JPS6314567B2 - - Google Patents

Description

[Detailed Description of the Invention] (a) Description of the Technical Field The present invention relates to a thyristor element failure diagnosis device for a high-voltage thyristor converter, and in particular uses a microprocessor to monitor all thyristor elements and make appropriate logical judgments. The present invention relates to an element failure diagnosis device for a high voltage thyristor converter that uses a microprocessor to diagnose failures.

(b) Description of the Prior Art A large number of thyristors connected in series or in series and parallel are used in high voltage thyristor valves for DC power transmission, frequency conversion, etc. Thyristors can be damaged due to failure of the thyristor itself or its gate circuit, so generally a predetermined number of thyristors are connected in series with a margin. Therefore, if a thyristor fails and the number of failed thyristors reaches this number of margin thyristors, it is necessary to stop the thyristor valve as inoperable and replace the failed thyristor. Therefore, it is important to monitor thyristor failures and know the number of failures at all times.

FIG. 1 is a configuration diagram of a detection circuit of a high voltage section of a conventional thyristor element failure (hereinafter referred to as element failure) diagnostic device. FIG. 1 shows an example in which one module consists of six thyristors 1 connected in series among a plurality of thyristors connected in series in a thyristor valve. Furthermore, two detection resistors 3 are connected in parallel to the six series-connected thyristors. A detection circuit 4 is connected to an intermediate point P between the six thyristors connected in series and an intermediate point Q between the detection resistor 3. A bridge circuit is formed by a voltage dividing circuit 2 divided into two groups A and B at an intermediate point P and a detection resistor 3 divided at an intermediate point Q. When all six thyristors are normal, the bridge circuit is balanced and no voltage v is generated. However, when one thyristor is damaged, the bridge circuit becomes unbalanced and a voltage v is generated. The detection circuit 4 causes a current to flow through the light emitting element 5 when the voltage v is equal to or higher than a predetermined value, and transmits an element failure signal to the element failure determination circuit 7 in the low potential section through the light guide 6. The element failure determination circuit 7 determines in which module an element is malfunctioning and in how many modules the element is malfunctioning in a module composed of six thyristors in a thyristor valve.

However, with this conventional device failure diagnosis, in the case of an element failure that causes the bridge circuit to become unbalanced, for example, 1
It is possible to detect damage to one thyristor or failure of two thyristors only on the A side, but if all six thyristors on the A and B sides fail or the same number of thyristors fail on the A and B sides, the bridge circuit is balanced. It is impossible to detect element failure. Further, even when the thyristors are normal, an imbalance occurs in the voltage applied to each thyristor due to an imbalance in the characteristics of the voltage dividing circuit and the thyristor. In addition, the peak value of the voltage applied to the thyristor during floating and when the thyristor valve is in operation is √3 times the peak value during operation, when voltage is applied to the thyristor valve in a thyristor converter connected to a three-phase bridge to block the gate. big. Therefore, the voltage v generated when one of the six thyristors is damaged is approximately √3 times larger during operation than during floating. Therefore, it is very difficult to set the detection level of the detection circuit 4 so as to avoid erroneous detection due to voltage imbalance of each thyristor and to ensure reliable detection both during floating and during operation. Furthermore, as the number of thyristors in the module increases, the voltage v generated when one thyristor fails becomes smaller and may be impossible to detect.
Furthermore, even if an element failure can be detected, it is completely unclear how many thyristors on the low voltage side have failed compared to one to three thyristors on the A side. Therefore, since the exact number of failed elements within the module is unknown, the total number of failed elements for the entire valve cannot be known. Therefore, it is also impossible to know the exact number of thyristor margins at that point in time.

As described above, it is impossible for the conventional element failure diagnosis apparatus to reliably determine element failure, and as a result, the number of failed thyristors may exceed the number of margin thyristors, and other thyristors may also be damaged.
The above is a disadvantage in the operation and maintenance of thyristor converters.

(c) Purpose of the Invention In view of the above-mentioned drawbacks, the purpose of the present invention is to provide a microcontroller which is capable of reliably detecting thyristor failures and further determining the number of failures, and which is extremely effective from the viewpoint of operation and maintenance of thyristor converters. An object of the present invention is to provide an element failure diagnosis device for a high voltage thyristor converter using a processor.

