JPWO2013111269A1 - Communications system - Google Patents

Abstract

A communication system including a plurality of communication devices built in a plurality of distributed control devices, wherein the plurality of distributed control devices are communication paths between the distributed control devices and communication paths with a central control device. By connecting with other central cells, it is connected with not only the control device of the adjacent cell but also the cell ahead of it. Even when some of the distributed control devices fail, the communication path is secured by detouring.

Description

  The present invention relates to a communication system in which a plurality of communication devices are connected in a daisy chain, and reliability is required. In particular, the present invention relates to a communication system provided in a power converter.

Controllers that control devices and plants are required to have a function that ensures the reliability of operation of devices and plants so that the effects of failures are minimized even when failures occur, depending on the application.
Moreover, it is necessary to control not only a single device but also a plant or system composed of a plurality of devices. In such a case, plants and systems are distributedly arranged, and in many cases, each device is controlled by communication from a central control device.
For this purpose, various configurations are considered for the communication path from the central control unit to each device. Among them, it is easy to install wiring on a communication path that connects devices in a daisy chain.
An example of utilizing such control communication is a modular multi-level cascade converter (MMCC).
The modular multi-level converter uses a switching element that can be turned on / off, such as an IGBT (Insulated Gate Bipolar Transistor), and is a circuit system that can output a voltage exceeding the breakdown voltage of the switching element. : High Voltage Direct Current System), reactive power compensator (STATCOM: Static synchronous Compensator), circuit system that is expected to be applied to motor drive inverters.

Non-Patent Document 1 discloses a technique regarding an MMCC circuit system.
According to Non-Patent Document 1, the MMCC is configured by connecting a plurality of unit converters (referred to as “cells” as appropriate) in series (cascade). Each cell is a bidirectional chopper circuit, for example, and includes a switching element and a DC capacitor. Each cell is connected to the outside through at least two terminals, and the voltage between the two terminals can be controlled to the voltage of the DC capacitor of the cell or zero.
When the MMCC performs PWM (Pulse Width Modulation) control on each cell, the output voltage waveform of the MMCC can be changed to a multilevel waveform by appropriately shifting the phase of the triangular wave carrier applied to each cell. As a result, harmonic components can be reduced as compared with the two-level converter.
As a feature of MMCC, the potential of each cell is different from each other, and a cell having a high ground potential exists. In particular, when MMCC is applied to HVDC, the ground potential of the cell ranges from several tens kV to several hundreds kV. Furthermore, the ground potential of each cell changes from moment to moment.
In Non-Patent Document 1, since a device at a laboratory level is targeted, consideration is not given to a withstand voltage between the control device of the MMCC and each switching element.

  Non-Patent Document 2 discloses a technology in which a signal processing circuit is mounted in the vicinity of the same potential as each cell, and the central control device at the ground potential and each signal processing circuit are connected by an optical fiber cable. Yes.

Makoto Sugawara and Yasufumi Akagi: “PWM control method and operation verification of modular multilevel converter (MMCC)”, IEEJ Transactions D, Vol. 957-965. B. Gemmell et al .: “Prospects of multilevel VSC technologies for power transmission”, IEEE / PES Transmission and Distribution Conference and Exposition 2008, pages 1-16.

However, in the case of the MMCC daisy chain disclosed in Non-Patent Document 1, there is a possibility that communication with a plurality of devices connected to the device may be impaired due to a failure of one device. Therefore, there is a problem that it may be difficult to continue operation of the plant or system.
In Non-Patent Document 2, at least one optical fiber cable is connected to one cell from the central controller. In other words, an optical fiber cable is connected to each cell from the central control unit, and a star connection is established. Therefore, at least the same number of optical fiber cables as the number of cells are required, and there are problems of cost and installation.
Further, in this case, all the optical fiber cables need to have a dielectric strength that can withstand a potential difference between the central control device at the ground potential and each cell. That is, it is necessary to make all the optical fiber cables special optical fiber cables having a dielectric strength against creeping discharge or the like (referred to as “high voltage optical fiber cables” as appropriate), and there is a problem of cost.
In addition, a high pressure-resistant optical fiber cable requires a special sheath material (Sheath material) to be used for the outer shell of the cable, and there is a problem that the manufacturing process is complicated and expensive.
Further, as in Non-Patent Document 2, when the control device and each cell are star-connected with an optical fiber cable, there is a problem that each optical fiber cable becomes long.

  Therefore, the present invention solves such problems, and its purpose is communication that can avoid the occurrence of a communication failure due to equipment failure and continue operation of the plant or system. It is to provide a system at a low cost.

In order to achieve the above object, each invention is configured as follows.
That is, the communication system of the present invention is a communication system including a plurality of communication devices built in a plurality of distributed control devices, the communication device having at least two communication input ends and two communication output ends, One communication input end of the two communication input ends of the kth communication device is connected to a communication output end of an adjacent (k−1) th communication device, and the two communication input ends of the kth communication device are connected. The other communication input terminal of the communication input terminal is further connected to the communication output terminal of the adjacent (k-2) th communication device, and is one communication output terminal of the two communication output terminals of the kth communication device. Is connected to a communication input terminal of an adjacent (k + 1) th communication device, and the other communication output terminal of the two communication output terminals of the kth communication device is further adjacent to the (k + 2) th communication. Connected to the communication input terminal of the device, the (k-2) th to (k) A plurality of communication devices including a communication device +2) is characterized by being linked together.
Other means will be described in the embodiment for carrying out the invention.

  ADVANTAGE OF THE INVENTION According to this invention, even if the malfunction of the communication accompanying a failure of an apparatus generate | occur | produces, this can be avoided and the communication system which can continue the operation | movement of a plant or a system can be provided at low cost.

It is a figure which shows the example of a connection with the system control apparatus of the central control apparatus of the communication path structure of the daisy chain connection of the some communication apparatus in the some distributed control apparatus of 1st Embodiment of this invention, and 2nd Embodiment. (A) shows a one-way communication path configuration by a plurality of distributed control devices and two system control devices of the first embodiment, and (b) shows a plurality of distributed control devices and four system controls of the second embodiment. The bidirectional | two-way communication path structure by the apparatus is shown. It is the figure which showed the mode of the cell failure avoidance corresponding to the structure of FIG. 1 of 1st Embodiment and 2nd Embodiment of the communication system of this invention, Fig.2 (a) is FIG. 1 which is 1st Embodiment. FIG. 2B shows the case where the direction of propagation of the signal corresponding to (a) is one-way. FIG. 2B shows the direction of propagation of the signal corresponding to FIG. 1B of the second embodiment. The case is shown in both directions. It is the block diagram which showed an example of the internal structure of the cell controller used in the case of the bidirectional | two-way communication in 2nd Embodiment of the communication system of this invention. It is the figure which showed an example of the outline | summary of a structure of a power converter device and a communication system when 3rd Embodiment of the communication system of this invention is applied to a power converter device. It is a figure which shows an example of a structure of the cell of a power converter device when 3rd Embodiment of the communication system of this invention is applied to a power converter device. It is a figure which shows an example of the function of the internal state of the control apparatus of a cell when 3rd Embodiment of the communication system of this invention is applied to a power converter device. In the cell rearrangement method when 3rd Embodiment of the communication system of this invention is applied to a power converter device, it is a figure which shows an example of the method of integrating the rearrangement result by four cells. Each cell provided in the arm of the power conversion device according to the third embodiment of the communication system of the present invention shows examples of various designation methods when a cell number is designated and turned on / off. ) Is the voltage waveform of the arm, (b), (c), (d) are the on / off designation methods of the first example, the second example, and the third example, respectively, (e) is the current waveform of the arm, (d ) Is an on / off designation method of the fourth example corresponding to the current. It is the figure which showed an example of the outline | summary of a structure of a power converter device and a communication system when 4th Embodiment of the communication system of this invention is applied to a power converter device. It is the figure which showed the example of the detail of the internal state register in connection with the ascending forward direction communication in 3rd, 4th embodiment of the communication system of this invention. It is the figure which showed the example of the detail of the register | resistor of the internal state in connection with the ascending forward direction communication in 3rd, 4th embodiment of the communication system of this invention. It is the figure which showed the example of the detail of the register | resistor of the internal state in connection with the kind of message | telegram of the ascending forward communication in 3rd, 4th embodiment of the communication system of this invention. It is the figure which showed the example of the detail of the register | resistor of the internal state in connection with the kind of message | telegram of the ascending forward communication in 3rd, 4th embodiment of the communication system of this invention.

  Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings.

(First embodiment)
A first embodiment of a communication system of the present invention will be described.
The first embodiment shows a communication system that can reduce the number of high-voltage optical fiber cables and can continue operation even when a cell failure occurs.

<Communication path configuration>
FIG. 1 shows an example of connection between a daisy chain communication path configuration by a cell controller (distributed controller) of a plurality of unit converters (cells) of the first embodiment and a system controller of a central controller. (A) is a one-way communication path configuration by a plurality of cell control devices (communication devices) C0 to Cm-1 (20400 to 20409) and two system control devices S0 and S1 (107S0 and 107S1). (B) shows a bidirectional communication path configuration by a plurality of cell control devices C0 to Cm-1 (20400 to 20409) and four system control devices S0, S1, S0, S1 (107S0, 107S1). .
The unit converter (cell) 105 (FIG. 5) is configured to include a part that handles power and a part that handles communication, as will be described later, and the cell control device controls communication of the unit converter. In addition, the cell control device includes a communication device and a communication device including a cell controller, but the cell control device is represented by a cell controller as appropriate, and is also referred to as a cell controller as appropriate.

<Communication channel / one-way>
In FIG. 1A, system control devices (referred to as “system controllers” as appropriate) 107S0 and 107S1 provided in the central control device (107, FIG. 4) are duplicated controllers and are identical in synchronization. Perform the process. The system controllers 107S0 and 107S1 and the cell controllers 20400 to 20409 constitute a daisy chain.
The plurality of cell controllers 20400 to 20409 are daisy chain-connected by a cell controller of a cell that is adjacent to or adjacent to the cell controllers by medium-voltage optical fiber cables (or low-voltage optical fiber cables) 11204 to 11228.

For example, the cell controller Ck (20404, the k-th communication device) is the next cell controller Ck-2 (20402, the (k-2) th communication device) next to the previous cell controller Ck-1 (20403, the (k-1) th communication device) are connected to input two outputs respectively, and the cell controller Ck + 2 (20406, the (k + 2) th communication device) adjacent to the next two and the next one respectively. It is connected to output to the adjacent cell controller Ck + 1 (20405, (k + 1) th communication device). Note that medium voltage optical fiber cables 11213, 11214, 11217, and 11216 are used for the connection between them.
Similarly, other cell controllers C1 to Ck-1 and Ck + 1 to Cm-3 are connected to adjacent cells with medium-voltage optical fiber cables 11204 to 11228 assigned with even numbers.
Further, among the medium-voltage optical fiber cables 11204 to 11228, an odd-numbered one is assigned and connected to the next cell.

The system controllers 107S0 and 107S1 are connected to cell controllers 20400, 20401, 20408, and 20409 by two cells from both ends of the daisy chain through high-voltage optical fiber cables 11101, 11103, 11129, and 11131, respectively.
High-voltage optical fiber cables 11102 and 11130 (shown by broken lines) connect between the terminals of the system controllers 107S1 and 107S0 and the cell controllers 20400 and 20409.

As described above, the configuration of the daisy chain in FIG. 1A represents a configuration of connection using an optical fiber in one direction (direction of arrow →), and a signal propagates from left to right in the drawing.
The operation, effect, and problem of the configuration of the daisy chain shown in FIG.

(Second Embodiment)
As a second embodiment, a case where the communication path is bidirectional will be described.

《Communication channel / bidirectional》
Next, the configuration of FIG. 1B will be described.
FIG. 1B shows a plurality of cell controllers (cell controllers) C0 to Cm-1 (20400 to 20409) and four system controllers (system controllers) S0, S1, S0, and S1 (107S0 and 107S1). The communication channel configuration is shown.
The number of cell controllers C0 to Cm-1 is the same as in FIG. 1A, but the number of inputs and outputs of each cell controller is increased to four.
In addition, the system controllers S0, S1, S0, and S1 are also increased to four.
The two inputs and outputs of each of the cell controllers C0 to Cm−1 and the two system controllers S0 and S1 constitute a first series of daisy chains, and the other two of the cell controllers C0 to Cm−1. A daisy chain of the second series is configured by the input, the output, and the other two system controllers S0 and S1.

In the first series of daisy chains, the signal propagates from left to right, and in the second series of daisy chains, the signal propagates from right to left.
In FIG. 1B, the signal flow in the high-voltage optical fiber cables 11141 to 11143, 11169 to 11171, and the medium-voltage optical fiber cables 11244 to 11268 is shown in the direction in which the signals of the second series daisy chain propagate. This is indicated by an arrow (←), which indicates that the direction of propagation of the first series of daisy chain signals is opposite.
With the above configuration, the cell controllers C0 to Cm-1 input and output bidirectional signals.
The operation, effect, and problem of the configuration of the daisy chain in FIG. 1B will be described later.

<Cell failure avoidance>
Next, in the configuration of the daisy chain shown in FIGS. 1A and 1B, how to avoid failure when a cell (cell controller or communication device or communication device in the cell control device) fails. Explain what can be done.
FIG. 2 is a diagram showing a state of cell failure avoidance corresponding to the configuration of FIG. 1, and FIG. 2 (a) shows a case where the signal propagation direction corresponding to FIG. 1 (a) is unidirectional. FIG. 2B shows the case where the signal propagation direction corresponding to FIG. 1B is bidirectional.

  In FIG. 2A, the failure mode is No. 1, no. Two cases are described. Failure mode No. 1, no. 2 will be described in order.

<Discontinuous multiple / one-way>
Failure mode is No. 1 is a case where the daisy chain signal of FIG. 1A propagates in one direction. In the plurality of cell control apparatuses (cell controllers) 20400 to 20409, the plurality of cell controllers 20401, 20404, and 20407 are Shows a failed case.
Even in such a case, if a plurality of cell controllers 20400, 20402, 20403, 20405, 20406, 20408, and 20409 are functioning normally, in the vicinity of the faulty cell controllers 20401, 20404, and 20407, Communication can be propagated to the previous cell via the pressure-resistant optical fiber cables 11205, 11215, and 11221. That is, even if a plurality of cell control devices (cell controllers) fail, communication can be continued if a plurality of non-adjacent discontinuous cells fail.

《Discontinuous multiple + one continuous location, one direction》
Failure mode is No. 2 is a case where the direction in which the signal of the daisy chain in FIG. 1A propagates is unidirectional, and in a plurality of cell control devices (cell controllers) 20400 to 20409 (cell numbers correspond to No. 1), This shows a case where a plurality of cell controllers 20401, 20402, 20404, and 20407 have failed. No. In this case, the cell controller 20402 further fails from the case of 1.
When the cell controller 20402 fails, the cell controller 20403 has no cell propagation path because the cell controllers 20401 and 20402 both fail. Therefore, an “island” state in which no signal is propagated to the plurality of cell controllers 20403 to 20409 located on the right side of the cell controller 20403 is obtained. At this time, the system controllers 107S0 and 107S1 cannot exchange communication with the plurality of cell controllers 20403 to 20409.

  From the above, the configuration of the daisy chain shown in FIG. 1 (a) is an effective configuration in which communication can be continued if there is a failure in a plurality of non-contiguous cells that are not adjacent to each other. If there is, it can be understood that the communication cannot be continued.

  Next, in FIG. 2B, No. 1 in FIG. Utilizing reverse communication to avoid problems when there are two consecutive cell failures. In FIG. 2B, the failure mode No. in the case where the daisy chain signal shown in FIG. 1, no. 2, no. 3 will be described in order.

《Discontinuous Plural ・ Bidirectional》
The failure mode in FIG. 1 is a case where the daisy chain signal of FIG. 1B propagates in one direction, and in the plurality of cell controllers 20400 to 20409, a plurality of cell controllers 20401, 20404, and 20407 have failed. This is possible even when the daisy chain signal propagates in one direction because the signal can be propagated even in one direction. Therefore, detailed description is omitted.

《Discontinuous multiple + one continuous location ・ bidirectional》
Failure mode is No. 2 is a case where the signal propagation direction of the daisy chain in FIG. 1B is bidirectional, and a plurality of cell controllers 20400 to 20409 (cell numbers correspond to No. 1 in FIG. 2A). , A case where a plurality of cell controllers 20401, 20402, 20404, and 20407 have failed is shown.
At this time, since the cell controllers 20401 and 20402 are continuously faulty, signals cannot be propagated through the cell controllers 20401 and 20402.
However, the right communication (11102) is switched to the left communication (11142) by the cell controller 20400, and the left communication (11259) is switched to the right communication (11215) by the cell controller 20403, so that the communication is continued. Controllers 107S0 and 107S1 can also receive communication results.
From the above, when the signal propagation direction is bidirectional, it is possible for a daisy chain signal to propagate even if there are consecutive cell failures even if the number of consecutive points is one. It shows that there is.

