JPS6229803B2 - - Google Patents

Description

[Detailed Description of the Invention] Industrial Application Field The present invention is applicable to electric power system control, such as control of electric motors used for generation active power control (ALR), reactive power control (AQR), power control (AVR), etc. Concerning a suitable method for doing so.

Conventional technology To describe the conventional technology using generated active power control (ALR) in power system control as an example,
The target generated active power value P ₀ given as shown in the figure,
The deviation ΔP of the measured value P of generated active power is determined, and the motor control device 1 continuously converts it into a control signal according to ΔP, and drives the governor motor 2.
It controlled the governor and changed the output of the generator 3. As shown in FIG. 2, the pulse period of the control signal output from the motor control device 1 differs depending on the deviation ΔP. The smaller the absolute value of ΔP, the larger T and the coarser the pulse train, and the larger the absolute value of ΔP, the smaller T and the denser the pulse train.

This control pulse train is continuously applied until the dead zone determined by the algorithm of the controller is reached. On the other hand, the response of the measured generated active power P, which is the feedback signal, has a delay of several seconds, although it varies depending on the system load. Furthermore, the governor motor and the governor connected to the governor motor also have large inertia, and the response rise characteristics are poor. On the other hand, when the response starts, the control pulse signal received during the response delay acts as an integral quantity and rises rapidly, causing an overshoot and making the control system unstable, as shown in Figure 3 a. was. In order to prevent this, a method was adopted in which the dead zone of ΔP was set large as shown in _FIG .

Problems to be Solved by the Invention According to the conventional method described above, control becomes unstable when the response delay of the controlled system is large, and it is difficult to reduce the dead zone of the control system.

SUMMARY OF THE INVENTION An object of the present invention is to solve the above-mentioned drawbacks, and to provide a highly accurate control method that is stable even with response delays in a controlled system and can reduce the dead zone of the control system.

Means for Solving the Problems In order to achieve the above object, the present invention is characterized by using a predetermined control table (CTB) to generate pulse signals to control the target equipment, as shown in FIGS. 7 and 8. The point is that it is prepared and used according to the procedure shown in .

In other words, the control information at the current point in time is stored in the CTB, and at the next control time, the contents of the CTB are shifted in the direction that indicates the past on the CTB, and at the same time, the amount of deviation between the target value and the actual measured value is calculated at each sampling time. The required pulse width, pulse period, and control duration are calculated, the content of the CTB is changed while comparing with the previous control information, and control is performed based on the content.

Effects According to the present invention, since the width, period, and duration of the pulse to be output to the controlled object are stored in the CTB for a certain period of time, the control system is stable even for a control system with a large response delay. Control becomes possible.

Embodiments Embodiments of the present invention will be described below with reference to FIGS. 4 to 10.

The device used in this embodiment compares the target generated active power P ₀ with the measured generated active power P measured by the power converter, as shown in FIG.
It has a digital motor control device 5 which has a microcomputer as its core which applies a control pulse according to P to an amplifier 6 or 7 that drives the governor motor 2. The logic used in the control device 5 consists of a control amount determination logic that samples the measured amount at regular intervals and determines the control amount based on it, and a pulse control logic that outputs control pulses based on the determined control amount. be done.

In general, in the case of sampling control by a digital control device, the activation period _T1 of the control amount determination logic is given as a time period of 1/5 to 1/10 of the required control speed. For each period T ₁ , the deviation ΔP between P ₀ and P is determined by the control amount determination logic, and from ΔP, the pulse width P _W and the pulse period T _P are determined.
A control time T _C for maintaining this control condition is determined. The pulse width P _W is determined from the step-shaped pulse width curve shown in FIG. This pulse width curve is such that the pulse width increases as the absolute value of ΔP increases.

Considering the stability of the control system, the pulse period T _P is calculated using a nonlinear curve that increases as the absolute value of ΔP decreases, and is determined by the logarithmic curve of the following formula (1) or (2).
Use the quadratic curve of

T _P =K・e−|ΔP| …(1) T _P =a・|ΔP| ² +b・|ΔP|+c
...(2) Note that k, a, b, and c are control characteristic constants.

The control time T _C determines how long T _P and P _W determined above are to be continued, and is given by the following equation (3).