(d) Structure of the invention The present invention will be explained below using the drawings.

FIG. 2 is a configuration diagram showing an embodiment of the present invention. A voltage dividing circuit 2 consisting of a capacitor and a resistor is connected in parallel to each thyristor 1, and a light emitting element 11 for detecting the voltage applied to each thyristor 1 (forward voltage in the example shown in Fig. 2) and a resistor for current limiting. 12, diodes 13 for preventing reverse voltage are connected respectively. The optical signal of each light emitting element 11 is transmitted to each light guide 1.
4 to each light receiving element 15, and each light receiving element 1
5, each voltage detection signal b1, b2... is an OR circuit 16
is input to the microprocessor 17. A voltage detection OR signal C, which is an output signal of the OR circuit 16, is also input to the microprocessor 17. The microprocessor 17 includes an I/O circuit 21 that performs an input/output interface, a latch circuit 22 that holds data for reading data, and a latch circuit 22 that stores the data.
It is composed of a RAM circuit 23, a PROM circuit 24 storing a program for reading data and a failure discrimination logic function for determining element failure, and a CPU circuit 25 for controlling each circuit of the microprocessor 17.

(e) Effect of the invention In the above configuration, each circuit in the microprocessor 17 can send and receive signals to and from the CPU circuit 25, and when the I/O circuit 21 receives the voltage detection OR signal C, the CPU circuit 25, a command is issued to the latch circuit 22, and the latch circuit 22 reads each voltage detection signal b1, b2, . . . as data.

Now, a voltage waveform a shown in FIG. 3a is applied to the thyristor valve, and if the thyristor T1 is normal, the light emitting element 11 emits a light signal during the forward voltage period, so the voltage detection signal b1 also remains during the forward voltage period. , generates an output "1". However, if the thyristor T2 is out of order, the output of the voltage detection signal b2 is "0". In addition, since the voltage detection OR signal C outputs “1” due to a normal thyristor, the I/O
The circuit 21 inputs this voltage detection OR signal C and generates a data read signal d to the latch circuit 22. The latch circuit 22 receives voltage detection signals b1 and b2 at the time when the data read signal d is generated.
Store the output of... Therefore, the voltage detection signal b1
is stored as “1” and voltage detection signal b2 is stored as “0”. The stored voltage detection signals b1, b2, . . . are stored in the RAM circuit 23 under program control.

Now assume that the RAM circuit 23 has n bits. RAM
There are various methods for storing voltage detection signal data in the circuit 23, and one example will be explained with reference to FIG. FIG. 3a is a time chart for explaining the operation of the present invention, and FIGS. 3b and 3c are RAM circuit 23.
FIG.

In the example of FIG. 3a, all thyristors are normal during the forward voltage period, and thyristor T is normal during the forward voltage period.
This is a case where only 2 is out of order.

FIG. 3b shows an example in which each bit of the RAM circuit 23 is stored in correspondence with data, which is a voltage detection logic signal of each thyristor. Data is stored in units of n in one memory address of the RAM circuit 23. When the number of thyristors is greater than n, data is stored in a plurality of memory addresses. Here, in order to simplify the explanation, the number of thyristors is assumed to be n. Therefore, the data of the voltage detection signal of the thyristor T1 is stored in the first bit of one memory address (RAM(i)) of the RAM circuit 23, and data of n thyristors are sequentially stored up to the nth bit.

The data of each thyristor during the forward voltage period of Fig. 3a is as shown in Fig. 3b.
~Tn is normal, so "1" is stored in all bits of RAM(i). In the next forward voltage period, since thyristor T2 is out of order, "0" is stored in the second bit corresponding to T2, and "1" is stored in all the bits corresponding to other healthy thyristors.

There are various methods for determining whether or not there is an element failure based on the stored data. For example, the data of each bit stored in RAM (i) is determined to be “1” or “0” by the failure judgment logic function stored in PROM 24 and its control program.
If it is "1", it is determined that the bit is normal, and if it is "0", it is determined that the thyristor corresponding to that bit is out of order.

