JPH0350489B2 - - Google Patents

Description

[Detailed Description of the Invention] (a) Description of the Technical Field The present invention provides a DC multi-terminal power transmission system consisting of a plurality of converters that can quickly switch the converter that determines the voltage. This invention relates to a control device for a multi-terminal power transmission system.

(b) Description of Prior Art In order to further utilize the advantages of conventional DC two-terminal power transmission, there is a strong desire to develop technology for DC multi-terminal power transmission.

FIG. 1 is a DC four-terminal power transmission system diagram to which the present invention can be applied, in which 1 to 4 are AC systems, 5 to 8 are conversion transformers, 9 and 10 are forward converters, and 11 and 12 are reverse converters. It is a converter.

Compared to DC two-terminal power transmission, DC multi-terminal power transmission has
More advanced cooperative control between each converter is required.
For this reason, in DC multi-terminal power transmission, mutual information from each converter is collected and monitored by a central control unit via a transmission system (not shown), and based on this information, optimal operating command values are given to each converter. Now, as a DC multi-terminal control method, the conventional two-terminal power transmission control method is extended to multi-terminal power transmission, and the characteristic diagram of one example of this control method is shown in Figure 2. Shown below. Below, for convenience of explanation,
In FIG. 1, the forward converter 9 is REC1, the forward converter 10 is REC2, the inverse converter 11 is INV1, and the inverse converter 12 is INV2, and the current setting values of the constant current control devices of REC1 and REC2 are respectively I _dpr1 , I _dpr2 and the current value actually flowing to REC1 and REC2 as I _dr1 ,
I _dr2 or the current setting values of the constant current control devices of INV1 and INV2 are abbreviated as I _dpi1 and I _dpi2 , and the current values actually flowing through INV1 and INV2 are abbreviated as I _di1 and I _di2 . Now, in FIG. 2a, the operating points of each converter in the steady state are P ₁ , P ₂ , P ₃ , and P ₄ . That is, RCE2 determines the voltage of the DC system by operating at a constant control angle, and the other REC1, INV1, and INV2 perform constant current control to determine the current. Also, the second
In Figure b, INV1 determines the voltage of the DC system by constant margin angle control, and the other REC1, REC2,
INV2 performs constant current control and determines the current. In order for this control method to operate stably,
The following conditions must be met:

(I _dpr1 + I _dpr2 ) - (I _dpi1 + I _dpi2 ) = ΔI>0 ... (I _dr1 + I _dr2 ) = (I _di1 + I _di2 ) ... In other words, the sum of the current setting values of the forward converter is must be greater than the sum of the current settings. In other words, the value obtained by subtracting the sum of the current settings of the inverse converters from the sum of the current settings of the forward converters (hereinafter referred to as
This is called current margin. ΔI in the formula. )
must be positive.

As mentioned above, as long as this method satisfies the formula,
Although stable operation is possible, it has the following drawbacks. First, when stopping the voltage determining terminal, the voltage determining function must be transferred to another converter and the converter must be set to constant current control mode before the current can be set to 0 and the converter cannot be stopped. This requires time-consuming control of the converter transformer tap and changing the no-load DC voltage of the converter.
Rapid control is not possible. Second, voltage fluctuations in the AC system at the voltage determining terminal directly become voltage fluctuations in the DC system, causing power fluctuations at other healthy terminals.

Another example of a control method to compensate for the above drawback is to equip each converter with a constant voltage control device in addition to a constant current control device, and to introduce the concept of voltage margin in addition to the concept of current margin described above. and
If a certain converter is determined as a voltage determining terminal, the voltage setting value of the constant voltage control device of that converter is set within a predetermined voltage margin (hereinafter referred to as ΔV) than the voltage setting values of all other converters. (abbreviated)), and the current setting value is set so that the formula holds true.

A characteristic diagram using this method is shown in FIG. 3a. In FIG. 3a, a voltage margin is given to REC2, and REC2 serves as a voltage determining terminal. When moving the voltage determining terminal to INV1, it is sufficient to move the voltage margin given to REC2 to INV1, and tap control of the transformer is not necessary. As a control device for realizing this, the one shown in FIG. 4 has been considered.

