JPH0628487B2 - Control method of reactive power compensator installed in power system - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a control system of a reactive power compensator installed in a power system for the purpose of suppressing flicker and the like.

[Prior Art and Problems] In order to suppress a voltage fluctuation caused by a fluctuating load connected to a power system, for example, a fluctuating load current having a steep delay current such as in an arc furnace, reactive power compensation as shown in FIG. 2 is performed. The device is being used.

In the figure, 1 is a power supply, 3 is a flicker countermeasure busbar generated by a fluctuating load 6, and 7 is a general load (base load). Further, 2 is a power source impedance.

10 is a reactive power compensator (hereinafter referred to as SVC),
An antiparallel connection thyristor 12 and a reactor 11 are connected in series, and are connected to the bus 3 in parallel with the variable load 6.

The voltage detecting transformer 4 is connected to the bus bar 3, the current detecting transformer 5 is coupled to the circuit of the variable load 6, and the transformers 4 and 4 are connected.
The outputs V _L and i _L of 5 are input to the Q detection circuit 20, the reactive power Q is calculated by calculation, and the output Q of the Q detection circuit 20 is
The bias signal Q _b from the fixed bias circuit 21 capable of adjusting the voltage is subtracted, input to the pulse generation unit 22, converted into an ignition phase pulse for controlling the thyristor switch 12, and the energization control of the reactor 11 is performed.

This method is for fixing the bias Q _b by the fixed bias circuit 21 and its operation is as shown in Figure 4. In this method, it is necessary to set in consideration of the long period fluctuation Q, and a large capacity SVC is required.

In order to improve such drawbacks, the system shown in Japanese Patent No. 999598 has been proposed. This method is expected Q
(Rapidly changing Q) detector and average Q detector are prepared, the latter Q is subtracted from the former Q, 1/2 of the predicted Q is added with a fixed bias, and the average Q fluctuation is calculated from all Q fluctuations. Subtract the minutes and control the SVC only with the sudden change.

In this method, half the expected Q is set to a fixed bias, so if the long-cycle fluctuation is faster than expected due to the actual fluctuating load, there is a time delay in the average Q fluctuation detector.
As shown in (d), SVC becomes insufficient compensation and residual flicker occurs in the system.

FIG. 5 is a diagram for explaining the actual operation of the above-mentioned conventional example, in which 40 is a load Q signal waveform, 41 is a base Q detection signal, and 42 is a bias Q _b waveform. Waveform (b) is the difference waveform of the base Q detection signal from the load Q signal,
(c) Figure shows a waveform 43 by adding a bias Q _b on the difference of the base Q detection signal from the load Q signals, (d) The figure shows the state of the Q operations particularly insufficient compensation.

[Object of the Invention] In order to solve the drawbacks of the above-described method, the present invention subtracts a fixed Q detected from a reactive power detection signal with high accuracy and controls an SVC only by a sudden change Q variation (ΔQ), and It is in the control system of the reactive power compensator that can reduce the compensation capacity of the device and enables more economical equipment, and in short, as a means for removing the long-cycle reactive power fluctuations that do not affect the voltage flicker, The average reactive power (average Q) is detected by the detector, and the sudden change Q fluctuation (ΔQ) is detected from this signal.
This is a control method characterized by subtracting a variation corresponding to 1/2 and automatically biasing a portion corresponding to 1/2 of the ΔQ so as to correspond to a sudden change in Q.

The present invention will be described below with reference to the drawings. The control circuit of the present invention is shown in FIG. 1, and this circuit can be replaced by the circuit after the Q detector 20 shown in FIG. The same parts as those in FIG. 2 are designated by the same reference numerals.

The low-pass filter 27 and the inverter 28 are connected to the output side of the Q detector 20, and the output Q of the adder 29 and the output Q _{av of the} inverter 28 are connected to the adder 29.
To enter. Further, the adder 29 has _a ΔQ _{a ε} obtained at the output side of the low-pass filter 30 connected to the output side of the adder 29.
Is added to the set value Q _s by the specified value 1/2 setter 32 to add a coefficient
The ΔQ _ε obtained through 31 and the Q _s signal from the setter 32 are input. The circuit for obtaining ΔQ _ε is an automatic bias correction circuit, which is designated as A _b . Then, the adder 29 is connected to the pulse generator 22 and outputs a pulse signal. The set value Q _s of the set value 1/2 setter 32 is defined by, for example, an average value of reactive power of sudden change.

(1) Now, a case where the ΔQ fluctuation matches the expected specified Q fluctuation will be described with reference to FIG.

In the figure, Q _max indicates the maximum reactive power value detected, ΔQ indicates the predicted reactive power for the sudden change, and the fixed Q
Indicates a reactive power fluctuation that does not become a member of the voltage flicker occurrence, Q _s is 1 / 2ΔQ, and Q _av is a fixed Q + 1 / 2ΔQ. Under the above conditions, ΔQ = Q _max − (Q _av −Q _s ) ± ΔQ _ε (1), where (Q _av −Q _s ) is a fixed Q and ΔO
_ε is the output of the automatic bias correction circuit A _b shown in FIG.

