JP2007115101A - Control device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To exhibit the control function to a control object including a saturated element to the limit without causing a wind-up phenomenon. <P>SOLUTION: A linear controller 5 generates a linear operation quantity Ueq for generating an operation quantity of the control object 2, and restricts the magnitude of the linear operation quantity Ueq, based on a nonlinear operation quantity Unl. Since it is not necessary to consider the magnitude of the nonlinear operation quantity Unl, a nonlinear controller 6 never causes wind-up. Consequently, since it is not necessary to consider the wind-up phenomenon in the nonlinear controller 6, the influence in saturation can be considered only in the linear controller 5. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a control device for controlling a controlled object having saturation characteristics.

In general, many control objects include one or more saturation elements (constraint conditions) such as the maximum driving force and torque of an actuator, the maximum current of an electric circuit, and the finite stroke length of a mechanism. If the control device is designed without taking such saturation elements into account, control output overshoot and system instability may occur. Against this background, the system designer considers the saturation value of the saturation element as a fixed value, and inserts an artificial saturation element using the fixed saturation value into the control compensator to control the saturation element. The apparatus is designed (see, for example, Patent Documents 1, 2, 3, and 4).
JP-A-5-143105 JP 2001-195102 A JP-A-1-173301 JP 2004-86858 A

  By the way, in general feedback control, a compensator (integral compensation element) including an integral operation is inserted in the control compensator to ensure followability, but the integral compensation element is inserted in the control compensator. When a saturation element is inserted into the control compensator, the integration calculation is continued even during saturation, so that the control variables such as position, speed, current, and temperature do not follow the target value for a long time. At the same time, a so-called wind-up phenomenon that overshoots or vibrates is likely to occur. From such a background, in a feedback control system, it is desired to provide a control device that can exhibit a control function to the limit for a controlled object having saturation characteristics without causing a windup phenomenon.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a control device capable of exerting a control function to the limit with respect to a control target including a saturated element without causing a windup phenomenon. Is to provide.

  In order to solve the above-described problem, a control device according to the present invention generates an operation amount of a control object based on control information including at least one of a target output value, an output value, and an operation amount of the control object. First intermediate value generating means for generating a first intermediate value for generating, a second intermediate value for generating an operation amount to be controlled based on the control information, and a first intermediate value based on the first intermediate value 2nd intermediate value generation means for limiting the size of the 2 intermediate value, and an operation amount generation that generates an operation amount of the control object based on the first intermediate value and the second intermediate value, and gives the operation amount generated to the control object Means.

  According to the control device of the present invention, the size of the second intermediate value for generating the control target operation amount is limited based on the first intermediate value for generating the control target operation amount. It is not necessary to consider the size of one intermediate value, and the windup phenomenon does not occur in the first intermediate value generating means. As a result, since it is not necessary to consider the windup phenomenon in the first intermediate value generating means, the influence at the time of saturation need only be considered in the second intermediate value generating means. Therefore, according to the control device of the present invention, the control function can be exerted to the limit with respect to the controlled object including the saturation element without causing the windup phenomenon.

  Hereinafter, the configuration of the control device according to the first to third embodiments of the present invention will be described with reference to the drawings.

[Configuration of control device]
As shown in FIG. 1, the control device 1 according to the first embodiment of the present invention outputs the difference between the target value and the output of the controlled object 2 as a control deviation, and is output from the comparator 3. An integration calculator (1 / s) that performs integral calculation of the control deviation, and an observer (observer) 4 that estimates the state variable vector using the target value, the integral calculation value of the control deviation, and the manipulated variable of the controlled object 2; Estimated by an observer 4 and a linear controller 5 that calculates a linear input for generating an operation amount of the control target 2 as a second intermediate value according to the target value, the integral calculation value of the control deviation, and the output of the control target 2. A non-linear controller 6 for calculating a non-linear input for generating an operation amount of the control object 2 according to the state variable vector as a first intermediate value, and an operation amount obtained by adding or subtracting the first intermediate value and the second intermediate value. Add to control object 2 as And a 7 gives the operation amount consisting of linear input and nonlinear input constituted by the feedback input to the controlled object 2 having a saturation characteristic. Note that the integral calculator (1 / s) may not be provided in the case of control in which the output of the control target 2 is desired to be 0, such as vibration control.

