JP2021042671A - Control device - Google Patents

Abstract

To actualize more efficient in-vehicle ticket inspection work by providing information of a site in which to inspect tickets on the basis of the reliability of each passenger.SOLUTION: A control device includes an operation part for calculating a desired output value y_d for a controlled object, an adjustment part for adjusting PID gain values for a PID controller so that a difference between the desired output value y_d and an output value y_0 for the controlled object is smaller, a prediction part for predicting an output value y_1 for the controlled object when using the adjusted PID gain values (Kp_1, Ki_1, Kd_1), and a determination part for prohibiting the adjusted PID gain values (Kp_1,Ki_1,Kd_1) from being applied to the controlled object when the predicted output value y_1 for the controlled object does not satisfy predetermined conditions.SELECTED DRAWING: Figure 1

Description

The present invention relates to a control device, and more particularly to a technique for adjusting parameters of the control device.

As a background technique of this technique, there is Japanese Patent Application Laid-Open No. 2018-112858 (Patent Document 1). In Patent Document 1, PID control is executed using a PID calculation unit that calculates a PID value, the PID value, a target value of a control target, and an actually measured value of the control target, and an operation on the control target is performed. A control including a PID control unit for calculating an amount, a combination of the target value and the actually measured value, and a determination unit for determining the stop of calculation of the PID value by using at least one of the PID values. The device is described (see claim 1).

Japanese Unexamined Patent Publication No. 2018-11285

However, in the above-mentioned prior art (Patent Document 1), the stop of calculation of the PID value is determined by using the result of actually applying the calculated PID value to the control target (actual measurement value of the output from the control target). However, before the calculated PID value is actually applied to the control target, the output from the control target is predicted, and the prediction result is not used to determine the stop of the calculation of the PID value.

A typical example of the invention disclosed in the present application is as follows. That is,
The PID gain value (Kp_1, Ki_1, Kd_1) of the PID controller so that the difference between the arithmetic unit that calculates the desired output value y_d of the control target and the desired output value y_d and the output value y_0 of the control target becomes small. ), A prediction unit that predicts the output value y_1 of the control target when the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are used, and a prediction unit that predicts the predicted output value of the control target. When y_1 does not satisfy a predetermined condition, it is characterized by including a determination unit for prohibiting the application of the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to the control target.

According to the present invention, it is possible to avoid adverse effects on the control system by verifying the PID gain value before application. Issues, configurations and effects other than those mentioned above will be clarified by the description of the following examples.

It is an overall view of the control apparatus in Examples 1-10. It is a system diagram of the control device in Examples 1-10. It is a figure which shows the control apparatus and the control object in Example 1. FIG. It is a figure which shows the processing of the desired output value calculation part in Examples 1-7. It is a figure which shows the process of the PID controller gain value adjustment part in Examples 1-7. It is a figure which shows the process of the control target output prediction part in Example 1. FIG. It is a figure which shows the process of the PID gain value application determination part in Examples 1 to 4. It is a figure which shows the processing of the PID controller in Examples 1-7. It is a figure which shows the control apparatus and the control object in Examples 2, 4 and 5. It is a figure which shows the processing of the control target output prediction part in Example 2. FIG. It is a figure which shows the control apparatus and the control object in Example 3. FIG. It is a figure which shows the process of the control target output prediction part in Example 3. FIG. It is a figure which shows the process of the control target output prediction part in Examples 4-5. It is a figure which shows the process of the PID gain value application determination part in Example 5. It is a figure which shows the control apparatus and the control object in Example 6. It is a figure which shows the process of the control target output prediction part in Example 6. It is a figure which shows the process of the PID gain value application determination part in Example 6. It is a figure which shows the control apparatus and the control object in Example 7. FIG. It is a figure which shows the process of the control target output prediction part in Example 7. It is a figure which shows the process of the PID gain value application determination part in Example 7. It is a figure which shows the control apparatus and the control object in Example 8. It is a figure which shows the processing of the desired output value calculation part in Example 8. It is a figure which shows the process of the PID controller gain value adjustment part in Example 8. It is a figure which shows the process of the control target output prediction part in Example 8. It is a figure which shows the process of the PID gain value application determination part in Example 8. FIG. It is a figure which shows the processing of the PID controller in Example 8. It is a figure which shows the control apparatus and the control object in Example 9. FIG. It is a figure which shows the processing of the desired output value calculation part in Example 9. It is a figure which shows the process of the PID controller gain value adjustment part in Example 9. It is a figure which shows the processing of the control target output prediction part in Example 9. It is a figure which shows the process of the PID gain value application determination part in Example 9. It is a figure which shows the processing of the PID controller in Example 9. It is a figure which shows the control apparatus and the control object in Example 10. It is a figure which shows the processing of the desired output value calculation part in Example 10. It is a figure which shows the process of the PID controller gain value adjustment part in Example 10. It is a figure which shows the process of the control target output prediction part in Example 10. It is a figure which shows the process of the PID gain value application determination part in Example 10. It is a figure which shows the processing of the PID controller in Example 10.

<Example 1>
In this embodiment, the PID control is performed so that the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target becomes small. Control target output that predicts the output value y_1 of the control target when the PID controller gain value adjustment unit 3 that adjusts the gain value of the device 6 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) are used. When the prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, it is prohibited to apply the adjusted PID gain values (Kp_1, Ki_1, Kd_1) (to the actual control). The control device 1 having the PID gain value application determination unit 5 is shown.

It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object.

Further, the control target output prediction unit 4 sets the output value y_0 of the control target and the control target r_0 estimated from the PID gain values (Kp_0, Ki_0, Kd_0) when the output value y_0 of the control target is obtained. The PID gain value (Kp_1) adjusted by calculating the transmission characteristic G of the control target and using an arbitrary control target r_a (for example, the control target r_0 when the output value y_0 of the control target is obtained) is calculated. , Ki_1) and the transmission characteristic G of the control target, and the output value of the control target when the adjusted PID gain value (Kp_1, Ki_1, Kd_1) at the control target r_a is used. Predict y_1. In each of the embodiments disclosed in the present specification, any control target may be in any form (for example, one represented by a step function or a ramp function). As long as the control target is represented by the same transfer function, the PID gain value adjusted by any type of control target can correspond to other types of control target.

Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control).

Further, the control device 1 is a device that controls the plant 7.

FIG. 1 is a diagram showing a central function of the control device 1.

In the control device 1, the desired output value calculation unit 2 calculates a desired profile of the output to be controlled. The PID controller gain value adjusting unit 3 adjusts the gain value of the PID controller 6. The control target output prediction unit 4 uses the PID gain value adjusted by the PID controller gain value adjustment unit 3 and the control target output used for adjusting the PID gain value to control the adjusted PID gain value. Predict the output from the controlled object before actually applying it. The PID gain value application determination unit 5 determines whether or not the control target output prediction value calculated by the control target output prediction unit 4 satisfies the predetermined condition, and if not, actually adjusts the adjusted PID gain value. The current PID gain value will continue to be used.

FIG. 2 is a system diagram of the control device 1.

The control device 1 has a storage device 11, a CPU 12, a ROM 13, a RAM 14, a data bus 15, an input circuit 16, an input / output port 17, and an output circuit 18 as hardware. The input circuit 16 processes a signal input from the outside. The signal input from the outside is, for example, a sensor output signal output from a sensor installed or connected to the control device 1. The signal input from the outside becomes an input signal via the input circuit 16 and is sent to the input / output port 17. Each input information sent to the input / output port 17 is stored in the RAM 14 or the storage device 11 via the data bus 15. At least one of the ROM 13 and the storage device 11 stores a program for executing a process described later, and the program is executed by the CPU 12. At that time, the values stored in at least one of the RAM 14 and the storage device 11 are appropriately used to perform the calculation. The information (value) to be sent to the outside of the calculation result is sent to the input / output port 17 via the data bus 15 and is sent to the output circuit 18 as an output signal. The output circuit 18 outputs an output signal to the outside. The output signal output to the outside is an actuator drive signal or the like transmitted to an actuator provided in the control target in order to make the control target move as desired.

A part of the processing performed by the CPU 12 by executing the program may be executed by another arithmetic unit (for example, hardware such as FPGA (Field Programable Gate Array) or ASIC (Application Specific Integrated Circuit)).

FIG. 3 is a diagram showing a control device 1 and a plant 7 controlled by the control device 1.

The PID controller 6 of this embodiment calculates an operation amount for controlling the temperature of the plant 7 (for example, a target valve opening degree for adjusting the steam temperature).

Further, the control target output prediction unit 4 obtains the PID gain value 1 adjusted by the PID controller gain value adjustment unit 3, the control target output used for adjusting the PID gain value 1, and the control target output. Using the PID gain value of 0, the output from the controlled object (the temperature of the plant 7) is predicted before the adjusted PID gain value 1 is actually applied to the controlled object. Here, the PID gain value 0 is a PID gain value adjusted immediately before the PID gain value 1, and is stored in at least one of the RAM 14 and the storage device 11.

The details of each process will be described below.

<Desired output value calculation unit (Fig. 4)>
The desired output value calculation unit 2 calculates a desired profile of the output to be controlled (temperature of the plant 7) based on the control target. Specifically, as shown in FIG. 4, a desired plant temperature profile De_Tdeg is calculated for the target plant temperature Tg_Tdeg, which is a control target, using, for example, a transfer function. The control target may be, for example, a step signal, a lamp signal, or the like.

<PID controller gain value adjustment unit (Fig. 5)>
The PID controller gain value adjusting unit 3 adjusts the gain value of the PID controller 6. Specifically, as shown in FIG. 5, the adjusted PID gain provisional value is determined so that the difference between the profile of the plant temperature Tdeg and the profile of the desired plant temperature De_Tdeg is minimized. The provisional values of the P gain value, the I gain value, and the D gain value of the obtained PID controller 6 are Kp_new, Ki_new, and Kd_new, respectively.

The evaluation function for evaluating the difference between the profile of the plant temperature Tdeg and the profile of the desired plant temperature De_Tdeg is, for example, the L2 norm of J2 = || Tdeg-De_Tdeg ||, or J1 = | Tdeg-De_Tdeg |. There is an L1 norm and so on. The PID gain value that minimizes J1 and J2 may be determined by, for example, IFT (Iterative Feedback Tuning), FRET (Fictitious Feedback Tuning), and the determination of the PID gain value results in an optimization problem. There are various methods for optimization such as Newton's method and Gauss-Newton's method, and there are many references, so they are not described in detail here.

This adjustment process may be performed either offline or online. In IFT, it is necessary to apply the PID gain value to the controlled object each time in the optimization process, but in FRIT, the optimization process can be processed offline.

