JP7089335B2 - Control device - Google Patents

Description

The present invention relates to a control device of a cascade control system in which a plurality of control units are connected in sequence.

Cascade control is known as an effective control system for controlled objects (processes) with large wasted time and time constant. The cascade control system is a control system configured as shown in FIG. 8 by a plurality of measured values (control quantities) from the process and one operation amount to the process. The primary control unit 100 ₁ calculates the manipulated variable MV ₁ so that the controlled variable PV ₁ of the primary process ₁₀ 11 matches the target value SP ₁ . The secondary control unit 100 ₂ sets the operation amount MV ₁ output from the primary control unit 100 ₁ as the target value SP ₂ , and operates so that the control amount PV ₂ of the secondary process 101 ₂ matches the target value SP ₂ . The quantity MV ₂ is calculated and output to the secondary process _{10 12} . In the example of FIG. 8, the primary control unit 100 ₁ and the secondary control unit 100 ₂ and the secondary process 101 ₂ and the primary process _{10 1} 1 form a control loop on the primary side, and the secondary control unit 100 ₂ And the secondary process 101 ₂ form a control loop on the secondary side.

However, in order to configure the cascade control system, a secondary process control amount PV ₂ for separating the primary process _{10 11} and the secondary process 10 ₁₂ is required as shown in FIG. Further, although the cascade control system shown in FIG. 8 has a two-stage configuration, it can also have a multi-stage configuration as shown in FIG.

In the control unit after n (n is an integer of 2 or more) in the practical cascade control system, the operation amount MV _n-1 of the immediately preceding control unit is used as the target value SP _n , so this operation amount MV _n-1 And scaling to match the static characteristics of the target value SP _n is required. Further, since cascade control is easy to hunt, a limit (limit processing) is often provided in the range of the target value SP _n of the control unit after the nth order. There are the following three methods (I) to (III) for scaling and limit processing. In the following description, the immediately preceding (n-1) next configuration is referred to as a main (primary), and the subsequent nth-order configuration is referred to as a sub (secondary).

(I) Fixed scaling limit of sub-control unit target value SP _n .
SP _n = (H _n -L _n ) / 100 x MV _n-1 + L _n ... (1)
In equation (1), MV _n-1 is the manipulated variable output from the main control unit, H _n (%) is the sub-control unit target value scaling limit upper limit value, and L _n (%) is the sub-control unit target value scaling. -The lower limit of the limit.

(II) Fixed scaling limit of the sub-control unit target value SP _n with the main control unit target value SP _n-1 as an offset.
SP _n = (H _n -L _n ) / 100 x MV _n-1 + L _n + SP _n-1 ... (2)

(III) Fixed scaling limit of the sub-control unit target value SP _n with the main control unit control amount PV _n-1 as an offset.
SP _n = (H _n -L _n ) / 100 x MV _n-1 + L _n + PV _n-1 ... (3)

In the above method (I), when the main control unit target value SP _n-1 is changed, the equilibrium point of the main control unit operation amount MV _n-1 before and after the change changes, so that the sub-control unit target value SP The range of change of _n expands. Therefore, the sub-control unit target value scaling limit upper and lower limit values H _n and L _n (%) should include the entire change range of the sub-control unit target value SP _n due to the change of the main control unit target value SP _n-1 . Must be set wider. As a result, control becomes easy to hunt, and the stability of the cascade control system may be impaired.

In the above method (II), the main control unit target value SP _n-1 is reflected as an offset of the sub-control unit target value SP _n . Therefore, when the main control unit target value SP _n-1 is changed, the sub-control Since the same scale ratio can be maintained before and after the change of the main control unit target value SP _{n -1} while maintaining the fluctuation range of the unit target value SP n, there is an advantage that the control can be easily stabilized. However, the method (II) can be applied when the main control unit target value SP _n-1 and the sub control unit target value SP _n have the same physical quantity (for example, the main control unit target value SP _n-1 ). There is a problem that it is limited to the case where both the sub-control unit target value SP _n and the temperature are the same).

In the above method (III), the control amount PV _n-1 of the main control unit obtained as a control response is more gradual than the change of the target value SP _n-1 of the main control unit, which may behave in a stepwise manner. Since it changes, the responsiveness of the control is impaired as compared with the method (II), but there is an advantage that the control is more easily stabilized. However, as in the method (II), there is a problem that the target value SP _n-1 of the main control unit and the target value SP _n of the sub control unit can be dealt with only when they have the same physical quantity. ..

On the other hand, Patent Document 1 discloses a technique for stabilizing control by operating an attenuator when converting the operation amount MV _n-1 of the main control unit to the target value SP _n of the sub control unit. The technique disclosed in Patent Document 1 appropriately replaces the attenuator (changes the attenuation rate) according to the control status (event), and causes an event such as a change in the target value SP _n-1 of the main control unit. Corresponds to the technique of switching between H _n and L _n by detecting. However, even in the technique disclosed in Patent Document 1, similarly to the methods (II) and (III), the main control unit target value SP _n-1 and the sub control unit target value SP _n can be dealt with. There was a problem that it was limited to the same kind of physical quantity.

Japanese Unexamined Patent Publication No. 7-219646

The present invention has been made to solve the above problems, and provides a control device for a cascade control system that can be applied even if the target value of the main control unit and the target value of the sub control unit are different types of physical quantities. The purpose is.

