supplied to the driving electromagnet 38. By the supply of the holding current, the armature 34 is attracted to the upper core 38C side by the electromagnetic force of the closing drive electromagnet 38. Due to this suction force, the valve body 2 is seated on the valve seat 16 against the elastic force of the upper spring 24, and the state in which the armature 34 is in contact with the upper core 38C is maintained.

Next, when the opening drive timing of the exhaust valve 1 arrives, the supply of the holding current is interrupted. As a result, the exhaust valve 1 is opened by the elastic force of the upper spring 24.
In other words, when the armature 34 moves to the lower core 36C side, the valve body 2 moves away from the valve seat 16 and moves toward the combustion chamber 17 side.

In the process of displacement of the exhaust valve 1 from the fully closed position to the fully open position, the energization control of the opening drive electromagnet 36 is performed. Then, when the exhaust valve 1 reaches the fully open position by the armature 34 coming into contact with the lower core 36C, for example, a holding current is supplied to the open drive electromagnet 36 so as to hold the state. By supplying this holding current, the armature 34 is moved by the electromagnetic force of the open drive electromagnet 36.
6C is sucked. Due to this suction force, the valve body 2 is held in the fully open position against the elastic force of the lower spring 14.

On the other hand, when the closing drive timing of the exhaust valve 1 arrives, the supply of the holding current for holding the exhaust valve 1 in the fully open position is interrupted. As a result, the exhaust valve 1 is displaced by the elastic force of the lower spring 14 toward the valve closing direction, in other words, the valve body 2 is displaced toward the valve seat 16.

In the process of displacement of the exhaust valve 1 from the fully open position to the fully closed position, the energization control of the closing drive electromagnet 38 is performed. Then, when the valve body 2 reaches the fully closed position where it is seated on the valve seat 16, a holding current is supplied to the closing drive electromagnet 38 in order to maintain this state.

In this way, when displacing the valve body 2 from one displacement end to the other displacement end, in the present embodiment, the sliding mode control is used to move each of the electromagnets 36, 38 as follows. Conduct energization control. In addition, in the displacement process from the fully closed position to the fully opened position and the displacement process from the fully opened position to the fully closed position, the control modes of the electromagnets 36 and 38 are the same. The process of displacement to the position will be described as an example.

In the present embodiment, the displacement of the valve body 2 to be controlled is a one-dimensional space (line segment) connecting the fully closed position and the fully open position, and the state quantity indicating the dynamic characteristic of the valve body 2 is obtained. Is the displacement and displacement speed of the valve body 2. Then, on a two-dimensional linear space having the displacement and displacement velocity of these valve bodies 2 as degrees of freedom,
The displacement and displacement velocity (state quantity) of these valve bodies 2 are calculated as
Control is performed so as to converge to a preset switching hyperplane (line segment) as a (one-dimensional) linear subspace in the dimensional space. Then, this switching hyperplane is variably set according to the displacement of the valve body 2 so as to satisfy a plurality of required elements concerning the control performance of the engine valve.

More specifically, the switching hyperplane is set as a hyperplane that contacts a preset reference model for the state quantity of the valve body 2 at its corresponding displacement point. Specifically, this normative model basically consists of disturbances and damping factors,
It is set based on the displacement mode of the valve body 2 (the locus of the displacement of the valve body 2 and the displacement speed) that is assumed under the assumption that there is no electromagnetic force. However, in the vicinity of the other displacement end, this reference model is set as a hyperplane. This hyperplane is set so that the displacement speed is "0" at the other displacement end. Further, this hyperplane is set in the vicinity of the other displacement end based on the displacement mode of the valve body 2 which is assumed under the assumption that there is no disturbance, damping element, or electromagnetic force. It is set to be smaller than that of the above reference model.

As a result, the switching hyperplane is set based on the displacement mode of the valve body 2 which is assumed under the assumption that there is no disturbance, damping element, or electromagnetic force except near the other displacement end. It will be set based on the normative model. On the other hand, in the vicinity of the fully open position, the displacement speed is set to "0" at the fully open position, and the hyperplane is set.

FIG. 2 shows a specific setting mode of the switching hyperplane according to this embodiment. As shown in FIG.
With respect to the displacement region of the valve body 2 other than the vicinity of the Low end face (fully open position), the reference model is as follows. That is, the locus of the displacement and the displacement speed of the valve body 2 when the valve body 2 is displaced from the fully closed position to the fully open position only by the urging force of the upper spring 24 and the lower spring 14 is shown.

This locus (reference model) is a quadratic curve calculated from a physical model in which the elastic body composed of the lower spring 14 and the upper spring 24 and the movable portion of the exhaust valve 1 connected to the elastic body is a physical system. Become. That is, the weight M of the movable part, the elastic constant K of the elastic body composed of the lower spring 14 and the upper spring 24, and the displacement (detection value) x of the exhaust valve 1 (valve body 2) based on the balanced position of the elastic body x The equation of motion of this physical system is
It becomes 1).

