JP2015115864A - Data transfer system and data transfer system control method - Google Patents

Abstract

Multicast communication is enabled without impairing real-time performance in control network communication. A master that generates and transmits a transfer frame is connected to a plurality of slaves to which control targets are connected through a communication network including at least one data transfer apparatus. The data transfer apparatus 10 has a communication port connected to the master 60 and a communication port connected to the slave 70. The data transfer device 10 copies the transfer frame received from the master 60 during a predetermined period set in advance according to a plurality of data transfer paths configured on the communication network, and performs communication on the slave 70 side. By transferring the data to the port, the data transfer is completed within a predetermined period. [Selection] Figure 5

Description

  The present invention relates to a data transfer system and a data transfer system control method.

  In the first half of the 1990s, Ethernet (registered trademark) was used as a local area network (LAN) communication medium in information systems, but it was difficult to apply to factory automation (FA) systems due to real-time performance problems. It had been. However, with the advent of switching hubs, this problem has been drastically improved, and the spread of the Internet has become a tailwind, and switching hubs can be obtained at a low cost. It has come to be used for industry (hereinafter referred to as “industrial Ethernet”).

  Industrial Ethernet has been applied to inter-controller networks since the mid-1990s, and since 2000, it has been extended to field networks that connect actuators, sensors, etc. installed on production sites via networks. Yes.

  In general, a control network system is composed of a control network such as an industrial Ethernet, a master connected to the control network, and a plurality of slaves. The master sends out a frame carrying a control command, and each slave transmits the frame. Receive and execute control instructions contained in it. Information necessary for each slave may be attached to the frame.

  Regarding the control system network system, for example, there are technologies as shown in Non-Patent Documents 1 to 3. EtherCAT (registered trademark) described in Non-Patent Document 1 is an Ethernet-based fieldbus standard. In this standard, the master writes control information for a plurality of slaves in a predetermined area on one Ethernet frame, and the slave that receives the frame reads the control information and controls the actuator and the like. The slave writes the sensor information in a predetermined area of the Ethernet frame, and transfers the information to the master. One Ethernet frame transmitted by the master is read and written by all the slaves, and the Ethernet frame is returned to the master again to complete the control for one cycle.

  As described above, a control network realized by EtherCat or the like generally realizes high-precision real-time communication or short-term cyclic communication using a fixed-length frame, and the master transmits a frame cyclically. By performing click control, the entire system including the network can be controlled synchronously.

  Non-Patent Document 2 describes an invention of a database that is composed of only keys and values and is mainly developed on a memory. The database is said to realize a high-speed access performance of two digits or more as compared with a database configured on a conventional hard disk or a database using SQL. This access performance may enable data processing within a predetermined time in real time required by the control network.

  Non-Patent Document 3 describes communication using a circuit switching system. The circuit switching system refers to a communication switching system in which a physical or virtual transmission path is set from the start to the end of communication and the transmission path is occupied as a communication line, or a data communication system based on that system. The circuit switching method stores data as in packet communication, and does not require flow rate / retransmission, so that the function of the switching facility is simple. In addition, since the communication line is occupied between the terminal devices, basically, the connection speed and QoS (Quality of Service) are guaranteed, and in principle, fluctuations in transmission time such as transmission delay due to congestion do not occur.

"EtherCAT Master Application Developers Manual Doc. No .: P.4500.21 / Rev. 1.4, Page 22 of 151", [online], [Search July 12, 2013], Internet <URL: http: // www. esd-electronics-usa.com/Shared/Handbooks/EtherCATMasterDevelopersManual.pdf> "Java development 2.0: Redis for the real world", developerWorks, [online], [searched July 12, 2013], Internet <URL: http://www.ibm.com/developerworks/library/j-javadev2 -22 /> "Introduction to WAN Technologies, 3 Circuit Switching", [online], [searched July 12, 2013], Internet <URL: http://docwiki.cisco.com/wiki/Introduction_to_WAN_Technologies#Circuit_switching>

  However, the inventions described in these documents are considered to have the following problems. First, in the invention described in Non-Patent Document 1, there is a problem that a dedicated device for transferring Ethernet frames in real time is required to realize real-time communication in the control system. Also, there is a problem that a control system that requires such real-time communication and an information system that does not require content distribution cannot be connected to the same device. The latter problem is because real-time communication in the control system is impaired by non-real-time communication in the information system.

  Further, the invention of the database described in Non-Patent Document 2 is for realizing high-speed data reading / writing. However, the real-time performance in the control system is not sufficient to simply realize high-speed access exceeding the real-time time. Absent. That is, it is not enough that the possibility of breaking the real-time property is close to 0, and it must be surely 0. In this regard, the technology of Non-Patent Document 2 does not completely guarantee reliable real-time performance.

  Further, in the circuit switching method of Non-Patent Document 3, since the transmission path or the like cannot be shared by a plurality of terminals regardless of the amount of data flowing through the communication line (multicast communication), the use efficiency of network resources is poor, and the speed of the network is different. Communication between terminal devices becomes relatively difficult. In addition, during communication, since the communication line connecting the terminal devices continues to be occupied, dynamic path switching is difficult, and even if switching is possible, it takes time.

  A typical example of the circuit switching system is a subscriber telephone network “Public Switching Telephone Network (PSTN)”. The PSTN uses a protocol called “SS7 (Signaling System No. 7) .The SS7 communicates a switching control command to the line control device, but the accuracy required by the control network The line cannot be switched at a speed of 1. Basically, a switching command is issued to the line control device every time a transmission destination address (telephone number or the like) transmitted by the transmission source is input.

  In view of the above problems, an object of the present invention is to provide a data transfer system or the like that enables multicast communication without impairing real-time performance in control network communication.