<Controlled output prediction unit (Fig. 6)>
The control target output prediction unit 4 is a control target output (plant temperature Tdeg_0 profile) used for adjusting the PID gain value 1 (Kp_1, Ki_1, Kd_1) adjusted by the PID controller gain value adjustment unit 3 and the PID gain value 1. And the PID gain value 0 (Kp_0, Ki_0, Kd_0) when the plant temperature Tdeg_0 profile is obtained, the output from the control target is predicted before the adjusted PID gain value is actually applied to the control target. To do. Specifically, the process shown in FIG. 6 is executed.
-From the Kp_0, Ki_0, Kd_0 (PID gain value 0) and Tdeg_0 (plant temperature) profiles, Tg_VO_0_est (target valve opening estimated value), which is an input signal to the control target, is estimated. The difference between the pre-adjustment PID gain value (PID gain value 0) and the input value of the PID controller 6 between the control target and the output T_deg0 from the control target is calculated, and the output value of the PID controller 6 (to the control target) is calculated. The input signal) Tg_VO_est is calculated.
-The transfer function G of the control target is identified by using the Tg_VO_est (input signal to the control target) profile and the Tdeg_0 (output from the control target) profile. Since there is a lot of literature on how to identify the transfer function of the controlled object from the input / output signal of the controlled object, it will not be described in detail here.
Tdeg_1_est (plant temperature predicted value), which is an output predicted value from the controlled target at the time of an arbitrary or predetermined control target, using the identified transfer function G of the controlled object and Kp_1, Ki_1, Kd_1 (PID gain value 1). ) Is calculated.

Since the PID controller 6 and the control target are connected in series and the output from the control target forms a closed loop system that feeds back to the PID controller 6, if there is an equation for the PID controller 6 and an equation for the control target, , Tdeg_1_est (plant temperature predicted value), which is an output predicted value from the control target at the time of an arbitrary or predetermined control target, can be calculated.

<PID gain value application determination unit (Fig. 7)>
The PID gain value application determination unit 5 determines whether or not the overshoot amount and undershoot amount of the control target output prediction value calculated by the control target output prediction unit 4 are within a predetermined range, and determines whether the overshoot amount and undershoot amount are within a predetermined range. If at least one of the quantities is not within a predetermined range, the actual application of the adjusted PID gain value is prohibited and the current PID gain value is continued to be used. Specifically, the process shown in FIG. 7 is executed.
-The overshoot amount OS_Tdeg_1 and the undershoot amount US_Tdeg_1 with respect to the control target of Tdeg_1_est are obtained.
-In the case of OS_Tdeg_1 ≧ K_OS_Tdeg_1 or US_Tdeg_1 ≧ K_US_Tdeg_1, f_Kpid_NG = 1.
-In other cases, f_Kpid_NG = 0.

<PID controller (Fig. 8)>
The PID controller 6 executes the PID control process. Further, the PID gain value is determined according to the determination result by the PID gain value application determination unit 5. Specifically, the process shown in FIG. 8 is executed.
・ When f_Kpid_NG = 0,
The PID gain value is Kp_1, Ki_1, Kd_1 (PID gain value 1).
・ When f_Kpid_NG = 1,
The PID gain value is Kp_0, Ki_0, Kd_0 (PID gain value 0).

In the control device 1 of this embodiment, the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target is small. , PID controller gain value adjusting unit 3 for adjusting the gain value of the PID controller 6 and the output value y_1 of the controlled object when the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are used are predicted. When the control target output prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are applied (for actual control). It is composed of a PID gain value application determination unit 5 that prohibits this. It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object. Further, the control target output prediction unit 4 sets the output value y_0 of the control target and the control target r_0 estimated from the PID gain values (Kp_0, Ki_0, Kd_0) when the output value y_0 of the control target is obtained. The PID gain value (Kp_1) adjusted by calculating the transmission characteristic G of the control target and using an arbitrary control target r_a (for example, the control target r_0 when the output value y_0 of the control target is obtained) is calculated. , Ki_1) and the transmission characteristic G of the control target, and the output value of the control target when the adjusted PID gain value (Kp_1, Ki_1, Kd_1) at the control target r_a is used. Predict y_1. Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control). Further, the control device 1 is a device that controls the plant 7.

Therefore, the control target output prediction unit 4 uses the control target output (plant temperature) used for adjusting the PID gain value 1 and the PID gain value 0 when the control target output is obtained to control the control target (plant 7). The transmission property G of is identified. Further, the control target output prediction unit 4 uses the adjusted PID gain value 1 and the control target transmission characteristic G to output the output from the control target before actually applying the PID gain value 1 to the control target. Can be predicted.

As described above, for determining the PID gain value 1, for example, methods such as IFT (Iterative Feedback Tuning) and FRET (Fictitious Feedback Tuning) can be used, and these methods result in an optimization problem. However, in IFT, it is necessary to apply a PID gain value to the controlled object each time in the optimization process, but in FRIT, the optimization process can be processed offline. In particular, in the case of FRIT in which the PID gain value 1 can be optimized offline, the result (performance) of applying the optimized PID gain value 1 can also be predicted offline according to this embodiment.

Further, since the validity of the PID gain value 1 is determined based on whether at least one of the predicted overshoot amount and undershoot amount of the output from the control target is equal to or more than a predetermined value, the output from the control target is from the target value. It is possible to prevent deviations beyond expectations.

As described above, according to this embodiment, the output from the control target can be predicted before the calculated PID gain value is actually applied to the control target, and the validity of the PID gain value can be determined in advance using the prediction result. Can be verified. As a result of the verification, if the PID gain value is not valid, it is possible to prohibit the actual application of the PID gain value. Furthermore, by verifying the validity of the PID gain value from the viewpoint of the output to be controlled before applying the gain value, it is possible to avoid adverse effects on the control system and further stabilize the performance of the control system. The reliability of the system can be increased. More specifically, it is possible to achieve both efficient operation and safe operation of the plant 7.

<Example 2>
In this embodiment, the PID control is performed so that the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target becomes small. Control target output that predicts the output value y_1 of the control target when the PID controller gain value adjustment unit 3 that adjusts the gain value of the device 6 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) are used. When the prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, it is prohibited to apply the adjusted PID gain values (Kp_1, Ki_1, Kd_1) (to the actual control). The control device 1 having the PID gain value application determination unit 5 is shown.

It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object.

Further, the control target output prediction unit 4 has a transmission characteristic of the control target based on the output value y_0 of the control target and the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained. The PID gain value (Kp_1, Ki_1, Kd_1) adjusted by calculating G and using an arbitrary control target r_a (for example, the control target r_0 when the output value y_0 of the control target is obtained) and the control target Based on the transmission characteristic G, the output value y_1 of the control target when the adjusted PID gain value (Kp_1, Ki_1, Kd_1) at the control target r_a is used is predicted.

Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control).

Further, the control device 1 is a device that controls the plant 7.

FIG. 1 is a diagram showing a central function of the control device 1, and is the same as that of the first embodiment, and is not described in detail. FIG. 2 is a system diagram of the control device 1, which is the same as that of the first embodiment, and will not be described in detail.

FIG. 9 is a diagram showing a control device 1 and a plant 7 controlled by the control device 1. The PID controller 6 of this embodiment calculates an operation amount for controlling the temperature of the plant 7 (for example, a target valve opening degree for adjusting the steam temperature).

Further, the control target output prediction unit 4 has a PID gain value 1 adjusted by the PID controller gain value adjustment unit 3, a control target output used for adjusting the PID gain value 1, and an operation amount (target valve opening degree). And are used to predict the output from the controlled object before the adjusted PID gain value 1 is actually applied to the controlled object. Here, the PID gain value 0 is a PID gain value adjusted immediately before the PID gain value 1, and is stored in at least one of the RAM 14 and the storage device 11.

The details of each process will be described below.

<Desired output value calculation unit (Fig. 4)>
The desired output value calculation unit 2 calculates a desired profile of the output to be controlled (temperature of the plant 7) based on the control target. Specifically, as shown in FIG. 4, since it is the same as that of the first embodiment, it will not be described in detail.

<PID controller gain value adjustment unit (Fig. 5)>
The PID controller gain value adjusting unit 3 adjusts the gain value of the PID controller 6. Specifically, as shown in FIG. 5, since it is the same as that of the first embodiment, it will not be described in detail.

<Controlled output prediction unit (Fig. 10)>
The control target output prediction unit 4 is a control target output (plant temperature Tdeg_0 profile) used for adjusting the PID gain value 1 (Kp_1, Ki_1, Kd_1) adjusted by the PID controller gain value adjustment unit 3 and the PID gain value 1. And the manipulated variable corresponding to the plant temperature Tdeg_0 profile (target valve opening TgVO profile) is used to predict the output from the controlled object before the adjusted PID gain value is actually applied to the controlled object. Specifically, the process shown in FIG. 10 is executed.
-The transfer function G to be controlled is identified using the Tg_VO (target valve opening) profile and the Tdeg_0 (plant temperature) profile. Since there is a lot of literature on how to identify the transfer function of the controlled object from the input / output signal of the controlled object, it will not be described in detail here.
Tdeg_1_est (plant temperature predicted value), which is an output predicted value from the controlled target at the time of an arbitrary or predetermined control target, using the identified transfer function G of the controlled object and Kp_1, Ki_1, Kd_1 (PID gain value 1). ) Is calculated.

Since the PID controller 6 and the control target are connected in series and the output from the control target forms a closed loop system that feeds back to the PID controller 6, if there is an equation for the PID controller 6 and an equation for the control target, , Tdeg_1_est (plant temperature predicted value), which is an output predicted value from the control target at the time of an arbitrary or predetermined control target, can be calculated.

<PID gain value application determination unit (Fig. 7)>
The PID gain value application determination unit 5 determines whether or not the overshoot amount and undershoot amount of the control target output prediction value calculated by the control target output prediction unit 4 are within a predetermined range, and determines whether the overshoot amount and undershoot amount are within a predetermined range. If at least one of the quantities is not within a predetermined range, the actual application of the adjusted PID gain value is prohibited and the current PID gain value is continued to be used. Specifically, as shown in FIG. 7, since it is the same as that of the first embodiment, it will not be described in detail.

<PID controller (Fig. 8)>
The PID controller 6 executes the PID control process. Further, the PID gain value is determined according to the determination result by the PID gain value application determination unit 5. Specifically, as shown in FIG. 8, since it is the same as that of the first embodiment, it will not be described in detail.