As shown in FIG. 1, the linear controller 5 includes a linear transfer function matrix N that generates a third intermediate value using the target value, the integral calculation value of the control deviation, and the output of the controlled object 2, and a limiter 8. A linear transfer function matrix R for generating a fourth intermediate value using the second intermediate value output from the second intermediate value, and an adder 9 for supplying a fifth intermediate value to the limiter 8 based on the third intermediate value and the fourth intermediate value. And a limiter 8 that outputs the second intermediate value while limiting the size based on the size of the fifth intermediate value, and is expressed by left irreducible decomposition, which is one method for constructing an anti-windup control system. ing. That is, when the linear controller 5 is expressed by a linear transfer function matrix C, the relationship shown in the following Expression 2 is established between the transfer function matrices C, N, and R.

  In the present embodiment, the nonlinear controller 6 and the linear controller 5 apply the sliding mode control theory, and the nonlinear input and the linear input calculated by the sliding mode control theory are the first intermediate value and the second intermediate value, respectively. Calculate as However, when the nonlinear control theory is applied to the nonlinear controller 6, the left irreducible decomposition expression applicable in the linear control theory cannot be used as it is.

In view of this, the control device 1 is provided with the following Equation 3 for calculating the equivalent control input Ueq, which is a linear operation amount in the sliding mode control, and Equation 4 below for calculating the nonlinear input Unl, which is a nonlinear operation amount. Only the linear equivalent control system is expressed by the left irreducible decomposition, and the anti-windup function is not added to the nonlinear control system.

  In this embodiment, the sliding mode control theory, which is one of nonlinear control theories, is applied. However, if the operation amount calculation method includes an operation amount calculated by a linear operation, the present invention provides: For example, when feed-forward control and linear feedback control are used together (in this case, the feed-forward control can use a map, a table, etc.), it can be applied to an arbitrary control system. Moreover, since this control does not ask | require continuous time and discrete time, it can be applied to either continuous time control theory or discrete time control theory.

  Further, the linear controller 5, the non-linear controller 6, and the adder 7 are respectively used as the second intermediate value generating means 12, the first intermediate value generating means 13, and the manipulated variable generating means 14 according to the present invention shown in FIG. Function. The transfer function matrix N, the transfer function matrix R, the limiter 8 and the adder 9 are respectively a third intermediate value generating means 21, a fourth intermediate value generating means 22 and a fifth intermediate value according to the present invention shown in FIG. It functions as the generating means 23 and the second intermediate value limiting means 24.

  And the control apparatus 1 which has such a structure demonstrates a control function to the limit with respect to the control object 2 which has a saturation characteristic, without producing a windup phenomenon by performing the operation amount production | generation process shown below. Make it possible. Hereinafter, with reference to the flowchart shown in FIG. 4, the operation of the control device 1 when executing the operation amount generation process will be described in detail.

[Operation amount generation processing]
The flowchart shown in FIG. 4 starts in response to the input of the target value, the control deviation, and the operation amount of the control target 2 to the observer 4, and the operation amount generation process proceeds to step S1.

  In the process of step S1, the observer 4 estimates the state variable vector X using the target value, the integral calculation value of the control deviation, the output of the controlled object 2, and the manipulated variable of the controlled object 2. Thereby, the process of step S1 is completed, and the operation amount generation process proceeds to the processes of step S2 and step S3.

  In the process of step S2, the linear controller 5 calculates the linear manipulated variable (equivalent control input) Ueq using the above-described equation 1. Thereby, the process of step S2 is completed, and the operation amount generation process proceeds to the process of step S5.

  In the process of step S3, the non-linear controller 6 calculates a non-linear manipulated variable (non-linear input) Unl using the above-described equation 2. Thereby, the process of step S3 is completed, and the operation amount generation process proceeds to the process of step S4. Note that the processing in step S2 and the processing in steps S3 and S4 are actually executed in parallel.