The present invention is configured to calculate an operation amount by inputting a target value and a control amount, respectively, in a control device of a cascade control system in which n (n is an integer of 2 or more) control units are connected in sequence. An operation provided between the n control units, the (i-1) order (i is an integer of 2 to n) control unit, and the i-order control unit, and calculated by the (i-1) order control unit. (N-1) target value conversion units configured to convert quantities to target values of the i-order control unit, and parameter setting units configured to set the parameters of these target value conversion units. Of the n control units, each of the primary to (n-1) next control units outputs the calculated operation amount to the target value conversion unit immediately after, and the nth control unit in the final stage. Outputs the calculated operation amount to the operation end of the control target, and each target value conversion unit controls the target value of the (i-1) primary control unit to SP _i-1 and (i-1) the primary control unit. The amount is PV _i-1 , the operation amount calculated by the (i-1) primary control unit is MV _i-1 , the target value of the i-order control unit is SP _i , and the preset i-order control unit target value scaling. Multiply the upper limit value by Hi , the preset i-order control unit target value scaling limit lower limit value by Li _, and ( _i -1) the target value SP _i-1 or control amount PV _i-1 of the next control unit. When the ratio as a parameter is R _i and the bias as a constant term is _Bi , SP _i = (H _i − Li ) / 100 × MV _{i -1} + SP _i-1 × R _i + _Bi , or SP _i . = (H _i − Li ) / 100 × MV _{i -1} + PV _i-1 × R _i + _Bi , so that the operation amount MV _i-1 calculated by the (i-1) primary control unit is the i- th control unit. Converted to the target value SP _i , the parameter setting unit changes the target value SP _i-1 of the (i-1) next control unit from the first value to the second value at the time of acquiring the past data. Based on the known data including the first and second values and the target value SP _i of the i-order control unit at the time of setting, which was acquired before and after the change of the target value SP _i-1 . A ratio bias calculation unit configured to calculate the ratio R _i and the bias B _i for each i-order target value conversion unit, and a ratio calculated for each i-th order target value conversion unit by this ratio bias calculation unit. It is characterized by including a ratio bias setting unit configured to set R _i and bias B _i to the corresponding i-order target value conversion units, respectively.

Further, the present invention is configured to calculate an operation amount by inputting a target value and a control amount, respectively, in a control device of a cascade control system in which n (n is an integer of 2 or more) control units are connected in cascade. The n control units are provided between the (i-1) order (i is an integer of 2 to n) control unit and the i-order control unit, and are calculated by the (i-1) order control unit. (N-1) target value conversion units configured to convert the operation amount to the target values of the i-order control unit, and parameter setting units configured to set the parameters of these target value conversion units. Of the n control units, each control unit from the primary to the (n-1) order outputs the calculated operation amount to the target value conversion unit immediately after the operation amount, and is the nth order in the final stage. The control unit outputs the calculated operation amount to the operation end of the control target, and each target value conversion unit sets the target value of the (i-1) primary control unit to SP _i-1 and the (i-1) primary control unit. The control amount of PV _i-1 , the operation amount calculated by the (i-1) primary control unit is MV _i-1 , the target value of the i-order control unit is SP _i , and the preset i-order control unit target value. The upper limit of the scaling limit is Hi , the preset target value of the i-order control unit is Li _, and ( _i -1) the target value of the next control unit is SP _i-1 or the control amount PV _i-1 . When the ratio that is the parameter to be multiplied by is R _i and the bias that is the constant term is _Bi , SP _i = ( Hi − _Li ) / 100 × MV i _{- 1} + SP _i-1 × R _i + _Bi , or By SP _i = (H _i − Li ) / 100 × MV _{i -1} + PV _i-1 × R _i + _Bi , the operation amount MV _i-1 calculated by the (i-1) next control unit is i- order controlled. Converted to the target value SP _i of the unit, the parameter setting unit changed the target value SP _i-1 of the (i-1) next control unit from the first value to the second value when the past data was acquired. Based on the known data including the first and second values of the time and the target value SP _i of the i-order control unit at the time of setting, which was acquired before and after the change of the target value SP _i-1 . , The ratio bias calculation unit configured to calculate the ratio R _i and the bias B _i for each i-th order target value conversion unit, and the ratio bias calculation unit calculates each i- th order target value conversion unit. Change of the ratio bias setting unit configured to set the ratio R _i and the bias B _i to the corresponding i-th target value conversion unit and the target value SP ₁ of the primary control unit of the first stage. On the other hand, on condition that the desired control response of the control amount PV _i-1 of the (i-1) order control unit can be obtained, the upper and lower limit widths of the target value scaling limit of the i-order control unit are made as narrow as possible. The upper and lower limits configured to determine the i-order control unit target value scaling limit upper limit value Hi and the _i -order control unit target value scaling limit lower limit value Li for each _i - order target value conversion unit. The value determination unit and the i-order control unit target value scaling limit upper limit value Hi and the _i -order control unit target value scaling limit lower limit value Lii _determined for each i-order target value conversion unit by the upper and lower limit value determination units. It is characterized by including an upper / lower limit value setting unit configured to set the above / lower limit value in the corresponding i-order target value conversion unit.

Further, in one configuration example of the control device of the present invention, the ratio bias calculation unit sets the first value of the target value of the (i-1) next control unit at the time of past data acquisition to SP _i-1 and the first. When the value of 2 is _SP'i-1 , (i-1) the target value of the next control unit is the first value SP _i-1 , the target value of the i-order control unit at the time of setting is SP _i , (i-). 1) When the target value of the i-order control unit at the time of setting is SP'i when the target value of the next control unit is changed from the first value SP _i - ₁ to the second value SP'i _-1 . , R _i = (SP _i -SP'i) / (SP _{i -1} -SP'i _-1 ), B _i = SP _i -SP _i-1 × R _i , the ratio R _i and the bias B _i It is characterized by calculating.

According to the present invention, it is possible to realize a control device that can be applied even if the target value of the main control unit and the target value of the sub control unit are different types of physical quantities. In the prior art, event detection was required for a change in the target value of the main control unit, but according to the present invention, it is not necessary to convert a coefficient by an event for a change in the target value of the main control unit, and it is more continuous. Since the behavior of the control can be realized, a smooth and stable control response can be obtained.