In FIG. 4, I _dp is a constant current setting value, and although this value may differ depending on each converter, the relationship in the equation must be satisfied. V _dp is a set value of the constant voltage control device and is a value common to all converters, and at the voltage determination terminal, the final set value is obtained by subtracting the voltage margin ΔV from this value. |V _d | is the absolute value of the DC voltage measurement value, I _d is the DC current measurement value, 13 is a constant current control device, 14 is a constant voltage control device for the forward converter, 15 is a constant voltage control device for the inverse converter, 16 is a constant margin angle control device, 17 is a maximum value selection circuit, and 18 is a minimum value selection circuit.
With such a control circuit, the stable operating point shown in FIG. 3a can certainly be obtained. However, as it is,
The characteristics during transients, such as when switching the voltage margin, cannot be said to be sufficient. In other words, since REC1 is not a voltage determining terminal, voltage margin ΔV and current margin ΔI are not given, and the output of the constant current control device is changed to the maximum value selection circuit 1 so that I _dpr1 = I _dr1 .
7. Although selected by the minimum value selection circuit 18, the constant voltage control device 14 outputs 14 to reduce the control angle because the set value V _dp is larger than the measured value |V _d | by approximately the voltage margin ΔV. It is saturated in the direction of In such a situation, even if you try to move the voltage margin from REC2 to REC1 and move the voltage determination terminal, REC1 will be changed due to the following reasons.
It cannot be moved until the constant voltage control device 14 is released from saturation.

In other words, in Figure 4, the REC before switching
1 is assumed to be operating with constant current control (ACR) at a control angle of 20 degrees. REC at this time
1, constant voltage control, AVRR, has a set value of Vdp.
1.05, whereas the measured value |Vd| is 1.0, so in an attempt to further increase the voltage, the voltage is saturated to a value corresponding to a control angle of 0 degrees. Therefore, before switching the voltage determining terminal, the maximum value selection circuit 17 selects the control angle of 20 degrees, which is the output of the constant current control circuit 13.

Note that the output of the constant margin angle control circuit 13 is normally 130
The minimum value selection circuit 18 also selects 20 degrees, which is the output of the constant current control circuit 13.
REC1 is operated at 20 degrees.

When the setting value is changed to switch the voltage determining terminal from REC2 to REC1, the operating points finally shift to P1', P2', P3', and P4' shown in FIG. 3b. To change the specific setting value, change Idpr1 of REC1.
The value is Idr1 plus current margin ΔI, and AVRR
For , subtract the voltage margin ΔV. vice versa
Set Idpr2 of REC2 equal to Idr2 and cancel ΔV that was subtracted from the input of AVRR.
But to reach this new stable operating point
The output of the AVRR must change from 0 degrees, where it was saturated, to 20 degrees. In this case, in contrast to high-speed constant current control, constant voltage control generally consists of a first-order delay circuit and a gain, and in order to ensure stability during steady state, its time constant is very slow, on the order of seconds.

Therefore, it is desired that the constant voltage control AVRR of REC1 quickly perform its original function and play the role of determining the voltage of the grid, but it takes time. During this time, there is a risk of falling into an unstable state in which all converters perform constant current control.

Similarly, when moving the voltage determining terminal from REC2 to INV1 or INV2, when the constant voltage control device 15 is saturated (the control angle is saturated in the direction of 180 degrees, the control angle is changed to around 140 degrees). It takes time to release, and the voltage determining terminal cannot be moved quickly.

(c) Purpose of the Invention The purpose of the present invention was to eliminate such drawbacks, and to provide a control device for a DC multi-terminal power transmission system that can quickly switch voltage determining converters and has good transient characteristics. It is about providing.

In order to achieve this objective, the present invention provides that all converters generate a signal corresponding to a control delay angle that generates a predetermined DC voltage from the measured values and voltage settings of the respective DC currents and voltages of the AC system. The present invention is characterized by comprising a constant voltage control means for generating a constant voltage, a constant current control means for controlling the DC current to a predetermined value, and a selection means for selecting one of these means.

(d) Structure of the invention The present invention will be explained below with reference to the drawings.