  In order to achieve the above and other objects, one aspect of the present invention includes a first communication unit that generates and transmits a transfer frame, and a plurality of data that receives the transfer frame and reads stored data. A second communication unit, a communication network that connects the first communication unit and the second communication unit in a communicable manner, the first communication unit and the second communication unit via the communication network, A data transfer system that is connected to the first communication unit and executes a data transfer between the first communication unit and the second communication unit, wherein the data transfer unit performs the data transfer A data transfer control unit for controlling, a storage unit for storing the transfer frame, a first communication port for executing transmission / reception of the transfer frame between the first communication unit, and at least one of the above Between the second communication department A second communication port for performing transmission / reception of the transfer frame, time-synchronized with the first communication unit, and the storage unit storing the transfer frame received by the first communication port And a second storage area for storing the transfer frame transmitted and received by the second communication port, and the data transfer control unit is configured to perform the first storage area in a predetermined period set in advance. When one communication port receives the transfer frame, the data transfer system stores the transfer frame in the first storage area, transfers the transfer frame to the second storage area, and transmits the transfer frame from the second communication port. It is. Another aspect of the present invention is a method for controlling the data transfer system.

  According to the data transfer system and the like of the present invention, multicast communication is possible without impairing real-time performance in control network communication.

It is a figure which shows the structural example of a general control network. It is a figure which shows the structural example of a general control network. It is a figure which shows an example of the outline | summary of operation | movement of the data transfer system 1 by one Embodiment of this invention. It is a figure which shows the structure of a general control system. It is a figure which shows the structural example of the information transfer system 1 by one Embodiment of this invention. It is a figure which shows the hardware structural example of the schedule server 20 provided in the data transfer system 1. FIG. It is a figure which shows the hardware structural example of the data transfer apparatus 10 provided in the data transfer system 1. FIG. 2 is a diagram illustrating a hardware configuration example of a master 60 provided in the data transfer system 1. FIG. 2 is a diagram illustrating a hardware configuration example of a communication interface 68 provided in the data transfer system 1. FIG. 2 is a diagram illustrating a configuration example of a general Ethernet frame and two types of transfer frames used in the data transfer system 1. FIG. It is an initial information table 1100 of the data transfer system 1. FIG. 2 is a schematic diagram illustrating a software configuration example of a schedule server 20 and its operation. 13 is a flowchart illustrating an example of data processing by the resource allocation module 233 of FIG. It is a figure which shows an example of the whole time table 235 produced | generated by the resource allocation module 233. FIG. FIG. 6 is a configuration diagram of the data transfer device TM_A for creating an individual time table of the data transfer device TM_A of FIG. 5. FIG. 6 is a diagram showing a matrix for creating an individual time table of the data transfer device TM_A of FIG. 5. It is a figure which shows the example of an individual time table structure of each data transfer apparatus TM_A-TM_D of FIG. It is a figure which shows the example of an individual time table structure of each master apparatus Cont_ (alpha) -Cont_ (gamma) of FIG. 2 is a diagram illustrating a software configuration example and an operation example of a microcomputer 19 provided in the data transfer apparatus 10. FIG. 5 is a flowchart showing the operation of a time base data transfer switching module 193. 3 is a diagram illustrating a software configuration example and an operation example of a communication interface 68 provided in a master 60. FIG. It is a figure which illustrates the memory area for an incoming transfer frame or an outgoing transfer frame comprised in the memory 689 of the communication interface 68. 12 is a flowchart illustrating an operation example of a time base frame transmission module 6863. It is a figure which shows the structural example of T-LAN comprised by construction information. It is a figure which shows the example of a structure change of T-LAN # 3 of a 1st Example. FIG. 26 is a diagram of an entire time table 235 reconstructed for the modified example of FIG. 25. FIG. 26 is a diagram illustrating an individual time table 236 of each data transfer device TM_B to TM_D and master device Cont_γ corresponding to FIG. 25. It is a figure which shows the example which applied the data transfer system 1 of this embodiment to the railway operation management system. FIG. 29 is a diagram showing a configuration example of an overall time table 235 in the system of FIG. 28. FIG. 29 is a diagram showing a configuration example of an individual time table 236 in the system of FIG. 28.

  Hereinafter, the present invention will be described in accordance with embodiments thereof with reference to the accompanying drawings as appropriate. First, in order to facilitate understanding of the configuration and operation of the present invention, a configuration example of a conventional control network will be described. 1 and 2 show a schematic configuration example of a conventional general control network. The control network shown in FIG. 1 is an Ethernet-based field bus as described with respect to Non-Patent Document 1. In the configuration example of FIG. 1, a control A network and a control B network for executing different controls are provided at a site S such as a production line. For example, in the control A network, a plurality of slaves 200 are connected to the control master 100 by a communication line 300 in a daisy chain configuration. The control master 100 sends a frame in which control information for each slave 200 is written to the communication line 300. Each slave 200 that has received the frame reads the control information and performs predetermined control, and writes information from a terminal device connected to each slave 200 into the frame and transfers it to the next slave 200. . One frame transmitted by the control master 100 is read and written by all the slaves 200 and returned to the control master 100 again, thereby completing the control for one cycle. In the control B network, the same communication control as described above is executed separately.

  The network shown in FIG. 2 is configured by connecting a plurality of LAN switches 250 to two control masters 100 via communication lines 300. As is well known, the LAN switch 250 is a device that performs packet switching using an ID (MAC address or the like) assigned to a terminal. As a result, the terminals connected to the network can be grouped. In FIG. 2, the terminals controlled by the control master 100 for control A form a virtual group identified by the code of VLAN # 1. doing. The abbreviation for VLAN represents Virtual LAN.

  On the other hand, it will be as follows if an embodiment of the present invention is illustrated. The data transfer system 1 according to the embodiment of the present invention has a time synchronization function and performs data transfer between predetermined ports according to a schedule described in a time table, or data according to a schedule. A master that transmits or receives the data, a schedule server that creates the time table and distributes the time table to the data transfer apparatus or the master, and a slave that transmits and receives the data.

  The schedule server converts information for connecting a predetermined master and slave at a predetermined time into a time table of each data transfer device, and transfers the data to each data transfer device. The data transfer device or the master always synchronizes the time by the time synchronization function, and switches the data transfer path between the ports according to the contents of the time table. Between the master and the slave, time synchronization is always performed by the time synchronization function, and a transfer frame is transmitted or received according to the contents of the time table.