FIG. 1 is a block diagram showing a configuration of a control device according to an embodiment of the present invention. FIG. 2 is a block diagram showing a configuration of a parameter setting unit of the control device according to the embodiment of the present invention. FIG. 3 is a flowchart illustrating a control operation of the control device according to the embodiment of the present invention. FIG. 4 is a flowchart illustrating a parameter setting operation of the control device according to the embodiment of the present invention. FIG. 5 is a flowchart illustrating the operation of the upper / lower limit value determination unit of the control device according to the embodiment of the present invention. FIG. 6 is a diagram showing an example of a controlled object according to an embodiment of the present invention. FIG. 7 is a block diagram showing a configuration example of a computer that realizes the control device according to the embodiment of the present invention. FIG. 8 is a block diagram showing a configuration of a conventional cascade control system. FIG. 9 is a block diagram showing another configuration of the conventional cascade control system.

[Principle of invention]
The present invention depends on the operation quantity MV i-1 of the main control unit and the target value SP _i of the sub-control unit when converting the operation quantity MV _{i -1} output from the main control unit into the target value SP _i of the sub-control unit. When there is a relationship, that is, when the main control unit operation amount MV _i-1 , which is an independent variable, is determined, the sub-control unit target value SP _i , which is a dependent variable, is determined. It is a control device that can absorb the difference in physical quantity between the value SP _i-1 and the sub-control unit target value SP _i by setting parameters.

For example, when there is a linear dependency between the main control unit operation amount MV _{i-1 and} the sub-control unit target value SP _i , the main control unit operation amount MV i-1 is converted to the sub-control unit target value SP _i . In addition to the conventional scaling, the main control unit and the sub control unit are different by adding the ratio and bias as set values to the main control unit target value SP _i-1 or the main control unit control amount PV _i-1 . Even when controlling for different types of physical quantities, appropriate conversion from the main control unit operation quantity MV _i-1 to the sub-control unit target value SP _i can be realized.

[Example]
Hereinafter, examples of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a control device according to an embodiment of the present invention. The control device includes n (n is an integer of 2 or more) control units 1 ₁ to 1 _n connected in cascade, and (i-1) order (i is an integer of 2 to n) control unit 1 _i-1 . (I-1) The operation amount MV _i-1 provided between the i-order control unit 1 _i and calculated by the (i-1) order control unit 1 _i-1 is converted into the target value SP _i of the i-order control unit 1 _i . It is composed of (n-1) target value conversion units 2 ₂ to 2 _n and a parameter setting unit 3 for setting the parameters of the target value conversion units 2 ₂ to 2 _n .

FIG. 2 is a block diagram showing the configuration of the parameter setting unit 3. The parameter setting unit 3 sets the ratio, which is a parameter multiplied by the target value SP _i-1 or the control amount PV _i-1 of the (i-1) next control unit 1 _i-1 , and the bias, which is a constant term of the conversion formula. The ratio / bias calculation unit 30 calculated for each target value conversion unit and the ratio and bias calculated for each target value conversion unit by the ratio / bias calculation unit 30 are converted into the corresponding target value conversion units 2 ₂ to 2 _n . The ratio bias setting unit 31 to be set and the upper and lower limit initial values of the i-order control unit target value scaling limit upper and lower limit values for setting the upper and lower limits of the target value SP _i of the i-order control unit 1 _i . It is determined that the desired control response of the control amount PV _i-1 of the (i-1) primary control unit 1 _i-1 can be obtained in response to the change of the target value SP ₁ of the setting unit 32 and the primary control unit ₁₁ . As a condition, the upper / lower limit value determination unit 33 that determines the i-order target value scaling limit upper / lower limit value for each target value conversion unit so that the i-order target value scaling limit upper / lower limit width becomes as narrow as possible, and the upper part. The upper and lower limit value setting units 34 that set the i-order target value scaling limit upper and lower limit values determined for each target value conversion unit 33 by the lower limit value determination unit 33 in the corresponding target value conversion units 2 ₂ to 2 _n , respectively. The control unit 1 ₁ to 1 _n is composed of a PID parameter adjustment unit 35 for adjusting PID parameters (proportional band, integration time, differentiation time).

The upper / lower limit value determination unit 33 includes a target value input unit 330, an arrival time measurement unit 331, a vibration amount measurement unit 332, and an upper / lower limit adjustment unit 333.

FIG. 3 is a flowchart illustrating a control operation of the control device of this embodiment. The target value SP ₁ of the primary control unit ₁₁ is set by the user of the control device and input to the _primary control unit 11 (step S100 in FIG. 3).
The controlled variable PV ₁ of the primary control unit ₁₁ is measured by a sensor (not shown) provided in the _primary process 41 and input to the _primary control unit 11 (step S101 in FIG. 3).

The primary control unit 1 ₁ inputs the target value SP ₁ and the control amount PV ₁ and calculates the operation amount MV ₁ by a well-known PID control operation so that the control amount PV ₁ matches the target value SP ₁ (. FIG. 3 step S102).

The secondary target value conversion unit 2 ₂ converts the manipulated variable MV ₁ output from the primary control unit 1 ₁ into the target value SP ₂ of the secondary control unit 1 ₂ by the conversion formula of the equation (4). Output to the next control unit 1 ₂ (step S103 in FIG. 3).
SP ₂ = (H ₂ -L ₂ ) / 100 x MV ₁ + SP ₁ x R ₂ + B ₂ ... (4)

In equation (4), H ₂ (%) is the upper limit of the target value scaling limit of the secondary control unit, L ₂ (%) is the lower limit of the target value of the secondary control unit, and R ₂ is the lower limit of the target value of the secondary control unit. The ratio, B ₂ , which is a parameter to be multiplied by the term of the primary control unit target value SP ₁ of the conversion formula, is a bias which is a constant term of the conversion formula of the formula (4).
The secondary target value conversion unit 2 ₂ uses the primary control unit control amount PV ₁ instead of the primary control unit target value SP ₁ by the following equation (5) to obtain the secondary control unit target value SP ₂ . It may be calculated.
SP ₂ = (H ₂ -L ₂ ) / 100 x MV ₁ + PV ₁ x R ₂ + B ₂ ... (5)

The control amount PV ₂ of the secondary control unit 1 ₂ is measured by a sensor (not shown) provided in the secondary process 4 ₂ and input to the secondary control unit 1 ₂ (step S104 in FIG. 3).
The secondary control unit 1 ₂ uses the target value SP ₂ and the control amount PV ₂ as inputs, and calculates the operation amount MV ₂ by a well-known PID control operation so that the control amount PV ₂ matches the target value SP ₂ (. FIG. 3 step S105).