T _C = (ΔP/P _F ) × T _F ...(3) Here, P _F : Band value from 0 to the maximum generated active power value T _F : Minimum pulse period, maximum pulse width, 0 The control table CTB shown in Fig. 6 is created using P _W , T _P , and T _C over the time required for control up to the maximum generated active power value. The CTB consists of a part that records past control history and a part that sets future control pulse flags. Also, by giving the maximum pulse width (P _W1 in Figure 5) as the time for one memory of this CTB (time from t to t+1), if T _P ≦ P _W , each memory of the CTB Control pulse flags are set in all cases so that continuous pulses can be obtained.

Now, when setting the control pulse flag in the control table CTB, compare T _P (t') when it was set in CTB last time with T _P (t) this time, and control as follows based on the result. Assumed to be performed.

1 When T _P (t')≦T _P (t): Since |ΔP| is decreasing under the current control conditions, according to the control table CTB, until all the flags in CTB are shifted to the past, C.T.B.
Control is performed without changing.

2 When T _P (t') > T _P (t): |ΔP| is in the divergent direction, so the current T
Write CTB by _P , P _W , and T _C .

In this case, in order to maintain continuity with the previous pulse output time, the previous pulse output time is
The control pulse flag is determined from t-1 to t-m of CTB, and the control pulse flag is sent to CTB from that position at a period _of T _P.
Set until the time of.

The flow of the control amount determination logic described above is shown in FIG. Further, the flow of the pulse control logic for outputting pulses is shown in FIG. The pulse control logic is based on one memory time (t~
If the control pulse flag is set at the t position of the CTB, a pulse signal with a pulse width of P _W is applied to the governor motor 2 via the signal amplifier 6, and the CTB is synchronized with the time t+1. This is to shift one memory to the left.

The flows shown in FIGS. 7 and 8 will be explained below using examples.

First, block 7 of the control amount determination logic in Fig. 7
1, target generated active power P ₀ and measured generated active power P
Find the deviation ΔP.

ΔP=|P ₀ −P| (4) In block 72, the pulse period T _P and pulse width P _W are determined from the curve shown in FIG. 5 based on the result of equation (4).

For example, when ΔP=3.0MW, T _P =1500ms,
Assume that P _W =500ms is obtained.

In block 73, the control time T _C is determined using the above-mentioned equation (3). For example, P _B =20.0MW, T _B
= 80 sec, T _C = (3.0 to 20.0) x 80 = 12 sec.

Next, in block 74, T _P (t') when the control table CTB was created at the previous time t' is compared with T _P (t) created from ΔP(t) at the current time t.

1 When T _P (t')≦T _P (t): Since ΔP is decreasing, the program jumps to block 75 which checks whether the control pulse flag still remains in the CTB.

2 When T _P (t') > T _P (t): ΔP becomes large and a dense pulse is required, so new T _P (t), P _W (t)
Jump to block 76 of the routine which rewrites CBT according to T _C (t).

In block 75, the contents of t to t+n in FIG. 6 are checked, and if the control pulse flag is set, it is determined that the previous control amount still remains, and the next sampling start time _T1 is set. Jump to processing block 77.

If the control pulse flag is not set, T _P (t), P _W (t), T _C at the current time t
It is necessary to create a CTB in block 78 (t).
Hejijump.

In block 76, in order to avoid discontinuity in the contents of CTB, it is checked whether the control pulse flag is set in the direction from t-1 to t-m in FIG. 6, and the elapsed time from there is checked. If the time when the latest pulse in the past was output is tm', the time _T3 after the pulse is output is t''=m'× _T2 .

However, T ₂ is the activation period of the pulse control logic.

1 When T _P > T ₃ Since the time T _P has not elapsed since the previous control pulse was output, the processing block 35 for creating CTB is jumped starting at t-m'.

2 When T _P ≦T ₃ Jumps from the current time t to processing block 78 for creating a CTB.

In block 35, starting from t-m' of CTB, the control pulse flag is set at a period of T _P until time T _C . An example of this is shown in FIG. 9b. Note that a in FIG. 9 shows the state before the control amount determination logic flow operates at time t, and T _P
= 2500 ms and P _W = 300 ms.