  FIG. 3 shows an example of an outline of the operation of the present invention. For example, in a configuration in which a master 60 that transmits a control frame and a data transfer apparatus 10 are connected to be communicable with each other via a communication line 300, they are time-synchronized so that the master 60 and the data transfer apparatus 10 can A route is configured for each predetermined period. That is, a transfer frame that arrives at a certain port of a data transfer device 10 at a certain time is once copied to a predetermined address in the memory of the data transfer device 10 and then transferred to a predetermined address in the memory for transfer to another port. Will be copied. A transfer frame that arrives outside the predetermined period can be prevented from being copied by the data transfer apparatus 10.

  As a result, in the data transfer device 10, by creating a logical data transfer device with one input and multiple outputs, a network path of an arbitrary shape directly connected by a logical path is constructed, and multicast communication that guarantees real-time performance It is possible to realize data transfer that enables

  In order to realize this, it is assumed that highly accurate time synchronization is possible between the master 60 and the data transfer device 10. Assuming a control system (power system management system, railway operation management system, etc.) that requires a range of 100 km square, it can be realized by, for example, time synchronization using GPS (Global Positioning System). However, as long as the accuracy of time synchronization required as a control system to which the data transfer system 1 according to the embodiment of the present invention is applied, such as NTP (Network Time Protocol) or IEEE 1588 is used, There is no limitation.

  With the configuration as described above, the data transfer system according to the embodiment of the present invention enables a plurality of multicast communications by changing the logical route in the network in units of time as shown in FIG. Real-time data transfer is possible.

  Next, the present invention will be described with reference to the configuration of an example according to the embodiment.

First Embodiment FIG. 4 is a diagram showing a configuration of a data transfer system 1 as a general control system. The master controller 60 (first communication unit; hereinafter referred to as “master 60”) is a controlled object such as a servo drive, I / O, thermo-hygrometer, temperature-humidity adjuster, inverter, stepping motor, or the like. Communication is possible by a slave device 70 (second communication unit; hereinafter referred to as “slave 70”) which is an interface device interposed between the target and a control Ethernet 80 (hereinafter referred to as “control LAN 80”). Cascade connection.

  The master 60 sends a transfer frame loaded with a request for a control command to the device to be controlled, and the slave 70 reads the data of the transfer frame area allocated to each slave unit, and is designated by the control command. The contents are executed, and the execution result and other sensor data are written in the transfer frame. The slave 70 sequentially transfers the transfer frame to the next slave 70 and finally returns to the master 60. In writing and reading each data, a transfer frame is transferred in units of frames, so that data transfer with guaranteed real time can be performed. This point will be specifically described later.

  The schedule server 20 is connected to the master 60 via a communication line such as Ethernet or RS-232C, and the data allocation area to the transfer frame and the transfer frame that the master 60 transfers cyclically to the slave 70. It has a function to set a cycle. The master 60, the slave 70, and the schedule server 20 are configured as a computer having a communication function as described later.

  Next, a configuration example of the control system via the communication network of this embodiment will be described. FIG. 5 is a diagram illustrating a configuration example of the data transfer system 1 configuring the control system via the communication network according to the present embodiment. Among the components, the master 60, the slave 70, and the schedule server 20 are the same as those in FIG.

  The data transfer device 10 (data transfer unit) is a network device that connects another data transfer device 10, a master 60, and a slave 70 so that they can communicate with each other. In the configuration example of FIG. 5, a star-shaped network is configured between the data transfer apparatuses 10, but the present invention is not limited to this network configuration. Further, the communication medium constituting the network is not particularly limited.

  The data transfer device 10 has a plurality of ports connected to the schedule server 20, the master 60, or the control LAN 80, and transfers data between two or more ports at a predetermined time according to its own time table. By performing only this, the ports are logically connected.

  The master 60 has a network interface, and transmits or receives a predetermined transfer frame at a predetermined time set in advance with respect to the data transfer system 1 of FIG.

  The control LAN 80 transfers the transfer frame received from the master 60 to the slave 70. The slave 70 reads out a control command included in the received transfer frame, writes control result information in the transfer frame, and transfers the control result to the slaves 70 sequentially connected via the control LAN 80. The last slave 70 transfers the transfer frame to the master 60 via the control LAN 80.

  In the present embodiment, Ethernet or wide area network (WAN) will be described as an example of a network medium for connecting the control LAN 80, the master 60, and the data transfer apparatus 10, but this is particularly true. It is not limited to network media. In addition, these networks include those having two or more physical or virtual communication lines that enable two-way communication, but in the figure, the description is made with one line in order to avoid complexity. It is simplified.

  In the configuration example of FIG. 5, it is assumed that highly accurate time synchronization is performed between the data transfer apparatus 10 and the master 60. The time synchronization method is not limited as long as the required time synchronization accuracy is ensured, for example, a method using GPS, a method using NTP, a method based on IEEE 1588, or the like. In this embodiment, a time synchronization method using GPS will be described as an example. For the time synchronization method using GPS, “The Role of GPS in Precise Time and Frequency Dissemination” [online] [searched on January 12, 2013] Internet, <URL: http: //ilrs.gsfc. It is described in detail in nasa.gov/docs/timing/gpsrole.pdf>.

  In the following description, in order to identify the plurality of data transfer apparatuses 10 and the master 60 from each other, codes such as TM_A and TM_B are used for the data transfer apparatus 10, and codes such as Cont_α and Cont_β are used for the master 60. It will be attached.

  Next, the schedule server 20 (schedule management unit) of this embodiment will be described. FIG. 6 is a diagram illustrating a hardware configuration example of the schedule server 20. The schedule server 20 is a storage device such as a CPU 22 that is a processor for executing various data operations, a program executed by the CPU 22, and a RAM (Random Access Memory) and a ROM (Read Only Memory) that provide storage areas for various data. A memory 23, an input / output controller 26 for controlling a data input device such as a keyboard 24 and a mouse 25, a monitor controller 27 for controlling a display monitor 271 for displaying various data, an Ethernet network interface card (NIC), and the like. The communication interface 21 is provided. These units are connected to each other via a bus 28 so that they can communicate with each other. The schedule server 20 may be provided with an auxiliary storage device such as a hard disk drive (HDD) or a solid state drive (SSD). The communication interface 21 is connected to the data transfer apparatus 10 via Ethernet, but is not limited thereto, and any communication medium or communication means such as RS-232C may be used.