《Discontinuous multiple + more than 2 consecutive locations ・ Bidirectional》
Failure mode is No. 3 is a case where the signal propagation direction of the daisy chain in FIG. 1B is bidirectional, and in a plurality of cell controllers 20400 to 20409 (cell numbers correspond to No. 1 in FIG. 2A). , A case where a plurality of cell controllers 20401, 20402, 20404, 20406, and 20407 have failed is shown.
At this time, since the cell controllers 20401 and 20402 are continuously faulty, signals cannot be propagated through the cell controllers 20401 and 20402. Accordingly, the right communication (11102) needs to be switched to the left communication (11142) by the cell controller 20400.
Further, since the cell controllers 20406 and 20407 are continuously malfunctioning, signals cannot be propagated through the cell controllers 20406 and 20407. Therefore, the left communication (11268) needs to be switched to the right communication (11228) by the cell controller 20408.

However, as described above, since the signals are returned by the cell controller 20400 and the cell controller 20408, the cell controllers 20403 to 20405 cannot exchange communication with the system controllers 107S0 and 107S1, and the “island” state is set. turn into.
However, this is the case when two or more consecutive faults occur in the same daisy chain. Moreover, it is a case where both occur separately, and the probability of occurrence is sufficiently smaller than a simple four-point failure.
The second embodiment has the effects described above.

As described above, by applying the communication system that performs the daisy chain connection of the first and second embodiments shown in FIGS. 1A and 1B to the cell control device (cell controller), only the cell control device of the adjacent cell. In addition, by connecting to a further cell, it is possible to perform communication bypassing the cell control device even when the cell control device fails.
Furthermore, the same connection is also implemented in the reverse daisy chain connection, so that even when two adjacent cell controllers fail, the operation is continued by turning back and communication from the opposite end.
Therefore, by applying the communication system according to the first and second embodiments of the present invention to an apparatus such as a power converter, for example, even when a communication failure occurs due to a device failure, this can be avoided, and the plant And system operation can be continued.

<Internal structure of cell controller>
Next, the cell controller 20404 (20400 to 20409) provided in the cell control device (distributed control device) 204 of the unit converter 105 (FIGS. 4 and 5) will be described.
FIG. 3 is a block diagram showing an example of the internal structure of the cell controller 20404 used in the case of bidirectional communication in the second embodiment.
In FIG. 3, the one-way communication is performed according to the configuration of reference numerals 30100 to 30109 of the blocks indicated by the notation (a). In addition, according to the configuration of the block codes 30140 to 30149 corresponding to the above-described configuration, another one-way communication is performed, and the overall configuration indicated by the notation (b) including the block codes 30120 and 30130 is described above. Communication in both directions.

First, the configuration of the portion indicated by the notation (a) will be described.
Communication telegrams (telegrams) input via the optical fiber cables 11213 and 11214 are clock-synchronized by clock synchronization (k-1) 30101 and clock synchronization (k-2) 30102. checking the content.
The message number collation unit 30107 records several message numbers (identification numbers) received in the past. When the numbers are the same, the message selection message merger 30108 stops the processing. If there is no same number, the number is recorded in the message number matching unit 30107 and the message is processed.
The message buffer (message storage unit) (k-2) 30106 temporarily stores an input from the optical fiber cable 11213, holds a message that needs to be synchronized with the message from the optical fiber cable 11214, and is an optical fiber. When a message from the cable 11214 is received, the message selection message merging unit 30108 merges the message into one message.

The instruction included in the electronic message is interpreted and executed by the electronic message interpretation unit 30109, and the result is reflected in the system internal state unit 30100, or the contents of the system internal state unit 30100 are added to the electronic message.
Thereafter, the telegram is broadcast to the cell controllers 20405 and 20406 (FIG. 1) via the optical fiber cables 11216 and 11217.
Non-reception timers (non-reception counting units) 30103 and 30104 are timers (timers) that measure and detect a state in which a message is not received for a predetermined period. The non-reception non-control determination unit 30105 causes the optical fiber cables 11214, The soundness of the communication path 11213 is confirmed.
These are recorded in the descending turn-back unit 30120 and, as necessary, a non-reception report message at the input terminal and a non-control state report message as a result of the non-reception of both inputs are generated.

If a plurality of messages stored in the message buffer (k-2) 30106 have a predetermined message type by the message selection message merging unit 30108 or the message interpretation unit 30109, the message is preferentially output (light May be transferred to the fiber cables 11216, 11217).
In addition, when the non-reception timers 30103 and 30104 determine that the communication device (cell controller 20404) has not received a further message for a predetermined time, the message stored in the message buffer (k-2) 30106 first. However, the message may be transferred to the output terminals (optical fiber cables 11216 and 11217) via the message selection message merging unit 30108 and the message interpretation unit 30109 without waiting for reception of a later message.

Also, optical fiber cables 11256, 11257, clock synchronization (k + 1) 30141, clock synchronization (k + 2) 30142, message number verification 30147, message selection message merger corresponding to the configuration of the part indicated by the notation (a) described above. 30148, message buffer (k + 2) 30146, message interpretation unit 30149, system internal status unit 30140, optical fiber cables 11254 and 11253, no reception timers 30143 and 30144, no reception no control determination unit 30145, and ascending order folding unit 30130 A reverse function that performs similar processing is implemented.
As described above, the bi-directional function is implemented with the entire configuration described in (b).

When detecting the no-control state in one-way communication, for example, when the no-reception / no-control determination unit 30105 determines that the communication states of the optical fiber cables 11214 and 11213 are not healthy, the cells 20402 and 20403 Since there is a possibility that there is a continuous failure, the descending forward message input from the optical fiber cables 11256, 11257 is changed to the ascending forward direction in the ascending turnback unit 30130, and the output end (optical fiber cable 11253, 11254).
In addition, when the communication states of the communication input from the input ends (the optical fiber cables 11257 and 11256) are not healthy, it is determined in the no-reception / non-control determination unit 30145 and corresponds to the above-described procedure in the same manner. .
Details of the system internal state unit 30100 related to ascending forward communication will be described later.

  In FIG. 3, the system internal state part is “internal state”, and the clock synchronization part is “clock synchronization (k−1), clock synchronization (k−2), clock synchronization (k + 1), clock synchronization (k + 2)”. The non-reception / non-control determination unit is “determination”, the message buffer is “message buffer (k−2), message buffer (k + 2)”, the message number verification unit is “message number verification”, and the message selection message merge unit is “message” “Selected message merge”, “message interpretation” as the message interpretation unit, “descending order folding” as the descending order folding unit, and “ascending folding” as the ascending order folding unit are described for convenience of description.

(Third embodiment)
Next, as a third embodiment of the communication system of the present invention, an example applied to a communication system power conversion device will be shown.
FIG. 4 is a diagram showing an example of an outline of the configuration of the power conversion device and the communication system when the third embodiment of the communication system of the present invention is applied to the power conversion device.
As an example, the power converter 103 is a three-phase MMCC linked to a three-phase power system.

<Overall configuration of power conversion device>
In FIG. 4, the power conversion device 103 is linked to the three-phase power system 101 via the transformer 102.
Further, the U point, the V point, and the W point in the power conversion device 103 are connected to the secondary side of the transformer 102. A load device (load) 115 is connected between the point P and the point N of the power conversion device 103.
With the configuration shown in the above outline, the power conversion device 103 exchanges AC power with the three-phase power system 101, and the power conversion device 103 exchanges DC power with the load device 115.
The load device 115 represents a direct current load, a direct current link of a motor drive inverter, a direct current power source, and the like.

《Configuration of each part of power conversion device》
The configuration of each part of the power conversion device will be described in more detail.
In FIG. 4, phase voltages (system phase voltages) of the three-phase power system 101 are denoted as VR, VS, and VT, respectively.
Moreover, the current of each phase flowing through the secondary side of the transformer 102 is expressed as IU, IV, and IW, respectively. The neutral point of the secondary side of the transformer 102 (the side to which the power converter 103 is connected) is grounded (G).
The power conversion device 103 is configured by cascading N cells (unit converters) 105. The converter arm 104 composed of a plurality of cells 105 connected in cascade has a configuration of three upper arms and three lower arms.