FIG. 5 is a control block diagram showing one embodiment of the present invention. In FIG. 5, 19 to 21 are adders, 22 is a constant margin angle control device, 23 to 25 are switches, 26 to 28 are limiter circuits, 29 is a level detector, 30 is a first-order delay circuit, and 31 is a proportional circuit. ,
32 is a constant current control device (hereinafter abbreviated as ACR), 33
34 is a maximum value selection circuit, and 35 is a minimum value selection circuit. The constant current control setting value I _dp , the constant voltage control setting value V _dp , and the constant voltage operation mode signal are transmitted from a central control device (not shown). Even without using the concept of voltage margin, since the converter with the lowest constant voltage setting value performs constant voltage control operation, the central control unit does not apply constant voltage control to the converter with the lowest constant voltage setting value. When the operation mode signal is set to "1", I _dp is given a value obtained by adding a current margin ΔI from the operating current in the case of a forward converter, or subtracting a value in the case of an inverse converter. For other converters, constant voltage operation mode signal “0”,
The same I _dp as the operating current is given. switch 2
3 selects the output of the adder 20 when the logic level of the constant voltage operation mode signal is "1", and selects zero when it is "0". The level detector 29 has a logic level output of "1" when the input is a positive value, at which time the switch 24 is closed and the switch 25 is open, and vice versa when the input to the level detector 29 is negative.

(e) Operation of the invention Now, the operation of the control block diagram of FIG. 5 will be explained by taking as an example the case of operation as shown in FIG. 3a.

The constant voltage control device 33 measures the voltage E _ac and the DC current I _d of the AC system, and determines the control delay angle (hereinafter abbreviated as α) so as to satisfy the following equation.

V _dp = V _dp (cosα−X _pu /2 i _pu / e _pu )... Here, V _dp : Constant voltage control setting value (converter direction is positive) Positive value; Forward converter Negative value ; Inverter V _dp : No-load DC voltage X _pu : Commutation reactance (pu value) based on transformer rating i _pu : DC current (pu
value) e _pu : AC system voltage (pu value) Therefore, whether the converter performs forward converter operation or inverse converter operation can be determined by the number attached to the constant voltage control setting value, especially when transmitting signals. do not have to.
Therefore, the measured value V _d of the DC voltage also takes the direction of the converter as positive, and is positive in the forward converter and negative in the inverse converter. It is assumed that the forward converter REC1 in FIG. 3a is operated by ARC at α=20°. Note that for convenience of explanation below, the set value and measured value will be expressed in pu. Since REC1 is a converter that performs constant current control during normal operation, the constant voltage operation mode signal is “0”.
The switch 23 selects zero, and the limiter circuit 26
The output of the proportional circuit 31 via the first-order lag circuit 30 is also zero, and the CVC 33 outputs an output α _cvc (for example, α _cvc =10°) such that V _dp =1.05 pu according to the formula.
In Figure 3, the constant current setting value and the measured DC current value match, and the output of the adder 19 becomes 0.
The output of ACR32 remains at α=20°. on the other hand,
Since REC1 is operating as a forward converter, V _dp is a positive value, the output of level detector 29 is "1", switch 24 is closed, and switch 25 is open. Therefore, the final output α of the control device is
Out of the output 20° of the ARC32 and the output 10° of the CVC33, 20° is selected by the maximum value selection circuit 34, and the value becomes 20° via the limiter circuit 27 (usually 10°≦α≦120° in the case of a forward converter). Performs constant current control.

Similarly, considering REC2 which is in constant voltage control operation, the open/close states of switches 24 and 25 are exactly the same as REC1 because it is a forward converter, but switch 23 has a logic level of constant voltage operation mode signal of "1'', the output of adder 20 is selected. If the measured value V _d of the output voltage of the converter is 1.0 pu, which is the same as the set value V _dp , the output of the adder 20 will be 0, and the output of the proportional circuit 31 via the limiter circuit 26 and the first-order lag circuit 30 will also be 0. but,
If V _d is not 1.0 pu due to a control error of the CVC 33, etc., the adder 20 detects the deviation and outputs it to the proportional circuit 31 via the limiter circuit 26 and the first-order delay circuit 30.
The set value inputted to the CVC 33 is slightly corrected by the amplification adder 21. In any case, CVC
33 outputs α such that V _d is 1.0 pu according to the formula (for example, α _cvc =20°). On the other hand, since the constant current set value is larger than the DC current by the current margin ΔI, the output of the adder 19 becomes −ΔI. accordingly
The ARC 32 attempts to increase the DC voltage in order to increase the current, and the output α _ACR of the ACR 32 reaches its lower limit (usually 10 degrees). The maximum value selection circuit 34 selects the larger of α _CVC and α _ACR (=20° ₎
is selected, and a final output α=20° is obtained through the limiter circuit 27 and switch 24, that is, constant voltage control operation is performed.

