JPS63311502A - Controller - Google Patents

Abstract

PURPOSE:To set an optimum PID value without noise influence by providing an integrating arithmetic part and a process characteristic detection part which obtains the time constant and the dead time of a process from an integrating arithmetic result in an optimum PID arithmetic part. CONSTITUTION:The integrating arithmetic part 7 and the process characteristic detection part 12 which obtains the time constant and the dead time of the process from the integrating arithmetic result are provided in the optimum PID arithmetic part. The integrating arithmetic part 7 integrates a controlled variable signal. Since the frequency of the noise component of the controlled variable signal is larger than a process equivalent response frequency, the noise component is canceled and the process characteristic can be extracted. With the extracted process characteristic, the optimum PID value is decided by using the method of Ziegler Nichols, for example, whereby the noise influence of the controlled variable signal can be reduced at the time of deciding the PID value.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a controller, and in particular outputs a step signal used for process control, obtains process characteristics from the integration result of the control response, and calculates the optimum PID value. The present invention relates to a controller suitable for setting.

[Conventional technology]

Conventionally, measuring device 986 Vo Q, 29, No.
As described in the literature entitled "Special Feature: Development Direction of Autotuning Controllers from the Perspective of Applicability" on pages 31 to 58 of LL, the process characteristics were determined and Ziegler-N1chols' The optimum PID value is determined based on the method. The step signal response method is used to measure process characteristics. A step signal is given to the process, the rate of change of the controlled variable signal is determined, and the process time constant is calculated from the maximum rate of change and the final value of the controlled variable signal. The process gain was determined from the dead time from the start to the rise of the controlled variable signal, the final value of the controlled variable signal, and the step signal width.

[Problem that the invention seeks to solve]

In the above-mentioned conventional technology, the process time constant is determined from the rate of change of the controlled variable signal, so when noise, etc. is added to the controlled variable signal, the process time constant is determined using the rate of change of the noise as the maximum rate of change, and the optimum PID value is determined. Therefore, the PID value may differ from the optimum value, and in order to prevent this, it is necessary to insert a noise filter or the like into the control amount signal.

An object of the present invention is to provide a controller that can determine a process time constant and dead time from an integrated value of a controlled variable signal and set an optimal PID value free from noise influence.

[Means for solving problems]

The above purpose is to provide the optimum PID calculation section with an integration calculation section and a process characteristic detection section that calculates the time constant and dead time of the process from the integration calculation results, and to calculate the optimum PID from the process characteristics.
This is achieved by reducing the influence of noise on the control amount signal when determining the D value.

[Effect]

The integration calculation unit integrates the control amount signal, whereby the noise component of the control amount signal is canceled out because the frequency of the noise component is higher than the process equivalent response frequency, and the process characteristic can be extracted. Since the optimum PID value is determined based on the extracted process characteristics using, for example, the Gee-Nichols method, it is possible to reduce the influence of noise on the control amount signal when determining the PID value.

〔Example〕

The present invention will be described in detail below with reference to the embodiments shown in FIGS. 1, 2, and 4, and FIG. 3.

FIG. 1 is an external view showing an embodiment of the controller of the present invention. In Fig. 1, reference numeral 1 denotes a controller body equipped with a microprocessor, which takes in a control amount signal from a process pressure transmitter 3, performs PID calculation on the received signal, and uses the result as an operation output signal. Output to the controller 4. 2 is the front panel of the controller, which includes a control amount and set value display 21, an operation output display 22. Operation output pushbutton switch 23. Set value pushbutton switch 24. It is equipped with an operation mode changeover switch 25 and an optimum PID tuning start switch 26.

FIG. 2 is a configuration diagram showing an example of the internal configuration of the controller shown in FIG. 1. In Figure 2, multiplexer 5. selects the analog input signal. 6. A/D converter that converts an analog input signal into a digital input signal. PI that takes in the digital input signal from the A/D converter 6 and performs PID calculation
D control calculation section 7, digital input section 8 which takes in signals from the pushbutton switches 23, 24, 25, 26 on the front panel 2 shown in FIG. A display section 9 that displays the controlled amount and set value on the controlled amount and set value display 21, a D/A converter 10 that converts the manipulated output value of the PID calculation result in the PID control calculation section 7 into an analog output signal, and an optimal An integration calculation unit 11 that performs @ calculation of the digitized control amount signal in response to a start command from the PID calculation unit 13. The process characteristic detecting section 12 calculates the gain, time constant, and dead time of the process from the integrated calculation value and the control amount signal, and calculates the optimum PID value from the determined process characteristics using, for example, the Ziegler-Nichols method.