The power converter 103 is configured by connecting six converter arms 104 and six reactors 106 as shown in FIG.
In addition, arm currents IUH, IVH, IWH, IUL, IVL, and IWL flow through each converter arm 104.
As will be described in detail later, each cell 105 is composed of a bidirectional chopper circuit having a DC capacitor (FIG. 5). Each cell 105 is connected to an external circuit through at least two terminals, and the voltage between the two terminals can be controlled to a DC capacitor voltage or zero.
The voltage between the two terminals will be referred to as the cell output voltage or cell voltage.

<Central control device (system control device)>
In addition, a central control device 107 is installed for the purpose of controlling the power conversion device 103. The central controller 107 includes a circuit that controls power conversion and a circuit that controls communication. The central control unit 107 is grounded to a potential represented by a point G.
AC voltage sensor 108 detects system phase voltages VR, VS, and VT, and transmits the instantaneous value signals to system control devices 107S0 and 107S1 (FIG. 1) provided in central control device 107. When communication is mainly considered, “system control device provided in central control device 107” is also simply referred to as “central control device” as appropriate.
The current sensor 109 detects each arm current IUH, IVH, IWH, IUL, IVL, IWL, and transmits the instantaneous value signal to the central control device 107.

In FIG. 4, the central controller 107 includes two optical transceivers 110 and communicates with each cell 105 via the optical transceiver 110 and the optical fiber cable 111.
The central controller 107 detects the system phase voltages VR, VS, VT, the arm currents IUH, IVH, IWH, IUL, IVL, IWL, and the DC capacitor voltage VC of each cell 105, and based on these information, The modulation rate MOD to be transmitted to each cell 105 is determined, and the modulation rate MOD is transmitted to each cell. The central controller 107 performs this series of operations approximately at a predetermined cycle. This cycle is called a control cycle.
Further, the central control device 107 performs the series of operations described above to control the arm currents IUH, IVH, IWH, IUL, IVL, and IWL, thereby transferring power to and from the three-phase power system 101. Further, the DC capacitor voltage VC of each cell 105 is maintained within an appropriate voltage range.
The central control device 107 transmits the modulation factor MOD to each cell 105 via the optical transceiver 110 and the optical fiber cables 111 to 114, and receives information on the DC capacitor voltage VC from each cell 105. Details of the communication will be described later.

<Optical fiber cable>
In the embodiment shown in FIG. 4, all the cells 105 are daisy chained (connected in a daisy chain) from the central controller 107 using optical fiber cables 111 to 114.
The optical fiber cable 111 that connects the central controller 107 and the cell 105 is an optical fiber cable having a dielectric strength that can withstand the sum of the output voltages of the plurality of cells 105.
An optical fiber cable 112 that connects two cells 105 adjacent to each other in the same converter arm 104 is an optical fiber cable 112 having a dielectric strength that can withstand the cell voltage of one cell 105.

An optical fiber cable 113 that connects two cells belonging to different converter arms 104 and connected to point P is an optical fiber cable 113 having a dielectric strength that can withstand the output voltage of one cell. It is.
An optical fiber cable 113 connecting two cells belonging to different converter arms 104 and connected to the N point is an optical fiber cable 113 having dielectric strength that can withstand the output voltage of one cell. It is.
An optical fiber cable 114 that connects two cells belonging to two converter arms 104 belonging to the same phase is an optical fiber cable having a dielectric strength that can withstand the sum of output voltages of a plurality of cells.

In the present embodiment, the optical fiber cables 111 and 114 are constituted by high-voltage optical fiber cables, and the optical fiber cables 112 and 113 are constituted by low-voltage optical fiber cables.
In FIG. 4, most of the optical fiber cables are low-voltage optical fiber cables 112 and 113, and the high-voltage optical fiber cables 111 and 114 are only five in total.
Further, the physical length of the low-voltage optical fiber cables 112 and 113 can be reduced to a length that is substantially equal to the physical dimension of the cell 105.

<Cell configuration>
Next, the configuration of a cell (unit converter) including a power unit and a communication unit will be described.
FIG. 5 is a diagram illustrating an example of a cell configuration.
In FIG. 5, the main circuit of the cell 105 is a bidirectional chopper circuit including a high-side switching element 201 serving as an upper arm, a low-side switching element 202 serving as a lower arm, and a DC capacitor 203. The voltage of the DC capacitor 203 is referred to as VC.
The high side switching element 201 and the low side switching element 202 are collectively referred to as switching elements.

Further, the cell 105 includes a cell control device 204. The cell controller 204 is connected to the two optical transceivers 205 via the optical fiber cables 111, 112, 113, or the optical fiber cable 114.
In addition, the cell control device 204 generates a gate pulse for the switching element 201 and the switching element 202 and transmits it to the gate driver 206.
The gate driver 206 applies an appropriate voltage between the gate and the emitter of the switching element 201 and the switching element 202 to turn on or off the switching element 201 and the switching element 202.

The DC voltage sensor 207 detects the DC capacitor voltage VC and transmits the instantaneous value signal to the cell control device 204.
Further, the temperature sensor (209, FIG. 6) detects the temperature of the cell, and transmits the instantaneous value signal to the cell controller 204.
The self-powered power supply 208 supplies power to the cell control device 204 and the gate driver 206.
The potential of the cell control device 204 is the same as that of the emitter terminal of the low-side switching element 202. In FIG. 5, this point is represented by a G (CELL) point. The potential of this G (CELL) point is common to each cell including other cells.
Note that a point G (CELL) in FIG. 5 is a point of a different potential isolated from the point G in FIG.

<Functions of cell control device>
Next, functions of the cell control apparatus will be described.
FIG. 6 is a diagram illustrating an example of functions of the system internal state unit 30100 or the system internal state unit 30140 of the cell control device 204 (FIG. 5) in the cell 105 (FIG. 5).
In FIG. 6, the system internal state unit 30100 or the system internal state unit 30140 stores registers 30161 to 30672 described later.
The correction calculation unit 30180 takes in the value of the DC voltage sensor 207 that measures the capacitor voltage of the cell 105 and the value of the temperature sensor 209 and separately uses a correction table 30181 that is provided with or holds a correction value. An operation for correcting the capacitor voltage of the cell 105 is performed, and the result is stored in a predetermined register address.
Here, the correction table 30181 may be set by a message, or may be set when a cell is manufactured, and the correction calculation unit 30180 may be a table lookup or a simple linear correction.

The capacitor voltage comparison & counting unit 30182 reads the corrected capacitor voltage of the cell 105 and the message that reads the register corresponding to the capacitor voltage, and the cell 105 located upstream of the communication extracted by the message interpretation units 30109 and 30149. The capacitor voltage value is compared, and a numerical value corresponding to the result is stored in the counting information sections 30183, 30184, and 30185.
In FIG. 6, the count information unit 30183 is expressed as “voltage is large”, the count information unit 30184 is expressed as “voltage is equal”, and the count information unit 30185 is expressed as “voltage is small”.
By doing in this way, the said cell 105 can determine the order of the other cell 105 of the communication upstream the magnitude | size of the capacitor voltage.
In addition, the relationship with the other cells 105 downstream of the communication can be obtained in the same way by the communication in the reverse direction. As a result, the position of the capacitor voltage of the cell 105 among all the cells 105 in the communication loop is determined. The information, or the count information of the count information sections 30183, 30184, and 30185 that are the basis of the information, can be obtained from the system controller 107S0, 107S1 (FIG. 1) of the central controller 107 (FIG. 4) by a register read message. ).

  In addition, the system control devices 107S0 and 107S1 (FIG. 1) of the central control device 107 (FIG. 4) use a method to be described later. On the basis of the counting information indicating the position, the plurality of cells 105 in the loop can be rearranged in the order of the capacitor voltage, and comparison information can be generated.

  In addition, the system control devices 107S0 and 107S1 of the central control device 107 acquire comparison information arranged in the order of capacitor voltages of a plurality of cells 105 configured in a daisy chain (communication loop) by a method described later. Integrated comparison information obtained by integrating comparison information can be generated.

  Further, the system control devices 107S0 and 107S1 of the central control device 107 are based on the integrated comparison information, the output voltage command of the cell 105, and the current flowing in the integrated cell 105 group separately measured by the current sensor 109 (FIG. 4). In addition, it is possible to determine the priority order for outputting the voltage of each cell in the cell 105 group.