[0053]

[Equation 1] The displacement x of the valve body 2, which is the solution of the equation (c1), is obtained as a periodic function, and from this, the displacement velocity is also obtained as a periodic function as its differential value. Then, these relational expressions are obtained as a quadratic curve shown in FIG. 2 by the displacement and the displacement velocity.

Further, as shown in FIG. 2, for the displacement region of the valve body 2 near the Low end face (fully open position), the reference model is used and the rate of change of displacement speed with respect to displacement is smaller than that of the quadratic curve. It is a one-dimensional hyperplane (line segment).

By variably setting the switching hyperplane according to the displacement of the valve body 2 based on such a reference model, a plurality of required elements required for controlling the exhaust valve 1 are satisfied. That is, except in the vicinity of the Low end face (fully open position), the actual displacement mode of the valve body 2 is controlled so as to follow the displacement mode of the valve body 2 when the valve body 2 is displaced only by the biasing force of the elastic body. As a result, the valve body 2 can be displaced by taking advantage of the natural vibration of this physical system. As a result, the time required for the displacement from the Up end face (fully closed position) to the Low end face (fully opened position) can be reduced.

On the other hand, in the vicinity of the Low end face (fully open position), the state quantity is bound to the hyperplane having a small change rate of the displacement velocity, whereby the lower core 36 of the armature 34 is bound.
C To reduce the impact when sitting on the upper surface.

By controlling the exhaust valve 1 so as to constrain the above state quantity on the switching hyperplane which is variably set according to the displacement of the valve body 2, the displacement time is shortened while seated. It is possible to perform control to mitigate the impact of. On the other hand, when the exhaust valve 1 is controlled using the switching hyperplane shown by the alternate long and short dash line in FIG. 2 in order to reduce the impact when seated, the displacement time increases.

The setting of the energization control mode for the exhaust valve 1 for performing the sliding mode control for restraining the state quantity on the switching hyperplane is performed as follows. First, a switching hyperplane that is in contact with the secondary reference model of FIG. 2 at the displacement x of the valve body 2 (tangent line at the displacement x of the secondary reference model)
And the first-order reference model of FIG. 2 is defined by the following expression (c2).

[0059]

[Equation 2] In this equation (c2), the coefficients a and b are actually functions of the displacement x of the valve body 2. Further, a switching function σ, which is a linear function that defines these switching hyperplanes, is defined by the following expression (c3).

[0060]

[Equation 3] As can be seen from the above equation (c3), the hyperplane whose switching function σ is zero is the switching hyperplane.

Next, in the actual physical system of the exhaust valve 1, in the case where the lower spring 14 and the upper spring 24 are connected to the movable part, the sliding resistance between the movable part and the fixed part, and the armature 34. It is a system to which electromagnetic force acting on is applied. The equation of motion of this system includes the weight M, the elastic constant K, the valve displacement x, the damping coefficient C between the movable portion and the fixed portion, and the sliding mode which is the electromagnetic force acting on the armature 34 in the sliding mode state. It is expressed by the following expression (c4) using the input Ul.

[0062]

[Equation 4] In the sliding mode state, the state quantity of the exhaust valve 1 is bound on the switching hyperplane, in other words, on the hyperplane where the switching function σ becomes zero. Therefore, the sliding mode input Ul is expressed by the following equation (c) using this equation of motion (c4) and the time derivative of the switching function σ being zero.
It is represented by 5).

[0063]

[Equation 5] An arrival mode input (feedback input) Unl that converges the state quantity on the switching hyperplane when the state quantity is separated from the switching hyperplane is defined by the following expression (c6).

[0064]

[Equation 6] Here, the feedback gain G is set so as to satisfy the condition for reaching the switching hyperplane, in other words, the condition for reaching the sliding mode. In this embodiment, the gain G satisfying this attainment condition is set using the Lyapunov function method. That is, for example, V = 1/2
The × σ × σ ^T as a Lyapunov function, and sets the gain G thus time derivative which is expressed by the following equation (c7) has a negative.

[0065]

[Equation 7] In the above equation (c7), by setting the positive / negative of the gain G while having a predetermined absolute value so that the time derivative of the Lyapunov function becomes negative, the switching function σ becomes zero using the arrival mode input Unl. Will converge to.

Here, the valve opening control of the exhaust valve 1 according to this embodiment will be described with reference to FIG. FIG. 4 is a flowchart showing a processing procedure of the same control. This process is repeatedly executed every predetermined period.

In this series of processes, first, at step 100, it is judged whether or not the displacement of the valve body 2 is below a predetermined threshold value, and if it is below the same threshold value, the routine proceeds to step 110. This threshold value is set, for example, when the attraction force of the armature 34 by the opening drive electromagnet 36 becomes a predetermined value or more.