In the process of step S4, the nonlinear controller 6 uses the maximum value Umax and the minimum value Umin of the operation amount of the control target 2 preset by the adder 7 and the linear operation amount Ueq calculated by the process of step S2 as follows. By substituting into Equations 5 and 6, the maximum value Ueq_max and the minimum value Ueq_min of the linear manipulated variable Ueq are calculated. Thereby, the process of step S4 is completed, and the operation amount generation process proceeds to the process of step S5.

  In the process of step S5, the linear controller 5 determines whether or not the linear operation amount Ueq calculated by the process of step S2 is greater than or equal to the maximum value Ueq_max of the linear operation amount Ueq calculated by the process of step S4. . As a result of the determination, when the linear operation amount Ueq calculated by the process of step S2 is equal to or larger than the maximum value Ueq_max of the linear operation amount Ueq, the linear controller 5 is calculated by the process of step S2 as the process of step S6. After the linear operation amount Ueq is set to the maximum value Ueq_max of the linear operation amount Ueq, the operation amount generation process proceeds to step S7. On the other hand, when the linear manipulated variable Ueq calculated by the process of step S2 is less than the maximum value Ueq_max of the linear manipulated variable Ueq, the linear controller 5 changes the manipulated variable generation process from the process of step S5 to the process of step S7. Proceed.

  In the process of step S7, the linear controller 5 determines whether or not the linear operation amount Ueq calculated by the process of step S2 is less than or equal to the minimum value Ueq_min of the linear operation amount Ueq calculated by the process of step S4. . As a result of the determination, when the linear operation amount Ueq calculated by the process of step S2 is less than or equal to the minimum value Ueq_min of the linear operation amount Ueq, the linear controller 5 is calculated by the process of step S2 as the process of step S8. After the linear operation amount Ueq is set to the minimum value Ueq_min of the linear operation amount Ueq, the operation amount generation process proceeds to step S9. On the other hand, when the linear manipulated variable Ueq calculated by the process of step S2 is not less than or equal to the minimum value Ueq_min of the linear manipulated variable Ueq, the linear controller 5 advances the manipulated variable generation process from the process of step S7 to the process of step S9. .

  In the process of step S9, the adder 7 gives a value obtained by adding the linear manipulated variable Ueq computed by the linear controller 5 and the nonlinear manipulated variable Unl computed by the nonlinear controller 6 to the controlled object 2 as the manipulated variable U. . Thereby, the process of step S9 is completed, and the operation amount generation process returns to the process of step S1.

  As is clear from the above description, according to the control device 1 according to the first embodiment of the present invention, the linear controller 5 generates the linear operation amount Ueq for generating the operation amount of the control target 2. At the same time, as shown in FIG. 5, since the magnitude of the linear manipulated variable Ueq is limited based on the nonlinear manipulated variable Unl, it is not necessary to consider the magnitude of the nonlinear manipulated variable Unl. No longer occurs. As a result, it is not necessary to consider the wind-up phenomenon in the non-linear controller 6, so that only the linear controller 5 needs to consider the influence at the time of saturation. Therefore, according to the control device 1 according to the first embodiment of the present invention, as shown in FIG. 6, unlike the conventional control device, the control object including the saturation element is controlled without causing the windup phenomenon. The function can be demonstrated to the limit.

  Further, according to the control device 1 according to the first embodiment of the present invention, the linear controller 5 performs a linear operation in which the state variable vector X and the linear manipulated variable Ueq have a linear relationship, so that the nonlinear manipulated variable Unl and Even when the linear manipulated variable Ueq is mixed, it is possible to reduce the deterioration of the control performance at the time of saturation. Further, since the nonlinear controller 6 performs a nonlinear operation in which the state variable vector X and the nonlinear manipulated variable Unl have a nonlinear relationship, the nonlinear controller 6 can generate an manipulated variable to be nonlinearly controlled and has no saturation element in the nonlinear manipulated variable. There is no need to take windup into account when computing. As a result, it is not necessary to consider the deterioration of the control performance at the time of saturation in the non-linear operation, so that the design and implementation can be easily performed.