Considering inverter INV1 operating at α = 140° under constant voltage control, V _dp is a negative value -
Since 1.05 pu is given, the output of the level detector 29 is "0", and the switch 24 is opened and the 25
is closed. Since the constant voltage operation mode signal of INV1 is "0", switch 23 selects zero,
The output of the proportional circuit 31 becomes zero as in the case of REC1. Therefore, the CVC33 is set to a control angle that outputs a voltage of -1.05pu according to the formula, for example, α _CVC
Output = 145° (-1.0pu at α = 140°.)
On the other hand, constant current control setting value I _dpi1 and DC current measurement value I _di1
Since they match, the output of the adder 19 is 0, and the output of the ACR 32 remains α _ACR =140°. Since the output of constant margin angle control No. 22 is normally larger than α _CVC unless the voltage of the AC system drops significantly, the minimum value selection circuit 35 selects the smaller of α _CVC and α _ACR , α _ACR (=140°). do. The limit value of the final stage limiter circuit 28 for the inverter is usually 85°≦α≦
It can be seen that α _ACR is passed through the switch 25 as it is and becomes the final output to perform constant current control.

Regarding the inverter INV2, constant current control is performed in the same way as INV1.

Now, while operating in this manner, consider moving the voltage determining terminal from REC2 to REC1 as shown in FIG. First, the central control device
Change the V _dp of REC1 to 1.0 and the V _dp of REC2 to 1.05, and move the converter whose constant voltage operation mode signal is "1" from REC2 to REC1.

Furthermore, I _dp (I _dpr1 ) of REC1 is increased by ΔI,
Decrease I _dp (I _dpr2 ) of REC2 by ΔI. The output _α of the constant voltage control device 33 of REC1 is V _dp before the change.
Since the temperature was 1.05 pu, α _CVC = 10°, but after the change, V _dp becomes 1.0 pu, so based on the formula, α _CVC = 20°. In addition, switch 23 of REC1
Since the constant voltage operation mode signal becomes "1", the output of the adder 20 is selected, but since both V _dp and V _d are 1.0 pu, it does not affect α _CVC . For example V _d
Even if it is not 1.0pu, the deviation from 1.0 is slight, the amount of correction is small, and α _CVC is still around 20°. On the other hand, regarding constant current control, since the set value is larger than the measured value by the current margin ΔI, the converter output voltage is increased in order to increase the current from α _ACR = 20°, that is, α _ACR =
Output decreases in the direction of 10°. Therefore, the maximum value selection circuit 34 selects α _CVC instead of α _ACR . That is, REC1 immediately shifts from constant current control operation to constant voltage control operation.

Regarding REC2, since V _dp changes from 1.0 to 1.05, the output α _CVC of the CVC 33 changes from 20° to 10°. Since the output α of ACR32, _ACR, was saturated at 10° immediately before the change, the REC immediately after the change is
The control angle of 1 is 10°. For this reason, the DC current begins to increase beyond the set value, but the ACR32 control generally reduces the output to α _ACR =
Since the current is increased from 10° to α _ACR = 20° immediately, the current fluctuation does not become too large. Therefore, immediately after changing the set value, the maximum value selection circuit 34 selects α _ACR and performs constant current control.

There is no change in INV1 and INV2. Therefore, although the position of the operating point does not change immediately from that shown in FIG. 3, the characteristic diagram immediately changes to that shown in FIG. 6 after changing the set value, and the voltage determining terminal moves from REC2 to REC1.

Change the voltage determination terminal from REC2 to INV1 or INV2
The same is true when moving to .

As described above, constant voltage control in a DC multi-terminal system can be performed using the CVC33 shown in Fig. 5, without using conventional feedback control that reduces the deviation between the DC voltage setting value and the DC voltage measurement value to 0. Based on the formula, the control angle that generates the DC voltage corresponding to the DC voltage setting value is calculated arithmetically from the measured value of DC current and the measured value of AC system voltage, so the DC voltage setting Even if there is a deviation between the DC voltage measurement value and the DC voltage measurement value, the constant voltage control circuit will not become saturated. Therefore, since the correct output can be output immediately upon changing the set value, the voltage determining terminal can be quickly switched. In addition, in FIG. 5, the adder 20, switch 23, limiter circuit 26, first-order lag circuit 30, and proportional circuit 31 are designed so that only the voltage determining terminal can be subjected to minute corrections, but the other current determining terminals are No delay element is added to the constant voltage control circuit.