The tertiary target value conversion unit 2 ₃ converts the manipulated variable MV ₂ output from the secondary control unit 1 ₂ into the target value SP ₃ of the tertiary control unit 1 ₃ by the conversion formula of the equation (6). Output to the next control unit 13 (step S106 in FIG. ₃ ).
SP ₃ = (H ₃ -L ₃ ) / 100 x MV ₂ + SP ₂ x R ₃ + B ₃ ... (6)

In equation (6), H ₃ (%) is the upper limit of the target value scaling limit of the tertiary control unit, L ₃ (%) is the lower limit of the target value of the tertiary control unit, and R ₃ is the lower limit of the target value of the tertiary control unit. The ratio, B ₃ , which is a parameter to be multiplied by the term of the target value SP ₂ of the secondary control unit of the conversion formula, is a bias which is a constant term of the conversion formula of the formula (5).
The tertiary target value conversion unit 2 ₃ uses the secondary control unit control amount PV ₂ instead of the secondary control unit target value SP ₂ according to the following equation (7) to obtain the tertiary control unit target value SP ₃ . It may be calculated.
SP ₃ = (H ₃ -L ₃ ) / 100 x MV ₂ + PV ₂ x R ₃ + B ₃ ... (7)

In this way, the processes are carried out in the order of primary, secondary, tertiary, and so on. Next, the nth-order target value conversion unit 2 _n converts the manipulated variable MV _n-1 output from the (n-1) -th order control unit (not shown) into the n-th order control unit 1 by the conversion formula of the equation (8). It is converted to the target value SP _n of _n and output to the nth control unit 1 _n (step S107 in FIG. 3).
SP _n = (H _n -L _n ) / 100 x MV _n-1 + SP _n-1 x R _n + B _n ... (8)

In equation (8), H _n (%) is the nth-order control unit target value scaling limit upper limit value, L _n (%) is the n-th order control unit target value scaling limit lower limit value, and R _n is the equation (8). The ratio and B _n , which are parameters to be multiplied by the term of the target value SP _n-1 of the (n-1) -order control unit of the conversion equation, are biases that are constant terms of the conversion equation of the equation (8).
In addition, the nth-order target value conversion unit _2n uses the (n-1) -th order control unit control amount PV _n-1 instead of the (n-1) -th order control unit target value SP _n-1 by the following equation (n-1). The nth-order control unit target value SP _n may be calculated according to 9).
SP _n = (H _n -L _n ) / 100 x MV _n-1 + PV _n-1 x R _n + B _n ... (9)

i (i is an integer of 2 to n) The following is a generalization of equations (4) to (9) for the target value SP _i of the next control unit 1 _i .
SP _i = (H _i -L _i ) / 100 x MV _i-1 + SP _i-1 x R _i + B _i ... (10)
SP _i = (H _i -L _i ) / 100 x MV _i-1 + PV _i-1 x R _i + B _i ... (11)

Next, the controlled variable PV _n of the nth-order control unit 1 _n is measured by a sensor (not shown) provided in the n-th order process 4 _n and input to the n-th order control unit 1 _n (step S108 in FIG. 3). ..
The nth-order control unit 1 _n inputs the target value SP _n and the control amount PV _n , and calculates the operation amount MV _n by a well-known PID control operation so that the control amount PV _n matches the target value SP _n . Output to the nth process 4 _n (step S109 in FIG. 3). The output destination of the operation amount MV _n is, for example, the operation end of a valve or the like.
The processes of steps S100 to S109 as described above are repeatedly executed every control cycle until the operation of the control device is completed (YES in step S110 of FIG. 3).

Next, FIG. 4 shows the setting operation of the parameters H ₂ to H _n , L ₂ to L _n , R ₂ to L _n , and B ₂ to B _n performed by the parameter setting unit 3 before the control operation as described above. Will be described with reference to.
Here, (i-1) a primary control unit 1 _i-1 (i is an integer of 2 to n) will be typically described as a main control unit, and a subsequent i-order control unit 1 _i will be typically described as a sub control unit.

When setting the parameters H _i , Li _i , R _i , B _i , the user changes the target value of the (i-1) next control unit from SP _{i -1} to SP'i-1 from the data acquired in advance. For each of the (i-1) primary control unit target values before and after the change, the (i-1) primary control unit operation amount MV _i-1 at the time of setting and the i-order control unit target value SP _i at the time of setting To get.

The ratio bias calculation unit 30 of the parameter setting unit 3 has known data acquired by the user, that is, the (i-1) primary control unit target value SP _i-1 before the change and the (i-1) primary control unit. When the target value is SP _i-1 , the i-order control unit target value SP _i at the time of setting, the changed (i-1) next control unit target value SP'i _-1 , and (i-1) next control Based on the _i -order control unit target value SP'i at the time of setting when the unit target value is changed from SP _i-1 to SP'i _-1 , the ratio R _i of the i-order target value conversion unit 2 _i and The bias B _i is calculated (FIG. 4, step S200).