  Further, according to the control device 1 according to the first embodiment of the present invention, the nonlinear controller 6 generates the nonlinear manipulated variable Unl based on the nonlinear input calculated by the sliding mode control theory, and the linear controller 5 Since the linear manipulated variable Ueq is generated based on the linear input calculated by the sliding mode control theory, a robust control system can be configured, and a control system that can reduce the windup phenomenon easily with respect to the non-linear control theory. Can be configured.

  Further, according to the control device 1 according to the first embodiment of the present invention, the adder 7 sets the maximum value Umax and the minimum value Umin of the operation amount of the controlled object 2, and the non-linear controller 6 sets the non-linear operation amount Unl. Since the nonlinear operation amount Unl is generated so that the size of the linear operation amount Ueq falls within the set operation amount range, the size of the linear operation amount Ueq falls within the range including the limit width of zero. This also reduces the frequency with which the linear manipulated variable Ueq is saturated, and can reduce deterioration in control performance.

  Further, according to the control device 1 according to the first embodiment of the present invention, the linear controller 5 includes the limiter 8 that limits the magnitude of the linear operation amount Ueq, and the limiter 8 has the maximum operation amount of the control target 2. Since the linear operation amount Ueq is limited to be equal to or less than the value Umax, it is only necessary to include one saturation element when limiting the linear operation amount Ueq based on the nonlinear operation amount Unl. Even in the case of performing computation using, the computation time does not become long and can be easily implemented.

  Further, according to the control device 1 according to the first embodiment of the present invention, the limiter 8 uses the calculated linear operation amount Ueq as the maximum value Ueq_max when the calculated linear operation amount Ueq is equal to or larger than the maximum value Ueq_max. When the calculated linear operation amount Ueq is less than or equal to the minimum value Ueq_min, the calculated linear operation amount Ueq is set to the minimum value Ueq_min, and thus the linear operation amount Ueq is unnecessarily limited. Therefore, a desired control performance can be obtained when it is not saturated, and a control system that restricts only when it is saturated can be configured.

  Further, according to the control device 1 according to the first embodiment of the present invention, the limiter 8 changes the maximum value Ueq_max of the linear operation amount Ueq so that the operation amount of the control target 2 is equal to or less than the maximum value Umax. The sum of the linear operation amount Ueq and the nonlinear operation amount Unl does not exceed the maximum value Umax of the operation amount of the control target 2. Thereby, an operation amount is not excessively applied to the controlled object 2 and the frequency of deterioration of the control performance can be reduced.

  Further, according to the control device 1 according to the first embodiment of the present invention, the linear controller 5 outputs the linear transfer function matrix N that generates the third intermediate value using the state variable vector X and the limiter 8. A linear transfer function matrix R that generates a fourth intermediate value using the second intermediate value, and an adder 9 that provides the fifth intermediate value to the limiter 8 based on the third intermediate value and the fourth intermediate value. Since the limiter 8 generates the second intermediate value based on the fifth intermediate value, the limited information is fed back even when the second intermediate value is limited, and the performance deterioration is reduced with a simple structure. can do.

  Further, according to the control device 1 according to the first embodiment of the present invention, the linear controller 5 is expressed by the left irreducible decomposition, which is one method constituting the antiwindup control system, and the left irreducible decomposition is performed. Since the second intermediate value can be generated by using it, theoretically considering the saturation of the manipulated variable, the performance deterioration can be reduced with a simple structure.

  Further, according to the control device 1 according to the first embodiment of the present invention, the adder 7 generates the operation amount of the controlled object 2 by adding or subtracting the nonlinear operation amount Unl and the linear operation amount Ueq. The control system can be configured while maintaining the characteristics.

  Further, according to the control device 1 according to the first embodiment of the present invention, the manipulated variable is calculated using the integral calculation value of the control deviation. Therefore, when the output of the control target 2 is desired to follow the target value. In addition, the operation amount that eliminates the steady-state deviation can be calculated, and the output of the controlled object 2 can be appropriately controlled with respect to the target value.