  Next, the data transfer apparatus 10 will be described. FIG. 7 is a diagram illustrating a hardware configuration example of the data transfer apparatus 10. The data transfer apparatus 10 includes a plurality of NICs 11, a memory 13, a GPS receiver / 1PPS generator 15, a PLL frequency synthesizer 17, and a microcomputer 19 (data transfer control unit; hereinafter referred to as “microcomputer 19”). The NIC 11, the memory 13 (storage unit), and the PLL frequency synthesizer 17 are respectively connected to the microcomputer 19. The NIC 11 is an interface card having a configuration corresponding to the control LAN 80, and is communicably connected to the other data transfer device 10, the slave 70, and the schedule server 20 via the control LAN 80.

  The GPS receiver and 1PPS generator 15 is connected to the GPS antenna ANT and generates a precise 1 PPS (Pulse Per Second) signal based on the GPS signal received from the GPS satellite via the antenna ANT. To the PLL frequency synthesizer 17. The PLL frequency synthesizer 17 generates a clock signal having a clock frequency required by the microcomputer 19 from the 1PPS signal. By using a clock signal synchronized with the 1PPS signal, each data transfer device 10 and the master 60 can be operated in time synchronization with each other.

  The microcomputer 19 is a program stored in the memory 13 having a function of outputting data from other devices received by each connected NIC 11 to the NIC 11 connected to the destination device according to a predetermined time schedule. This is realized using This data transfer function is a main function for ensuring real-time data transfer in the control system of this embodiment.

  Next, the master 60 will be described. FIG. 8 is a diagram illustrating a hardware configuration example of the master 60. Similar to the schedule server 20, the master 60 is a CPU 62 that is a processor for executing various data operations, a program executed by the CPU 62, and a RAM (Random Access Memory) that provides storage areas for various data, and a ROM (Read Only). A memory 63 such as a memory), an input / output controller 65 that controls a data input device such as a keyboard 66 and a mouse 67, a monitor controller 64 that controls a display monitor 641 that displays various data, and an Ethernet network interface card ( NIC) or the like. These units are connected to each other via a bus 69 so as to communicate with each other. The communication interface 68 is connected to the data transfer apparatus 10 via the control LAN 80.

  Next, the communication interface 68 provided in the master 60 will be described. FIG. 9 is a diagram illustrating a hardware configuration example of the communication interface 68 described above. The communication interface 68 includes a NIC 688, a memory 689, a GPS receiver / 1PPS generator 682, a PLL frequency synthesizer 684, and a microcomputer 686 (hereinafter “microcomputer 686”). The NIC 688, the memory 689, and the PLL frequency synthesizer 684 are connected to the microcomputer 686, respectively. The NIC 688 is an interface card having a configuration corresponding to the control LAN 80 and is communicably connected to the data transfer apparatus 10 via the control LAN 80.

  The microcomputer 686 and the NIC 688 have two connections for data transmission to the data transfer apparatus 10 and data reception from the data transfer apparatus 10. The functions and operations of the frequency synthesizer 684 and the GPS signal / 1PPS generator 682 are the same as those of the master 60 described with reference to FIG. In the present embodiment, the communication interface 68 is described as being controlled using the microcomputer 686, but the present invention is not limited to this. For example, even if the function of the communication interface 68 is realized using an FPGA or the like, it does not affect the effects and operations of the present embodiment.

  Next, the structure of the data frame used in the control system in this embodiment will be described. FIG. 10 is a diagram illustrating a general Ethernet frame and two types of transfer frames. The Ethernet frame includes a preamble (PA) 611, a transmission destination address (DA) 612, a transmission source address (SA) 613, a type 614, data 615, and a frame check sequence (FCS) 616. The structure of this general Ethernet frame is described in detail in, for example, “Ethernet Technologies” [online] [searched on January 12, 2013] Internet, <http://docwiki.cisco.com/wiki/Ethernet_Technologies> Therefore, it is not specifically described here.

  In this embodiment, two types of transfer frames indicated by #A and #B in FIG. 10 are used. These frames are stored in the Ethernet frame data 615. The #B frame may be freely used as long as the format is defined between the master 60 and the slave 70, and as many as necessary can be prepared according to the type of control. In the #A transfer frame, the header area 6151 is a header area used for the master 60 to inform the slave 70 of the frame type, and the data areas 6152 and 6153 are on the frame prepared for the slave 70. It is a data area. In these areas, the master 60 or the slave 70 mounts data such as control commands and sensor information in bit units. Regarding this frame configuration, for example, EtherCAT Master Application Developers Manual.Doc: P4500.21 /Rev.1.4 Page20 of 151 [online] [searched on January 12, 2013] Internet, <URL: http: // www. More details can be found at esd-electronics-usa.com/Shared/Handbooks/EtherCATMasterDevelopersManual.pdf>.

  Next, based on the above configuration, the operation of the data transfer system of the present embodiment will be sequentially described. First, initial information of the data transfer system 1 of the present embodiment will be described. FIG. 11 is an initial information table 234 indicating initial information regarding the real-time data transfer system 1 according to the present embodiment, which is input from the schedule server 20. The initial information is information that should also be referred to as a basic specification representing a communication network to be realized in the data transfer system 1, and is designed and created in advance according to specifications required for the data transfer system 1. The initial information table 234 includes items of a T-LAN 2341, a master 2342, a control LAN 2343, a communication period 2344, and a communication cycle 2345. The T-LAN 2341 is a data transfer system whose form changes in units of time in this embodiment, and an entry number of a time-multiplexed control LAN (hereinafter referred to as “T-LAN”) is input. The priority is assumed to be the highest entry number. That is, when the T-LAN 2341 competes with the data transfer system # 1 and # 2, the configuration # 1 has priority.

  In the columns of the master 2342 and the control LAN 2343, a character string for identifying the master 60 and the control LAN 80 controlled by the master 60 is input as a pair. A T-LAN that enables multicast communication at a predetermined time is constructed by the master 60 and the control LAN 80. In the communication period 2344 and the communication period 2345, a period in which the T-LAN is valid and a period in which the period is repeated are input. Explaining T-LAN # 1 as an example, the master 60 (Cont_α) and the control LAN 80 (CLAN_1) constitute the wide-area real-time data transfer system 1 of this embodiment in a period of 5 ms, which is 10 ms. Indicates that it is performed in a cycle. Note that any number of devices and networks may be included in the T-LAN.