  In addition, the system control devices 107S0 and 107S1 of the central control device 107, as a priority order for outputting the voltage of each cell 105 in the cell 105 group, if the current flowing through the cell 105 group is in the charge direction based on the integrated comparison information. For example, the cell 105 having a low DC capacitor voltage provided in the cell 105 can be selected, and the cell 105 having a high DC capacitor voltage provided in the cell 105 can be selected in the discharge direction.

  Further, the system control devices 107S0 and 107S1 of the central control device 107 can rearrange them in the order of voltage using the integrated comparison information, and set the priority order of outputting the voltages of the cells.

  In addition, the system control devices 107S0 and 107S1 of the central control device 107 classify each cell 105 into a predetermined voltage range using the integrated comparison information, and set the priority order in which the voltages are output in this classification unit. Can do.

  In the above description, one arm is described as being composed of one communication loop (a daisy chain). However, even when one arm is composed of a plurality of loops, the loops are rearranged. Then, the entire arm can be rearranged at a higher speed by a known algorithm.

<Cell sorting method>
Next, a method for rearranging a plurality of cells 105 (FIG. 4) having voltage variations will be described.
FIG. 7 is a diagram illustrating an example of a method for integrating the rearrangement results 10700, 10710, 10720, and 10730 obtained by importing the results of rearrangement by four cells into the integrated rearrangement result 10740 in the cell rearrangement method. It is.
In FIG. 7, the elements in each line (cell voltages V00... V02... V34... V39) are sequentially taken into the comparison table 10750. In the comparison table 10750, there are six large and small comparison results 10751 to 10756, and in the arrows shown in FIG. 7, the element where three arrows gather is the largest or smallest.

The element so pointed, in this case, V33 (10733) is copied to the integrated rearrangement result 10740, and then V34 (10734) is newly incorporated.
V34 (10734) is compared with the three elements V02 (10702), V13 (10713), and V21 (10721) on the comparison table 10750, and the magnitude relations 10553, 10755, and 10756 are updated. The integrated rearrangement result 10740 is completed by repeating this.
In FIG. 7, the rearrangement results 10700, 10710, 10720, and 10730 are numbered according to the rule that the number increases by one (lower one digit) when moved to the right, and the lower two digits are ( V00... V02... V34... V39) are numbered according to a rule corresponding to a two-digit number.

Next, an example of how to use the cell information obtained by the above-described method will be described.
FIG. 8 shows examples of various designation methods when each cell provided in the arm is turned on / off with a cell number designated. (A) is a voltage waveform of the arm. , (B), (c), and (d) correspond to the on / off designation method of the first example, the second example, and the third example, respectively, (e) corresponds to the current waveform of the arm, and (d) corresponds to the current. This is an on / off designation method of the fourth example.

FIG. 8A shows the voltage waveform of the arm as described above, the horizontal axis is the voltage angle (electrical angle) corresponding to time, and the vertical axis is the combined voltage of the arm.
Waveforms TV1, TV2, TV3, and TV4 indicate voltage waveforms at 0 to π, π to 2π, 2π to 3π, and 3π to 4π, respectively, and generally indicate a negative cosine waveform (the waveform is equivalent to a sine wave). Yes. It is desirable to output this cosine waveform or equivalent sine waveform.

FIG. 8B shows a first example of the cell designation method, where the horizontal axis represents a voltage angle (electrical angle) corresponding to time, and the vertical axis represents N cell numbers # 0 to # (N− 1), the waveforms S11 and S13 represent the timing when the corresponding cell number is turned on, and the waveforms S12 and S14 represent the timing when the cell is turned off.
For example, at the voltage angle 0, all the cells # 0 to # (N-1) are in an off state, so no voltage is output from the upper arm, and the output from the lower arm becomes dominant. Voltage (−Vdc) is output. Note that when the cell is turned off, the output voltage becomes zero.
Then, with time (voltage angle), as the cells # 0, # 1,... Turn on sequentially in accordance with the waveform S11, the voltage rises like the waveform TV1. When the voltage angle (electrical angle) is π, all the cells # 0 to # (N−1) are turned on, so that the maximum voltage (+ Vdc) is output.
When the voltage is positive, it means that the cell is discharged, and when the voltage is negative, it means that the cell is charged.

Also, when the voltage angle goes beyond π and goes to 2π, the cells # 0, # 1,... Are sequentially turned off according to the timing of the waveform S12. Then, as the number of cells to be turned off increases, the voltage decreases to a voltage waveform TV2.
When the voltage angle is 2π, all cells # 0 to # (N−1) are turned off, so that the lowest voltage (−Vdc) is output.
As described above, since the output voltage needs to output a cosine waveform or an equivalent sine waveform, the ON / OFF timings of the cells # 0 to # (N-1) are also changed to the waveforms S11 and S12. It is necessary to follow the waveform as shown.
Also, in the section where the voltage angle is 2π to 4π, the cells of # 0 to # (N−1) are turned on / off according to the timing of similar waveforms S13 and S14.
As described above, in the first example of the cell designating method in FIG. 8B, the on / off of each cell is switched once per cycle, and the period is half of the cycle.

FIG. 8C shows a second example of the cell designation method, in which the horizontal axis represents a voltage angle (electrical angle) corresponding to time, and the vertical axis represents N cell numbers # 0 to # (N− 1), waveforms S21 and S22 indicate that the corresponding numbered cell is on, and waveforms S23 and S24 indicate that the cell is off.
In the section where the voltage angle is 0 to π, the cell numbers # 0 to # (N−1) are sequentially turned on according to the timing of the waveform S21, and the maximum voltage (+ Vdc) is output when the voltage angle is π. .
Then, in the section where the voltage angle is π to 2π, the ON state is canceled in order of # (N−1) and # (N−2) according to the timing of the waveform S22. When the voltage angle is 2π, there is no cell in the on state, so that the lowest voltage (−Vdc) is output.

The waveform S23 shows a state in which the cells are off. When the voltage angle is 2π, all the cells are off. However, in the interval where the voltage angle is 2π to 3π, the cell is in accordance with the timing of the waveform S23. Since the OFF state of is eliminated, the voltage increases, and the maximum voltage (+ Vdc) is output when the voltage angle is 3π.
In the section where the voltage angle is 3π to 4π, the cell off state increases according to the timing of the waveform S24, so that the voltage decreases. When the voltage angle is 4π, the voltage is the lowest voltage (−Vdc). ) Is output.
As described above, in the second example of the cell designating method in FIG. There is also a control method based on the above method and viewpoint.

FIG. 8D shows a third example of the cell designation method, where the horizontal axis represents a voltage angle (electrical angle) corresponding to time, and the vertical axis represents N cell numbers # 0 to # (N− 1), waveforms S31 and S32 indicate that the corresponding numbered cell is on, and waveforms S33A, S33B, S34A, and S34B indicate that the cell is off.
The section where the voltage angle is 0 to 2π is the same as the section where the voltage angle is 0 to 2π in FIG.
In the section where the voltage angle is 2π to 3π, the designated number is shifted and designated. That is, in FIG. 8C, # (N−1), # (N−2)... Are sequentially cleared from the OFF state, but FIG. The OFF state is canceled from the cell number considerably smaller than the cell number according to the waveform S33A.

At this time, in the waveform S33A, since the small cell number disappears before the voltage angle reaches 3π, the cell numbers remain large according to the timing of the waveform S33B # (N−1), # (N− 2).. At this time, the maximum voltage (+ Vdc) is also output when the voltage angle is 3π.
Further, in the section where the voltage angle is 3π to 4π, the cell number designation is changed in accordance with the waveforms S34A and S34B to turn it off.
As described above, in the third example of the cell designating method in FIG. 8D, the cell number is circulated at the timing when all the cells are turned off, and the effect of the periodic replacement is increased.

FIG. 8E shows the current waveform of the arm, the horizontal axis is the voltage angle (electrical angle) corresponding to time, and the vertical axis is the current flowing through the arm.
Waveforms TI1, TI2, TI3, and TI4 indicate voltage waveforms at 0 to π, π to 2π, 2π to 3π, and 3π to 4π, respectively, and are generally cosine waves or sine waves. However, since an electric circuit usually includes a reactance component, there is generally a phase difference between a current waveform and a voltage waveform. Therefore, the current waveform in FIG. 8 (e) has a phase difference from the voltage waveform in FIG. 8 (a), and the current (A) section in which the current is positive and the negative section (B) are in the negative section. FIG. 8A is shifted from π / 2 to 3π / 2 where the voltage is a positive interval and 3π / 2 to 5π / 2 where the voltage is a negative interval.