In the manipulated variable generation process, the value of the constant K in Equation 2 is desirably set so that the constant K and the maximum value Umax and the minimum value Umin of the manipulated variable of the controlled object 2 satisfy Equation 7 shown below. . By setting the value of the constant K in this way, the maximum value Ueq_max and the minimum value Ueq_min of the linear manipulated variable calculated by the formulas 5 and 6 satisfy the following formula 8, and the linear manipulated variable Ueq is linear. Since the probability that the manipulated variable Ueq is greater than or equal to the maximum value Ueq_max or less than the minimum value Ueq_min is lowered, the manipulated variable of the control target 2 is less likely to be saturated, and the amount of deterioration in control performance due to saturation of the manipulated variable can be reduced. it can.

The control device according to the second embodiment of the present invention calculates an operation amount composed of a linear input composed of a feedback input and a nonlinear input composed of a feedforward input by a method similar to the above-described operation amount generation processing. Then, the calculated operation amount is given to the control object 2. In this embodiment, the non-linear controller 6 calculates a non-linear input using a neural network or fuzzy control theory. In this embodiment, when the nonlinear feedforward input by the neural network and the linear inputs calculated by a known linear feedback control system are Unn and Ufb, respectively, the nonlinear manipulated variable Unl and the linear manipulated variable in the first embodiment. The operation amount generation process described above is executed by replacing Ueq with Unn and Ufb, respectively. When calculating a nonlinear input using a neural network, the maximum value Usg_max and the minimum value of the sigmoid function are set so that the maximum value Usg_max and the minimum value Usg_min of the sigmoid function of the neural network satisfy the following formulas 9 to 11. By setting the value Usg_min, it is possible to configure a neural network that satisfies Equation 8 described above, and to reduce the amount of deterioration in control performance that accompanies saturation of the operation amount.

  The control device according to the third embodiment of the present invention is similar to the above-described manipulated variable generation processing in the manipulated variable composed of a linear input composed of feedback inputs and a nonlinear input composed of feedback inputs and feedforward inputs. The calculated operation amount is given to the control object 2. In the present embodiment, the non-linear controller 6 calculates a non-linear input using a neural network. The linear controller 5 calculates a linear input by applying a sliding mode control theory. In this embodiment, when the feedforward input by the neural network, the non-linear feedback input by the sliding mode control theory, and the linear feedback input by the sliding mode control theory are Unn ′, Unl ′, and Ueq ′, respectively, the first implementation is performed. The nonlinear manipulated variable Unl and the linear manipulated variable Ueq in the embodiment are replaced with Unn ′ + Unl ′ and Ueq ′, respectively, and the aforementioned manipulated variable generation process is executed.

  As mentioned above, although embodiment which applied the invention made | formed by this inventor was described, this invention is not limited with the description and drawing which make a part of indication of this invention by this embodiment. As described above, it is a matter of course that all other embodiments, examples, operation techniques, and the like made by those skilled in the art based on the above embodiments are included in the scope of the present invention.

It is a block diagram which shows the structure of the control apparatus used as embodiment of this invention. It is a functional block diagram which shows the structure of the control apparatus concerning this invention. It is a functional block diagram which shows the internal structure of the 2nd intermediate value production | generation means shown in FIG. It is a flowchart figure which shows the flow of the operation amount calculation process used as embodiment of this invention. It is a figure for demonstrating the outline | summary of the operation amount calculation process used as embodiment of this invention. It is a figure which shows the result of having simulated the relationship between the control amount of a control object, and the operation amount about the control apparatus used as embodiment of this invention, and the conventional control apparatus.

Explanation of symbols

1: Control device 2: Control object 3: Comparator 4: Observer 5: Linear controller 6: Non-linear controller 7, 9: Adder 8: Limiter N, R: Transfer function matrix

Claims (15)