In the control device 1 of this embodiment, the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target is small. , PID controller gain value adjusting unit 3 for adjusting the gain value of the PID controller 6 and the output value y_1 of the controlled object when the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are used are predicted. When the control target output prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are applied (for actual control). It has a PID gain value application determination unit 5 that prohibits this. It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object. Further, the control target output prediction unit 4 has a transmission characteristic of the control target based on the output value y_0 of the control target and the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained. The PID gain value (Kp_1, Ki_1, Kd_1) adjusted by calculating G and using an arbitrary control target r_a (for example, the control target r_0 when the output value y_0 of the control target is obtained) and the control target Based on the transmission characteristic G, the output value y_1 of the control target when the adjusted PID gain value (Kp_1, Ki_1, Kd_1) at the control target r_a is used is predicted. Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control). Further, the control device 1 is a device that controls the plant 7.

Therefore, the control target output prediction unit 4 uses the control target output (plant temperature) and the operation amount (target valve opening) used for adjusting the PID gain value 1 to transmit the transmission characteristic G of the control target (plant 7). To identify. Further, the control target output prediction unit 4 uses the adjusted PID gain value 1 and the control target transmission characteristic G to predict the output from the control target before actually applying the PID gain value 1 to the control target. it can.

As described above, for determining the PID gain value 1, for example, methods such as IFT and FRET can be used, and these methods result in an optimization problem, but IFT is used in the process of optimization. It is necessary to apply the PID gain value to the controlled object each time, but FRIT can process the optimization process offline. In particular, in the case of FRIT in which the PID gain value 1 can be optimized offline, the result (performance) of applying the optimized PID gain value 1 can also be predicted offline according to this embodiment.

Further, since the validity of the PID gain value 1 is determined based on whether at least one of the predicted overshoot amount and undershoot amount of the output from the control target is equal to or more than a predetermined value, the output from the control target is from the target value. It is possible to prevent deviations beyond expectations.

As described above, according to this embodiment, by verifying the validity of the PID gain value before applying the gain value from the viewpoint of the output to be controlled, it is possible to avoid adverse effects on the control system and further stabilize the performance of the control system. The reliability of the system can be improved. More specifically, it is possible to achieve both efficient operation and safe operation of the plant 7.

<Example 3>
In this embodiment, the PID control is performed so that the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target becomes small. Control target output that predicts the output value y_1 of the control target when the PID controller gain value adjustment unit 3 that adjusts the gain value of the device 6 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) are used. When the prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, it is prohibited to apply the adjusted PID gain values (Kp_1, Ki_1, Kd_1) (to the actual control). The control device 1 having the PID gain value application determination unit 5 is shown.

It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object.

Further, the control target output prediction unit 4 is adjusted at an arbitrary control target r_a by using the physical model representing the transmission characteristic G of the control target and the PID gain values (Kp_1, Ki_1, Kd_1). The output value y_1 of the control target when the PID gain values (Kp_1, Ki_1, Kd_1) are used is predicted.

Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control).

Further, the control device 1 is a device that controls the plant 7.

FIG. 1 is a diagram showing a central function of the control device 1, and is the same as that of the first embodiment, and is not described in detail. FIG. 2 is a system diagram of the control device 1, which is the same as that of the first embodiment, and will not be described in detail.

FIG. 11 is a diagram showing a control device 1 and a plant 7 controlled by the control device 1. The PID controller 6 of this embodiment calculates an operation amount for controlling the temperature of the plant 7 (for example, a target valve opening degree for adjusting the steam temperature).

Further, the control target output prediction unit 4 uses the PID gain value 1 adjusted by the PID controller gain value adjustment unit 3 to control the control target before actually applying the adjusted PID gain value 1 to the control target. Predict the output from. Here, the PID gain value 0 is a PID gain value adjusted immediately before the PID gain value 1, and is stored in at least one of the RAM 14 and the storage device 11.

The details of each process will be described below.

<Desired output value calculation unit (Fig. 4)>
The desired output value calculation unit 2 calculates a desired profile of the output to be controlled (temperature of the plant 7) based on the control target. Specifically, as shown in FIG. 4, since it is the same as that of the first embodiment, it will not be described in detail.

<PID controller gain value adjustment unit (Fig. 5)>
The PID controller gain value adjusting unit 3 adjusts the gain value of the PID controller 6. Specifically, as shown in FIG. 5, since it is the same as that of the first embodiment, it will not be described in detail.

<Controlled output prediction unit (Fig. 12)>
The controlled object output prediction unit 4 is adjusted by using the PID gain value 1 (Kp_1, Ki_1, Kd_1) adjusted by the PID controller gain value adjusting unit 3 and the transfer function G of the controlled object based on the physical model. Before actually applying the PID gain value to the control target, the output from the control target is predicted. Specifically, the process shown in FIG. 12 is executed.
-The transfer function G to be controlled is identified using the Tg_VO (target valve opening) profile and the Tdeg_0 (plant temperature) profile. Since there is a lot of literature on how to identify the transfer function of the controlled object from the input / output signal of the controlled object, it will not be described in detail here.
Tdeg_1_est (plant temperature prediction), which is an output prediction value from the control target at the time of an arbitrary or predetermined control target, using the transfer function G of the control target based on the physical model and Kp_1, Ki_1, Kd_1 (PID gain value 1). Value) is calculated.

The transfer function G is, for example, based on a physical model representing the heat balance. The physical formula for the heat balance is generally known and will not be described in detail here.

Since the PID controller 6 and the control target are connected in series and the output from the control target feeds back to the PID controller 6, a closed loop system is formed. , Tdeg_1_est (plant temperature predicted value), which is an output predicted value from the control target at the time of an arbitrary or predetermined control target, can be calculated.

<PID gain value application determination unit (Fig. 7)>
The PID gain value application determination unit 5 determines whether or not the overshoot amount and undershoot amount of the control target output prediction value calculated by the control target output prediction unit 4 are within a predetermined range, and determines whether the overshoot amount and undershoot amount are within a predetermined range. If at least one of the quantities is not within a predetermined range, the actual application of the adjusted PID gain value is prohibited and the current PID gain value is continued to be used. Specifically, as shown in FIG. 7, since it is the same as that of the first embodiment, it will not be described in detail.

<PID controller (Fig. 8)>
The PID controller 6 executes the PID control process. Further, the PID gain value is determined according to the determination result by the PID gain value application determination unit 5. Specifically, as shown in FIG. 8, since it is the same as that of the first embodiment, it will not be described in detail.

In the control device 1 of this embodiment, the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target is small. , PID controller gain value adjusting unit 3 for adjusting the gain value of the PID controller 6 and the output value y_1 of the controlled object when the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are used are predicted. When the control target output prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are applied (for actual control). It has a PID gain value application determination unit 5 that prohibits this. It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object. Further, the control target output prediction unit 4 is adjusted at an arbitrary control target r_a by using the physical model representing the transmission characteristic G of the control target and the PID gain values (Kp_1, Ki_1, Kd_1). The output value y_1 of the control target when the PID gain values (Kp_1, Ki_1, Kd_1) are used is predicted. Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control). Further, the control device 1 is a device that controls the plant 7.

Therefore, the control target output prediction unit 4 uses the adjusted PID gain value 1 and the physical model-based transmission characteristic G of the control target to control the control target before actually applying the PID gain value 1 to the control target. You can predict the output from.

As described above, for determining the PID gain value 1, for example, methods such as IFT and FRET can be used, and these methods result in an optimization problem, but IFT is used in the process of optimization. It is necessary to apply the PID gain value to the controlled object each time, but FRIT can process the optimization process offline. In particular, in the case of FRIT in which the PID gain value 1 can be optimized offline, the result (performance) of applying the optimized PID gain value 1 can also be predicted offline according to this embodiment.

Further, since the validity of the PID gain value 1 is determined based on whether at least one of the predicted overshoot amount and undershoot amount of the output from the control target is equal to or more than a predetermined value, the output from the control target is from the target value. It is possible to prevent deviations beyond expectations.

As described above, according to this embodiment, by verifying the validity of the PID gain value before applying the gain value from the viewpoint of the output to be controlled, it is possible to avoid adverse effects on the control system and further stabilize the performance of the control system. The reliability of the system can be improved. More specifically, it is possible to achieve both efficient operation and safe operation of the plant 7.

<Example 4>
In this embodiment, the PID control is performed so that the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target becomes small. Control target output that predicts the output value y_1 of the control target when the PID controller gain value adjustment unit 3 that adjusts the gain value of the device 6 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) are used. When the prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, it is prohibited to apply the adjusted PID gain values (Kp_1, Ki_1, Kd_1) (to the actual control). The control device 1 having the PID gain value application determination unit 5 is shown.

It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object.

Further, the control target output prediction unit 4 uses the PID from the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). When the input value e_1 of the controller 6 is obtained and the control target r_1 which is the sum of the output value y_0 of the control target and the input value e_1 of the PID controller 6 is set as the control target of the desired output value calculation unit 2. The desired output value yd_1 of the control target is calculated, the difference d_1 between the output value y_0 of the control target and the desired output value yd_1 is calculated, and the desired output of the control target at an arbitrary control target r_a is calculated. The value obtained by adding the calculated difference d_1 to the value yd_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a.

Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control).

Further, the control device 1 is a device that controls the plant 7.

FIG. 1 is a diagram showing a central function of the control device 1, and is the same as that of the first embodiment, and is not described in detail. FIG. 2 is a system diagram of the control device 1, which is the same as that of the first embodiment, and will not be described in detail.

FIG. 9 is a diagram showing a control device 1 and a plant 7 controlled by the control device 1, and is the same as in the second embodiment, and is not described in detail.

The details of each process will be described below.

<Desired output value calculation unit (Fig. 4)>
The desired output value calculation unit 2 calculates a desired profile of the output to be controlled (temperature of the plant 7) based on the control target. Specifically, as shown in FIG. 4, since it is the same as that of the first embodiment, it will not be described in detail.

<PID controller gain value adjustment unit (Fig. 5)>
The PID controller gain value adjusting unit 3 adjusts the gain value of the PID controller 6. Specifically, as shown in FIG. 5, since it is the same as that of the first embodiment, it will not be described in detail.

<Controlled output prediction unit (Fig. 13)>
The control target output prediction unit 4 is a control target output (plant temperature Tdeg_0 profile) used for adjusting the PID gain value 1 (Kp_1, Ki_1, Kd_1) adjusted by the PID controller gain value adjustment unit 3 and the PID gain value 1. And the manipulated variable corresponding to the plant temperature Tdeg_0 profile (target valve opening TgVO profile) is used to predict the output from the controlled object before the adjusted PID gain value is actually applied to the controlled object. Specifically, the process shown in FIG. 13 is executed.
-Using Tg_VO (target valve opening) and Kp_1, Ki_1, Kd_1 (PID gain value 1), e_1 (input value of PID controller 6) is obtained from the inverse function of the PID controller.
-Add e_1 (input value of PID controller 6) to Tdeg_0 (plant temperature) to obtain Tg_Tdeg_1 (target plant temperature).
-The transfer function used in the desired output value calculation unit 2 is applied to Tg_Tdeg_1 (target plant temperature) to obtain Tdeg_1_tilde (provisional plant temperature).
-Calculate d_1 (prediction error), which is the difference between Tdeg_0 (plant temperature) and Tdeg_1_tilde (provisional plant temperature).
-The transfer function used in the desired output value calculation unit 2 is applied to Tg_Tdeg_a (target plant temperature), which is an arbitrary or predetermined control target, to obtain De_Tdeg_a (desired plant temperature).
-Add d_1 (prediction error) to De_Tdeg_a (desired plant temperature) to calculate Tdeg_1_est (plant temperature predicted value), which is the output predicted value from the control target at the time of an arbitrary or predetermined control target.