Here, ΔQ _ε will be described. When the output ΔQ of the adder 29 of FIG. 1 is averaged by the low-pass filter 30, ΔQ _{a ε}
Is obtained, and in this case, Q _s = ΔQ _{a ε} . If the coefficient unit K = 1 now, ΔQ _ε = K (Q _s −ΔQ _{a ε} ) = 0.

Therefore, the equation (1) becomes ΔQ = Q _max −fixed Q (Q _b ). That is, each input of the adder 29 is summed and ΔQ is output.

However, there are cases where ΔQ fluctuates more than expected and cases where it does not, and these cases will be described.

(2) ΔQ fluctuation is smaller than the Q fluctuation specified by prediction. If the automatic bias correction circuit _Ab is not provided, then FIG.
As shown in. When expanded in the same manner as in (1), if the output as the control signal is ΔQ ′, then ΔQ ′ = ΔQ + ΔQ _ε
Therefore, controlling the SVC with this is not efficient.

In this case, the ΔQ _ε generated by the automatic bias correction circuit A _b will be described. The output ΔQ _{a ε} of the low pass filter 30 shown in FIG.
_s, and the difference is ΔQ _ε = Q _s + ΔQ _{a ε} , which is added by the adder 29. Therefore, the equation (1) becomes ΔQ = Q _max − (Q _av −Q _s ) + ΔQ _ε ... (2).

In other words, the automatic bias correction circuit A _b
It means that the SVC can be controlled only by ΔQ by raising Q _ε .

(3) When the ΔQ fluctuation is larger than the Q fluctuation defined by the prediction, it becomes as shown in FIG. SVC as in (1)
Is, Delta] Q that there is no automatic bias compensation circuit A _b '= ΔQ-
Since it is controlled by ΔQ _ε , the accuracy is reduced only by ΔQ _ε .

Explaining ΔQ _ε as in (2), the output of the low-pass filter 30 in FIG. 1 is compared with Q _{a ε} and Q _s , and the difference is ΔQ _ε = Q _s −ΔQ _{a ε} , which is added by the adder 29. To be done. That is, the specified 1/2 set value Q _s is corrected by ΔQ _ε .

Therefore, the equation (1) becomes ΔQ = Q _max − (Q _av −Q _s ) −ΔQ _ε ... (3).

In reality, the states described in (2) and (3) above will occur during operation.

As shown in FIG. 3 (b), the base Q detector signal detects the average reactive power Q _av passing through the center of ΔQ in the entire Q signal, and therefore, when ΔQ is subtracted from Q, Q _av is obtained. A reactive power signal of ± ΔQ / 2 is obtained at the center. This is SV
Because C can only Delta] Q / 2 compensation, the automatic bias compensation circuit A _b of FIG. 1, there is further correcting the Delta] Q / 2, accurately detects Delta] Q sudden change content, SVC in the signal To control.

[Effects of the Invention] As described above, according to the present invention, by adding the automatic bias circuit, the reactive power sudden change (ΔQ) can be accurately detected, the equipment capacity can be the same as ΔQ, and the high performance. Equipment can be expected, efficient and economical equipment can be realized.

[Brief description of drawings]

FIG. 1 is a block diagram showing a control circuit of the present invention. FIG. 2 is a block diagram showing a reactive power device to which the present invention is applied. 3 (A), (B), and (C) are operation explanatory diagrams by the control circuit of the present invention. FIG. 4 is an operation explanatory diagram when the conventional control circuit is used in the apparatus of FIG. 5 (a), (b), (c), and (d) are operation waveform diagrams of respective parts of the conventional reactive power compensator. 3 ... Flicker countermeasure bus bar, 4 ... Transformer for voltage detection,
5 ... Transformer for current detection, 20 ... Q detector, 21 ... Fixed bias circuit, 22 ... Pulse generator, 27 ... Low pass filter, 28 ... Inverter, 29 ... Adder, 30 ... … Low pass filter, 31 …… Coefficient multiplier, 32 …… Specified value 1/2 setter, 33
…… Comparator.

Claims (1)

[Claims] 1. A reactive power compensator for automatically subtracting a long-cycle reactive power fluctuation from a reactive power of a grid, detecting only a reactive power signal component of a sudden change, and controlling based on the reactive power signal. As a means for removing long-cycle reactive power fluctuations that do not affect flicker, a reactive power signal detected by a Q detector and an average reactive power signal are calculated by a low-pass filter connected to the Q detector, and both are added to make a sudden change Q. A signal is detected, and a signal corresponding to 1/2 of the expected reactive power fluctuation is added to this signal to form a combined reactive power signal, and the combined reactive power is corrected in order to correct the fixed 1 / 2Q. A signal obtained by averaging the first-order lag, and the specified half
Control of the reactive power compensator installed in the power system, characterized in that the reactive power compensator is controlled only by the reactive power signal component which is input to the adder in comparison with the set value by the setter and suddenly changes. method.