From the above known data, the following simultaneous equations hold.
SP _i = SP _i-1 x R _i + B _i ... (12)
SP'i = _{SP'i -1} x R _i + B _i ... (13)

The following equations can be obtained from the equations (12) and (13).
R _i = (SP _i -SP'i) / (SP _{i -1} -SP'i _-1 ) ... (14)
B _i = SP _i -SP _i-1 x R _i ... (15)

The ratio bias calculation unit 30 may calculate the ratio R _i by the equation (14) and the bias B _i by the equation (15). The unit of the ratio R _i is the unit of the i-order control unit target value SP _i / the unit of the (i-1) order control unit target value SP _i-1 . On the other hand, the unit of the bias B _i is the same as the unit of the target value SP _i of the i-order control unit.

Next, the ratio bias setting unit 31 of the parameter setting unit 3 sets the ratio R _i and the bias B _i calculated by the ratio bias calculation unit 30 for the i-order target value conversion unit 2 _i (). FIG. 4 Step S201).

The ratio bias calculation unit 30 and the ratio bias setting unit 31 (i is an integer of 2 to n) may perform the processes of steps S200 and S201 for each i-order target value conversion unit 2 _i .

Next, the upper / lower limit initial setting unit 32 of the parameter setting unit 3 sets the ratio R _i and the bias B _i of the i-th order target value conversion unit 2 _i as described above, and then the i-th order target value conversion unit 2 _{For i} , the initial values of the i-order control unit target value scaling limit upper limit value H _i and the i-order control unit target value scaling limit lower limit value L _i are set (FIG. 4, step S202).

In consideration of shortening the trial time and safe operation of the device, the initial value of the i-order control unit target value scaling limit upper limit value H _i is set as the known upper limit value of the control amount PV _i of the i-th order control unit 1 _i . The initial value of the i-order control unit target value scaling limit lower limit value L _i is set as the known lower limit value of the control amount PV _i of the i-order control unit 1 _i .
The upper and lower limit initial setting units 32 (i is an integer of 2 to n) may perform the process of step S202 for each i-order target value conversion unit 2 _i .

Next, the PID parameter adjusting unit 35 of the parameter setting unit 3 adjusts the PID parameters (proportional band, integration time, differential time) of each control unit 1 ₁ to 1 _n (FIG. 4, step S203). Since the PID parameter adjustment method of each control unit 1 ₁ to 1 _n is a well-known technique, detailed description thereof will be omitted. When adjusting the PID parameters, the nth control unit 1 _n → (n-1) primary control unit 1 _n-1 → ... → secondary control unit 1 ₂ → primary control unit 1 ₁ It is necessary to make adjustments in order. Further, in this embodiment, the PID parameter adjusting unit 35 is provided inside the control device, but it may be provided outside.

Next, the upper / lower limit value determination unit 33 of the parameter setting unit 3 obtains a desired control response of the (i-1) primary control unit control amount PV _i-1 to the change of the primary control unit target value SP ₁ . The i-order control unit target value scaling limit upper and lower limit width (difference between Hi and L _i ) is as narrow as possible so that the _i -th order control unit target value scaling limit upper and lower limit width H _i And the i-order control unit target value scaling limit lower limit value L _i (FIG. 4, step S204).

Narrowing the upper and lower limit widths of the i-order control unit target value scaling limit makes it easier to stabilize the (i-1) primary control unit control amount PV _i-1 , while (i-1) the primary control unit control amount. It acts in the direction of weakening the potential of shortening the time to reach the target value of PV _i-1 . Therefore, the upper and lower limit widths of the i-order control unit target value scaling limit are narrowed as much as possible within the range in which the desired control response of the (i-1) next control unit control amount PV _i-1 can be obtained.

The i-order control unit target value scaling limit upper limit value H _i and the i-order control unit target value scaling limit lower limit value L _i are the control amounts of the (i-1) next control unit 1 _i-1 which is the main control unit. It is a value that should be determined by a trade-off between the stabilization of PV _i -1 and the control characteristics such as the time to reach the target value of the control amount PV _i-1 and the allowable characteristics such as environmental changes such as changes over time. It is difficult to set the i-order control unit target value scaling limit upper limit value H _i and the i-order control unit target value scaling limit lower limit value L _i unless the controller gain is determined to some extent. For example, when the controller gain of the (i-1) primary control unit 1 _i-1 becomes a strong adjustment result, the control response of the (i-1) primary control unit control amount PV _i-1 tends to be oscillating. In this situation, if the i-order control unit target value scaling limit upper / lower limit width is selected to be wider, the control response of the control amount PV _i-1 becomes more oscillating. Therefore, it is necessary to determine the i-order control unit target value scaling limit upper limit value H _i and the i-order control unit target value scaling limit lower limit value L _i based on the control response of the control amount PV _i-1 .

FIG. 5 is a flowchart illustrating the operation of the upper / lower limit value determination unit 33. The target value input unit 330 of the upper / lower limit value determination unit 33 inputs a predetermined target value SP ₁ to the primary control unit ₁₁ (step S300 in FIG. 5), and after a certain period of time has elapsed (for example, the control system is settled). After a sufficient time has elapsed), the target value SP ₁ is changed by a predetermined target value change range (FIG. 5, step S301).

In the arrival time measurement unit 331 of the upper / lower limit value determination unit 33, the (i-1) primary control unit control amount PV _i-1 is (i-1) next from the time when the primary control unit target value SP ₁ is changed. The target value arrival time _Ti-1 until the control unit target value SP _i-1 is reached is measured (FIG. 5, step S302).

In the vibration amount measurement unit 332 of the upper / lower limit value determination unit 33, after the primary control unit target value SP ₁ is changed, the (i-1) primary control unit control amount PV _i-1 is (i-1) next-order controlled. (I-1) Next control unit target value SP _i-1 and (i-1) Next control unit control amount from the time when the target value SP _i-1 is reached until the predetermined vibration amount measurement time elapses. The vibration amount D _i-1 , which is the integrated value of the absolute values of the deviations from PV _i-1 (SP _i-1 − PV _i-1 ), is measured (FIG. 5, step S303).