  Next, the function of the schedule server 20 will be described. FIG. 12 is a diagram for explaining the software configuration stored in the memory 23 of the schedule server 20 and the operation thereof. First, the initial system information is input to the initial information setting module 232 via the external interface driver (hereinafter referred to as “external I / F driver”) 231 through the keyboard 24 and mouse 25 to the schedule server 20. The The external I / F driver 231 is a functional unit that manages data input / output to the schedule server 20.

  Here, the input initial system information is a switching period (1 ms in this embodiment) of the data transfer apparatus 10 or a repetition maximum value period (20 ms in this embodiment). When a 1518 byte long frame is used in 100 Mbps Ethernet, the switching period can be set to a minimum of about 125 μsec. The repetition period is set depending on the transmission period of the transfer frame of the control system that uses the network of this embodiment. For example, if there is a control system that needs to transmit a transfer frame at a cycle of 1 second, the maximum value needs to be 1000 ms.

  The initial information setting module 232 receives these initial system information from the external I / F driver 231 and stores them in the initial information table 234 illustrated in FIG. 11 as a switching period 2346 and a repetition period 2347, for example. The items of the T-LAN 2341, the master 2342, and the control LAN 2343 recorded in the initial information table 234 are also input via the keyboard 24 and the like.

  Next, in the schedule server 20, the resource allocation module 233 refers to the initial information table 234 and allocates network resources to each network (T-LAN # 1 etc.). The success or failure of resource allocation is notified to the schedule server 20 via the external I / F driver 231 and output to the monitor 271 or the like. When the resource allocation is successful, the resource allocation module 233 creates files for an overall time table 235 (entire configuration information storage unit), an individual time table 236 (individual schedule information), and a distribution time schedule 237 described later. These files are stored in the memory 23.

  When the time table distribution module 238 monitors the time information of the schedule server 20 and determines that the distribution time described in the distribution time schedule 237 has come, the individual time table 236 is transmitted via the NIC I / F driver 239. And distributed to the data transfer apparatus 10 or the master 60. This distribution method is not particularly limited, and may be performed via a network or offline. As described above, the schedule server 20 sets the network resource necessary for the data transfer system used by each control system and the time schedule for configuring each data transfer system.

  Next, resource allocation processing by the resource allocation module 233 described above will be described. FIG. 13 is a flowchart illustrating data processing by the resource allocation module 233. The resource allocation process starts, for example, after the schedule server 20 is activated (S100), and waits for a new resource allocation request to be received through the keyboard 24 or the like during operation (S101, No).

  When the resource allocation module 233 confirms a request for a new T-LAN (S101, Yes), the initial information table 234 is referred to, then the request is input (S102), and an entry is added to the overall time table 235 (S103). ). Next, the resource allocation module 233 determines whether resource allocation is possible for the requested T-LAN (S104).

  In the determination processing in S104, for example, when two or more different T-LANs use the same network resource at the same time, it is determined that resource allocation is impossible. However, when the T-LAN can be set by a different route, or when the resource can be allocated by shifting the time, the resource can be allocated to the competing T-LAN.

  Various algorithms can be used as a method for determining whether or not resource allocation is possible. Here, the Dijkstra method will be described as an example. Considering a network composed of four data transfer devices 10 of TM_A, TM_B, TM_C, and TM_D shown in FIG. 5, it is only necessary to allocate resources so that the network between the data transfer devices 10 is not used at the same time. When applying the Dijkstra method, the initial value of the network cost between the data transfer apparatuses 10 is set to 1. Of course, the use frequency of the network may be changed using a value other than 1. Further, when a network between the data transfer apparatuses 10 is used at a certain time, the cost is infinite (hereinafter referred to as “∞”).

  When the T-LAN # 1 shown in the initial information table 234 of FIG. 11 is configured, it is necessary to calculate a route from TM_A to TM_B. In this case, if a network directly connecting TM_A and TM_B is used, the minimum cost is 1 Therefore, this route is fixed. Next, when configuring T-LAN # 2, the minimum cost for directly connecting TM_A and TM_B is ∞, so this cannot be used. The cost of directly connecting TM_A and TM_C is 1, and the cost of directly connecting TM_A and TM_D is also 1. Since the cost of directly connecting TM_C and TM_B is 1, the cost via TM_A, TM_C, and TM_B is 2 in total. Next, since the cost for directly connecting TM_D and TM_C is 1, the cost via TM_A, TM_D, and TM_C is 2 at this stage, and becomes larger than 1 that is the cost for directly connecting TM_A and TM_C, and thus cannot be adopted. Since the cost via TM_A, TM_D, and TM_B is 2, in this case, either a route that passes through TM_A, TM_C, or TM_B or a route that passes through TM_A, TM_D, or TM_B may be adopted. Here, if the cost is the same, an algorithm that adopts the first path found is adopted. Finally, when configuring T-LAN # 3, the path directly connecting TM_A and TM_B and the path directly connecting TM_A and TM_C are cost ∞ and cannot be adopted. Therefore, since a route directly connecting TM_A and TM_D is established at a cost of 1, this is adopted. Since the above three routes are established at any time, they are fixed as they are. In the case of a modification as shown in FIG. 24 described later, the use of network paths between TM_B and TM_C overlaps, so that resource allocation becomes impossible at the same time. In this case, the resource allocation time is shifted and it is checked whether the allocation is possible (S105). This method will be described later.

  If the allocation is possible in S104 (S104 or S105, Yes), the resource allocation module 233 confirms the entry addition to the entire time table 235 and creates the individual time table 236 (S106). Thereafter, the resource allocation module 233 returns to S101 and waits for a new request. On the other hand, if it is determined that resource allocation is not possible (S105, No), the resource allocation module 233 deletes the entry in the overall time table 235 and indicates that the resource allocation is impossible. The data is output through 20 monitors 271 and the like (S107). If it is determined in S101 that there is no further T-LAN addition request (S101, No), the resource allocation module 233 ends the process (S108). Through the resource allocation process described above, efficient resource allocation can be performed for a plurality of data communication systems (control system) that intend to use the network.