On the other hand, in step 110, the switching function is calculated based on the detected displacement x of the valve body 2. This is realized by, for example, having a storage function of storing the reference model shown in FIG. 2 in advance in the electronic control unit 40 and calculating the tangent line of the reference model at the displacement x of the valve body 2. be able to. Further, instead of this, the electronic control unit 40 is provided with a storage function for holding data (maps and the like) regarding switching functions such as the coefficients a and b of the above equation (c3) at each displacement x of the valve body 2. You may do it.

In step 120, the sliding mode input Ul is calculated based on the switching function thus calculated. That is, here, from the switching function at the displacement x, the sliding mode input U is calculated using the above equation (c5).
Calculate l. Further, at step 130, the arrival mode input Unl is calculated from the switching function at the displacement x based on the above equation (c6).

From the sliding mode input Ul and the reaching mode input Unl calculated in this way, step 14
At 0, the control input U, which is the electromagnetic force acting on the armature 34, is calculated. Then, in step 150, the gap between the armature 34 and the lower core 36C is calculated based on the displacement x detected by the displacement amount sensor 42, and in step 160, the energization control current amount to the exhaust valve 1 is calculated using this gap. calculate. This calculation is performed as follows.

That is, the control current is calculated by providing the electronic control unit 40 with the function of storing the physical model formula that determines the amount of current that is passed through the electromagnet 36 from the gap and the electromagnetic force. By the way, the electromagnetic force acting on the armature 34 is determined by the gap between the armature 34 and the lower core 36C and the amount of current supplied to the opening drive electromagnet 36. Therefore, this gap and the electromagnet 3
A physical model formula that determines the electromagnetic force acting on the armature 34 can be defined based on the amount of current passed through the armature 6. And from now on, the gap and the electromagnetic force (control input U
, U2 ...), it is possible to obtain a physical model formula that determines the amount of current that is applied to the electromagnet 36.

Further, as shown in FIG. 3, the electronic control unit 40 may be provided with a function of storing a map that defines the relationship between the gap and the electromagnetic force and the amount of current supplied to the electromagnet 36. When the control input U is negative, it is necessary to apply a force in the fully closing direction.
Since the electromagnetic force acting on the armature 34 by 6 is a force in the fully opening direction, the energization control amount is set to "0".

When the control current is calculated in this manner, the exhaust valve 1 based on the control current is calculated based on the calculated control current in step 170.
Energization control is performed. In the present embodiment described above, the following effects can be obtained.

(1) By variably setting the switching hyperplane according to the displacement of the valve body 2, it becomes possible to satisfy a plurality of requirements regarding the control of the exhaust valve 1. (2) A reference model based on the displacement mode of the valve body 2 only by the urging force of the upper spring 24 and the lower spring 14 is set, and a hyperplane contacting this at each displacement point is a switching hyperplane at the corresponding displacement point. And As a result, the exhaust valve 1 can be controlled while making use of the natural vibration obtained by the upper spring 24, the lower spring 14, and the mass of the movable portion, and the time required for the displacement of the valve body 2 can be reduced. .

(3) A hyperplane, which has a displacement velocity of zero at the Low end face and has a smaller rate of change in displacement velocity than the quadratic normative model near the Low end face, is used as the normative model. In other words, the switching hyperplane is set such that the displacement velocity becomes zero at the Low end face and the change rate of the displacement velocity is smaller in the vicinity of the Low end face than the quadratic reference model. As a result, it is possible to perform control for suitably reducing vibration when seated on the Low end surface.

(Second Embodiment) The second embodiment, in which the sliding mode control device according to the present invention is applied to a control device for an electromagnetically driven engine valve, is different from the first embodiment. Will be mainly described with reference to the drawings.

In the present embodiment, when the exhaust valve 1 is controlled (model reference adaptive control) so as to follow the dynamic characteristics of the reference model shown in FIG. The mode of convergence of the state quantity is controlled by sliding mode control. Hereinafter, this will be described in detail.

First, the physical model of the actual exhaust valve 1 is defined. This is because when the lower spring 14 and the upper spring 24 are connected to the movable part, the sliding resistance between the movable part and the fixed part, the electromagnetic force acting on the armature 34, the in-cylinder pressure, etc. A system to which an external force exerted on is applied. The equation of motion of this system is
In addition to the elastic constant K and the valve displacement x, the damping coefficient C between the movable part and the fixed part, the external force f, and the electromagnetic force acting on the armature 34 to follow the dynamic characteristics of the reference model are input u. ,

[0079]

[Equation 8] This equation of motion can be expressed as follows using a matrix.

[0080]

[Equation 9] On the other hand, the equation of motion of the physical system in which the state quantity of the exhaust valve 1 follows a predetermined reference model is shown below.

[0081]

[Equation 10] In this equation, by setting the matrix Am, the vector Bm, and the input r, the physical system of the reference model having desired characteristics can be expressed. In particular, in the reference model shown in FIG. 2, the quadratic model part is expressed using the following matrices Am and Bm obtained by removing the damping term and the electromagnetic force input term from the above equation (c8), for example. be able to.