FIG. 8F shows a fourth example of the cell designation method, in which the horizontal axis represents a voltage angle (electrical angle) corresponding to time, and the vertical axis represents N cell numbers # 0 ′ to # (N -1) ′, and the waveforms S41C, S41D, 42A, 42B, S43C, S43D, S44A, and S44B indicate that the cells with the corresponding numbers are on, and the waveforms S41A, S41B, S42C, 42D, S43A, S43B, S44C, and S44D show a state where the cell is off.
Waveforms S41C, S41D, 42A, 42B, S43C, S43D, S44A, S44B, S41A, S41B, S42C, 42D, S43A, S43B, S44C, which designate ON / OFF of the cell number of the fourth example in FIG. S44D considers the timing at which the current waveform in FIG. 8E switches from negative to positive or from positive to negative.

Note that the cell designation method of the first example, the second example, and the third example of FIGS. 8B, 8C, and 8D is the timing at which the voltage waveform of FIG. 8A is switched to increase or decrease. Is taken into account.
In FIG. 8 (f), the cell numbers # 0 'to # (N-1)' and "'(dash)" are added to the numbers after the cells are rearranged as described above. It is shown that.
For example, the waveform S41A, the waveform S41B, and the waveforms S41C and S41 are designated locations where numbers are jumped on / off for the rearranged cells.
The two characteristic lines of the waveform S41A and the waveform S41B exist because the number is skipped and the cell off state is specified, so the relationship between time and cell number may not be on one curve. For it to happen.

In addition, the waveform S41A, the waveform S41B, and the waveforms S41C and S41 exist in the same time zone because the designation of ON and the designation of OFF may be mixed.
The waveforms 42A, 42B, S43C, S43D, S44A, S44B and the waveforms S42C, 42D, S43A, S43B, S44C, S44D in the other sections have the same relationship.
As described above, there is a method of specifying ON / OFF of a complicated cell number as in the fourth example in FIG. 8F, and there are various other cell specifying methods.

The method of FIG. 8F is compared with the methods of FIGS. 8B, 8C, and 8D as follows.
In section (A) shown in FIG. 8E, the direction of current is positive, and in section (B), the direction of current is negative. In this case, since each cell in the method of FIG. 8B is always turned on at a specific current pattern, a cell having a long period (B) is discharged and a cell having a long period (A) is continuously charged. The cell voltage fluctuates and the voltage variation between the cells also increases.
Further, in the methods of FIGS. 8C and 8D, the voltage fluctuation of the cell is suppressed, but the effect of suppressing the voltage variation between the cells may be small.
In contrast, in FIG. 8F, the vertical axis indicates the cell numbers # 0 ′ to # (N−1) ′ rearranged in order of voltage as described above. Further, the cells to be turned on are replaced according to the sections (A) and (B). By doing so, the low voltage cell is in the charging direction, and the high voltage cell is in the discharging direction, voltage fluctuation is suppressed, and voltage variation between the cells is also reduced.

Also, as shown in (f), # 0 ′ and # (N−1) ′, # 1 ′ and # 1 are used for the purpose of preventing switching of many cells at the time of switching between sections (A) and (B). (N-2) ′, # 2 ′ and # (N-3) ′ are sequentially replaced.
Furthermore, as a method of switching the rearrangement, in order to prevent simultaneous switching and manage the number of switching, it is performed at the timing when all the cells are turned on or all the cells are turned off, that is, at the timing of the voltage angle nπ.
Further, the rearrangement calculation may be performed with the predicted voltage corrected by the current flowing after the voltage measurement.
Further, these rearrangements may not be rearranged according to the total cell voltage of the arm, and may be divided into several stages such as large, medium and small.
Moreover, you may implement by these combinations partially.

(Fourth embodiment)
Next, a fourth embodiment of the present invention will be described.
FIG. 9 is a diagram illustrating an example of a schematic configuration of the power conversion device and the communication system when the fourth embodiment of the communication system of the present invention is applied to the power conversion device.
In FIG. 9, the basic structure as a power converter is the same as that of the third embodiment of FIG.
The difference is that the optical fiber cable 111 is daisy chain connected for each phase (three phases) of the power converter. The central control unit 107 has six optical transceivers 110 and is connected to the three daisy chains according to the phases.
In the third embodiment (FIG. 4), all cells are daisy chain connected by optical fiber cables.

By adopting the above-described configuration shown in FIG. 9, the number of cells communicating with one optical transceiver 110 is N / 3, where N is the third embodiment.
Therefore, the length of the optical serial signal frame can also be shortened to about 1/3 compared with the third embodiment.
In the fourth embodiment, the number of high-voltage optical fibers is increased compared to the third embodiment, but most of the optical fiber cables connecting adjacent cells have a higher voltage resistance than the medium-voltage optical fiber cable. A low-voltage optical fiber cable that requires less can be used.
Moreover, in 4th Embodiment, the effect that communication time can be shortened compared with 3rd Embodiment is acquired.

  In addition, the overlapping description about the same structure, function, and effect as 3rd Embodiment is abbreviate | omitted.

<Details of System Internal State Unit 30100 Related to Ascending Forward Direction Communication>
Next, details of the system internal state unit 30100 (FIG. 3) related to ascending forward communication in the second, third, and fourth embodiments will be described with reference to FIGS. 10A, 10B, 11A, and 11B.
FIGS. 10A and 10B are diagrams illustrating detailed examples of registers in the system internal state unit 30100 (FIG. 3) related to ascending forward communication. Note that the system internal state unit 30140 (FIG. 3) of the descending forward communication has the same structure.
In the following, not all functions are frequently used, but those that may be used are listed below.

<< Register 30160 >>
In FIG. 10A, reference numeral 30160 denotes a legend for register explanation. The upper 0 to 7 indicate 8-bit positions.
“CSNOR (CSNOR)” is an example, and indicates the name of a function that occupies a register field.
Further, W0, R1, etc. are examples, and “W” is dedicated to writing, and “R” is dedicated to reading. When both writing and reading are possible, “WR” is used.
The numerical values 0 and 1 in “W0” and “R1” indicate initial values.
The following registers are described according to the above rules.
In the following, alphabetic names and numerical values are described by mixing full-width and half-width as appropriate.

<< Register 30161 >>
The register 30161 is CSN [0: 1] R (Command Serial Number Zero / One Register). This holds the value of the number field of the received message.
The new value is held in CSN0 and the previous value is held in CSN1. Since transfer of the distributed message is controlled according to this numerical value, a value different from the value held in this register is described in the number area of the message to be delivered.
RFF R represents read-only (READ ONLY), and FF represents an initial value (1111 1111), but also represents that the initial value FF is invalid. Therefore, control is performed so as not to transmit a message whose number area is FF.
Note that the operation when such a message is received is undefined.

<< Register 30162 >>
The register 30162 is an RXSR (Receive Transaction Status Register). This maintains the state of the two receiving ends. Writing to register RXSR is ignored and writing to RXCR resets the value. Also,
RXN [1: 0]: 1: Indicates that there was no reception from the input end for a certain period indicated by CTTR.
0: Indicates that a message has been received in the past shorter than a certain period indicated by CTTR.
RXD [1: 0]: 1 indicates that the input terminal is closed.
Set when the same report message is received continuously from Ck-1 / k-2.
0: The input end is not closed.
RXE [1: 0]: 1 indicates that the input terminal is not healthy.
Set by evaluation of the confirmation area of the message.
0 indicates a healthy input end.
NOC [0: 1]: 1: Indicates no control status.
Set when both inputs are RXN, RXE or RXD.
0: Indicates that there is no uncontrolled state.

<< Register 30163 >>
The register 30163 is an RXCR (Receive Transaction Clear Register). Writing 1 to each bit resets the corresponding RXSR state to 0.
The read value of RXSR is undefined.

<< Register 30164 >>
The register 30164 is CTT [1: 0] [1: 0] R (Command Timeout Time Register). This sets the input end determination waiting time. This register is common to bidirectional communication. Holds the value set later. The unit is 256 cycles (2.5 μs at 100 MHz). CTT1 [1: 0] R corresponds to the input terminal 1 and CTT0 [1: 0] R corresponds to the input terminal 0.