  Next, the entire time table 235 created by the above resource allocation process will be described. FIG. 14 is a diagram illustrating a configuration example of the entire time table 235 in the present embodiment. Based on the initial information of the data transfer system 10 that guarantees the real-time property recorded in the initial information table 234 illustrated in FIG. 11, first, information on T-LAN # 1 is input. Then, the resource allocation module 233 illustrated in FIG. 12 searches for a path that connects the Cont_α that is the master 60 of the T-LAN # 1 and the CLAN_1 that is the LAN to be controlled. In this case, referring to the network configuration in FIG. 5, since the route of Cont_α-TM_A-TM_B-CLAN_1 is available, this route is determined. Next, in consideration of the T-LAN # 1 communication period 1140, communication period 1150, and repetitive maximum value period (20 ms), it is recorded in the entire time table 235 (reference numeral 2351). In the overall time table 235, communication times 000 to 019 are times synchronized between the respective data transfer apparatuses 10 and the respective masters 60. Specifically, for example, “011” is 12:34. It shows the time repeated at intervals of 20 ms, such as 56.011 seconds, same minute 56.031 seconds, and same minute 56.051 seconds.

  Next, when T-LAN # 2 information is input as the initial information, the resource allocation module 233 searches for a path connecting the master 60 Cont_β and the control target LAN CLAN_2. In this case, referring to the network configuration of FIG. 5, this route cannot be used because T-LAN # 1 uses TM_A-TM_B at the same time. Therefore, the resource allocation module 233 determines the path Cont_β-TM_A-TM_C-TM_B-CLAN_1 by the Dijkstra method and records it in the entire time table 235 (reference numeral 2352).

  Next, when information on T-LAN # 3 is input, the resource allocation module 233 similarly searches for a path connecting Cont_γ and CLAN_4. As a result, the resource allocation module 233 determines the path Cont_β-TM_D-CLAN_4 and records it in the entire time table 235 (reference numeral 2353).

  Next, the resource allocation module 233 creates an individual time table 236 for the data transfer apparatus 10 or the master 60 based on the entire time table 235. As described above, T-LANs # 1 to # 3 satisfying the conditions of the communication period 2344 and the communication period 2345 of the initial information table 234 are set in the overall time table 235.

  Next, the individual time table 236 will be described. FIG. 15 is a configuration diagram of the data transfer apparatus 10 connected to TM_A for creating the individual time table 236 for the data transfer apparatus 10 specified by TM_A. In the figure, reference numerals 0 to 5 indicate port numbers of the NIC 11 included in the data transfer device TM_A. FIG. 16 is a diagram showing the configuration diagram of FIG. 15 in a matrix. The resource allocation module 233 reads a description with a description of TM_A from the overall time table 235. In the entire time table 235 of this embodiment illustrated in FIG. 14, # 1 (Cont_α-TM_A-TM_B-CLAN_1) and # 2 (Cont_β-TM_A-TM_C-TM_B) are assigned as T-LANs to which TM_A is allocated. -CLAN_2) and TM_A relays a pair of Cont_α and TM_B and a pair of Cont_β and TM_C. By applying this information to the matrix shown in FIG. 16, the port numbers to be paired in FIG. 15 are clarified. Here, the port pairs “1-2” and “3-5” are extracted.

  FIG. 17 illustrates an individual time table 236 in the present embodiment. In FIG. 17, (for TM_A) represents a pair of port numbers corresponding to the individual time table 236 for TM_A. The expression “1-2” means that TM_A port 1 and port 2 are logically connected between time 000 and 004 and between 010 and 014 in order to realize T-LAN # 1. Show. Similarly, for TM_A, the expression “3-5” indicates that port 3 and port 5 are logically connected between time 000 and 003 in order to realize T-LAN # 2. ing.

  Similarly, (for TM_B) in FIG. 17 is a diagram illustrating port number pairs corresponding to the individual time table 236 for TM_B. The expression “5-1” indicates that port 5 and port 1 are logically connected between time 000 and 004 and between 010 and 014 in order to realize T-LAN # 1. Yes. The expression “3-2” indicates that the port 3 and the port 2 are logically connected between the time 000 and 003 in order to realize the T-LAN # 2.

  FIG. 17 (for TM_C) is a diagram showing port number pairs corresponding to the individual time table 236 for TM_C. The expression “2-3” indicates that port 2 and port 3 are logically connected between time 000 and 003 in order to realize T-LAN # 2.

  FIG. 17 (for TM_D) is a diagram showing port number pairs corresponding to the individual time table 236 for TM_D. The expression “5-4” indicates that port 5 and port 4 are logically connected between time 000, 005, 010, and 015 to 1 ms in order to realize T-LAN # 3. Yes. Each of the data transfer devices 10 of TM_A to TM_D connects the ports as set in the individual time table 236 illustrated in FIG. 17, thereby specifying the data transfer system specified by the T-LANs # 1 to # 3. Can be set to satisfy a desired condition.

  Next, the individual time table 236 related to the master 60 will be described. FIG. 18 is an individual time table 236 indicating periods in which the Cont_α, Cont_β, and Cont_γ as the master 60 illustrated in FIG. 5 are allowed to transmit the transfer frame to the CLAN_1, CLAN_2, and CLAN_3 that are control targets. This is an example of the configuration. For each master 60, an individual time table 236 is created so that ports are opened corresponding to the communication periods set for each data transfer apparatus 10 illustrated in FIG. The individual time table 236 is distributed to each master 60 by the schedule server 20 at a predetermined timing.

  Next, a software configuration for realizing the function of the data transfer apparatus 10 and its operation will be described. FIG. 19 is a diagram showing a software configuration and operation of the microcomputer 19 installed in the data transfer apparatus 10. The functions of the data transfer apparatus 10 in the present embodiment are mainly realized by the individual time table reception module 192 and the time base transfer switching module 193. The individual time table receiving module 192 receives the individual time table 236, the distribution time schedule 237, or the individual time table activated at the timing described in the distribution time schedule 237 from the schedule server 20 via the network or offline. . The time-base transfer switching module 193 executes a process of reading a bit of a predetermined port using a time synchronized with the individual time table 236 and writing it to another port. When the time described in the distribution time schedule 237 arrives, the individual time table 237 is changed to a new individual time table 237, and processing for discarding the old one is executed. The distribution time schedule 237 may be stored in the memory 13 as time information for the microcomputer 19 to periodically update the individual time table 236, for example.