[0082]

[Equation 11] In the reference model shown in FIG. 2, the first-order model portion can be represented by the following matrices Am and Bm, where γ is the rate of change of the displacement speed with respect to the valve displacement.

[0083]

[Equation 12] A similar reference model may be expressed by appropriately setting the input r of the above equation (c10) instead of the above equations (c11) and (c12).

In the above equations (c9) and (c10), the deviation between the state quantity showing the dynamic characteristics of the reference model and the actual state quantity of the exhaust valve 1 is represented by the vector e = Xm-X. By setting this vector e on the switching hyperplane, the state quantity of the exhaust valve 1 is made to follow the state quantity showing the dynamic characteristics of the reference model in a desired manner. Hereinafter, the setting mode of the switching hyperplane according to the present embodiment will be described with reference to FIG.

FIG. 5A shows an area A and an area B in which different switching hyperplanes are set. FIG. 5 (a) shows a transition example of valve displacement when the exhaust valve 1 is displaced from the Up end face to the Low end face. The Up end face side is the region A and the Low end face side is the boundary near the Low end face. Area B
And are set respectively. Then, FIG. 5B shows a setting manner of the switching hyperplane in the area A, and FIG. 5C shows a setting manner of the switching hyperplane in the area B. As shown in FIGS. 5B and 5C, in the present embodiment, the slope of the time derivative with respect to the deviation xm-x in the region B is set to be larger than that in the region A. By setting in this way, the armature 3 by the electromagnetic force
In the vicinity of the Low end face where the attracting ability of 4 increases,
The convergence speed of the switching hyperplane to the origin can be increased. Therefore, the convergence speed to the reference model can be increased in the vicinity of the Low end surface.

The data regarding the switching function that defines the switching hyperplanes corresponding to the areas A and B is held in the electronic control unit 40. Here, in order to follow the dynamic characteristics of the reference model, the armature 34
A setting mode of the electromagnetic force applied to the will be described.

First, from the above equations (c9) and (c10),
The time derivative of the vector e is as follows.

[0088]

[Equation 13] Here, the switching function σ = S · e is defined. In the sliding mode state, since the time derivative of the switching function σ is zero, the following equation holds.

[0089]

[Equation 14] Accordingly, the input in the sliding mode state (sliding mode input) is given by the following expression (c15).

[0090]

[Equation 15] In this equation (c15), the above equations (c9) and (c1
1), the switching hyperplane S shown in FIG. 5 is used, and the external force f
Is set to "0" to set the sliding mode input Ul. Here, since the external force f changes depending on the operating state of the engine, in the present embodiment, the sliding mode input Ul is set without adding the external force f, and the influence of the external force f reaches It will be absorbed by mode input.

In order to set the switching hyperplane so that the actual state quantity converges to the state quantity showing the dynamic characteristics of the reference model, the following restrictions are imposed on this setting. First, by substituting the above equation (c15) into the above equation (c14),
The following formula (c16) is obtained.

[0092]

[Equation 16] In this equation (c16), the matching condition Am-A =
Substituting B · K1, Bm = B · K2, and f = B · K3, the following formula (c17) is obtained.

[0093]

[Equation 17] In this equation (c17), if the switching function (switching matrix S) is set so that the coefficient matrix related to the deviation e between the state quantities is stable, the deviation e will converge to zero.

Here, the valve opening control of the exhaust valve 1 according to this embodiment will be described with reference to FIG. FIG. 6 is a flowchart showing a processing procedure of the control. This process is repeatedly executed every predetermined period.

In this series of processing, step 20
0, as in step 100 of FIG.
It is determined whether or not the displacement of is less than or equal to a predetermined threshold value, and if it is less than or equal to the threshold value, the process proceeds to step 210.

On the other hand, in step 210, it is judged whether the detected displacement of the valve body 2 belongs to the area A or the area B in FIG. 5, and the switching function of the corresponding area is electronically controlled. It is read from the memory or the like in the device 40.

Next, at step 220, a target displacement and a displacement velocity (target state amount) which are state amounts showing the dynamic characteristics of the reference model during the processing are calculated. These target state quantities are calculated based on the time from the start of valve opening control of the exhaust valve 1 (valve body 2) or the start of displacement of the exhaust valve 1 to the processing time. This calculation may be performed in the electronic control unit 40 based on the above formula (c10), and the electronic control unit 40 has a function of storing the target state quantity for each sampling time as a map. It may be configured.

Then, in step 230, the sliding mode input Ul is calculated from the above equation (c15) using the state quantity indicating the dynamic characteristics of the reference model, the switching function, and the detected displacement and displacement velocity. .

Further, in step 240, step 21
The arrival mode input Unl is calculated from the above equation (c6) using the switching function read at 0. Then, the processes of steps 250 to 280, which are the same as the processes of steps 140 to 170 in FIG. 4 described above, are performed, and this process is temporarily terminated.