The upper / lower limit adjusting unit 333 of the upper / lower limit value determining unit 33 is the _{i-order when the vibration amount Di-1} measured by the vibration amount measuring unit 332 is equal to or more than a predetermined vibration amount index value (YES in step S304 of FIG. 5). The i-order control unit target value scaling limit upper limit value H _i of the target value conversion unit 2 _i is lowered by a predetermined upper / lower limit value change width, and the i-order control unit target value scaling limit lower limit value L _i is set to a predetermined upper / lower limit value. By increasing the value change width, the i-order control unit target value scaling limit upper / lower limit width is narrowed (FIG. 5, step S306), and the process returns to step S300. The vibration amount index value is preset according to the purpose of the controlled target device.

However, when the vibration amount D _i-1 is equal to or greater than the vibration amount index value, the upper / lower limit adjustment unit 333 determines whether the i-order control unit target value scaling limit upper limit value H _i reaches the allowable minimum value or the i-th order. When at least one of the control unit target value scaling limit lower limit value L _i reaches the allowable maximum value is satisfied and the i-order control unit target value scaling limit upper and lower limit values are out of the allowable range. (NO in step S305 of FIG. 5), the i-order control unit target value scaling limit upper and lower limit width cannot be further narrowed.

Therefore, the upper / lower limit adjusting unit 333 lowers the controller gain of the (i-1) primary control unit 1 _i-1 (step S307 in FIG. 5), and returns to step S300. Specifically, the upper / lower limit adjusting unit 333 raises the proportional band by multiplying the proportional band set in the (i-1) order control unit 1 _i-1 by a predetermined coefficient larger than 1. i-1) Next control unit 1 The controller gain of _i-1 may be lowered. The permissible minimum value of the i-order control unit target value scaling limit upper limit value H _i and the permissible maximum value of the i-order control unit target value scaling limit lower limit value L _i are environmental changes such as changes over time. It is preset from the allowable characteristics.

Further, in the upper / lower limit adjusting unit 333, when the vibration amount D _i-1 is less than the vibration amount index value, the target value arrival time _Ti-1 measured by the arrival time measurement unit 331 is equal to or larger than the predetermined time index value. In the case (YES in step S308 of FIG. 5), the i-order control unit target value scaling limit upper limit value H _i of the i-order target value conversion unit 2 _i is increased by a predetermined upper / lower limit value change width, and the i-order control unit target value is increased. By lowering the scaling limit lower limit value L _i by a predetermined upper / lower limit value change width, the i-order control unit target value scaling limit upper / lower limit width is widened (FIG. 5, step S310), and the process returns to step S300. The time index value is preset according to the purpose of the controlled device.

However, when the target value arrival time Ti _-1 is equal to or longer than the time index value, the upper / lower limit adjustment unit 333 determines whether the i-order control unit target value scaling limit upper limit value H _i reaches the allowable maximum value or i. Next control unit target value scaling limit lower limit value L _i reaches at least one of the allowable minimum values, and at least one of them is satisfied, and the i-order control unit target value scaling limit upper and lower limit values are out of the allowable range. (NO in step S309 of FIG. 5), the i-order control unit target value scaling limit upper and lower limit width cannot be further widened.

Therefore, the upper / lower limit adjusting unit 333 (i-1) raises the controller gain of the next control unit 1 _i-1 (step S311 in FIG. 5), and returns to step S300. Specifically, the upper / lower limit adjusting unit 333 lowers the proportional band by multiplying the proportional band set in the (i-1) order control unit 1 _i-1 by a predetermined coefficient smaller than 1. -1) The controller gain of the next control unit 1 _i-1 may be increased. The maximum allowable value of the i-order control unit target value scaling limit upper limit value H _i is the same as the above initial value of the upper limit value H _i , and the allowable maximum value of the i-order control unit target value scaling limit lower limit value L _i . The minimum possible value is the same as the above initial value of the lower limit value L _i .

In this way, the processes of steps S300 to S311 are repeatedly executed until the vibration amount D _i-1 is less than the vibration amount index value and the target value arrival time T _i-1 is less than the time index value, and the vibration amount D _i-1 is obtained. When the vibration amount index value is less than the target value arrival time Ti _{-1 is less than the time index value, the i-} order control unit target value scaling limit upper limit H _i and the i-order control unit target value scaling limit lower limit The determination of the value L _i is completed, and the processing of the upper / lower limit value determination unit 33 is completed.

The upper / lower limit value setting unit 34 of the parameter setting unit 3 has the i-order target value conversion unit 2 _i and the i-order control unit target value scaling limit upper limit values H _i and i determined by the upper / lower limit value determination unit 33. Next Control unit Target value Scaling limit Lower limit value L _i is set (FIG. 4, step S205).

The upper / lower limit value determination unit 33 and the upper / lower limit value setting unit 34 perform the above steps S204 (steps S300 to S311) and S205 in the order of i = 2, 3, ..., N (secondary target value). Conversion unit 2 ₂ → 3rd target value conversion unit 2 ₃ → nth target value conversion unit 2 _n ) is performed for each i-order target value conversion unit 2 _i . The reason for carrying out in this order is that it is considered difficult to stabilize the control unless the vibration of the control unit on the lower order side is suppressed.
When the processing of steps S204 and S205 is completed for all _i -order target value conversion units 2i (YES in step S206 of FIG. 4), the processing of the parameter setting unit 3 is completed.

In this embodiment, the problems of the above methods (II) and (III) and the technique disclosed in Patent Document 1 can be solved, and the main control unit target value SP _i-1 and the sub control unit target value can be solved. It can be applied even if the physical quantity is different from that of SP _i . In the technique disclosed in Patent Document 1, the (i-1) order control unit operation amount is converted by a ratio coefficient called the attenuation factor, but since a simple ratio alone cannot cope with the conversion of different types of physical quantities. (I-1) Coefficient conversion is required triggered by some event such as a change in the target value of the next control unit. On the other hand, in this embodiment, the conversion of the coefficient by the event is not required, and the ratio R _i , the bias B _i , and the target value scaling limit upper and lower limit values H _i and L _i are used to obtain the (i-1) order. It is possible to convert the control unit target value SP _i-1 according to the physical quantity of the i-order control unit target value SP _i .

