JPS593100B2 - Thyristor failure detection method for high voltage thyristor valve - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides DC power transmission, AC variable speed power supply, AC switch,
The present invention relates to a method for detecting a thyristor failure in a high voltage thyristor valve, such as a static starter, which includes a large number of thyristors connected in series and parallel.

High-voltage thyristor valves for DC power transmission, etc., are made by connecting many thyristors in series.
For 00KV to 200-V valves, 100 to 400 thyristors are connected in series.

Therefore, testing of thyristor valves (industrial tests, field tests)
If any of the thyristors is damaged during operation or during operation, it is extremely important to accurately understand the state of the damage from the viewpoint of improving valve reliability and reducing costs. This is especially effective for thyristor valves housed in sealed tanks, such as oil-cooled thyristor valves and gas-insulated thyristor valves, since normally only two terminals (i.e., an anode terminal and a cathode terminal) are led out to the outside. . Conventionally, methods for detecting the presence or absence of damage to the elements of this type of thyristor valve, that is, the thyristor, have been roughly divided into the following two methods.

The first method is to connect an extra number of thyristors in series, which is calculated from the inspection and repair interval (usually several years) and the failure rate of the thyristors, in addition to the number of thyristors that should originally be connected in series. There is no special monitoring for thyristor failure.

The extra thyristor is hereinafter referred to as a margin thyristor. In the second method, only the number of thyristors that should originally be connected in series is provided, and constant monitoring of thyristor failures is performed accurately. The first of these conventional methods is
In this method, a thyristor with a margin of about 10 to 20% is usually provided, which complicates the overall valve structure and increases the price.

In addition, the second method requires a special installation of a monitoring device for thyristor failure, which not only makes the overall valve configuration complicated and expensive, but also currently poses problems in the reliability of the monitoring operation. be. In order to compensate for this lack of reliability, a thyristor with a margin of about 5 to 10% is actually provided, and therefore, in practice, the complexity and cost of the configuration cannot be avoided. Generally, in such high voltage thyristor valves, when a thyristor deteriorates or completely breaks down, the impedance of the thyristor portion decreases or a complete short circuit occurs, so a thyristor failure can be detected as a change in the applied voltage or impedance.

Further, if the thyristor is continued to be used even after such a failure occurs, the deterioration of the thyristor progresses over time, and in many cases, the thyristor eventually becomes completely destroyed (impedance is O state). Conventional high-voltage valves utilize this phenomenon to detect thyristor failure using a detection method as shown in FIG. 1 or 2.

Figure 1 shows monitoring methods for detecting element failure during operation, including the bridge method (Fig. 1 A), the direct detection method (Fig. 1 B), and the bushing voltage ratio method (Fig. 1 B). C) There is.

For explanation purposes, Fig. 1 shows a specific N example in which N thyristors are connected in series and the number of parallel connections is 1. It is assumed that the thyristor is composed of n thyristors connected in series.

In Fig. 1, L is an anode reactor, Th is a thyristor, Z1 is a voltage divider for the thyristor, Z2 is a damping circuit or surge voltage dividing circuit, and Z3 (represents the element failure detection impedance). The failure detection device compares the voltages of adjacent units using a bridge type circuit. If the element, that is, the thyristor, is healthy, no voltage is generated between points A and B; When the thyristor included in the unit breaks down, a voltage is generated between points A and B.

This voltage is converted into light by a light emitting diode placed between points A and B, and converted into an electrical signal by a light emitting diode PD at case potential (or ground potential) via a light pipe LG made of a photoconductive material. will be reconverted. An isolation transformer may be used as a means for transmitting voltage between points A and B. The disadvantage of the method shown in Fig. 1A is that even if the thyristor is healthy, the difference between the points A and B due to variations in the circuit constants of the voltage divider circuit, etc.
Detection sensitivity is not very good because a voltage is generated between the points, and if the same number of thyristors in the two units being compared fail, no voltage will be generated between points A and B. Detection device Z3, signal transmission device (Light-emitting diode LDl
Optical Nll pipe LGl photoreceptor diode PD) 1.--1a) Only N22 is required and expensive, and the impedance of the detection device Z3 is lower than that of the voltage divider Z for the thyristor, so even if the thyristor breaks, there is no connection between AB. The voltage does not increase proportionally.