In this embodiment described above, the following effects can be obtained. (4) In the model reference adaptive control, the deviation between the state quantity indicating the dynamic characteristics of the reference model and the actual state quantity is bound on the switching hyperplane, and the switching hyperplane is changed based on the displacement of the valve body 2. Set. As a result, it is possible to variably set the manner of convergence of the actual state quantity to the dynamic characteristics of the reference model according to the displacement of the valve body 2. Therefore, for example, energization control can be performed in consideration of the fact that the power applied to the exhaust valve 1 by the upper spring 24, the lower spring 14, the opening drive electromagnet 36, and the closing drive electromagnet 38 differs depending on the displacement.

(Third Embodiment) A third embodiment, in which the sliding mode control device according to the present invention is applied to a control device for an electromagnetically driven engine valve, will be described below as the first and second embodiments. The differences will be mainly described with reference to the drawings.

In each of the above embodiments, the arrival mode input (feedback input) Unl expressed by the above equation (c6) is used.
The feedback gain G of was set as a constant. In contrast,
In the present embodiment, this gain G is set to the engine speed NE,
It is variably set according to the engine load Q and the displacement x of the valve body 2. That is, the larger the engine speed NE and the engine load Q, the larger the disturbance applied to the valve body 2 and the like.
The larger the rotational speed NE and the engine load Q, the larger the gain G is set. Further, in the vicinity of the fully open position, the upper spring 24 and the lower spring 14 cause the valve element 2 to
Is increased in the direction of full closing, the gain G
To increase the attraction force toward the fully open position.

Further, in the present embodiment, when the actual state quantity is in the area near the switching hyperplane, the arrival mode input U is made to reach the switching hyperplane as compared with when it is in other areas.
Set the size of nl small. As a result, the arrival mode input Un is determined depending on whether the switching function is positive or negative.
By switching l, it is possible to suppress the occurrence of so-called chattering in which the actual state quantity causes high frequency vibration in the vicinity of the switching hyperplane. Specifically, in this embodiment, a smoothing function as shown in FIG. 7 is used.

The arrival mode input Unl of this embodiment set in this way is shown in the following expression (c18).

[0105]

[Equation 18] According to this embodiment described above, the following effects can be obtained.

(5) Gain G of arrival mode input Unl
Was variably set according to the engine rotation speed, the load, and the displacement of the valve body 2. As a result, the external force such as the in-cylinder pressure applied to the valve body 2 changes depending on the engine operating state, and the armature 34
The control can be performed in consideration of the electromagnetic force acting on the exhaust valve 1 and the urging force of the upper spring 24 and the lower spring 14 acting on the exhaust valve 1 depending on the displacement.

(6) When the actual state quantity is in the area near the switching hyperplane, the magnitude of the arrival mode input Unl is set smaller than the other areas to suppress the occurrence of chattering. You can

The above embodiments may be modified and implemented as follows. -In FIG. 5 above, two different switching hyperplanes are set according to the displacement of the valve body 2, but three or more switching hyperplanes may be set.

The manner of setting the reference model is not limited to that shown in FIG. For example, as shown in FIG. 8, there is no external force or damping term from the fully open position to the fully closed position, there is no electromagnetic force, and the state quantity of the valve body 2 is assumed to be transitioned only by the urging force of the upper spring 24 and the lower spring 14. Good.

Further, the valve body 2 assumed by the urging force of the upper spring 24 and the lower spring 14 is used.
The normative model set based on the transition of the state quantity is not limited to that shown in FIG. In short, the reference model may be set based on the transition of the state quantity of the valve body 2 under the assumption that there is no external force such as electromagnetic force or cylinder pressure. For example, the damping term between the movable part and the fixed part in the above equation (c8) may be added. FIG. 9 shows a reference model based on the transition of the valve body 2 due to the urging forces of the upper spring 24 and the lower spring 14 and the damping term except for the vicinity of the Low end face, and FIG. 2 for the Low end face and its vicinity. An example in which the same hyperplane is set as a reference model will be shown.

In the third embodiment, the arrival mode input Unl may be set small in at least the area near the area where the switching hyperplane is zero in the area where the arrival mode input Unl is positive. As a result, it is possible to prevent the control input U from taking a negative value when controlling the actual state quantity on the switching hyperplane.

By the way, the exhaust valve 1 shown in FIG.
It is controlled by the attraction control of the armature 34 by the opening drive electromagnet 36. That is, in the control input U shown in FIGS. 4 and 6, the opening drive electromagnet 36 is the armature 3
4 is generated by the electromagnetic force. Therefore, the electromagnetic force exerted on the armature 34 by the opening drive electromagnet 36 cannot apply a force in the direction opposite to the operation direction of the armature 34, so that the control input U is set to zero when the control input U is negative. . In this regard, by suppressing the control input U from becoming negative, the switching control by the electromagnetic force can be appropriately performed.