(f) Other Embodiments The constant current control device 32 in FIG. 5 is saturated in a converter performing constant voltage control operation. As explained above, constant current controllers generally perform fast control, so the saturation limit value is set, for example.
Even if fixed at 10°≦α≦160°, it takes less time to recover from saturation than with conventional feedback-type constant voltage control devices, and there is less negative impact on the system, but the saturation of constant current control devices By making the limit value variable, the transient characteristics can be further improved. A specific example is shown in FIG. In FIG. 7, the same elements as in FIG. 5 are indicated by the same reference numbers. 36,
37 is an adder, 38 and 39 are switches, both 2
When the output of the level detector LD of No. 9 is "1", that is, in the case of forward converter operation, the switch 38 is α _nax
(for example, 120°), and switch 39 selects the output of adder 37. Conversely, when the inverter is in operation, the switch 38 selects the output of the adder 36,
Switch 39 selects α _nio (eg 85°).
Note that Δα, which is the input to the adders 36 and 37, is a small value of, for example, about 3°. 40 is an absolute value circuit,
41 is a level detection circuit whose input is a predetermined value (for example,
0.05pu) or less, outputs logic output “1”,
42 is an AND circuit.

Normally, when the forward converter operates under constant voltage control
Output α of CVC33 _CVC is around 20°. In the case of forward converter operation, the output of the level detector 29 becomes "1", the switch 24 is closed, 25 is open, and 38 is α _nax
(=120°), switch 39 selects adder 37
Select the output of α _nax is selected to be the same as the upper limit value of the limiter circuit 27 at the final stage of the forward converter. The adder 37 has a value of 17°, which is obtained by subtracting Δα=3° from α _CVC =20°. In other words, the limit of ACR32 is 17°≦
α≦120°, which is narrower than 10≦α≦160° in the case of Fig. 5. Output α of ACR32 _ACR is CVC
If the output α of 33 is smaller than _CVC , it will not be selected by the maximum value selection circuit 34.
If the minimum value of the output α _ACR of 32 is as large as possible without interfering with the CVC 33, it is possible to quickly respond to a sudden increase in DC current and suppress overshoot as much as possible. Similarly, the upper limit value of α _ACR does not need to be 160° in the case of a forward converter, and is preferably about 120°, which is the same as the upper limit of the limiter circuit 27.

On the other hand, when the inverter operates under constant voltage control,
Normally, the output α _CVC of the CVC 33 is around 140°. In the case of reverse converter operation, the output of the level detector 29 becomes "0", and the switch 24 is opened, the switch 25 is closed, and the switch 38 is closed.
selects the output of adder 36, and 39 selects α _nio (=85°)
Select. Adder 36 converts α _CVC = 140° to Δα = 3°
The total angle is 143°. i.e. ACR32
The limit is 85°≦α≦143°, which is narrower than 10°≦α≦160° in the case of FIG. ACR32
If the output α _ACR of CVC33 is larger than the output α _CVC , it will not be selected by the minimum value selection circuit, so the maximum value of the output α _ACR of ACR32 is
The smaller one is as small as possible without interfering with 3.
It can respond quickly to sudden decreases in DC current and suppress undershoot as much as possible. Similarly, the lower limit value of α _ACR does not need to be 10° in the case of an inverter, and is preferably about 85°, which is the same as the lower limit value of the limiter circuit 28 at the final stage of the inverter.

In addition to the limiter circuit of the ACR32 mentioned above, the difference between FIG. 7 and FIG. 5 is that the absolute value circuit 40,
A level detector 41 and an AND circuit 42 are added,
The only difference is that the 26 limiter circuits in FIG. 5 are omitted. The purpose of the circuits from adder 20 to adder 21 is to minutely correct the measurement error of DC current and AC system voltage, the error of X _pu in the formula (which varies somewhat depending on the tap position of the transformer), and the control error of the entire CVC 33. The amount of correction was not very large, and the correction was applied slowly in a steady state. Therefore, during transient times such as startup or line ground fault, if the correction circuit is used, the deviation between the set value and the measured value will cause the proportional circuit 31 and the first-order lag circuit 3 to
0 may result in saturation. In order to prevent saturation, the limiter circuit 26 shown in FIG. 5 can achieve the purpose to some extent, but the absolute value circuit 40, as shown in FIG.
The switch 23 activates the adder only when the deviation between the voltage setting value V _dp and the measured DC voltage value V _d is determined by the level detector 41 as |V _dp −V _d |≦0.05 and the constant voltage operation mode signal is "1". Select output of 20, V _dp
The correction circuit can be prevented from becoming saturated by applying a minute correction of , and setting the switch 23 to zero otherwise.