<< Register 30165 >>
The register 30165 is a CTC [1: 0] [1: 0] R (Command Timeout Counter Register). This adds 1 to CTC [1: 0] 0 every 256 cycles (2.5 μs at 100 MHz). When CTC [1: 0] 0 is incremented by 254, 1 is added to CTC [1: 0] 1, and CTC [1: 0] 0 returns to 00. This is to avoid 255 representing an interruption.
CTC1 [1: 0] corresponds to the input terminal 1 and CTC0 [1: 0] corresponds to the input terminal 0. When a message is received at each input terminal, these values are 00.
When the value of CTC [1: 0] [1: 0] exceeds the corresponding CTT [1: 0] [1: 0], the corresponding NOC [1: 0] is set to 1. If NOC [1: 0] is already 1, a report message to that effect is generated.

  In FIG. 10B, the continuation of the same register will be described.

<< Register 30166 >>
The register 30166 is an RMTR (Report Meeting Register). This indicates the waiting time for a report-type message. This register is common for bidirectional communication. Holds the value set later. The RMT is set to a value smaller than the CTT.
When one of the CTC [1: 0] 0 exceeds the RMT, the waiting is discarded and the transmission of the contents of the BUF is started in order from the BUF0.

<< Register 30167 >>
The register 30167 is an RIR (Report Interval Register). This indicates the non-reception report repetition interval. The unit is 4 cycles, and the default value is 16.
RN (Report Number) represents the number of report message transmissions. Report the report message four times, and stop sending the report message when it returns to 00.

<< Register 30168 >>
The register 30168 is a BUFR (Buffer Register). This temporarily stores the collected telegrams in full for waiting for the collected telegrams. Message storage starts from BUF0. When the waiting message arrives, the message is taken out from BUF00 in order, synthesized, and transferred.

<< Register 30169 >>
The register 30169 is a BUFE (Buffer Enable). This indicates that BUF0 is valid.

<< Register 30170 >>
The register 30170 is a gate pulse designation (GATER). GP [1: 0] and GN [1: 0] hold P-side and N-side gate pulse signal command values, respectively. GPP [1: 0] and GNP [1: 0] hold the previous GP [1: 0] and GN [1: 0]. GN1 and GP1 are control bits for the bridge circuit. In the chopper circuit, GN1 represents a gate signal of a commutation thyristor, and GP1 represents a closing signal of a short-circuit switch (fixed type).

<< Register 30171 >>
Resistor 30171 represents the cell state and capacitor voltage. This register is common for bidirectional communication.
TMP (Temperature) indicates that the temperature is high. PE (P-side Error) and NE (N-side Error) indicate that each is abnormal. VC1 / 0 (Voltage of Capacitor) R holds the upper 4 bits and lower 8 bits of the capacitor voltage.

<< Register 30172 >>
Register 30172 is CN [3: 0]. This is common in bidirectional communication and records the cell number. The number is recorded in the nonvolatile memory area at the time of factory shipment.

Next, a register related to the type of electronic message (communication message) will be described with reference to FIGS. 11A and 11B.
FIGS. 11A and 11B are diagrams illustrating detailed examples of registers in the system internal state unit 30100 (FIG. 3) related to the type of electronic message (communication message).
The registers in FIG. 11A will be described in order.

<< Register 40100 >>
A basic structure is shown in the register 40100.
CHK [7: 0] is a confirmation area and describes the parity of the message. Here, odd parity is used. This area may have a value of FF.
END [7: 0] is the message end area, and the value is always FF. When the cell controller detects FF in the message, it recognizes that the message ends in the next CHK area. Since the CHK area itself may become FF, FF following FF is not recognized as END.
END may be used for forced termination due to message contention. Forcibly terminated messages are discarded. At that time, the CHK area is checked, and if it is abnormal, a report-type message is generated and reported.
MID [7: 0] is a message identifier. The first 8 bits of the message.

<< Register 40101 >>
The register 40101 is CMD [3: 0]. Indicates the instruction type of this message.
CN [3: 0] are the lower 4 bits of the cell number. Identify the issuer of the report-type message. Since only 16 cells can be identified by this alone, the remaining cell identification information is included in the following 8 bits. However, only the first 8 bits are used to determine whether or not the message transfer is possible.
Therefore, there is a possibility that a report message from a cell 16n away from the cell with the same CID [5: 4] is interpreted as the same message and the upstream message is discarded. For this reason, the CID [5: 4] is hashed with the cell number [5: 4] so that the CIDs are different.
CID [5: 4] is an instruction identifier. The instruction identifier is incremented each time an instruction is issued. As a result, the cell controller recognizes continuously issued instructions as separate instructions.

  Next, details of the distribution-type and collection-type messages are shown. For simplicity, CID [5: 4] = ’b00’.

<< Register 40102 >>
The register 40102 is an on / off command for the in-cell IGBT.
L [3: 0]: Message length. The length of the message is expressed in units of 4B (bytes). When LEN = 1, the whole is 4B. When the communication loop is composed of 40 cells, L = 6. In this case, since 8L-7 = 41, the 4L-3rd byte is not used. When the communication loop is composed of 32 cells, L = 5. In this case, since 8L-7 = 33, the 4L-3rd byte is not used.
GP / GN: P / N side gate control command. Each control is commanded with 2 bits. 1 is on and 2 is off. In the case of a chopper cell, only even bits are valid. In the case of a bridge cell, odd bits are also effective. A command in which the even bit of GP and the even bit of GN, and the odd bit of GP and the odd bit of GN are both “1” is prohibited. This is because it is a setting that causes a DC short circuit and it is necessary to prevent the message from being forcibly terminated due to the 0xFF byte being mixed in the message.

<< Register 40103 >>
The register 40103 writes to the registers of all cells in units of 2 bytes.

  FIG. 11B is a continuation of the message (message format) in FIG. 11A. FIG. 11B will be described.

<< Register 40104 >>
The register 40104 writes the register value of the designated cell in units of 64 bytes.

<< Register 40105 >>
The register 40105 reads register values of all cells in units of 2 bytes.

<< Register 40106 >>
Register 40106 is the RA [7: 0] register address. The least significant bit RA [0] is always “b0”. For reading, two bytes are read in the order of even number and odd number.
CnRD0 / 1: Register data 0 and 1 of cell n.
2 ≦ L. When L = 2, the whole is 8B. In this case, only the data C0RD0 and C0RD1 of the first cell are returned. When the communication loop is composed of 40 cells, L = 22. In this case, since 2L-4 = 40, the 4L-4th and 4L-3th 2 bytes are not used. When the communication loop is composed of 32 cells, L = 18. In this case, since 2L-4 = 32, the 4L-4th and 4L-3th 2 bytes are not used.
When this message is transmitted, all zeros are set in the area. Each cell incorporates the value of the register only in the part that the cell is in charge of and transfers it to the downstream cell.

<< Register 40107 >>
The register 40107 reads the register value of the designated cell in units of 64 bytes.

  Although not shown, there are the following registers.

<< Register 40108 >>
The register 40108 specifies the address of the target cell with CA [7: 0]: cell address.
When a value larger than the minimum power of 2 that exceeds the number of cells in the communication loop is specified, the upper bits corresponding to the power are ignored.
RA [7: 0]: Register address. The register read unit is 64B. Therefore, L = 18. Since 2L-4 = 32, the 2nd byte of 4L-4th and 4L-3th is undefined and not changed. Note that RA [5: 0] is 'b000000'.
When this message is transmitted, all zeros are set in the area. Each cell incorporates the register value only in the part that the cell is in charge of and transfers it to the downstream cell.
No reception report
MID [7: 0] = '0x40': RXS
Report that a no-reception condition has been detected by sending out the RXSR value.

(Other embodiments)
The present invention is not limited to the embodiment described above. Here are some examples:

  In this embodiment, the three-phase MMCC linked to the three-phase power system has been described as the power converter, but the present invention is also applicable to a single-phase MMCC linked to the single-phase grid and an MMCC that drives the motor. Is possible. The present invention is also applicable to CMC (Cascade Multi-level Converters).

  Further, in the plurality of control devices having the configuration of daisy chain (daisy chain connection) used in the present embodiment, the communication system that connects not only the control device of the adjacent cell but also the further cell thereof is the above-described power conversion device. In addition, the present invention can be applied to a general device having a plurality of control device configurations.

  In the present embodiment, the switching elements 201 and 202 have been described using IGBTs. However, the switching elements 201 and 202 are not limited to IGBTs, but are not limited to IGBTs, but include GTOs (Gate-Turn-Off Thyristors), GCTs (Gate-Commutated turn-off Thyristors), MOSFETs (Metal- It can also be applied to on / off control elements (switching elements) such as Oxide-Semiconductor Field-Effect Transistor).

  In addition, the communication between the system control device and the cell control device has been described using the optical fiber cable. However, in the electric communication using the electric cable, the “adjacent control devices in the daisy chain connection” are adjacent to each other. A communication system that connects not only to the cell control apparatus, but also to further cells can be applied.