The direct detection type fault detection device shown in FIG. 1B is a type in which the thyristor voltage is directly and constantly observed (in the case of this embodiment, the observation is made for each n thyristor of 1.1 nits).

The disadvantages of this method are that the impedance of the surge voltage dividing circuit Z2 is lower than that of the voltage divider Z1 for the psi and j star, so the detection sensitivity is poor;
Although the optical system described above for Figure A is used, there is a problem with reliability because the optical system (which may also be an isolation transformer) is always in operation, and the number of detection devices and signal transmission devices N is large (this example (in the case of 1 piece), it is expensive, and fluctuations in power supply voltage directly affect accuracy.

FIG. 1C shows a bushing voltage comparison method, in which the presence or absence of a failure is detected by comparing the voltages between points a and b.

On the other hand, FIG. 2 shows a failure detection device similar to a system in which element failure detection is performed only when the operation is stopped, and elements having the same functions as those described in FIG. 1 are denoted by the same reference numerals.

Figure 2A shows the unit impedance method, and in order to measure the impedance of each unit without applying voltage between terminals A and K, contact C on the unit side is used.
s (fixed) is provided with two movable contacts CM so as to be slidable up and down, and the movable contacts CM are slidably controlled from outside the case.

This movable contact is kept separate from the thyristor components during operation. This method inevitably requires a large-scale configuration for slidingly controlling the movable contact from outside the case. Figure 2B (valve A-K impedance type)
Compared to FIG. 2A, the measurement unit is changed from unit to valve terminal.

Although this method makes it easy to detect element failures, it
This has the disadvantage that the output sensitivity is lowered by An to -(a=1) compared to the case shown in FIG. 2A.

In consideration of the problems of the conventional failure detection method described above, the present invention proposes a failure detection method that improves the drawbacks based on the method of performing element failure detection operation when the operation is stopped as described above with reference to FIG. It is something to do.

An example of the present invention will be described in detail below with reference to the drawings. In FIG. 3, 1 indicates a high voltage thyristor valve as a device to be detected, and 2 indicates a device for implementing the thyristor failure detection method according to the present invention.

In this case, the high-voltage thyristor valve 1 is, for example, an oil-cooled or gas-insulated thyristor valve (other wind-cooled type thyristor valves may be used), and an anode terminal A and a cathode terminal K are each led out to the outside of the case C1 through a bushing Bu. There is. One valve has N=NXa thyristors Th
A resistor R2 and a series circuit of a resistor R1 and a capacitor C1 are connected in parallel to each thyristor to form a voltage dividing circuit.

In addition, each unit is connected in parallel with a capacitor C2 and a resistor R.
3 and a series circuit of capacitor C3 are respectively connected in parallel, thereby performing a surge voltage suppressing effect and an overvoltage suppressing effect.

Incidentally, instead of inserting the series circuit of the resistor R3 and the capacitor C3 in each unit, only one set may be inserted between the anode terminal A and the cathode terminal K. In valve 1, the series circuit of resistor R2, resistor R, and capacitor C1 corresponds to the thyristor voltage divider Z1 in FIG. 1, and the series circuit of capacitor C1, capacitor C3, and resistor R3 corresponds to This corresponds to the damping circuit Z2.
In FIG. 3, elements unnecessary for explaining the present invention, such as a gate system and a cooling system, are omitted. On the other hand, the thyristor failure detection device 2 includes a DC high voltage generator 3 and a bridge circuit 4.

The DC high voltage generator 3 varies the AC input using a voltage regulator VS, and then performs insulation and step-up using a transformer Tr.

The output of the transformer Tr is connected to the AC terminals of the single-phase full-wave rectifiers D1 to D4 via the protective resistor Rs, and
One end of the output terminal of is connected to both ends of a smoothing capacitor CO, which is grounded through the case. The positive and negative terminals of the capacitor CO are connected to power supply side terminals a and b of the bridge circuit 4, and a high sensitivity, low resistance current detector G is connected to the other terminal pair c and d of the bridge circuit 4.

In this way, the terminals a and d of the bridge circuit 4 constitute both ends of the proportional side, which is one side of the bridge, and are taken out of the case C2 of the detection device 2 by the bushings BUl and BU2 and connected to the object to be measured. are connected to the anode terminal A and the cathode terminal K of the bulb 1, respectively.