<PID gain value application determination unit (Fig. 7)>
The PID gain value application determination unit 5 determines whether or not the overshoot amount and undershoot amount of the control target output prediction value calculated by the control target output prediction unit 4 are within a predetermined range, and determines whether the overshoot amount and undershoot amount are within a predetermined range. If at least one of the quantities is not within a predetermined range, the actual application of the adjusted PID gain value is prohibited and the current PID gain value is continued to be used. Specifically, as shown in FIG. 7, since it is the same as that of the first embodiment, it will not be described in detail.

<PID controller (Fig. 8)>
The PID controller 6 executes the PID control process. Further, the PID gain value is determined according to the determination result by the PID gain value application determination unit 5. Specifically, as shown in FIG. 8, since it is the same as that of the first embodiment, it will not be described in detail.

In the control device 1 of this embodiment, the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target is small. , PID controller gain value adjusting unit 3 for adjusting the gain value of the PID controller 6 and the output value y_1 of the controlled object when the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are used are predicted. When the control target output prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are applied (for actual control). It has a PID gain value application determination unit 5 that prohibits this. It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object. Further, the control target output prediction unit 4 uses the PID from the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). When the input value e_1 of the controller 6 is obtained and the control target r_1 which is the sum of the output value y_0 of the control target and the input value e_1 of the PID controller 6 is set as the control target of the desired output value calculation unit 2. The desired output value yd_1 of the control target is calculated, the difference d_1 between the output value y_0 of the control target and the desired output value yd_1 is calculated, and the desired output of the control target at an arbitrary control target r_a is calculated. The value obtained by adding the calculated difference d_1 to the value yd_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a. Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control). Further, the control device 1 is a device that controls the plant 7.

Therefore, in the control target output prediction unit 4, the PID gain value 1 is used by using the control target output (plant temperature) and the operation amount (target valve opening degree) used for adjusting the PID gain value 1 and the PID gain value 1. Can be predicted for output from the controlled object before it is actually applied to the controlled object.

As described above, for determining the PID gain value 1, for example, methods such as IFT and FRET can be used, and these methods result in an optimization problem, but IFT is used in the process of optimization. It is necessary to apply the PID gain value to the controlled object each time, but FRIT can process the optimization process offline. In particular, in the case of FRIT in which the PID gain value 1 can be optimized offline, the result (performance) of applying the optimized PID gain value 1 can also be predicted offline according to this embodiment.

Further, since the validity of the PID gain value 1 is determined based on whether at least one of the predicted overshoot amount and undershoot amount of the output from the control target is equal to or more than a predetermined value, the output from the control target is from the target value. It is possible to prevent deviations beyond expectations.

As described above, according to this embodiment, by verifying the validity of the PID gain value before applying the gain value from the viewpoint of the output to be controlled, it is possible to avoid adverse effects on the control system and further stabilize the performance of the control system. The reliability of the system can be improved. More specifically, it is possible to achieve both efficient operation and safe operation of the plant 7.

<Example 5>
In this embodiment, the PID control is performed so that the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target becomes small. Control target output that predicts the output value y_1 of the control target when the PID controller gain value adjustment unit 3 that adjusts the gain value of the device 6 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) are used. When the prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, it is prohibited to apply the adjusted PID gain values (Kp_1, Ki_1, Kd_1) (to the actual control). The control device 1 having the PID gain value application determination unit 5 is shown.

It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object.

Further, the control target output prediction unit 4 uses the PID from the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). When the input value e_1 of the controller 6 is obtained and the control target r_1 which is the sum of the output value y_0 of the control target and the input value e_1 of the PID controller 6 is set as the control target of the desired output value calculation unit 2. The desired output value yd_1 of the control target is calculated, the difference d_1 between the output value y_0 of the control target and the desired output value yd_1 is calculated, and the desired output of the control target at an arbitrary control target r_a is calculated. The value obtained by adding the calculated difference d_1 to the value yd_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a.

Further, when the response time until the predicted output value y_1 of the controlled object reaches the predetermined value is not within the predetermined range, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1, Kd_1). ) Is prohibited (for actual control).

Further, the control device 1 is a device that controls the plant 7.

FIG. 1 is a diagram showing a central function of the control device 1, and is the same as that of the first embodiment, and is not described in detail. FIG. 2 is a system diagram of the control device 1, which is the same as that of the first embodiment, and will not be described in detail.
FIG. 9 shows the control device 1 and the plant 7 controlled by the control device 1, but since it is the same as that of the second embodiment, it will not be described in detail.

The details of each process will be described below.

<Desired output value calculation unit (Fig. 4)>
The desired output value calculation unit 2 calculates a desired profile of the output to be controlled (temperature of the plant 7) based on the control target. Specifically, as shown in FIG. 4, since it is the same as that of the first embodiment, it will not be described in detail.

<PID controller gain value adjustment unit (Fig. 5)>
The PID controller gain value adjusting unit 3 adjusts the gain value of the PID controller 6. Specifically, as shown in FIG. 5, since it is the same as that of the first embodiment, it will not be described in detail.

<Controlled output prediction unit (Fig. 13)>
The control target output prediction unit 4 is a control target output (plant temperature Tdeg_0 profile) used for adjusting the PID gain value 1 (Kp_1, Ki_1, Kd_1) adjusted by the PID controller gain value adjustment unit 3 and the PID gain value 1. And the manipulated variable corresponding to the plant temperature Tdeg_0 profile (target valve opening TgVO profile) is used to predict the output from the controlled object before the adjusted PID gain value is actually applied to the controlled object. Specifically, as shown in FIG. 13, since it is the same as in the fourth embodiment, it will not be described in detail.

<PID gain value application determination unit (FIG. 14)>
The PID gain value application determination unit 5 determines whether or not the overshoot amount and undershoot amount of the control target output prediction value calculated by the control target output prediction unit 4 are within a predetermined range, and determines whether the overshoot amount and undershoot amount are within a predetermined range. If at least one of the quantities is not within a predetermined range, the actual application of the adjusted PID gain value is prohibited and the current PID gain value is continued to be used. Specifically, the process shown in FIG. 14 is executed.
-If the response time until the Tdeg_1_est reaches the predetermined value K1_Tdeg_1_est is not within the predetermined range, f_Kpid_NG = 1 is set.
-In other cases, f_Kpid_NG = 0.

<PID controller (Fig. 8)>
The PID controller 6 executes the PID control process. Further, the PID gain value is determined according to the determination result by the PID gain value application determination unit 5. Specifically, as shown in FIG. 8, since it is the same as that of the first embodiment, it will not be described in detail.

In the control device 1 of this embodiment, the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target is small. , PID controller gain value adjusting unit 3 for adjusting the gain value of the PID controller 6 and the output value y_1 of the controlled object when the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are used are predicted. When the control target output prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are applied (for actual control). It has a PID gain value application determination unit 5 that prohibits this. It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object. Further, the control target output prediction unit 4 uses the PID from the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). When the input value e_1 of the controller 6 is obtained and the control target r_1 which is the sum of the output value y_0 of the control target and the input value e_1 of the PID controller 6 is set as the control target of the desired output value calculation unit 2. The desired output value yd_1 of the control target is calculated, the difference d_1 between the output value y_0 of the control target and the desired output value yd_1 is calculated, and the desired output of the control target at an arbitrary control target r_a is calculated. The value obtained by adding the calculated difference d_1 to the value yd_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a. Further, when the response time until the predicted output value y_1 of the controlled object reaches the predetermined value is not within the predetermined range, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1, Kd_1). ) Is prohibited (for actual control). Further, the control device 1 is a device that controls the plant 7.

Therefore, in the control target output prediction unit 4, the PID gain value 1 is used by using the control target output (plant temperature) and the operation amount (target valve opening degree) used for adjusting the PID gain value 1 and the PID gain value 1. Can be predicted for output from the controlled object before it is actually applied to the controlled object.

As described above, for determining the PID gain value 1, for example, methods such as IFT and FRET can be used, and these methods result in an optimization problem, but IFT is used in the process of optimization. It is necessary to apply the PID gain value to the controlled object each time, but FRIT can process the optimization process offline. In particular, in the case of FRIT in which the PID gain value 1 can be optimized offline, the result (performance) of applying the optimized PID gain value 1 can also be predicted offline according to this embodiment.

Further, since the validity of the PID gain value 1 is judged based on whether or not the response time until the predicted output from the control target reaches a predetermined value is within the predetermined range, the responsiveness of the output from the control target is assumed. It is possible to prevent it from shifting more.

As described above, according to this embodiment, by verifying the validity of the PID gain value before applying the gain value from the viewpoint of the output to be controlled, it is possible to avoid adverse effects on the control system and further stabilize the performance of the control system. The reliability of the system can be improved. More specifically, it is possible to achieve both efficient operation and safe operation of the plant 7.

<Example 6>
In this embodiment, the PID control is performed so that the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target becomes small. Control target output that predicts the output value y_1 of the control target when the PID controller gain value adjustment unit 3 that adjusts the gain value of the device 6 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) are used. When the prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, it is prohibited to apply the adjusted PID gain values (Kp_1, Ki_1, Kd_1) (to the actual control). The control device 1 having the PID gain value application determination unit 5 is shown.

It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object.

Further, the control target output prediction unit 4 obtains the control target r_0 when the control target output value y_0 and the control target output value y_0 are obtained, and the PID gain when the control target output value y_0 is obtained. Based on the values (Kp_0, Ki_0, Kd_0), the transmission characteristic G of the control target is calculated, and the arbitrary control target r_a and the control target r_0 (with the control target r_0 when the output value y_0 of the control target is obtained) are adjusted. When the adjusted PID gain value (Kp_1, Ki_1, Kd_1) at the time of the control target r_a is used based on the obtained PID gain value (Kp_1, Ki_1, Kd_1) and the transmission characteristic G of the control target. The output value y_1 of the control target of the above is predicted.