  Next, processing executed by the time base transfer switching module 193 of the data transfer apparatus 10 will be described. FIG. 20 is a flowchart showing the operation of the time base transfer switching module 193. The time-base transfer switching module 193 starts processing when the data transfer apparatus 10 is activated (S1930), and confirms the timer interrupt of the microcomputer 19 activated by the clock synchronized with the GPS signal (S1931). When the timer interruption of the microcomputer 19 is detected, the time base transfer switching module 193 acquires the current time (S1932) and refers to the individual time table 236. The time base transfer switching module 193 determines whether the frame arrives at the port assigned in the individual time table 236, in contrast to the contents of the individual time table 236 (S1935). When it is determined that the port is an assigned port (S1935, Yes), the time-based transfer switching module 193 checks whether one or more transfer frames created from one designated port are completed (S1934). ). When it is determined that one or more transfer frames created from the designated one port are completed (S1934, Yes), the time-based transfer switching module 193 transmits the transfer frame to the memory for the designated other port. Copying is performed (S19341). This is repeated until the next timer interrupt occurs (S1931). On the other hand, if it is determined that the data is not received from the port assigned in the individual time table 236 (S1935, No), the time-based transfer switching module 193 discards the received transfer frame without copying it to the memory. (S1936), the process returns to the timer interruption confirmation step (S1931).

  The memory 13 of the data transfer apparatus 10 has a memory area (first storage area, second storage area) for incoming transfer frames and outgoing transfer frames for each NIC 11. The microcomputer 19 stores the address of each transfer frame memory area in its built-in memory. Thereby, the microcomputer 19 can establish the connection between the ports by copying the transfer frame that arrives between the ports to be connected during the communication period specified by the individual time table 236.

  Next, a software configuration for realizing the function of the communication interface 68 installed in the master 60 and its operation will be described. FIG. 21 is a diagram showing the configuration and operation of software executed by the microcomputer 686 included in the communication interface 68 of the master 60. The function of the communication interface 68 in this embodiment is realized mainly by the individual time table receiving module 6862 and the time base transfer switching module 6863.

  The individual time table receiving module 6862 receives from the schedule server 20 the individual time table 236, the distribution time schedule 237, or the individual time table 236 activated at the timing described in the distribution time schedule 237 via the network. The time base frame sending module 6863 reads the transfer frame stored in a predetermined memory area using the time synchronized with the individual time table 236 and sends it to the NIC 688 as a transfer frame. The NIC 688 is also transferred in units of transfer frames. Send to the network. Also, when the time described in the distribution time schedule 237 arrives, the time base frame transmission module 6863 updates the individual time table 236 to the new individual time table 236 and discards the old one. The distribution time schedule 237 may be stored in the memory 689, for example, as time information for the microcomputer 686 to periodically update the individual time table 236.

  FIG. 22 is a schematic diagram showing a memory area 6891 for incoming transfer frames and outgoing transfer frames configured in the memory 689. Information is transferred in units of transfer frames from the CLAN outgoing transfer frame memory area 68911 at the timing when the CLAN port is opened. Similarly, information is stored in units of transfer frames from the memory area 68912 for incoming transfer frames for CLAN.

  FIG. 23 is a flowchart illustrating the operation of the time base frame transmission module 6863. The time base frame transmission module 6863 starts the processing together with the activation of the communication interface 68 (S68630), and confirms the timer interrupt of the microcomputer 686 activated by the clock synchronized with the GPS signal (S68631). When detecting the timer interruption of the microcomputer 686, the time base frame transmission module 6863 acquires the current time (S68632) and refers to the individual time table 236 (S68633).

  When the current time and the contents of the individual time table 236 are compared and it is determined that the time is assigned by the individual time table 236 (S68634, Yes), the time base frame transmission module 6863 has one or more transfer frames. Is determined to be completed (S68635), and if it is determined to be completed (S68635, Yes), the transfer frame stored in the memory area 68691 for the outgoing transfer frame for CLAN is transferred to the NIC688. Alternatively, the transfer frame transferred from the NIC 688 is stored in the memory area 68692 for the incoming transfer frame for CLAN (S68636). The time base frame transmission module 6863 repeats this process until the next timer interrupt (S68631) occurs. On the other hand, when it is determined that the time is not allocated in the individual time table 236 (S68634, No), the time base frame transmission module 6863 discards the transfer frame (S68637) and returns to the processing of S68631. In this way, in the communication interface 68 of the master 60, a network connection is established with a predetermined data transfer apparatus 10 during a predetermined period set in the individual time table 236.

  FIG. 24 is a diagram illustrating T-LANs # 1 to # 3 configured by construction information (initial information) in the present embodiment. In this embodiment, since the routes of any T-LAN do not overlap, it is not necessary to consider the adjustment of the time when the T-LAN is set. In the present embodiment, the use of GPS is assumed as a time synchronization method, but the present invention is not limited to this, and a network synchronization method such as NTP or IEEE 1588 may be used. Alternatively, time synchronization using the individual time table 236 of the present embodiment may be performed. For example, in the communication time 000 to 1 ms from the schedule server 20, for example, the time synchronization trigger signal is described in advance in the individual time table 236 so that all the data transfer apparatuses 10 and the master 60 can receive the trigger signal. To leave. Each data transfer device 10 and the master 60 can correct the time so as to be synchronized with each other by receiving the trigger.

Second Embodiment Next, a data transfer system 1 according to a second embodiment of the present invention will be described. FIG. 25 is a diagram illustrating the configuration of the data transfer system 1 according to the present embodiment, in which the configuration of T-LAN # 3 of the first embodiment is changed.

  The T-LAN # 3 of this configuration is obtained by adding CLAN5 and CLAN6 as control targets to the T-LAN # 3 in the first embodiment, and is configured by Cont_γ-CLAN4-CLAN5-CLAN6. It is. In this case, since the route between TM_C and TM_B is shared by T-LAN # 2 and T-LAN # 3, both networks cannot share network resources at the same time. Different from the first embodiment. Further, since the transfer frame of T-LAN # 3 reaches from CLAN4 to CLAN6 via CLAN5, CLAN4 and CLAN5 are different from the first embodiment in that two ports are used.