In this embodiment, the upper / lower limit value determination unit 33 determines the i-order control unit target value scaling limit upper limit value H _i and the i-order control unit target value scaling limit lower limit value L _i . However, the upper and lower limit values _Hi and Li _{predetermined} by the user may be used.

FIG. 6 is a diagram showing one example of the controlled object of this embodiment. In the cascade control system shown in FIG. 6, the temperature (primary control unit control amount PV ₁ ) measured by the thermometer 201 provided in the combustion furnace 200 is input to the _primary control unit 11. The primary control unit 1 ₁ calculates and outputs the operation amount MV ₁ . The secondary target value conversion unit 2 ₂ converts the primary control unit operation amount MV ₁ into the secondary control unit target value SP ₂ . The fuel flow rate (secondary control unit control amount PV ₂ ) measured by the flow rate transmitter 202 is input to the secondary control unit 1 ₂ . The secondary control unit 1 ₂ calculates the operation amount MV ₂ and outputs it to the valve 203. The fuel whose flow rate is adjusted by the valve 203 is supplied to the burner 204 of the combustion furnace 200.

When 0 to 100% of the primary control unit operation amount MV ₁ corresponds to 0 to 40 m ³ / h of the secondary control unit target value SP ₂ , 1% of the primary control unit operation amount MV ₁ is 0.4 m ³ It corresponds to the change of the secondary control unit target value SP ₂ corresponding to / h. On the other hand, if 0 to 100% of the primary control unit operation amount MV ₁ corresponds to 0 to 400 m ³ / h of the secondary control unit target value SP ₂ , 1% of the primary control unit operation amount MV ₁ is 4 m. It will be a change of the secondary control unit target value SP ₂ equivalent to ³ / h. When the target value SP ₂ of the secondary control unit becomes wide in this way, the control system becomes easy to hunt.

For example, in the cascade control system shown in FIG. 6, it is assumed that the following relationship is known between the combustion furnace set temperature (primary control unit target value SP ₁ ) and the required fuel flow rate.

When the target value SP ₁ of the primary control unit is selected only at 200 ° C., the range of the target value SP ₂ of the secondary control unit may be set to 15 to 25 m ³ / h even if there is some margin. However, in reality, for example, when the primary control unit target value SP ₁ is 3000 ° C., the required range of the secondary control unit target value SP ₂ is 15 to 25 m ³ / h, and the primary control unit target value SP ₁ is set. At 5000 ° C., the required range of the secondary control unit target value SP ₂ is 35 to 45 m ³ / h. Therefore, the range of the target value SP ₂ of the secondary control unit must be set to 15 to 45 m ³ / h. It is assumed that the user who is aware of such a situation has preset the secondary control unit target value scaling limit upper limit values H ₂ and L ₂ .

Then, the ratio bias calculation unit 30 of the present embodiment has known data acquired by the user, that is, the primary control unit target value SP ₁ = 3000 ° C. before the change, and the primary control unit target value is SP ₁ = 3000. Secondary control unit target value SP ₂ = 20m ³ / h at the time of setting in the case of ° C, primary control unit target value SP ' ₁ = 5000 ° C after change, and primary control unit target value SP ₁ = 3000 Second-order target value conversion by solving the following simultaneous equations based on the quadratic control unit target value _SP'2 = 40 m ³ / h at the time of setting when _SP'1 = 5000 ° C is changed from ° C. The ratio R ₂ and the bias B ₂ of the part 2 ₂ are calculated.
20 = 3000 x R ₂ + B ₂ ... (16)
40 = 5000 x R ₂ + B ₂ ... (17)

As a result, a ratio R ₂ = 0.01 m ³ / h ° C. and a bias B ₂ = −10 m ³ / h can be obtained. If the scaling is set so that the primary control unit operation amount MV ₁ = 0 to 100% and ± 5 m ³ / h is set, the primary control unit operation amount MV ₁ is a physical quantity that is appropriately different from the secondary control unit target value. It becomes possible to convert to SP ₂ .

The control device described in this embodiment can be realized by a computer provided with a CPU (Central Processing Unit), a storage device, and an interface, and a program for controlling these hardware resources. An example of the configuration of this computer is shown in FIG. The computer includes a CPU 300, a storage device 301, and an interface device (hereinafter, abbreviated as I / F) 302. For example, a sensor such as a thermometer or a flow rate transmitter, an operating end such as a valve, or the like is connected to the I / F 302. In such a computer, a program for realizing the parameter setting method of the present invention is stored in the storage device 301. The CPU 300 executes the process described in this embodiment according to the program stored in the storage device 301.

The cascade control function is installed as a well-known function of the multi-loop control device (regulator). That is, by using a controller having a plurality of control units inside, it is possible to realize a cascade control system in which a plurality of control units are connected in series as in the present invention. Further, each control unit may be realized by a regulator, and a plurality of regulators may be connected in series to form the cascade control system of the present invention.

The present invention can be applied to a cascade control system.

1 ₁ to 1 _n ... Control unit, 2 ₂ to 2 _n ... Target value conversion unit, 3 ... Parameter setting unit, 4 ₁ to 4 _n ... Process, 30 ... Ratio / bias calculation unit, 31 ... Ratio bias setting unit, 32 ... Upper and lower limit initial setting unit, 33 ... Upper and lower limit value determination unit, 34 ... Upper and lower limit value setting unit, 35 ... PID parameter adjustment unit, 330 ... Target value input unit, 331 ... Arrival time measurement unit, 332 ... Vibration amount measurement Unit, 333 ... Upper and lower limit adjustment unit.