The optimum PID calculation unit 13 sends a start command to the integration calculation unit 11 and the step signal generation unit 14 with the tuning start signal from the PID control calculation unit 7, and upon receiving the start command, switches the operation output changeover switch 15 and steps to the operation output side. It consists of a step signal generating section 14 that outputs a signal.

FIG. 3 is a diagram showing the operation of the integration calculation section 7 in FIG.
(a) is a normal waveform, and (b) is a waveform with noise added. Now, when we give a step signal to the process,
The response characteristics of the process are −order delay + dead time ke−”/
1 + T S (K: process gain, T: time constant, L: dead time). Integration calculation takes time t
When the control amount signal at is P V t and the sampling time is Δt, the integrated value St is.

St= St-mut+ PV tXAt
”・(1) Here, St−Δt
;Accumulated value from the previous sample. In other words, Fig. 3(a), (
This is the area of the shaded part S in b). The spring constant T of the process is given by PVs, where the end value of the controlled variable signal is PVse, the end time is ts, and the dead time is L.

It can be easily calculated using PVs. Controlled amount signal PV before step signal
When t is not zero, take the moving average of the control amount signal P V t before applying the step signal, and let that value be PVR. Equations (1) and (3) are respectively S t = S t-
Δt+ (P Vt-P VR)
4). During dead time, the integrated value Sc is ΔSX0.0
Let the time exceeding 5/2 (ΔS; step signal) be L', and find it as L=L'-0.05T. For process gain, K=PVs PVR/ASt'.

Even if noise is applied to PVs, since the integrated value is obtained, the effect may disappear when the time constant of the noise (process time constant) is the same.

FIG. 4 is an operation flow diagram of the controller of this embodiment. In step S1, input processing is performed and an analog input signal is input. Next, in step S2, key switch capture processing is performed to capture the operation output of the operation output pushbutton switch 23 on the front panel 2 in FIG. 1.Next, in step S3, the optimum PID tuning start switch 26 is pressed If it is off, PID calculation is performed in step S4 to create a manipulated output value. If it is on, the optimum PID setting operation begins in step S5. After the PID calculation is completed, display processing is performed in step S6, and the control amount and set value are displayed on the front panel 2. Next, in step S7, output processing is performed,
Outputs the manipulated output value in analog form. Next, in step S8, a calculation period wait (wait for Δ) is performed, and if the calculation period is found, the operation is performed again from the input processing in step S1. In the optimum PID setting operation, first, it is determined in step S5 whether or not the step signal output has been executed, and if not, the operation output changeover switch 15 is switched in step S9 to output the step output signal. If the output has been completed, Pv integration is performed in step S10. Next, in step S1, it is determined whether the cumulative value up to now is 5% or more, and when it becomes 5% or more, the dead time is calculated in step S12. , -Δt, and whether they are within the threshold for 0.1% integration is determined in step S13. If they are equal, T and L are determined in step 814. Next, in step S15, P, I, and D optimum values are calculated and set.

〔Effect of the invention〕

According to the present invention described above, the time constant and dead time of the process are easily determined from the integrated value of the control amount signal, and the optimum PID is calculated.
A controller that sets values can be realized.

[Brief explanation of the drawing]

FIG. 1 is an external view showing one embodiment of the controller of the present invention, and FIG.
The figure is a configuration diagram showing an example of the internal configuration of the controller in Figure 1, Figure 3 is a diagram showing the operation of the integration calculation section in Figure 2, and Figure 4 is a diagram showing the operation of the integration calculation section in Figure 2.
The figure is an operation flow diagram of the controller of the present invention. 1... Controller main body, 2... Front panel, 5... Multiplexer, 6... A/D conversion section, 7... PID
Control calculation section, 8... Digital input section, 9... Display section, 1o... D/A conversion section, 11... Integration calculation section,
12... Process characteristic detection order, ia... Optimal PID
Arithmetic unit, 14... Step signal generation unit, 15... Operation output changeover switch, 26... Optimal PID tuning start switch. Figure 3 (α) Rate 4 days

Claims (1)

[Claims] 1. An input unit that inputs a control amount signal from a process, and a PI that performs a predetermined control calculation on the input control amount signal.
D control calculation unit, an output unit that outputs a manipulated output signal, and an optimal PID calculation unit that applies a step signal to the manipulated output signal, determines the characteristics of the process from the response of the control amount signal, and calculates an optimal PID value. In the controller configured as above, the optimum PID calculation section is provided with an integration calculation section and a process characteristic detection section that calculates the time constant and dead time of the process from the integration calculation results, and the optimum PID calculation section is configured to calculate the optimum PID from the process characteristics.
A controller characterized by having a configuration that reduces the influence of noise on a control amount signal on a value.