A control device for controlling the output of a controlled object having saturation characteristics,
First intermediate value generation for generating a first intermediate value for generating an operation amount of the control target based on control information including at least one of the target output value, output value, and operation amount of the control target Means,
Second intermediate value generation for generating a second intermediate value for generating the manipulated variable of the control object based on the control information and limiting the size of the second intermediate value based on the first intermediate value Means,
A control device comprising: an operation amount generating unit that generates an operation amount of a control target based on the first intermediate value and the second intermediate value and gives the operation amount generated to the control target. The control device according to claim 1,
The control apparatus according to claim 2, wherein the second intermediate value generation means performs a linear operation in which the control information and the second intermediate value have a linear relationship. The control device according to claim 1 or 2,
The control apparatus according to claim 1, wherein the first intermediate value generating means performs a non-linear operation in which the control information and the first intermediate value have a non-linear relationship. It is a control device given in any 1 paragraph among Claims 1 thru / or 3, Comprising:
The first intermediate value generating means generates the first intermediate value based on a nonlinear input calculated by a sliding mode control theory, and the second intermediate value generating means is a linear input calculated by a sliding mode control theory. The second intermediate value is generated based on the control device. It is a control device given in any 1 paragraph among Claims 1 thru / or 4, Comprising:
The manipulated variable generation means generates an manipulated variable so that the magnitude of the manipulated variable is less than or equal to a predetermined saturated manipulated variable, and the first intermediate value generating means saturates the magnitude of the first intermediate value. A control device that generates a first intermediate value so as to be equal to or less than an operation amount. It is a control device given in any 1 paragraph among Claims 1 thru / or 5, Comprising:
The second intermediate value generating means includes second intermediate value limiting means for limiting the size of the generated second intermediate value, and the second intermediate value limiting means is configured such that the magnitude of the manipulated variable is a saturated manipulated variable. A control device that limits the size of the second intermediate value so as to have the following size. The control device according to claim 6,
The second intermediate value limiting means generates a maximum value of the second intermediate value as a saturated second intermediate value based on the saturation operation amount and the first intermediate value, and generates a second intermediate value. Is greater than the saturation second intermediate value, the second intermediate value is the saturation second intermediate value, and when the second intermediate value is less than the saturation second intermediate value, the generated second intermediate value A control device that outputs a value. The control device according to claim 7,
The control device characterized in that the second intermediate value limiting means changes the saturation second intermediate value so that the magnitude of the manipulated variable is equal to or less than the magnitude of the saturated manipulated variable. It is a control device given in any 1 paragraph among Claims 6 thru / or 8, Comprising:
The second intermediate value generating means, based on the control information, a third intermediate value generating means for generating a third intermediate value for generating the second intermediate value, and the second intermediate value generating means based on the second intermediate value. Fourth intermediate value generation means for generating a fourth intermediate value for generating a second intermediate value, and a fifth intermediate value generator for generating a fifth intermediate value based on the third intermediate value and the fourth intermediate value And the second intermediate value limiting means generates the second intermediate value based on the fifth intermediate value. The control device according to claim 9,
When the second intermediate value generating means is an arithmetic means expressed as a linear transfer function matrix C, the third intermediate value generating means is expressed using the left irreducible decomposition of the second intermediate value generating means expressed by Equation 1 below. And the fourth intermediate value generating means are a linear transfer function matrix N and a transfer function matrix R, respectively.

It is a control device given in any 1 paragraph among Claims 1 thru / or 10, Comprising:
The control apparatus, wherein the control information includes a calculation result obtained by integrating a difference between the target output value and the output value. It is a control device given in any 1 paragraph among Claims 1 thru / or 11, Comprising:
The control apparatus according to claim 1, wherein the first intermediate value generating means performs an operation based on a neural network or a fuzzy control theory and sets the operation result as the first intermediate value. It is a control device given in any 1 paragraph among Claims 1 thru / or 11, Comprising:
The control apparatus according to claim 1, wherein the first intermediate value generating means performs a feedforward calculation according to a neural network or a fuzzy control theory, and sets the calculation result as the first intermediate value. The control device according to claim 13,
The first intermediate value generating means generates the first intermediate value based on a feedforward calculation result calculated by a neural network or fuzzy control theory and a non-linear calculation result calculated by sliding mode control. Control device. It is a control device given in any 1 paragraph among Claims 1 thru / or 14, Comprising:
The operation amount generating means generates an operation amount by adding or subtracting the first intermediate value and the second intermediate value.

Images