Claims (3)

In a control device of a cascade control system in which n (n is an integer of 2 or more) control units are connected in sequence.
The n control units configured to calculate the operation amount by inputting the target value and the control amount, respectively, and
(I-1) The operation amount provided between the next (i is an integer of 2 to n) control unit and the i-order control unit, and the operation amount calculated by the (i-1) order control unit is the target of the i-order control unit. (N-1) target value conversion units configured to convert to values,
It is provided with a parameter setting unit configured to set the parameters of these target value conversion units.
Of the n control units, each of the primary to (n-1) next control units outputs the calculated operation amount to the target value conversion unit immediately after, and the nth-order control unit in the final stage is The calculated operation amount is output to the operation end of the control target, and
Each target value conversion unit uses SP _i-1 for the target value of the (i-1) primary control unit, PV _i-1 for the control amount of the (i-1) primary control unit, and (i-1) primary control unit. The calculated operation amount is MV _i-1 , the target value of the i-order control unit is SP _i , the preset i-order control unit target value scaling limit upper limit value is Hi , and the preset i-order control unit target _. The lower limit of the value scaling limit is Li, the ratio (i-1), which is a parameter multiplied by the target value SP i-1 of the next control unit or the control amount PV i-1, is R _{i ,} and the _{bias that} is a constant term is _Bi . When SP _i = (H _i - L _i ) / 100 x MV _i-1 + SP _i-1 x R _i + _Bi , or SP _i = (H _i - L _i ) / 100 x MV _i-1 + PV By _i-1 × R _i + B _i , the operation amount MV _i-1 calculated by the (i-1) primary control unit is converted into the target value SP _i of the i-order control unit.
The parameter setting unit is
(I-1) Target value SP of the next control unit at the time of past data acquisition The first and second values when _i-1 is changed from the first value to the second value, and this target value SP . Based on the known data including the target value SP _i of the i-order control unit at the time of settling, which was acquired before and after the change of _i-1 , the ratio R _i and the bias B _i were converted to the i-order target value. A ratio / bias calculation unit configured to calculate for each unit,
A ratio bias setting unit configured to set the ratio R _i and the bias B _i calculated for each i -th target value conversion unit by the ratio bias calculation unit to the corresponding i-th target value conversion units. A control device characterized by including. In a control device of a cascade control system in which n (n is an integer of 2 or more) control units are connected in sequence.
The n control units configured to calculate the operation amount by inputting the target value and the control amount, respectively, and
(I-1) The operation amount provided between the next (i is an integer of 2 to n) control unit and the i-order control unit, and the operation amount calculated by the (i-1) order control unit is the target of the i-order control unit. (N-1) target value conversion units configured to convert to values,
It is provided with a parameter setting unit configured to set the parameters of these target value conversion units.
Of the n control units, each of the primary to (n-1) next control units outputs the calculated operation amount to the target value conversion unit immediately after, and the nth-order control unit in the final stage is The calculated operation amount is output to the operation end of the control target, and
Each target value conversion unit uses SP _i-1 for the target value of the (i-1) primary control unit, PV _i-1 for the control amount of the (i-1) primary control unit, and (i-1) primary control unit. The calculated operation amount is MV _i-1 , the target value of the i-order control unit is SP _i , the preset i-order control unit target value scaling limit upper limit value is Hi , and the preset i-order control unit target _. The lower limit of the value scaling limit is Li, the ratio (i-1), which is a parameter multiplied by the target value SP i-1 of the next control unit or the control amount PV i-1, is R _{i ,} and the _{bias that} is a constant term is _Bi . When SP _i = (H _i - L _i ) / 100 x MV _i-1 + SP _i-1 x R _i + _Bi , or SP _i = (H _i - L _i ) / 100 x MV _i-1 + PV By _i-1 × R _i + B _i , the operation amount MV _i-1 calculated by the (i-1) primary control unit is converted into the target value SP _i of the i-order control unit.
The parameter setting unit is
(I-1) Target value SP of the next control unit at the time of past data acquisition The first and second values when _i-1 is changed from the first value to the second value, and this target value SP . Based on the known data including the target value SP _i of the i-order control unit at the time of settling, which was acquired before and after the change of _i-1 , the ratio R _i and the bias B _i were converted to the i-order target value. A ratio / bias calculation unit configured to calculate for each unit,
A ratio bias setting unit configured to set the ratio R _i and the bias B _i calculated for each i -th target value conversion unit by the ratio bias calculation unit to the corresponding i-th target value conversion units. ,
Target value of the primary control unit in the first stage (i-1) Target value of the primary control unit on condition that the desired control response of the control amount PV _i-1 of the primary control unit can be obtained in response to the change of SP ₁ . The i-order control unit target value scaling limit upper limit value Hi and the _i -order control unit target value scaling limit lower limit value Li are set to the _i - order target value so that the scaling limit upper and lower limit widths are as narrow as possible. An upper / lower limit value determination unit configured to be determined for each conversion unit,
The i-order control unit target value scaling limit upper limit value Hi and the _i -order control unit target value scaling limit lower limit value Li, _which are determined for each i-order target value conversion unit by the upper and lower limit value determination units, correspond to each other. A control device including an upper / lower limit value setting unit configured to be set in the i-order target value conversion unit. In the control device according to claim 1 or 2 .
The ratio bias calculation unit sets the first value of the target value of the (i-1) next control unit at the time of past data acquisition to SP _i-1 , the second value to SP'i _-1 , and (i-). 1) When the target value of the next control unit is the first value SP _i-1 , the target value of the i-order control unit at the time of setting is SP _i , and (i-1) the target value of the next control unit is the first value. When the target value of the i-order control unit at the time of setting is SP'i when the second value is changed from SP _i-1 to _SP'i-1 , R _i = (SP _i - _{SP'i )} / (SP _i-1- SP'i _-1 ), Bi = SP _i -SP _{i -1} × R _i , a control device characterized in that the ratio R _i and the bias B _i are calculated.

Images