Further, the PID gain value application determination unit 5 calculates the desired output value yd_b of the control target using the control target r_b when the predicted output value y_1 of the control target is obtained, and the predicted output value yd_b is calculated. When the error d_2 between the output value y_1 of the control target and the desired output value yd_b of the control target is calculated and the maximum value of the error d_2, the integrated value of the error d_2, or the sum of squares of the errors is equal to or greater than a predetermined value. It is prohibited to apply the adjusted PID gain values (Kp_1, Ki_1, Kd_1) (for actual control).

Further, the control device 1 is a device that controls the plant 7.

FIG. 1 is a diagram showing a central function of the control device 1, and is the same as that of the first embodiment, and is not described in detail. FIG. 2 is a system diagram of the control device 1, which is the same as that of the first embodiment, and will not be described in detail.

FIG. 15 is a diagram showing a control device 1 and a plant 7 controlled by the control device 1. The PID gain value application determination unit 5 of the present embodiment determines whether or not the controlled target output predicted value satisfies a predetermined condition by using the control target used for the output prediction in addition to the controlled target output predicted value. , When the controlled output predicted value does not satisfy the predetermined condition, it is prohibited to actually apply the adjusted PID gain value, and the current PID gain value is continued to be used. Other than that, it is the same as that of the first embodiment, and will not be described in detail.

The details of each process will be described below.

<Desired output value calculation unit (Fig. 4)>
The desired output value calculation unit 2 calculates a desired profile of the output to be controlled (temperature of the plant 7) based on the control target. Specifically, as shown in FIG. 4, since it is the same as that of the first embodiment, it will not be described in detail.

<PID controller gain value adjustment unit (Fig. 5)>
The PID controller gain value adjusting unit 3 adjusts the gain value of the PID controller 6. Specifically, as shown in FIG. 5, since it is the same as that of the first embodiment, it will not be described in detail.

<Controlled output prediction unit (Fig. 16)>
The control target output prediction unit 4 is a control target output (plant temperature Tdeg_0 profile) used for adjusting the PID gain value 1 (Kp_1, Ki_1, Kd_1) adjusted by the PID controller gain value adjustment unit 3 and the PID gain value 1. And the PID gain value 0 (Kp_0, Ki_0, Kd_0) when the plant temperature Tdeg_0 profile is obtained is used to predict the output from the control target before actually applying the adjusted PID gain value to the control target. To do. Specifically, it is the same as that of the first embodiment except for the process of outputting Tg_Tdeg_b (control target used for output prediction), which is shown in FIG. 16, and is not described in detail.

<PID gain value application determination unit (FIG. 17)>
The PID gain value application determination unit 5 calculates a desired output using the control target when the output predicted value from the control target is obtained, and the error between the value and the output predicted value from the control target is a predetermined condition. If the above is not satisfied, the actual application of the adjusted PID gain value is prohibited, and the current PID gain value is continued to be used. Specifically, the process shown in FIG. 17 is executed.
-The transfer function used in the desired output value calculation unit 2 is applied to Tg_Tdeg_b (control target used for output prediction) to obtain De_Tdeg_b (desired plant temperature).
The difference between Tdeg_1_est (predicted plant temperature) and De_Tdeg_b (desired plant temperature) is defined as d_2 (error).
When | d_2 | ≧ K_d_2, f_Kpid_NG = 1.
-In other cases, f_Kpid_NG = 0.
Alternatively, when the sum of squares of d_2 is equal to or greater than a predetermined value, f_Kpid_NG = 1 may be set.

<PID controller (Fig. 8)>
The PID controller 6 executes the PID control process. Further, the PID gain value is determined according to the determination result by the PID gain value application determination unit 5. Specifically, as shown in FIG. 8, since it is the same as that of the first embodiment, it will not be described in detail.

In the control device 1 of this embodiment, the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target is small. , PID controller gain value adjusting unit 3 for adjusting the gain value of the PID controller 6 and the output value y_1 of the controlled object when the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are used are predicted. When the control target output prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are applied (for actual control). It has a PID gain value application determination unit 5 that prohibits this. It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object. Further, the control target output prediction unit 4 uses the PID from the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). When the input value e_1 of the controller 6 is obtained and the control target r_1 which is the sum of the output value y_0 of the control target and the input value e_1 of the PID controller 6 is set as the control target of the desired output value calculation unit 2. The desired output value yd_1 of the control target is calculated, the difference d_1 between the output value y_0 of the control target and the desired output value yd_1 is calculated, and the desired output of the control target at an arbitrary control target r_a is calculated. The value obtained by adding the calculated difference d_1 to the value yd_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a. Further, the PID gain value application determination unit 5 calculates the desired output value yd_b of the control target at the control target r_b when the output value y_1 of the predicted control target is obtained, and the predicted control target. The error d_2 between the output value y_1 of the above and the desired output value yd_b of the control target is calculated, and when the maximum value of the error d_2, the integrated value of the error d_2, or the sum of squares of the errors is equal to or greater than a predetermined value, the adjustment is made. It is prohibited to apply the PID gain values (Kp_1, Ki_1, Kd_1) (for actual control). Further, the control device 1 is a device that controls the plant 7.

Therefore, in the control target output prediction unit 4, the PID gain value 1 is used by using the control target output (plant temperature) and the operation amount (target valve opening degree) used for adjusting the PID gain value 1 and the PID gain value 1. Can be predicted for output from the controlled object before it is actually applied to the controlled object.

As described above, for determining the PID gain value 1, for example, methods such as IFT and FRET can be used, and these methods result in an optimization problem, but IFT is used in the process of optimization. It is necessary to apply the PID gain value to the controlled object each time, but FRIT can process the optimization process offline. In particular, in the case of FRIT in which the PID gain value 1 can be optimized offline, the result (performance) of applying the optimized PID gain value 1 can also be predicted offline according to this embodiment.

Further, since the validity of the PID gain value 1 is determined based on the error between the predicted output from the control target and the desired output, it is possible to prevent the control error from becoming larger than expected over the entire control region. ..

As described above, according to this embodiment, by verifying the validity of the PID gain value before applying the gain value from the viewpoint of the output to be controlled, it is possible to avoid adverse effects on the control system and further stabilize the performance of the control system. The reliability of the system can be improved. More specifically, it is possible to achieve both efficient operation and safe operation of the plant 7.

<Example 7>
In this embodiment, the PID control is performed so that the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target becomes small. Control target output that predicts the output value y_1 of the control target when the PID controller gain value adjustment unit 3 that adjusts the gain value of the device 6 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) are used. When the prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, it is prohibited to apply the adjusted PID gain values (Kp_1, Ki_1, Kd_1) (to the actual control). The control device 1 having the PID gain value application determination unit 5 is shown.

It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object.

Further, the control target output prediction unit 4 uses the PID from the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). When the input value e_1 of the controller 6 is obtained and the control target r_1 which is the sum of the output value y_0 of the control target and the input value e_1 of the PID controller 6 is set as the control target of the desired output value calculation unit 2. The desired output value yd_1 of the control target is calculated, the difference d_1 between the output value y_0 of the control target and the desired output value yd_1 is calculated, and the desired output of the control target at an arbitrary control target r_a is calculated. The value obtained by adding the calculated difference d_1 to the value yd_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a.

Further, the PID gain value application determination unit 5 prohibits the application of the adjusted PID gain value (Kp_1, Ki_1, Kd_1) (for actual control) when the difference d_1 is equal to or greater than a predetermined value.

Further, the control device 1 is a device that controls the plant 7.

FIG. 1 is a diagram showing a central function of the control device 1, and is the same as that of the first embodiment, and is not described in detail. FIG. 2 is a system diagram of the control device 1, which is the same as that of the first embodiment, and will not be described in detail.

FIG. 18 is a diagram showing a control device 1 and a plant 7 controlled by the control device 1. The PID gain value application determination unit 5 of this embodiment also uses the control target output prediction value and the prediction error to determine whether or not the control target output prediction value satisfies a predetermined condition, and the control target output prediction value is predetermined. If the conditions are not met, the actual application of the adjusted PID gain value is prohibited and the current PID gain value is continued to be used. Other than that, it is the same as in Example 4, so it will not be described in detail.

The details of each process will be described below.

<Desired output value calculation unit (Fig. 4)>
The desired output value calculation unit 2 calculates a desired profile of the output to be controlled (temperature of the plant 7) based on the control target. Specifically, as shown in FIG. 4, since it is the same as that of the first embodiment, it will not be described in detail.

<PID controller gain value adjustment unit (Fig. 5)>
The PID controller gain value adjusting unit 3 adjusts the gain value of the PID controller 6. Specifically, as shown in FIG. 5, since it is the same as that of the first embodiment, it will not be described in detail.

<Controlled output prediction unit (Fig. 19)>
The control target output prediction unit 4 is a control target output (plant temperature Tdeg_0 profile) used for adjusting the PID gain value 1 (Kp_1, Ki_1, Kd_1) adjusted by the PID controller gain value adjustment unit 3 and the PID gain value 1. And the manipulated variable corresponding to the plant temperature Tdeg_0 profile (target valve opening TgVO profile), the output from the controlled object is predicted before the adjusted PID gain value is actually applied to the controlled object. Specifically, it is shown in FIG. 19, but since it is the same as in the fourth embodiment except for the process of outputting the error d_1, it will not be described in detail.

<PID gain value application determination unit (FIG. 20)>
When the error d_1 does not satisfy the predetermined condition, the PID gain value application determination unit 5 prohibits the actual application of the adjusted PID gain value, and continues to use the current PID gain value. Specifically, the process shown in FIG. 20 is executed.
When | d_1 | ≧ K_d_1, f_Kpid_NG = 1.
-In other cases, f_Kpid_NG = 0.

Further, the PID gain value application determination unit 5 may set f_Kpid_NG = 1 when the sum of squares of d_1 is equal to or greater than a predetermined value.

<PID controller (Fig. 8)>
The PID controller 6 executes the PID control process. Further, the PID gain value is determined according to the determination result by the PID gain value application determination unit 5. Specifically, as shown in FIG. 8, since it is the same as that of the first embodiment, it will not be described in detail.

In the control device 1 of this embodiment, the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target is small. , PID controller gain value adjusting unit 3 for adjusting the gain value of the PID controller 6 and the output value y_1 of the controlled object when the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are used are predicted. When the control target output prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are applied (for actual control). It has a PID gain value application determination unit 5 that prohibits this. It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object. Further, the control target output prediction unit 4 uses the PID from the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). When the input value e_1 of the controller 6 is obtained and the control target r_1 which is the sum of the output value y_0 of the control target and the input value e_1 of the PID controller 6 is set as the control target of the desired output value calculation unit 2. The desired output value yd_1 of the control target is calculated, the difference d_1 between the output value y_0 of the control target and the desired output value yd_1 is calculated, and the desired output of the control target at an arbitrary control target r_a is calculated. The value obtained by adding the calculated difference d_1 to the value yd_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a. Further, the PID gain value application determination unit 5 prohibits the application of the adjusted PID gain value (Kp_1, Ki_1, Kd_1) (for actual control) when the difference d_1 is equal to or greater than a predetermined value. Further, the control device 1 is a device that controls the plant 7.