In addition, two registers in the CPU are used as the first and second counters, and the first counter counts by 1 when the result of the judgment of normality or failure is a failure. The total number can be stored, and the number corresponding to the thyristor Ti can be stored in the second counter by counting by 1 each time the above fault determination operation is performed for each bit of RAM (i). . Therefore, when the result of failure determination is failure, the contents of counter 2 can be called to know the number of the failed thyristor, and the contents of counter 1 can be called to know the total number of failed thyristors in the thyristor valve. Note that the first and second counters are initialized to 0 prior to the failure determination operation for new data in RAM(i).

In addition, if the number of margin thyristors is m and you want to stop the operation of the thyristor converter when this number of thyristors fail, compare the contents of the first counter with m, and when this number m is reached, the I/O A stop signal e is output from the O circuit 21 to the outside. In this embodiment, the data of the voltage detection logic signal is read using the voltage detection OR signal c, and element failure can be reliably determined without reading data outside the forward voltage period.

The above explanation is based on one part of the data of the voltage detection logic signal.
This is a case where an element failure is determined by reading the number of times.

In the mode in which the thyristor converter performs inverter operation, some of the thyristors connected in series may not turn off. (Hereinafter, this phenomenon will be referred to as a partial commutation failure.) Since no forward voltage is applied to the thyristor in which partial commutation has failed, if an element failure is determined by reading data once, it will be erroneously determined.

If such a phenomenon of partial commutation failure is possible, a method of determining element failure based on data from several consecutive times is suitable. For example, using the data of three successive times in FIG. 3a, if the second bit of thyristor T2 is "0" all three times, it is determined that the element has failed. Generally, a situation where partial commutation fails does not continue for several cycles, so the above method can distinguish between partial commutation failure and element failure. In addition, when determining element failure based on data obtained several times, it is necessary to secure RAM addresses for storing data accordingly, as shown in FIG. 3c.

That is, the memory address RAM(i) of the RAM circuit 23,
RAM (i+1) and RAM (i+2) hold the data for the three times detected in , and the fourth data is configured to rewrite the contents of RAM (i) so that the latest three times' worth of data is always retained. do.

For example, the following operation is performed using the fault discrimination logic function stored in the PROM circuit 24 and its control program.

The sum of the bits corresponding to each thyristor Ti of the latest three times of data held in the RAM circuit 23 is calculated, and when the result is "0", it is determined that there is a failure. Also, the result of the operation is “1” to “3”
In the case of , it is judged as normal, and the sampling data of the next cycle is collected, the sum of each bit of data, , is calculated, and the same failure determination operation is repeated, and the data that is subsequently collected in succession is monitored. .

In the case of the embodiment shown in FIG. 3c, the data,
, performs failure determination operation on thyristor T
When the sum of the bits corresponding to 2 is calculated, it becomes "0", which is determined to be a failure.

In addition, when increasing the reliability of data, there is a method of taking majority vote. In this case as well, for example, if the number of times data is sampled is three times, if the calculation result of the sum of the three times is "2" or "3", it can be determined to be normal. The number of times this data is sampled is
If the number of data samples exceeds 10, and therefore the number of data samples exceeds 10, the circuit using conventional logic elements becomes extremely complex, but if a microprocessor is used, the data can be processed even if the number of data sampling increases. All you need to do is secure a RAM circuit to store it.

(f) Effects of the Invention As described above, according to the present invention, even if a thyristor failure occurs in a thyristor valve consisting of a plurality of thyristors connected in series, it can be reliably detected for each thyristor. This allows the total number of failures within the thyristor valve to be accurately known at all times. Therefore, unlike the conventional method, it is not possible to know the exact number of failures, and other healthy thyristors are also damaged, leading to an increase in accidents. Since thyristor failures are determined immediately, the number of failures can be accurately determined. Therefore, when the number of faulty thyristors reaches the number of margin thyristors, the operation of the thyristor converter can be stopped and protected, so there is no problem such as damage to healthy thyristors, and the operation or maintenance of the thyristor converter can be improved. It is also an effective device failure diagnosis device.