On the other hand, on the other proportional side between terminals a and c of bridge 4 (
A resistor Rv' close to the valve DC resistance Rv is connected, and a variable resistor with a low resistance value is connected to the variable side between terminals d and b, and the variable side between terminals C and b, respectively.

Note that these variable resistors Rp and Rp'' may be constructed by connecting a fixed resistor and a variable resistor in series.In practice, the voltage between terminals c and b, between d and b, and between c and d In order to suppress the voltage to a low voltage of several V to several hundred, a variable resistor R is used.
The values of p and Rp'' are determined in relation to the valve DC resistance Rv and Pm = k (1) RvRvk
=0.01 to 0.001 It is desirable to select (2).

Here, Rpm and R'Pm are the maximum resistance values of variable resistors Rp and R'p. Others In FIG. 3, PS is a push button switch, which is normally closed and opened when measuring with ammeter G. Arr is an arrester, between terminals c and d, between c and b
This protects the elements between d and b. RG is a variable resistor for changing the sensitivity of the ammeter G. A is DC voltage V
A DC ammeter for DC indication, RA is its protective resistance.
It would be ideal if the maximum value of the output voltage VDC of the DC high voltage generator 3 could be around the rated voltage of the valve 1, but since this would make the detection device 2 large,
For a 200K valve, a range of a few K to a few +KV is appropriate.

Resistors R'V, Rp, and R'p have high precision, low temperature coefficient,
Use one with a low voltage coefficient.

In the thyristor failure detection device 2 having the above configuration, measurement is performed as follows.

That is, first, the terminals Tl, t2 are connected to the anode terminal A of the thyristor valve 1 via the bushings BUl, BU2.
and cathode terminal K to apply a positive voltage between these terminals and therefore to the thyristor. Thereafter, the output voltage DC of the DC high voltage generator 3 is increased stepwise by the voltage regulator VS, and each time the output voltage DC is increased by the variable resistor Rp.
Alternatively, adjust R″p so that the output of the current detector G becomes 0. When the output of the detector G becomes O by this balancing method, the DC resistance Rv of the thyristor valve 1 becomes the resistance From the values of Rp and R'p, Rp RV=-・1e'V...
(3) Obtained as Wn.

The value of this DC resistance Rv is determined by the output of the high voltage generator 3 as Dc=0
~Dcmax (for example, several KV to Atsushi+KV), and as shown in FIG. 4, a DC voltage-valve DC resistance curve is determined in a positive DC voltage range.

Next, the terminals Tl and t2 of the detection device 2 are switched and connected to the cathode terminal K1 and the anode terminal A, respectively.

At this time, a negative voltage is applied between the anode terminal A and the cathode terminal K, and thus to the thyristor. In this connection state, measurement is performed using the same balance method as described above for the positive polarity connection, and a DC voltage-bulb DC resistance curve is determined in the negative DC voltage range as shown in FIG.

However, in the curve diagram of FIG. 4, in the first case where the thyristor is normal, the curves in the positive and negative regions are indicated by symbols 1 and 12, and the voltage dependence of the valve resistance Rv is very small, and the polarity effect is do not have.

Next, one or more of the N thyristors of the valve 1 has a positive polarity voltage (VDCl).
In the case where the blocking ability decreases as described above, the positive polarity curve bends at the voltage (VDC1) as shown by reference numeral 13. However, the characteristics in the negative voltage region match curve 12 as in the normal case. Next, if one or more thyristors out of the N thyristors of valve 1 are completely destroyed (short-circuit failure) in both positive and negative polarities from an area where the voltage is approximately 0, then The curves in the region, as shown by reference numerals 14 and T5, exhibit a reduced resistance value compared to the normal curves 11 and 12.

The following judgments can be made from each of the above curves.

That is, the number of failed thyristors among the N thyristors of the valve 1 is determined from the difference between curves 1 and 12 during normal operation and curves 14 and 8 during complete failure. Further, the number of failures of the forward direction (positive polarity) thyristor can be determined from the curves 11 and 12 in the normal state and the curve 13 and T2 in the case where the blocking ability is lost in the positive polarity region.