Block 78 is the control table CTB of FIG.
Is there a control pulse flag between t and t+m?
(state c in Figure 9), or if more than T _P has already elapsed from the time t-m' when the previous pulse was output to the current time t (state d in Figure 9)
It is. Therefore, the control pulse flag is set in CTB starting from the current time t. This example is the 9th
Shown in figure e. In addition, T _P and P _W in c and d of Fig. 9
is the same as in a.

After the above processing is completed, in block 77, the cycle time _T1 for the next activation is set, and the processing flow of the control amount determination logic ends.

The pulse control logic flow in FIG. 8 is activated every _T2 cycles. In block 50, it is determined whether a control pulse flag is set at the current time t in the control table CTB.

If the control pulse flag is not set, a jump is made to block 70.

When the control pulse flag is set,
Since the pulse width of P _W requires control, the process jumps to block 55.

In block 55, the sign of the deviation between the target generated active power _P0 and the measured generated active power P is determined to determine whether the output is increasing or decreasing.

In case of an increase in output, jump to block 60.
In block 60, the control pulse with the pulse width P _W calculated by the control amount determining logic shown in FIG.
It is output to the output increasing direction signal amplifier 6 shown in the figure. In the case of power reduction, block 65 outputs a control pulse having the same pulse width P _W as in block 60 to power reduction direction signal amplifier 7 in FIG.

In block 70, since the control that started from time t has ended, the control table CTB1 memory is shifted in the past direction (leftward in FIG. 6) to prepare for activation after _T2 .

In block 75, the activation period _T2 of the next pulse control logic flow is set, and the entire flow ends.

FIG. 10 shows another embodiment of the present invention, in which a plurality of electric motors are controlled by this method. In this case, by having control tables CTB for the number of motors to be controlled, control can be performed in the same process by looping the control amount determination logic and pulse control logic for the number of motors.

Furthermore, even when control systems including multiple motors have different characteristics, according to this method, the constants that determine the control pulse period T _P and pulse width P _W shown in Fig. 5 can be changed for each control system with different characteristics. A compact and economical control system can be constructed without adding any control logic, and the only parameters to be adjusted are the constants that determine T _P and P _W in Figure 5. Adjustments can be made easily.

Effects of the Invention As explained above, according to the present invention, the predictive control function determines the absolute value of the controlled variable according to the amount of deviation between the target value and the measured value, and maintains that state for a certain period of time. , it is stable even with large response delays in the controlled system, and the dead zone of the control system can be kept small, allowing for highly accurate control.

In addition, this method can be incorporated into a digital control device with a memorandum function such as a digital microcomputer, rather than a conventional wire logic control device that handles analog quantities, resulting in a compact and economical control device. can be provided.

[Brief explanation of the drawing]

Figure 1 is a generated active power control block diagram, Figure 2 is a signal output state diagram using the conventional control method, Figure 3 is a control characteristic diagram of the conventional control method, and Figure 4 is a diagram showing the control method according to the present invention. Generated active power control system block diagram, Figure 5 is a pulse period and pulse width determination characteristic diagram, Figure 6 is a control table diagram, Figure 7 is a diagram of the control table.
8 is a logic flow diagram for determining a control amount, FIG. 8 is a diagram for explaining a pulse control logic flow, FIG. 9 is a diagram for explaining a control table creation method, and FIG. 10 is a diagram for explaining another embodiment of the present invention. 1... Motor control device, 2... Governor motor, 3... Generator, 4... System, P ₀ ... Target active power value.

Claims (1)

[Claims] 1. A method of applying a controlled variable to a controlled device in the form of a pulse signal having a predetermined width, period, and duration, thereby controlling the operation of the controlled device, in which the control is performed at a predetermined sampling period. Calculate the deviation between the target value and the actual measured value of the quantity, and based on this, calculate the control information including the width, period, and duration of the pulse signal to be given to the controlled device at the first control time, and store it in the storage means. The (+1)th
At the time control time, the storage position of the control information at the time point stored on the control table is moved in the direction indicating the past on the control table, and the above control information at the time (+1) time is moved based on the deviation. A digital device that calculates the width, period, and duration of a pulse signal, changes the contents of the control table while comparing it with the control information from the first time, and controls the operation of the controlled equipment according to the contents. Control method.