The control target output prediction unit 4 controls the PID gain value 1 by using the control target output (plant temperature) and the operation amount (target valve opening) used for adjusting the PID gain value 1 and the PID gain value 1. The output from the controlled object can be predicted before it is actually applied to.

As described above, for determining the PID gain value 1, for example, methods such as IFT and FRET can be used, and these methods result in an optimization problem, but IFT is used in the process of optimization. It is necessary to apply the PID gain value to the controlled object each time, but FRIT can process the optimization process offline. In particular, in the case of FRIT in which the PID gain value 1 can be optimized offline, the result (performance) of applying the optimized PID gain value 1 can also be predicted offline according to this embodiment.

Further, since the validity of the PID gain value 1 is determined based on the error between the predicted output from the control target and the desired output, it is possible to prevent the control error from becoming larger than expected over the entire control region. ..

As described above, according to this embodiment, by verifying the validity of the PID gain value before applying the gain value from the viewpoint of the output to be controlled, it is possible to avoid adverse effects on the control system and further stabilize the performance of the control system. The reliability of the system can be improved. More specifically, it is possible to achieve both efficient operation and safe operation of the plant 7.

<Example 8>
In this embodiment, the PID control is performed so that the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target becomes small. Control target output that predicts the output value y_1 of the control target when the PID controller gain value adjustment unit 3 that adjusts the gain value of the device 6 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) are used. When the prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, it is prohibited to apply the adjusted PID gain values (Kp_1, Ki_1, Kd_1) (to the actual control). The control device 1 having the PID gain value application determination unit 5 is shown.

It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object.

Further, the control target output prediction unit 4 uses the PID from the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). When the input value e_1 of the controller 6 is obtained and the control target r_1 which is the sum of the output value y_0 of the control target and the input value e_1 of the PID controller 6 is set as the control target of the desired output value calculation unit 2. The desired output value yd_1 of the control target is calculated, the difference d_1 between the output value y_0 of the control target and the desired output value yd_1 is calculated, and the desired output of the control target at an arbitrary control target r_a is calculated. The value obtained by adding the calculated difference d_1 to the value yd_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a.

Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control).

Further, the control device 1 is a device that controls the autonomous driving vehicle 8.

FIG. 1 is a diagram showing a central function of the control device 1, and is the same as that of the first embodiment, and is not described in detail. FIG. 2 is a system diagram of the control device 1, which is the same as that of the first embodiment, and will not be described in detail.

FIG. 21 is a diagram showing a control device 1 and an autonomous driving vehicle 8 controlled by the control device 1. The PID controller 6 of this embodiment calculates an operation amount (for example, a target throttle opening degree, a target steering angle, etc.) for controlling the speed of the autonomous driving vehicle 8.

Further, the control target output prediction unit 4 has a PID gain value 1 adjusted by the PID controller gain value adjustment unit 3, a control target output (speed) used for adjusting the PID gain value 1, and an operation amount (target throttle). (Opening) is used to predict the output from the controlled object before the adjusted PID gain value 1 is actually applied to the controlled object. Here, the PID gain value 0 is a PID gain value adjusted immediately before the PID gain value 1, and is stored in at least one of the RAM 14 and the storage device 11.

The details of each process will be described below.

<Desired output value calculation unit (FIG. 22)>
The desired output value calculation unit 2 calculates a desired profile of the output to be controlled (speed of the autonomous driving vehicle 8) based on the control target. Specifically, as shown in FIG. 22, a desired velocity profile De_VSP is calculated for the target velocity Tg_VSP, which is a control target, by using, for example, a transfer function. The control target may be, for example, a step signal, a lamp signal, or the like.

<PID controller gain value adjustment unit (Fig. 23)>
The PID controller gain value adjusting unit 3 adjusts the gain value of the PID controller 6. Specifically, as shown in FIG. 23, the adjusted PID gain provisional value is determined so that the difference between the velocity VSP_0 profile and the desired velocity De_VSP profile is minimized. The provisional values of the P gain value, the I gain value, and the D gain value of the obtained PID controller 6 are Kp_1, Ki_1, and Kd_1, respectively.

The evaluation function for evaluating the difference between the velocity VSP_0 profile and the desired velocity De_VSP profile includes, for example, the L2 norm in which J2 = || VSP_0-De_Tdeg ||, or the L1 norm in which J1 = | VSP_0-De_Tdeg | ..

As described above, the PID gain value that minimizes J1 and J2 may be determined by, for example, IFT, FRIT, etc., and the determination of the PID gain value results in an optimization problem. There are various methods for optimization such as Newton's method and Gauss-Newton's method, and there are many references, so they are not described in detail here.

This adjustment process may be performed either offline or online. In IFT, it is necessary to apply the PID gain value to the controlled object each time in the optimization process, but in FRIT, the optimization process can be processed offline.

<Controlled output prediction unit (Fig. 24)>
The control target output prediction unit 4 includes a PID gain value 1 (Kp_1, Ki_1, Kd_1) adjusted by the PID controller gain value adjustment unit 3 and a control target output (speed VSS_0 profile) used for adjusting the PID gain value 1. Using the manipulated variable corresponding to the velocity VSP_0 profile (target velocity TgVSP profile), the output from the controlled object is predicted before the adjusted PID gain value is actually applied to the controlled object. Specifically, the process shown in FIG. 24 is executed.
-Using Tg_VO (target throttle opening) and Kp_1, Ki_1, Kd_1 (PID gain value 1), e_1 (input value of PID controller 6) is obtained from the inverse function of the PID controller.
-Add e_1 (input value of PID controller 6) to VSP_0 (speed) to obtain Tg_VSP_1 (target speed).
-The transfer function used in the desired output value calculation unit 2 is applied to Tg_VSP_1 (target speed) to obtain VSP_1_tilde (provisional speed).
-Calculate d_1 (prediction error), which is the difference between VSP_0 (speed) and VSP_1_tilde (provisional speed).
-The transfer function used in the desired output value calculation unit 2 is applied to Tg_VSP_a (target speed), which is an arbitrary or predetermined control target, to obtain De_VSP_a (desired speed).
-Add d_1 (prediction error) to De_VSP_a (desired speed) to calculate VSP_1_est (speed prediction value), which is the output prediction value from the control target at the time of an arbitrary or predetermined control target.

<PID gain value application determination unit (FIG. 25)>
The PID gain value application determination unit 5 determines whether or not the overshoot amount and undershoot amount of the control target output prediction value calculated by the control target output prediction unit 4 are within a predetermined range, and determines whether the overshoot amount and undershoot amount are within a predetermined range. If at least one of the quantities is not within a predetermined range, the actual application of the adjusted PID gain value is prohibited and the current PID gain value is continued to be used. Specifically, the process shown in FIG. 25 is executed.
-The overshoot amount OS_VSP_1 and the undershoot amount US_VSP_1 with respect to the control target of VSP_1_est are obtained.
-In the case of OS_VSP_1 ≧ K_OS_VSP_1 or US_VSP_1 ≧ K_US_VSP_1, f_Kpid_NG = 1.
-In other cases, f_Kpid_NG = 0.

<PID controller (Fig. 26)>
The PID controller 6 executes the PID control process. Further, the PID gain value is determined according to the determination result by the PID gain value application determination unit 5. Specifically, the process shown in FIG. 26 is executed.
・ When f_Kpid_NG = 0,
The PID gain value is Kp_1, Ki_1, Kd_1 (PID gain value 1).
・ When f_Kpid_NG = 1,
The PID gain value is Kp_0, Ki_0, Kd_0 (PID gain value 0).

In the control device 1 of this embodiment, the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target is small. , PID controller gain value adjusting unit 3 for adjusting the gain value of the PID controller 6 and the output value y_1 of the controlled object when the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are used are predicted. When the control target output prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are applied (for actual control). It has a PID gain value application determination unit 5 that prohibits this. It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object. Further, the control target output prediction unit 4 uses the PID from the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). When the input value e_1 of the controller 6 is obtained and the control target r_1 which is the sum of the output value y_0 of the control target and the input value e_1 of the PID controller 6 is set as the control target of the desired output value calculation unit 2. The desired output value yd_1 of the control target is calculated, the difference d_1 between the output value y_0 of the control target and the desired output value yd_1 is calculated, and the desired output of the control target at an arbitrary control target r_a is calculated. The value obtained by adding the calculated difference d_1 to the value yd_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a. Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control). Further, the control device 1 is a device that controls the autonomous driving vehicle 8.

Therefore, the control target output prediction unit 4 uses the adjusted PID gain value 1 and the control target output (speed) and the operation amount (target throttle opening degree) used for adjusting the PID gain value 1 to obtain the PID gain. The output from the controlled object can be predicted before the value 1 is actually applied to the controlled object.

As described above, for determining the PID gain value 1, for example, methods such as IFT and FRET can be used, and these methods result in an optimization problem, but IFT is used in the process of optimization. It is necessary to apply the PID gain value to the controlled object each time, but FRIT can process the optimization process offline. In particular, in the case of FRIT in which the PID gain value 1 can be optimized offline, the result (performance) of applying the optimized PID gain value 1 can also be predicted offline according to this embodiment.

Further, since the validity of the PID gain value 1 is determined based on whether at least one of the predicted overshoot amount and undershoot amount of the output from the control target is equal to or more than a predetermined value, the output from the control target is from the target value. It is possible to prevent deviations beyond expectations.

As described above, according to this embodiment, by verifying the validity of the PID gain value before applying the gain value from the viewpoint of the output to be controlled, it is possible to avoid adverse effects on the control system and further stabilize the performance of the control system. The reliability of the system can be improved. More specifically, it is possible to achieve both efficient control and safe control of the autonomous driving vehicle 8.

<Example 9>
In this embodiment, the PID control is performed so that the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target becomes small. Control target output that predicts the output value y_1 of the control target when the PID controller gain value adjustment unit 3 that adjusts the gain value of the device 6 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) are used. When the prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, it is prohibited to apply the adjusted PID gain values (Kp_1, Ki_1, Kd_1) (to the actual control). The control device 1 having the PID gain value application determination unit 5 is shown.

It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object.

Further, the control target output prediction unit 4 uses the PID from the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). When the input value e_1 of the controller 6 is obtained and the control target r_1 which is the sum of the output value y_0 of the control target and the input value e_1 of the PID controller 6 is set as the control target of the desired output value calculation unit 2. The desired output value yd_1 of the control target is calculated, the difference d_1 between the output value y_0 of the control target and the desired output value yd_1 is calculated, and the desired output of the control target at an arbitrary control target r_a is calculated. The value obtained by adding the calculated difference d_1 to the value yd_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a.

Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control).

Further, the control device 1 is a device that controls the robot 9.

FIG. 1 is a diagram showing a central function of the control device 1, and is the same as that of the first embodiment, and is not described in detail. FIG. 2 is a system diagram of the control device 1, which is the same as that of the first embodiment, and will not be described in detail.

FIG. 27 is a diagram showing a control device 1 and a robot 9 controlled by the control device 1. The PID controller 6 of this embodiment calculates an operation amount (for example, a target motor current amount, etc.) for controlling the position of the robot 9.

Further, the control target output prediction unit 4 has a PID gain value 1 adjusted by the PID controller gain value adjustment unit 3, a control target output (position) used for adjusting the PID gain value 1, and an operation amount (target motor). The amount of current) is used to predict the output from the controlled object before the adjusted PID gain value 1 is actually applied to the controlled object.

The details of each process will be described below.

<Desired output value calculation unit (FIG. 28)>
The desired output value calculation unit 2 calculates a desired profile of the output (position of the robot 9) to be controlled based on the control target. Specifically, as shown in FIG. 28, a desired position profile De_POS is calculated for the target position Tg_POS, which is a control target, by using, for example, a transfer function. The control target may be, for example, a step signal, a lamp signal, or the like.

<PID controller gain value adjustment unit (Fig. 29)>
The PID controller gain value adjusting unit 3 adjusts the gain value of the PID controller 6. Specifically, as shown in FIG. 29, the adjusted PID gain provisional value is determined so that the difference between the position POS_0 profile and the desired position De_POS profile is minimized. The provisional values of the P gain value, the I gain value, and the D gain value of the obtained PID controller 6 are Kp_1, Ki_1, and Kd_1, respectively.

The evaluation function for evaluating the difference between the position POS_0 profile and the desired position De_POS profile includes, for example, the L2 norm of J2 = || POS_0-De_POS ||, or the L1 norm of J1 = | POS_0-De_POS |. ..

As described above, the PID gain value that minimizes J1 and J2 may be determined by, for example, IFT, FRIT, etc., and the determination of the PID gain value results in an optimization problem. There are various methods for optimization such as Newton's method and Gauss-Newton's method, and there are many references, so they are not described in detail here.

This adjustment process may be performed either offline or online. In IFT, it is necessary to apply the PID gain value to the controlled object each time in the optimization process, but in FRIT, the optimization process can be processed offline.

<Controlled output prediction unit (Fig. 30)>
The control target output prediction unit 4 has a control target output (position POS_0 profile) and a control target output (position POS_0 profile) used for adjusting the PID gain value 1 (Kp_1, Ki_1, Kd_1) adjusted by the PID controller gain value adjustment unit 3 and the PID gain value 1. Using the manipulated amount corresponding to the position POS_0 profile (target motor current amount Tg_Ic profile), the output from the controlled object is predicted before the adjusted PID gain value is actually applied to the controlled object. Specifically, the process shown in FIG. 30 is executed.
-Using Tg_Ic (target motor current amount) and Kp_1, Ki_1, Kd_1 (PID gain value 1), e_1 (input value of PID controller 6) is obtained from the inverse function of the PID controller.
-Add e_1 (input value of PID controller 6) to POS_0 (position) to obtain Tg_POS_1 (target position).
-The transfer function used in the desired output value calculation unit 2 is applied to Tg_POS_1 (target position) to obtain POS_1_tilde (provisional position value).
-Calculate d_1 (prediction error), which is the difference between POS_0 (position) and POS_1_tilde (provisional position).
-The transfer function used in the desired output value calculation unit 2 is applied to Tg_POS_a (target position), which is an arbitrary or predetermined control target, to obtain De_POS_a (desired position).
-Add d_1 (prediction error) to De_POS_a (desired position) to calculate POS_1_est (position prediction value), which is an output prediction value from the control target at the time of an arbitrary or predetermined control target.

<PID gain value application determination unit (FIG. 31)>
The PID gain value application determination unit 5 determines whether or not the overshoot amount and undershoot amount of the control target output prediction value calculated by the control target output prediction unit 4 are within a predetermined range, and determines whether the overshoot amount and undershoot amount are within a predetermined range. If at least one of the quantities is not within a predetermined range, the actual application of the adjusted PID gain value is prohibited and the current PID gain value is continued to be used. Specifically, the process shown in FIG. 31 is executed.
-The overshoot amount OS_POS_1 and the undershoot amount US_POS_1 with respect to the control target of POS_1_est are obtained.
-In the case of OS_POS_1 ≧ K_OS_POS_1 or US_POS_1 ≧ K_US_POS_1, f_Kpid_NG = 1.
-In other cases, f_Kpid_NG = 0.

<PID controller (Fig. 32)>
The PID controller 6 executes the PID control process. Further, the PID gain value is determined according to the determination result by the PID gain value application determination unit 5. Specifically, the process shown in FIG. 32 is executed.
・ When f_Kpid_NG = 0,
The PID gain value is Kp_1, Ki_1, Kd_1 (PID gain value 1).
・ When f_Kpid_NG = 1,

The PID gain value is Kp_0, Ki_0, Kd_0 (PID gain value 0).

In the control device 1 of this embodiment, the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target is small. , PID controller gain value adjusting unit 3 for adjusting the gain value of the PID controller 6 and the output value y_1 of the controlled object when the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are used are predicted. When the control target output prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are applied (for actual control). It has a PID gain value application determination unit 5 that prohibits this. It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object. Further, the control target output prediction unit 4 uses the PID from the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). When the input value e_1 of the controller 6 is obtained and the control target r_1 which is the sum of the output value y_0 of the control target and the input value e_1 of the PID controller 6 is set as the control target of the desired output value calculation unit 2. The desired output value yd_1 of the control target is calculated, the difference d_1 between the output value y_0 of the control target and the desired output value yd_1 is calculated, and the desired output of the control target at an arbitrary control target r_a is calculated. The value obtained by adding the calculated difference d_1 to the value yd_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a. Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control). Further, the control device 1 is a device that controls the robot 9.

Therefore, the control target output prediction unit 4 uses the adjusted PID gain value 1 and the control target output (position) and the operation amount (target motor current amount) used for adjusting the PID gain value 1 to obtain the PID gain. The output from the controlled object can be predicted before the value 1 is actually applied to the controlled object.

As described above, for determining the PID gain value 1, for example, methods such as IFT and FRET can be used, and these methods result in an optimization problem, but IFT is used in the process of optimization. It is necessary to apply the PID gain value to the controlled object each time, but FRIT can process the optimization process offline. In particular, in the case of FRIT in which the PID gain value 1 can be optimized offline, the result (performance) of applying the optimized PID gain value 1 can also be predicted offline according to this embodiment.

As described above, according to this embodiment, by verifying the validity of the PID gain value before applying the gain value from the viewpoint of the output to be controlled, it is possible to avoid adverse effects on the control system and further stabilize the performance of the control system. The reliability of the system can be improved. More specifically, both efficient operation and safe operation of the robot 9 can be achieved at the same time.

<Example 10>
In this embodiment, the PID control is performed so that the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target becomes small. Control target output that predicts the output value y_1 of the control target when the PID controller gain value adjustment unit 3 that adjusts the gain value of the device 6 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) are used. When the prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, it is prohibited to apply the adjusted PID gain values (Kp_1, Ki_1, Kd_1) (to the actual control). The control device 1 having the PID gain value application determination unit 5 is shown.

It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object.

Further, the control target output prediction unit 4 uses the PID from the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). When the input value e_1 of the controller 6 is obtained and the control target r_1 which is the sum of the output value y_0 of the control target and the input value e_1 of the PID controller 6 is set as the control target of the desired output value calculation unit 2. The desired output value yd_1 of the control target is calculated, the difference d_1 between the output value y_0 of the control target and the desired output value yd_1 is calculated, and the desired output of the control target at an arbitrary control target r_a is calculated. The value obtained by adding the calculated difference d_1 to the value yd_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a.

Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control).

Further, the control device 1 is a device that controls the drone 10.

FIG. 1 is a diagram showing a central function of the control device 1, and is the same as that of the first embodiment, and is not described in detail. FIG. 2 is a system diagram of the control device 1, which is the same as that of the first embodiment, and will not be described in detail.

FIG. 33 is a diagram showing a control device 1 and a drone 10 controlled by the control device 1. The PID controller 6 of this embodiment calculates an operation amount (for example, the rotation speed of each rotor) for controlling the position of the drone 10.

Further, the control target output prediction unit 4 includes a PID gain value 1 adjusted by the PID controller gain value adjustment unit 3, a control target output (position) used for adjusting the PID gain value 1, and an operation amount (each rotor). The output from the controlled object is predicted before the adjusted PID gain value 1 is actually applied to the controlled object.

The details of each process will be described below.

<Desired output value calculation unit (FIG. 34)>
The desired output value calculation unit 2 calculates a desired profile of the output to be controlled (flying speed of the drone 10) based on the control target. Specifically, as shown in FIG. 34, a desired flight speed profile De_FSP is calculated for the target flight speed Tg_FSP, which is a control target, by using, for example, a transfer function. The control target may be, for example, a step signal, a lamp signal, or the like.

<PID controller gain value adjustment unit (Fig. 35)>
The PID controller gain value adjusting unit 3 adjusts the gain value of the PID controller 6. Specifically, as shown in FIG. 35, the adjusted PID gain provisional value is determined so that the difference between the flight speed FSP_0 profile and the desired position De_FSP profile is minimized. The provisional values of the P gain value, the I gain value, and the D gain value of the obtained PID controller 6 are Kp_1, Ki_1, and Kd_1, respectively.

The evaluation function for evaluating the difference between the flight speed FSP_0 profile and the desired position De_FSP profile is, for example, the L2 norm in which J2 = || FSP_0-De_FSP ||, or the L1 norm in which J1 = | FSP_0-De_FSP | is there.

As described above, the PID gain value that minimizes J1 and J2 may be determined by, for example, IFT, FRIT, etc., and the determination of the PID gain value results in an optimization problem. There are various methods for optimization such as Newton's method and Gauss-Newton's method, and there are many references, so they are not described in detail here.

This adjustment process may be performed either offline or online. In IFT, it is necessary to apply the PID gain value to the controlled object each time in the optimization process, but in FRIT, the optimization process can be processed offline.