Note that when the characteristic in either the positive or negative direction is actually damaged, the thyristor is determined to be a damaged thyristor. In this way, according to the present invention, it is possible to detect a failure in a thyristor, but the principle is that a thyristor exhibits a polarity effect and a voltage effect depending on the type of failure, whereas a thyristor exhibits a polarity effect and a voltage effect depending on the type of failure. Circuit elements (resistors, capacitors, reactors) have almost no polarity or voltage effects, and there are no polarity or voltage effects in the measurement system. The purpose of this method is to measure thyristor failure components on a valve-by-valve or unit-by-unit basis, and to extract and determine only the failure component of the thyristor from the measurement results.

The reason for setting the power supply voltage of the bridge circuit 2 to a high voltage is not only to examine the voltage dependence from low voltage to high voltage as described above, but also for the following reason to ensure measurement accuracy.

Generally, it is assumed that the forward and reverse resistances of the thyristor during normal operation are sufficiently larger than the parallel voltage divider circuit resistance R2 (Fig. 3). value). However, in reality, in the region S where the voltage value is small (i.e., 0 to several V), the thyristor exhibits low resistance characteristics as shown in Figure 5A, and the above assumption may not be satisfied (this is especially noticeable at high temperatures). ). However, in this case, as shown in FIG. 5B, the combined resistance of the resistance R2 and the thyristor resistance is almost the same as the curve of the resistance R2 in the high voltage region, as shown by the broken line, but in the low voltage region, the combined resistance of the thyristor resistance Th is will be affected by the decline. Therefore, it is necessary to avoid measurements in this low voltage region. From this point of view, it is understood that it is not appropriate in principle to use existing measuring instruments such as digital testers and Wheatstone bridges whose power supply voltage is on the order of O~ several as a means for measuring thyristor failure. As described above, according to the present invention, it is possible to obtain a thyristor failure detection method that exhibits the following effects compared to conventional detection methods.

First, the detection sensitivity improves, so that what was conventionally about 3 to 5% can be reduced to 0.05 to 0.1% or less according to the present invention.

As a result, in the case of a valve made of 200 thyristors connected in series, it was possible to detect the occurrence of a failure in about 6 to 10 thyristors, whereas this has been reduced to about 0.1 to 0.2. can be reduced. The reason for this is that by adopting a bridge, the effects of meter errors, reading errors, and power fluctuations are reduced, and by adopting a DC system, the valve impedance (resistance) increases in the event of a thyristor failure. It is in.

Incidentally, in the latter case, for example, when an AC system is adopted as shown in Fig. 2, resistor R2, resistor R1 and capacitor C1
Since the impedance of the parallel circuit consisting of capacitor C2 and the series circuit of resistor R3 and capacitor C3 is lower than the impedance of the parallel circuit consisting of the series circuit of
Even if a thyristor fails, its failure output is not proportional to the number of failures, and its impedance change is also smaller than in the case of direct current. On the other hand, in the case of the DC system, the value of the resistor R2 is the lowest (the other resistors are almost infinite). Second aspect of the present invention
The advantage of this is that the device itself for implementing the failure detection method has high reliability, and as described above with reference to FIG. 3, it has a configuration with a low failure rate.

Furthermore, since the failure detection device 2 is placed outside the case C1 of the thyristor valve 1, the effort required for repair is simple. Thirdly, since the failure detection method according to the present invention can accurately measure valve failures, the number of margin thyristors can be reduced to 0 or significantly less than the conventional method, making it possible to obtain a valve that is correspondingly cheaper.

Further, since failures in the valve can be detected with high accuracy, the reliability of the valve can be further improved, and since it is not necessary to provide a failure detection device inside the valve, the reliability of the valve can be further improved. Fourthly, the apparatus for carrying out the fault detection method according to the present invention is as described above with reference to FIG.
It has a relatively simple configuration using ordinary elements, and therefore can be lower in cost than conventional devices. Fifth, according to the present invention, in addition to measuring the valve DC resistance, the voltage effect and polarity effect are also used for failure determination, thereby preventing thyristor deterioration.
Failure detection can be ensured.

The device shown in FIG. 6 can also be used as the device 2 for carrying out the failure detection method according to the present invention.

This device 2 has a polarity reversal function in the DC high voltage generator 3 instead of switching the connections of the output terminals t1 and T2 of the failure detection device 2 in the process of measurement in FIG. In FIG. 6, the rectifiers D1 to D4 of the DC high voltage generator 3 are connected to an insulating shaft 1S by a motor M around an imaginary line connecting the opposing connection points H and G (constituting the input end) of the diode bridge. through ±180
This allows for zero rotation, and this allows the other opposing connection point E
From the movable contacts DP and DN provided at positions DP and DN (constituting the output end), DC outputs having different polarities before and after rotation can be sent out.