  Further, the communication between the system control device and the cell control device has been described in the form of a communication telegram (telegram) using a register, but it can also be applied to the case of general telecommunication that does not depend on the “telegram” form.

  The control circuit of the system control device and the cell control device may have a software configuration, a hardware configuration, or a configuration in which software and hardware are mixedly mounted.

(Supplement of the present invention and this embodiment)
In a plurality of control devices having a configuration of daisy chain connection (daisy chain connection) used in the present embodiment, not only the control device of the adjacent cell but also the further cell is connected, so that the control device of the cell breaks down. Even in such a case, communication is possible by bypassing the control device of the cell.
Furthermore, by applying the same connection to the daisy chain connection in the reverse direction, communication and operation can be continued by communication by loopback or communication from the output terminal in the reverse direction even when two adjacent cells fail. Is.

DESCRIPTION OF SYMBOLS 101 Three-phase electric power system 102 Transformer 103 Power converter 104 Arm, converter arm 105 Cell, unit converter 106 Reactor 107 Central controller 108 AC voltage sensor 109 Current sensor 110 Optical transceiver 111, 114, 1111-1111171 Optical fiber cable , High voltage optical fiber cable 11204-11268 optical fiber cable, medium voltage optical fiber cable (low voltage optical fiber cable)
112, 113 Optical fiber cable, low-voltage optical fiber cable 115 Load, load device 10700, 10710, 10720, 10730 Rearrangement result 10740 Integrated rearrangement result 10750 Comparison table 10751-10756 Comparison result
107S0, 107S1 System controller, system controller 201 Switching element, high-side switching element 202 Switching element, low-side switching element 203 DC capacitor 204 Cell controller (distributed controller)
205 Optical Transceiver 206 Gate Driver 207 DC Voltage Sensor 209 Temperature Sensor 20400-20409 Cell Controller, Cell Control Device (Communication Device)
30100, 30140 Internal state, internal state part, system internal state part 30101, 30102, 30141, 30142 Clock synchronization, clock synchronization part 30103, 30104, 30144, 30145 No reception timer, no reception counting part 30105, 30145 No reception, no control determination Part, (judgment)
30106, 30146 Message buffer, message storage unit 30107, 30147 Message number matching unit 30108, 30148 Message selection message merging unit 30109, 30149 Message interpretation execution unit (message interpretation)
30100, 30140 System internal state section 30120 Descending order folding section 30130 Ascending order folding section 30160-30172 Details of registers and system internal status section 30180 Correction calculation section (correction calculation section)
30181 Correction table 30182 Capacitor voltage comparison & counting unit 30183 Counting information unit (large voltage)
30184 Count information part (equal voltage)
30185 Count information part (voltage is small)
40100-40107 Details of the message

Claims (20)

A communication system comprising a plurality of communication devices built in a plurality of distributed control devices,
The communication device has at least two communication input ends and two communication output ends;
One communication input end of the two communication input ends of the kth communication device is connected to a communication output end of an adjacent (k-1) th communication device,
The other communication input terminal of the two communication input terminals of the kth communication device is further connected to the communication output terminal of the next (k-2) th communication device,
One communication output terminal of the two communication output terminals of the kth communication device is connected to a communication input terminal of an adjacent (k + 1) th communication device,
The other communication output terminal of the two communication output terminals of the k-th communication device is further connected to a communication input terminal of an adjacent (k + 2) -th communication device,
A communication system, wherein a plurality of communication devices including the (k−2) to (k + 2) th communication devices are connected in a daisy chain. In the communication system according to claim 1,
Furthermore, a system control device built in a central control device that performs overall control of the plurality of distributed control devices is provided,
The system control device is connected to two communication devices located at both ends or in front of the plurality of communication devices connected in a daisy chain. In the communication system according to claim 1,
Furthermore, each system has two system control devices that perform the same processing in duplicate.
One system control device of the two system control devices is connected to a communication device located at one end of the plurality of communication devices connected in a daisy chain and a communication device located immediately before the other end,
The other system control devices of the two system control devices are connected to a communication device located immediately before one end of the plurality of communication devices connected in a row and a communication device located at the other end. Communications system. In the communication system according to claim 2,
A plurality of the system control devices;
A plurality of daisy chain connections of the plurality of communication devices, and
A communication system, wherein a plurality of daisy chained communication devices are connected to different system control devices. In the communication system according to claim 3,
A plurality of four or more system control devices;
A plurality of daisy chain connections of the plurality of communication devices, and
A communication system, wherein a plurality of daisy chained communication devices are connected to different system control devices. In the communication system according to claim 5,
Each of the plurality of communication devices further has two input ends and two output ends, and enables bidirectional communication by connecting so that communication is transmitted in the opposite direction from the previous input end and output end. A communication system. In the communication system according to claim 5,
Each of the plurality of communication devices has a non-reception counter that measures the time when there is no input,
When the non-reception counting unit detects that there is no input for a predetermined time, it generates and outputs a message for notifying the system control device of the detection result. In the communication system according to claim 6,
Each of the plurality of communication devices has a non-reception counter that measures the time when there is no input,
When the non-reception counting unit detects that there is no input for a predetermined time at the input end of one communication direction of the two-way communication, a message for notifying the detection result is sent to the one communication direction. A communication system, wherein output is performed from an output end in the opposite communication direction. In the communication system according to claim 6,
Each of the plurality of communication devices sends the same communication message to an output end in each direction of the bidirectional communication,
The communication message has an individual identification number;
The input terminal of each of the plurality of communication devices has a message number verification unit that records and determines the identification number;
A communication system, wherein when the message number matching unit of one communication device receives a communication message having the same identification number, a communication message received later is not transferred to the output terminal of the communication device. In the communication system according to claim 9,
Each of the plurality of communication devices has a message storage unit that temporarily stores the communication message.
A communication system characterized in that, in a plurality of communication messages stored in the message storage unit, a communication message of a predetermined message type is preferentially transferred to an output terminal. In the communication system according to claim 10,
The communication message received first by the communication device and stored in the message storage unit is output without waiting for reception of a later communication message when the communication device does not receive a further communication message for a predetermined time. A communication system characterized by being transferred to an end. The communication system according to any one of claims 1 to 11,
A communication system, wherein the plurality of communication devices are provided in a power conversion device. The communication system according to any one of claims 1 to 11,
The plurality of communication devices is a power conversion device including a distributed control device in which a plurality of unit converters are cascade-connected and installed in the vicinity of the same potential as the unit converter, and a central control device that performs overall control of the distributed control device. ,
A communication system characterized by being provided. In the communication system according to claim 2,
The plurality of communication devices are
A distributed control device in which a plurality of unit converters are cascade-connected and installed in the vicinity of the same potential as the unit converter, a central control device that performs overall control of the distributed control device, and the distributed control device that stores data of the unit converter A power converter comprising: a means for measuring; a means for correcting the data; and a means for transmitting the data to the system controller of the central controller by a communication function.
A communication system characterized by being provided. The communication system according to claim 14,
A communication device downstream in the transmission direction in the communication device of the plurality of distributed control devices configured in a daisy chain,
When transmitting the corrected measurement data to the system control device of the central control device, the correction measurement data and the result of comparing with the measurement data of the upstream communication device in the transmission direction are held. Communication system. The communication system according to claim 15,
The system control device of the central control device acquires the comparison information in the communication devices of the plurality of distributed control devices configured to connect the beads from the communication device group connected to the beads, and integrates the plurality of comparison information. A communication system that generates integrated comparison information. The communication system according to claim 16,
Based on the integrated comparison information, the output voltage command of the unit converter, and the current flowing in the unit converter group separately measured and integrated, the system controller of the central control unit A communication system, wherein a priority order for outputting the voltage of the unit converter is determined. The communication system according to claim 16,
The priority order is based on the integrated comparison information, if the current is in the charging direction, the unit converter having a low DC capacitor voltage provided in the unit converter is selected, and if the current is in the discharging direction, the unit converter is provided. And selecting a unit converter having a high DC capacitor voltage. The communication system according to claim 18,
A communication system, wherein the plurality of unit converters are rearranged in the order of voltage using the integrated comparison information, and the priority order of outputting the voltages of the unit converters is set. The communication system according to claim 18,
A communication system, wherein the plurality of unit converters are classified into a predetermined voltage range using the integrated comparison information, and a priority order in which the voltages of the unit converters are output in the classification unit.

Images