(Fourth Embodiment) The fourth embodiment in which the sliding mode control device according to the present invention is applied to a control device for an electromagnetically driven engine valve will be described below. The difference from the modification will be mainly described with reference to the drawings.

In each of the above-described embodiments and the modifications thereof, control is performed to displace from one displacement end to the other displacement end, in other words, from the fully closed position to the fully open position. On the other hand, in the present embodiment, the displacement control is performed from one displacement end to the other displacement end, in other words, from the fully closed position to the vicinity of the fully opened position, and the fixed control is performed near the fully opened position. In this way, by controlling the valve body 2 to be fixed near the fully open position, the armature 34 can be fixed before the armature 34 contacts the lower core 36C, and seating noise can be avoided. Furthermore, since the armature 34 does not contact the lower core 36C during each valve opening operation, the durability of these is also improved.

FIG. 10 shows a control mode of the exhaust valve 1 according to this embodiment. FIG. 10A is a time showing an example of a displacement mode of the valve body 2 from the fully closed position (Up end surface) to the displacement point (−xm) on the fully opened position (Low end surface) side. It is a chart. Further, FIG. 10 (b) shows the energizing current to the closing drive electromagnet 38, and FIG. 10 (c) shows the energizing current to the open drive electromagnet 36. That is, when the valve opening control is performed by stopping the supply of the holding current for holding the valve body 2 to the fully closed position to the closing drive electromagnet 38 at time t1, the valve body 2 is moved to the fully open position side. It will be displaced. Then, from time t2, the opening drive electromagnet 3
6 controls to attract the valve body 2 (armature 34). When the valve body 2 reaches the predetermined displacement point at time t3, the holding drive current for fixing the valve body 2 at the displacement point is supplied to the open drive electromagnet 36.

In order to perform such control, the reference model is set as shown in FIG. 11 in this embodiment. That is, here, the displacement point (-xm) for fixedly controlling the valve body 2
Up to the displacement point (-xl) on the fully closed side of the upper spring 24 and the lower spring 1 as shown in FIG.
The trajectory (curve) in which the state quantity of the valve element 2 changes due to the biasing force of 4 and the damping term is a reference model. And
In the vicinity of the displacement point (-xl), the displacement speed becomes zero at the displacement point (-xm) to be fixedly controlled, and the rate of change of the displacement speed with respect to the displacement is larger than that of the reference model of the curve. ) Is set as a reference model.

The valve opening control of the exhaust valve 1 using such a reference model may be performed in the same manner as in the second embodiment. That is, when the exhaust valve 1 is controlled so as to follow the dynamic characteristics of the reference model shown in FIG. 11 (model reference adaptive control), the mode of convergence of the actual state quantity to the state quantity of the reference model is set in the sliding mode. It may be controlled by control. Instead of this, as in the first embodiment, the sliding mode control may be performed by setting the hyperplane in contact with the reference model shown in FIG. 11 at the displacement point as the switching hyperplane.

Not limited to the reference model illustrated in FIG. 11, an appropriate reference model in which the displacement velocity is zero at the displacement point (-xm) to be fixedly controlled may be used. For example, as a reference model on the side of the fully closed position, as shown in FIG. 2 and the like, the reference model may be set as a locus in which the state quantity of the valve body 2 is transitioned by the urging forces of the upper spring 24 and the lower spring 14. Good.

According to this embodiment described above, the following effects can be obtained. (7) Since the exhaust valve 1 is fixedly controlled in the vicinity of the fully open position, the armature 34 can be fixed before the armature 34 contacts the lower core 36C, and seating noise can be reduced. Further, since the armature 34 does not contact the lower core 36C during each valve opening operation, it is possible to improve the durability of these.

The fourth embodiment may be modified and implemented as follows. The valve body 2 is not limited to one in which the valve body 2 is displaced from one displacement end side to the other displacement end side, in other words, from the fully closed position to the vicinity of the fully opened position, and the fixed control is performed near the fully opened position. For example, the control shown in FIG. 12 may be used. In this FIG. 12, the armature 34 is located at the upper core 38 in the fully closed position defined as the position where the valve body 2 is seated on the valve seat 16.
It is set so that it does not touch C. Therefore, the armature 34 and the upper core 3 when fully closed.
It is possible to avoid tapping sound between 8C.

Further, as shown in the example of FIG. 13, variable lift amount control may be performed. FIG. 13 shows an example in which the displacement of the valve element 2 is fixedly controlled on the side of the fully closed position with respect to the neutral position. Therefore, in this case, the closing drive electromagnet 38
Performs sliding mode control by the force of attracting the armature 34.

(Other Embodiments) In addition, there are the following elements that can be changed in common with each of the above embodiments.

While only the attraction force of the armature 34 by the opening drive electromagnet 36 is used when the valve is opened, the attraction force of the armature 34 by the closing drive electromagnet 38 may be used together.