According to the configuration shown in FIG. 6, the polarity effect of the valve DC resistance can be achieved without changing the connection relationship between the anode terminal A and cathode terminal K of the thyristor valve to be measured and the terminals tl and T2 of the failure detection device 2. can be obtained, making the measurement work easier and more efficient.

Also, in FIG. 6, a contact S1 is connected between the variable side connection point 1 connected to one end of the output of the bridge circuit 4 and the case C2.
is provided, and the connection point d between the proportional side and the variable side and case C2
An interlocking contact S2 that operates in the opposite direction to the contact S1 is provided between the contact point C and the case C2 (this contact point S2 may be provided between the connection point c and the case C2).

This is necessary to ensure the safety of the person performing the measurement and to protect the bridge low voltage circuit (i.e. the elements between terminals c and d, between d and b, and between c and b). The diagram of FIG. 8 shows separation points when measuring a valve having a three-phase thyristor bridge configuration as shown in FIG. 7 using the failure detection device 2 of FIG. 6. For example, when measuring bulbs X, Y, Z, U, . . . W, the anode terminal A1 and cathode terminal K of each bulb must be separated from the circuit. That is, points A, B, C,
It is necessary to separate at D, E, F, J, K, L, M, O, P (12 points in total). However, according to the fault detection method using the fault detection device shown in FIG. 6, the number of separation points can be reduced as shown in FIG. FIG. 9 shows yet another embodiment, and in this case, FIG. The resistor Rv' is omitted, the terminals Tl, d, and d' are led out, and both of the two proportional sides are connected to the thyristor valve 1.

According to the configuration shown in FIG. 9, there is no need for the resistance Rv'' in the failure detection device 2, and the configuration can be simplified and miniaturized accordingly.

In addition, measurement accuracy is improved because the same general meter bulb is measured under the same conditions (same temperature and voltage). Furthermore, it is possible to measure two valves at the same time. In addition to these, the number of separation points for busses can be reduced. Incidentally, using the failure detection device 2 shown in FIG.
When measuring a valve as shown in Figure 7, the separation point is the first
This is shown in the diagram in Figure 0.

[Brief explanation of drawings]

1 and 2 are connection diagrams showing a conventional thyristor failure detection device, and FIG. 3 is a connection diagram showing an example of a device for implementing the thyristor failure detection method for a high voltage thyristor valve according to the present invention. 4 and 5 are curve diagrams for explaining its operation, FIG. 6 is a connection diagram of a device for implementing a failure detection method showing another embodiment of the present invention, and FIG. 7 is a thyristor valve to be measured. FIG. 8 is a diagram showing separation points when the thyristor failure detection device of FIG. 6 is applied to the thyristor valve of FIG. FIG. 10, a connection diagram of a device for carrying out the detection method, is a chart showing separation points when the thyristor failure detection device of FIG. 9 is applied to the thyristor valve of FIG. 7.

Claims (1)

[Claims] 1. A DC voltage generator with a variable output voltage, and two variable sides that are supplied with the output of the DC voltage generator as a power source, are connected to one end of the output of the DC voltage generator, and have a variable low resistance value; A bridge consisting of one proportional side made of high resistance and the remaining proportional side led out of the case via two measurement terminals and connected between the anode and cathode terminals of the thyristor valve to be measured. Using a circuit, first apply the output voltage of the DC voltage generator to the bridge circuit so that the voltage between the anode terminal and cathode terminal of the thyristor valve to be measured has a predetermined polarity, and vary this output voltage. Then, measure the DC resistance of the thyristor valve to be measured corresponding to the variable range by the balance method, then reverse the voltage polarity between the anode terminal and the cathode terminal of the thyristor valve to be measured, and then apply the DC voltage generator again. The output voltage is varied and the DC resistance of the thyristor valve under test corresponding to the variable range is measured using the balance method, and the thyristor under test is determined from the DC voltage-valve DC resistance characteristics during normal and abnormal conditions based on the measurement results. A method for detecting a thyristor failure in a high voltage thyristor valve, the method comprising determining a failure in a thyristor of the valve.