<Controlled output prediction unit (Fig. 36)>
The control target output prediction unit 4 is a control target output (flight speed FSP_0 profile) used for adjusting the PID gain value 1 (Kp_1, Ki_1, Kd_1) adjusted by the PID controller gain value adjustment unit 3 and the PID gain value 1. And the manipulation amount corresponding to the flight speed FSP_0 profile (target rotor rotation speed Tg_RPM profile) is used to predict the output from the controlled object before the adjusted PID gain value is actually applied to the controlled object. Specifically, the process shown in FIG. 36 is executed.
-Using Tg_RPM (target rotor rotation speed) and Kp_1, Ki_1, Kd_1 (PID gain value 1), e_1 (input value of PID controller 6) is obtained from the inverse function of the PID controller.
-Add e_1 (input value of PID controller 6) to FSP_0 (flight speed) to obtain Tg_FSP_1 (target flight speed).
-The transfer function used in the desired output value calculation unit 2 is applied to Tg_FSP_1 (target flight speed) to obtain FSP_1_tilde (provisional speed value).
-Find d_1 (prediction error), which is the difference between FSP_0 (flight speed) and FSP_1_tilde (provisional flight speed).
-The transfer function used in the desired output value calculation unit 2 is applied to Tg_FSP_a (target flight speed), which is an arbitrary or predetermined control target, to obtain De_FSP_a (desired flight speed).
-Add d_1 (prediction error) to De_FSP_a (desired flight speed) to calculate FSP_1_est (prediction value of flight speed), which is an output predicted value from a controlled object at an arbitrary or predetermined control target.

<PID gain value application determination unit (FIG. 37)>
The PID gain value application determination unit 5 determines whether or not the overshoot amount and undershoot amount of the control target output prediction value calculated by the control target output prediction unit 4 are within a predetermined range, and determines whether the overshoot amount and undershoot amount are within a predetermined range. If at least one of the quantities is not within a predetermined range, the actual application of the adjusted PID gain value is prohibited and the current PID gain value is continued to be used. Specifically, the process shown in FIG. 37 is executed.
-The overshoot amount OS_FSP_1 and the undershoot amount US_FSP_1 with respect to the control target of FSP_1_est are obtained.
-In the case of OS_FSP_1 ≧ K_OS_FSP_1 or US_FSP_1 ≧ K_US_FSP_1, f_Kpid_NG = 1.
-In other cases, f_Kpid_NG = 0.

<PID controller (Fig. 38)>
The PID controller 6 executes the PID control process. Further, the PID gain value is determined according to the determination result by the PID gain value application determination unit 5. Specifically, the process shown in FIG. 38 is executed.
・ When f_Kpid_NG = 0,
The PID gain value is Kp_1, Ki_1, Kd_1 (PID gain value 1).
・ When f_Kpid_NG = 1,
The PID gain value is Kp_0, Ki_0, Kd_0 (PID gain value 0).

In the control device 1 of this embodiment, the difference between the desired output value calculation unit 2 that calculates the desired output value yd of the control target and the desired output value yd and the output value y_0 of the control target is small. , PID controller gain value adjusting unit 3 for adjusting the gain value of the PID controller 6 and the output value y_1 of the controlled object when the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are used are predicted. When the control target output prediction unit 4 and the predicted output value y_1 of the control target do not satisfy the predetermined conditions, the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are applied (for actual control). It has a PID gain value application determination unit 5 that prohibits this. It also has a storage unit that stores the output value y_0 of the control target. The control target output prediction unit 4 uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1, Kd_1). ) Is used to predict the output value y_1 of the controlled object. Further, the control target output prediction unit 4 uses the PID from the output value u_0 of the PID controller 6 when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). When the input value e_1 of the controller 6 is obtained and the control target r_1 which is the sum of the output value y_0 of the control target and the input value e_1 of the PID controller 6 is set as the control target of the desired output value calculation unit 2. The desired output value yd_1 of the control target is calculated, the difference d_1 between the output value y_0 of the control target and the desired output value yd_1 is calculated, and the desired output of the control target at an arbitrary control target r_a is calculated. The value obtained by adding the calculated difference d_1 to the value yd_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a. Further, when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value, the PID gain value application determination unit 5 determines the adjusted PID gain value (Kp_1, Ki_1). , Kd_1) is prohibited from being applied (for actual control). Further, the control device 1 is a device that controls the drone 10.

Therefore, the control target output prediction unit 4 controls the PID gain value 1 by using the control target output (speed) and the operation amount (target speed) used for adjusting the PID gain value 1 and the PID gain value 1. The output from the controlled object can be predicted before it is actually applied to.

As described above, for determining the PID gain value 1, for example, methods such as IFT and FRET can be used, and these methods result in an optimization problem, but IFT is used in the process of optimization. It is necessary to apply the PID gain value to the controlled object each time, but FRIT can process the optimization process offline. In particular, in the case of FRIT in which the PID gain value 1 can be optimized offline, the result (performance) of applying the optimized PID gain value 1 can also be predicted offline according to this embodiment.

As described above, according to this embodiment, by verifying the validity of the PID gain value before applying the gain value from the viewpoint of the output to be controlled, it is possible to avoid adverse effects on the control system and further stabilize the performance of the control system. The reliability of the system can be improved. More specifically, it is possible to achieve both efficient operation and safe operation of an air vehicle such as a drone 10.

The present invention is not limited to the above-described embodiment, and includes various modifications and equivalent configurations within the scope of the attached claims. For example, the above-described examples have been described in detail in order to explain the present invention in an easy-to-understand manner, and the present invention is not necessarily limited to those having all the described configurations. Further, a part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Further, the configuration of another embodiment may be added to the configuration of one embodiment. In addition, other configurations may be added / deleted / replaced with respect to a part of the configurations of each embodiment.

Further, each of the above-described configurations, functions, processing units, processing means, etc. may be realized by hardware by designing a part or all of them by, for example, an integrated circuit, and the processor realizes each function. It may be realized by software by interpreting and executing the program to be executed.

Information such as programs, tables, and files that realize each function can be stored in a memory, hard disk, storage device such as SSD (Solid State Drive), or recording medium such as IC card, SD card, DVD, BD, etc. it can.

In addition, the control lines and information lines indicate those that are considered necessary for explanation, and do not necessarily indicate all the control lines and information lines necessary for implementation. In practice, it can be considered that almost all configurations are interconnected.

1 Control device 2 Desired output value calculation unit 3 PID controller Gain value adjustment unit 4 Control target output prediction unit 5 PID gain value application judgment unit 6 PID controller 7 Power generation plant 8 Self-driving car 9 Robot 10 Drone 11 Control device Storage device 12 Controller CPU
13 Control device ROM
14 Controller RAM
15 Control device data bus 16 Control device input circuit 17 Control device input / output port 18 Control device output circuit

Claims (11)

An arithmetic unit that calculates the desired output value y_d to be controlled, and
An adjusting unit that adjusts the PID gain value (Kp_1, Ki_1, Kd_1) of the PID controller so that the difference between the desired output value y_d and the output value y_0 to be controlled becomes small.
A prediction unit that predicts the output value y_1 of the control target when the adjusted PID gain values (Kp_1, Ki_1, Kd_1) are used.
When the predicted output value y_1 of the control target does not satisfy the predetermined condition, the determination unit for prohibiting the application of the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to the control target is provided. A control device characterized by. The control device according to claim 1.
A storage unit for storing the output value y_0 to be controlled is provided.
The prediction unit uses at least the stored output value y_0 and the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to use the adjusted PID gain value (Kp_1, Ki_1, Kd_1). A control device for predicting an output value y_1 of the control target when the control target is present. The control device according to claim 1.
The prediction unit
Using the output value y_0 of the control target and the control target r_0 estimated from the PID gain value when the output value y_0 is obtained, the transmission characteristic G of the control target is calculated.
The adjusted PID gain value (Kp_1, Ki_1, Kd_1) at the control target r_a was used based on the PID gain value adjusted using the arbitrary control target r_a and the calculated transmission characteristic G. A control device for predicting the output value y_1 of the controlled object in the case. The control device according to claim 1.
The prediction unit
Using the output value y_0 of the control target and the output value u_0 of the PID controller when the output value y_0 is obtained, the transmission characteristic G of the control target is calculated.
Based on the PID gain value (Kp_1, Ki_1, Kd_1) adjusted using an arbitrary control target r_a and the transmission characteristic G of the control target, the adjusted PID gain value (Kp_1,) at the time of the control target r_a. A control device for predicting an output value y_1 of the control target when Ki_1, Kd_1) is used. The control device according to claim 1.
The prediction unit uses a physical model representing the transmission characteristics of the controlled object and the PID gain values (Kp_1, Ki_1, Kd_1) to adjust the PID gain value (Kp_1, Ki_1) at an arbitrary control target. , Kd_1), a control device for predicting the output value y_1 of the controlled object. The control device according to claim 1.
The prediction unit
The input value e_1 of the PID controller is obtained by using the output value u_0 of the PID controller when the output value y_0 of the control target is obtained and the adjusted PID gain value (Kp_1, Ki_1, Kd_1). ,
The desired output of the control target when the control target r_1, which is the sum of the output value y_0 and the input value e_1 of the PID controller, is set as the control target of the calculation unit with the desired output value y_d of the control target. Calculate the value y_d1 and
The difference d_1 between the output value y_0 and the desired output value yd_1 is calculated.
The value obtained by adding the calculated difference d_1 to the desired output value yd_a of the control target at the time of the arbitrary control target r_a is set as the output value y_1 of the control target at the time of the arbitrary control target r_a. Characteristic control device. The control device according to claim 1.
The determination unit controls the adjusted PID gain value (Kp_1, Ki_1, Kd_1) when at least one of the overshoot amount and the undershoot amount of the predicted output value y_1 of the controlled object is equal to or greater than a predetermined value. A control device characterized in that its application to an object is prohibited. The control device according to claim 1.
When the response time until the predicted output value y_1 of the control target reaches the predetermined value is not within the predetermined range, the determination unit sends the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to the control target. A control device characterized by prohibiting the application of. The control device according to any one of claims 3 to 5.
The determination unit
Using the control target r_b when the predicted output value y_1 of the control target is obtained, the desired output value yd_b of the control target is calculated.
The error d_2 between the predicted output value of the control target and the desired output value yd_b of the control target is calculated.
When the maximum value of the error d_2, the integrated value of the error d_2, or the sum of squares of the error d_2 is equal to or greater than a predetermined value, the adjusted PID gain value (Kp_1, Ki_1, Kd_1) is applied to the control target. A control device characterized by being prohibited. The control device according to claim 6.
When the difference is equal to or greater than a predetermined value, the determination unit prohibits the application of the adjusted PID gain value (Kp_1, Ki_1, Kd_1) to the control target. The control device according to claim 1.
A control device, which is at least one of a device for controlling the temperature of a plant, a device for controlling an autonomous vehicle, a device for controlling a robot, and a device for controlling an air vehicle.

Images