The setting of the gain satisfying the reaching condition is not limited to the one using the Lyapunov function method, but the reaching law method or the like may be used.
Any method may be used, and the arrival mode input is not limited to those exemplified in the above equations (c6) and (c18).

-Instead of controlling the state quantity of the valve body 2 to converge on the switching hyperplane defined at each displacement point,
Control may be performed such that the state quantity converges on the switching hyperplane at the displacement point on the fully open position side by a predetermined amount from this displacement point. That is, for example, in the first embodiment,
As schematically shown in FIG. 14, sliding mode control is performed using a switching hyperplane at a displacement point xz that is a predetermined amount after the detected displacement xy of the valve body 2. This is because, for example, in the procedure shown in FIG. 4 above, after adding a predetermined displacement amount Δx to the actually detected displacement xy, the process from step 110 to step 140 is performed using this. Good. This makes it possible to perform control in consideration of the response delay of the electromagnetic force that accompanies actual control.

The sliding mode input may be calculated in consideration of the external force acting on the valve element 2. This is for example
In the exhaust valve 1, the in-cylinder pressure and the exhaust pressure are detected, and the force acting on the valve body 2 is calculated according to the pressure difference between them, and in the intake valve, the in-cylinder pressure and the intake pressure are detected. However, by calculating the force acting on the valve body according to these pressure differences, it is possible to take them into consideration. Alternatively, this disturbance may be estimated from the detectable state quantity, in other words, each of the differential pressures may be estimated. As this estimation method,
For example, it is possible to set an observer. This observer is set, for example, as a spring / mass vibration model that simulates the opening / closing operation of the valve body 2 excluding the external force f in the above equation (c8), and observes the internal state based on this model. . Thus, the external force f can be estimated from the detectable state quantity.

The structure of the engine valve is not limited to that shown in FIG. For example, permanent magnets may be provided in the electromagnets 36 and 38, and the magnetic force of these permanent magnets may be used to control the valve body 2 to be held in the fully open position or the fully closed position. In this case, at the time of valve opening control or valve closing control, a magnetic flux in the direction opposite to the magnetic flux of the permanent magnet that fixes the valve element 2 is generated in each electromagnet 36 and 38, and the permanent magnet that fixes and controls this valve element 2 is generated. Try to cancel the magnetic flux of. Also,
Even in such a configuration, the setting of the reference model based on the transition of the state quantity of the valve body 2 due to the urging force of the upper spring 24 and the lower spring 14 can be set as in the above embodiment and its modification. That is, (a) Displacement mode of the valve body 2 when the urging forces of the upper spring 24 and the lower spring 14 act. In other words, the displacement mode of the valve body 2 when the electromagnetic force of the permanent magnet, the external force f in the above equation (c8), or the damping term does not exist. (A) Displacement mode of the valve body 2 when the biasing forces of the upper spring 24 and the lower spring 14 and the damping term act. In other words, the displacement mode of the valve body 2 when the electromagnetic force of the permanent magnet or the external force f in the above equation (c8) does not exist.

Further, for example, for the lower spring 14 and the upper spring 24, these may be used as appropriate elastic members. Further, the exhaust valve 1 may be configured to include only an elastic member that urges one of the displacement ends of the exhaust valve 1. Further, the exhaust valve 1 may be drive-controlled by exerting a repulsive force between the armature and the electromagnet instead of exerting an attractive force between the armature and the electromagnet.

The means for detecting the displacement of the exhaust valve 1 is not limited to the above displacement amount sensor. For example, a sensor that detects the operating speed of the valve body 2, the valve shaft 4, the armature 34, or the like may be used. In this case, the valve displacement can be grasped by integrating the detected speed.

The switching hyperplane may be appropriately variably set according to the displacement of the valve body 2 without using the reference model. Further, not only the engine valve, but when displacing an arbitrary controlled object to which a biasing force is applied by an elastic member from one displacement end side to the other displacement end side, the state quantity of the controlled object is a switching hyperplane. The application of the present invention is effective when controlling the same control target so as to converge upward. At this time, the displacement region when the displacement is performed from one displacement end side to the other displacement end side does not necessarily have to be a one-dimensional space. Furthermore, at this time, the equation of motion of the physical system, the reference model, the control input, etc. are not limited to those that can be expressed by linear functions.

The setting means for variably setting the switching hyperplane according to the displacement of the controlled object is not limited to the one provided in the electronic control device.

[Brief description of drawings]

FIG. 1 is a diagram showing an overall configuration of a first embodiment in which a sliding mode control device according to the present invention is applied to an engine valve control device.

FIG. 2 is a diagram showing a setting manner of a switching hyperplane in the same embodiment.

FIG. 3 is a diagram showing a map defining a relationship between a gap, a control input, and an energization current amount.