Further, a single microprocessor can perform element failure diagnosis for a plurality of thyristor valves. For example, when performing element failure diagnosis for all thyristors of six bridge-connected thyristor valves, memory addresses in the RAM circuit are reserved for the six thyristor valves, and each thyristor valve is usually Since it operates while maintaining a phase difference of 60 degrees in electrical angle, if element failure diagnosis for one thyristor valve is performed within 60 degrees in electrical angle, element failure will occur for all thyristors of six thyristor valves in one cycle. Diagnosis can be made.

As mentioned above, despite the large number of thyristors, 1
It is possible to diagnose all thyristors within the cycle time, and furthermore, it is possible to provide an element failure diagnosis device for a high voltage thyristor converter using a microprocessor that is more compact than conventional element failure diagnosis devices. .

(g) Modification In the embodiment of the present invention, the forward voltage of each thyristor is detected to perform element failure diagnosis, but the same effect can be obtained by configuring the device to detect reverse voltage or forward/reverse voltage. can.

Also, the time chart of the embodiment of the present invention is shown in Fig. 3a.
Although the explanation has been given using an example in which data is read once within the forward voltage period of each cycle, if necessary, data can be read a plurality of times and failure determination can be performed in the same manner as described above.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a conventional device failure diagnosis device, FIG. 2 is a block diagram of a device failure diagnosis device according to the present invention,
FIG. 3a is a time chart of each signal for explaining the present invention, and FIGS. 3b and 3c are memory block diagrams of the RAM circuit. 1... Thyristor, 2... Voltage dividing circuit, 3... Detection resistor, 4... Detection circuit, 5, 11... Light emitting element, 6, 14... Light guide, 7... Element failure determination circuit, 12 ...Resistance, 13...Diode,
15... Light receiving element, 16... OR circuit, 17...
Microprocessor, 21...I/O circuit, 22
...Latch circuit, 23...RAM circuit, 24...
PROM circuit, 25...CPU circuit.

Claims (1)

[Claims] 1. In an element failure diagnosis device for a high voltage thyristor converter configured with a thyristor valve in which a plurality of thyristors are connected in series, the voltage of the thyristor is determined for each thyristor of the plurality of thyristors. A first voltage detection circuit is provided with a voltage detection circuit for detecting the voltage, and the voltage detection logic signal detected by each of the voltage detection circuits is held inside the element failure diagnosis device.
a second memory circuit holding a control program for determining a failure from the respective voltage detection logic data held in the memory circuit and the first memory circuit;
a microprocessor that performs a logic operation for fault determination; and an OR circuit that ORs the respective voltage detection logic signals, and provides an output signal of the OR circuit as an input signal to the microprocessor; The respective voltage detection logic signals are read into the first memory circuit as the respective voltage detection logic data by a data read signal outputted from the microprocessor at a predetermined timing, and the respective voltages corresponding to the respective thyristors are read. The logic of the detection logic data is sequentially determined by the control program held in the second memory circuit and the microprocessor, and a failure is determined when a predetermined logic state is found. An element failure diagnosis device for a high voltage thyristor converter using a microprocessor, which outputs an element failure signal when the number reaches a predetermined number. 2. In an element failure diagnosis device for a high voltage thyristor converter configured with a thyristor valve in which a plurality of thyristors are connected in series, a voltage detection circuit for detecting the voltage of each thyristor of the plurality of thyristors is provided. a first voltage detection logic signal detected by each voltage detection circuit;
a second memory circuit that holds a control program that performs failure determination from the respective voltage detection logic data held in the first memory circuit, and a microprocessor that performs the logical operation of the failure determination; An OR circuit is provided to OR the respective voltage detection logic signals, and the output signal of this OR circuit is provided as an input signal to the microprocessor, and the data is read to be output from the microprocessor at a predetermined timing. The respective voltage detection logic signals are read into the first memory circuit as the respective voltage detection logic data at least once or more by a signal, and the respective voltage detection logic data are read at least once or more corresponding to the respective thyristors. is sequentially determined by the control program held in the second memory circuit and the microprocessor, and when a predetermined number of the at least one voltage detection logic data corresponding to a predetermined thyristor is in a predetermined logic state. An element failure diagnosis device for a high voltage thyristor converter using a microprocessor, which outputs an element failure signal when the number of thyristors determined to be failure reaches a predetermined number.