In the explanations so far, the output of constant current control devices and constant voltage control has been expressed as a control angle for convenience, but normally it is a control voltage from 0V to 10V, which corresponds to a control angle of 0° to 180°. do.
Further, the same explanation holds not only when the DC line voltage is positive as shown in FIG. 1, but also when the converter is oriented in the opposite direction and the DC line voltage is negative.

Furthermore, the present invention can be applied not only to the method disclosed in JP-A-51-66455, but also to a DC multi-terminal control device having constant voltage control and constant current control, and to a conventional two-terminal control device, and similar effects can be obtained. It will be done.

(g) Overall effect As explained above, according to the present invention, in a DC transmission system, whichever converter becomes the voltage determining terminal can be quickly switched, increasing the flexibility of system operation. It can be expanded significantly. Moreover, since whether the converter is in forward converter operation or inverse converter operation is determined by the polarity of the voltage setting value, there is no need to transmit any particular signal.

[Brief explanation of drawings]

Figure 1 is a DC 4-terminal power transmission system diagram, Figures 2 and 3 are conventional control characteristic diagrams, Figure 4 is a conventional control block diagram for obtaining the characteristics shown in Figure 3, and Figure 5 is a diagram of conventional control characteristics. FIG. 6 is a control block diagram showing one embodiment of the present invention, FIG. 6 is a characteristic diagram according to the present invention, and FIG. 7 is a block diagram showing another embodiment of the present invention. 1-4... AC system, 5-8... Conversion transformer, 9-10... Forward converter, 11-12... Inverse converter, 13... Constant current control device, 14-15...
Constant voltage control device, 17... Maximum value selection circuit, 18
... Minimum value selection circuit, 19-21 ... Adder, 2
2...Constant margin angle control, 23-25...Switch,
26-28...Limiter circuit, 29...Level detector, 30...First-order delay circuit, 31...Proportional circuit, 32...Constant current control device, 33...Constant voltage control device, 34...Maximum value Selection circuit, 35... Minimum value selection circuit, 36-37... Adder, 38-39
... switch, 40 ... absolute value circuit, 41 ... level detector, 42 ... AND circuit.

Claims (1)

[Claims] 1. In a DC multi-terminal power transmission system consisting of a plurality of converters, all converters correspond to a predetermined control delay angle from the measured values and voltage settings of the DC current and AC system voltage. an open-loop constant voltage control means that generates a signal, a constant current control means that controls the DC current to a predetermined value, and the converter operates as a forward converter based on the polarity of the voltage setting value of the constant voltage control means. a discriminating means for discriminating whether to operate the converter as a forward converter or an inverse converter, and a signal corresponding to a large control delay angle is selected when the converter is operated as a forward converter based on the determination by this discriminating means; a selection means for selecting a signal corresponding to a small control delay angle when operating; and a voltage setting that forms a closed loop and is input to the constant voltage control means when a converter for determining the voltage of the DC system is selected. 1. A control device for a DC multi-terminal power transmission system, comprising: a correction means for correcting the value using a signal obtained by multiplying the deviation between the voltage setting value and the measured value of the DC voltage by a predetermined value. 2. In a DC multi-terminal power transmission system consisting of multiple converters, all converters are open circuits that generate signals corresponding to a predetermined control delay angle from the measured values and voltage settings of each DC current and AC system voltage. A constant voltage control means of the loop, a constant current control means for controlling the DC current to a predetermined value, and a voltage setting value of the constant voltage control means to determine whether the converter operates as a forward converter or an inverse converter. A determining means for determining whether to operate the converter, and a signal corresponding to a large control delay angle is selected when the converter is operated as a forward converter, and a signal corresponding to a small control delay angle is selected when the converter is operated as an inverse converter. a selection means for selecting a signal corresponding to a control delay angle; and when the selection means selects a signal corresponding to a large control delay angle, a control angle smaller than the output of the constant voltage control means is provided to the constant current control means; When a signal corresponding to a small control delay angle is selected by the selection means, the constant voltage control means When a limiting means that limits saturation to a value equal to or greater than the sum of a predetermined value corresponding to a control angle smaller than the output and a converter that determines the voltage of the DC system are selected, a closed loop is formed to perform the constant voltage control. A DC multi-terminal power transmission system comprising a correction means for correcting a voltage setting value inputted to the means by a signal obtained by multiplying a deviation between the voltage setting value and a measured value of DC voltage by a predetermined value. Control device.