FIG. 4 is a flowchart showing an exhaust valve opening control procedure according to the embodiment.

FIG. 5 is a diagram showing a setting mode of a switching hyperplane in a second embodiment in which the sliding mode control device according to the present invention is applied to a control device for an engine valve.

FIG. 6 is a flowchart showing an exhaust valve opening control procedure according to the embodiment.

FIG. 7 is a diagram showing a setting mode of arrival mode input in the third embodiment in which the sliding mode control device according to the invention is applied to an engine valve control device.

FIG. 8 is a diagram showing a reference model used in the modified examples of the first to third embodiments.

FIG. 9 is a view showing a reference model used in the modified examples of the first to third embodiments.

FIG. 10 is a time chart showing an exhaust valve control mode in a fourth embodiment in which the sliding mode control device according to the present invention is applied to an engine valve control device.

FIG. 11 is a view showing a reference model used in the same embodiment.

FIG. 12 is a time chart showing a control mode of an exhaust valve in a modified example of the same embodiment.

FIG. 13 is a time chart showing a control mode of an exhaust valve in a modified example of the same embodiment.

FIG. 14 is a view showing a further modified example of the first to fourth embodiments and their modified examples.

[Explanation of symbols]

1 ... Exhaust valve, 2 ... Valve body, 4 ... Valve shaft, 10 ... Cylinder head, 12 ... Lower retainer, 14 ... Lower spring, 16
... valve seat, 17 ... combustion chamber, 18 ... exhaust port, 20 ... upper cap, 22 ... upper retainer, 24 ... upper spring, 30 ... electromagnetic drive part, 34 ... armature, 36 ...
Open drive electromagnet, 36c ... lower coil, 36C ... lower core, 36h ... groove, 38 ... Close drive electromagnet, 38c ... upper coil, 38C ... upper core, 38h ... groove, 40 ... electronic control device, 42 ... displacement amount sensor.

   ─────────────────────────────────────────────────── ─── Continued front page    F-term (reference) 3G018 AB09 AB19 BA38 CA12 DA41                       DA42 DA70 EA02 EA11 EA22                       FA01 FA06 GA02                 3G084 BA23 DA04 EB02 EB13 EC04                       EC08 FA10 FA33                 3G092 AA11 DA01 DA02 DA07 DF05                       DG09 DG10 EA01 EA02 EA03                       EA04 EC02 EC06 FA09 HA11Z                       HA13Z HE01Z                 5H004 GB12 GB20 HA14 HB07 JB01                       KA74 KC39 LA12 LB06

Claims (10)

[Claims] 1. A switching hyperplane is set when a control target to which a biasing force is applied by an elastic member is displaced from one displacement end side to the other displacement end side, and the state quantity of the control target is set by this. A sliding mode control device for controlling the controlled object so as to converge on the switched hyperplane, characterized by comprising setting means for variably setting the switched hyperplane according to the displacement of the controlled object. Sliding mode controller. 2. The setting means variably sets the switching hyperplane as a hyperplane in contact with a preset reference model for the state quantity of the controlled object at its corresponding displacement point. Sliding mode controller. 3. The switching hyperplane is defined with respect to the deviation between the state quantity indicated by the dynamic characteristic and the actual state quantity of the controlled object when the controlled object follows the dynamic characteristic of the reference model. The sliding mode control device according to claim 1. 4. The sliding mode control device according to claim 2, wherein the reference model is set based on a transition of the state quantity of the control target by the urging force of the elastic member. 5. The sliding mode control device according to claim 4, wherein the reference model is set as a hyperplane in the vicinity of the other displacement end of the controlled object. 6. The sliding mode according to claim 2, wherein the reference model is set as a model derived from different physical characteristics for each displacement region divided into a plurality of the controlled objects. Control device. 7. The control of the controlled object based on the switching hyperplane variably set by the setting means is performed at a displacement point on the other displacement end side by a predetermined amount from a displacement point detected each time of the controlled object. The sliding mode control device according to any one of claims 2 to 6, which is performed based on a switching hyperplane that is variably set. 8. The control of the controlled object based on the switching hyperplane variably set by the setting means, when the state quantity of the controlled object is in the area near the variably set switching hyperplane, the other area is controlled. The sliding mode control device according to any one of claims 1 to 7, wherein an external force for reaching the switching hyperplane is made smaller than that in the above case. 9. The sliding mode control device according to claim 1, wherein the control target is an engine valve of an internal combustion engine including a valve body connected to an armature, and an electromagnetic force acting on the armature is used. A sliding mode control device, wherein the state quantity of the engine valve is controlled by sucking the valve body to the one displacement end side and the other displacement end side, respectively. 10. The sliding mode control device according to claim 9, wherein a feedback gain for causing the state quantity of the valve element to reach the switching hyperplane that is variably set is set to a rotational speed of the engine and an engine load, And a sliding mode control device further comprising means for variably setting in accordance with at least one of the displacements of the valve element.