  FIG. 26 shows an overall time table 235 according to this embodiment, which is reconfigured using the algorithm for creating the overall time table 235 in the first embodiment illustrated in FIG. Thus, by shifting the configuration timing of T-LAN # 3 by 4 ms, the problem of sharing the paths of TM_C and TM_B can be avoided. Various algorithms can be used as a method for determining whether assignment is possible. Here, the Dijkstra method described in connection with the first embodiment will be described as an example.

  In this embodiment, in the network composed of the TM_A, TM_B, TM_C, and TM_D data transfer devices 10 shown in FIG. 25, the resources may be allocated so that the network between the data transfer devices 10 is not used at the same time. . When applying the Dijkstra method, the calculation of the network cost between the data transfer apparatuses 10 is as described in the first embodiment.

  First, when configuring the T-LAN # 1 shown in FIG. 26, it is necessary to calculate a route from TM_A to TM_B. In this case, the minimum cost is 1 when a network directly connecting TM_A and TM_B is used. Confirm with this route.

  Next, when configuring T-LAN # 2, the minimum cost for directly connecting TM_A and TM_B is ∞, so this cannot be used. The cost of directly connecting TM_A and TM_C is 1, and the cost of directly connecting TM_A and TM_D is also 1. Since the cost of directly connecting TM_C and TM_B is 1, the cost via TM_A, TM_C, and TM_B is 2 in total. Next, since the cost of directly connecting TM_D and TM_C is 1, when passing through TM_A, TM_D, and TM_C, it becomes 2 at this stage and becomes larger than 1, which is the cost of directly connecting TM_A and TM_C, and cannot be adopted. Since the cost via TM_A, TM_D, and TM_B is 2, in this case, either a route that passes through TM_A, TM_C, or TM_B or a route that passes through TM_A, TM_D, or TM_B may be adopted. Here, if two routes have the same cost, an algorithm that adopts the first route found is adopted.

  Next, consider the configuration of T-LAN # 3. Since it is necessary to connect in order of TM_D, TM_C, and TM_B in T-LAN # 3, TM_D and TM_C are connected and TM_C and TM_B are connected at the same time. In order to connect TM_D and TM_C, since the cost of directly connecting them is 1, it is definitely determined by this route. However, for the route from TM_C to TM_B, the cost of TM_C and TM_B is ∞, the cost of TM_C and TM_A is ∞, and the cost of TM_A and TM_B is also ∞. There is no method for coexisting # 3 with T-LAN # 1 and T-LAN # 2.

  Therefore, a search for a time at which T-LAN # 3 can be established is started while the schedules of T-LAN # 1 and T-LAN # 2 that have already been established are fixed. When the communication time of T-LAN # 3 is shifted by 1 ms and a time when the corresponding T-LAN # 1 and T-LAN # 2 configurations are not set is searched, T-LAN # 2 is set at the time of 004. Can be detected. At this time, for the route from TM_C to TM_B, the costs of TM_C and TM_B are 1, and the costs of TM_C and TM_A are also 1. However, the costs of TM_A and TM_B remain ∞, so TM_C and The cost for directly connecting TM_B becomes 1, and is determined by this route. If T-LAN # 3 is established at time 004, whether or not T-LAN # 3 is established is checked as described above even if it is repeatedly set at the required 5 ms cycle. As a result, when it is determined that T-LAN # 3 is established without competing with T-LAN # 1 and T-LAN # 2, entries in the entire time table 235 are determined. If the network resource contention cannot be resolved even if the T-LAN communication time is shifted as described above, the schedule server 20 is informed that the T-LAN cannot be set. Can be output through the display monitor 271 or the like.

  By setting the entire time table 235 as described above, in this embodiment, the individual time table 236 illustrated in FIG. 27 is created. That is, in the individual time table 236 for the data transfer device TM_B, an entry for connecting the port 3 and the port 6 is added for the T-LAN # 3. Further, as the individual time table 236 for the data transfer device TM_C, entries connecting port 1 and port 4 and port 3 and port 6 are added. In addition, an entry for connecting the port 3 and the port 6 is added to the individual time table 236 for the data transfer device TM_D. Furthermore, Cont_γ, which is the master 60 of T-LAN # 3, is permitted to send a transfer frame at the timing when T-LAN # 3 is configured in the overall time table 235, as illustrated in FIG. (The port is opened) is recorded.

  According to this embodiment, even when a plurality of control systems compete with each other in time and resources cannot be allocated, the configuration timing of the control system having a low priority is shifted with respect to other control systems. As a result, it is possible to configure as many control system systems as possible within the limits of network resources while ensuring real-time performance. As a result, a network as hardware can be shared by a plurality of real-time control systems, and operation costs can be reduced.

Third Embodiment Next, a third embodiment of the present invention will be described. The third embodiment is an example in which the data transfer system 1 of the present invention is applied to a railway operation management system. FIG. 28 is a diagram illustrating a configuration example of a railway operation management system according to the present embodiment. As shown in the figure, T-LAN # 1 including master 60 (Cont_α) is in charge of information transmission for managing the operation of the ring-shaped railway line, and includes T-LAN # 1 including master 60 (Cont_β). LAN # 2 is in charge of information transmission for performing operation management of another railway line sharing a part of the ring-shaped railway line. Here, the master 60 controls the signal 71 and the point 72 installed on each railway line.

  In this embodiment, the network is laid regardless of the route shape of the railway line. However, all the data transfer devices 10 included in the network are one of the data transfer devices 10 from the schedule server 20 and the master 60. The transfer frame can be delivered to the signal 71 and the point 72 via. Since the transfer frame is time-divided in units of routes and is delivered to a predetermined route signal 71 or pointer 72, it is not necessary to install a network for each route, and the route resources of the network can be used effectively. . Further, even when a new route is created, it is not necessary to construct a dedicated network for the route, and a desired operation management network can be configured by setting. The difference between this embodiment and the first and second embodiments is that there is no need to return the transfer frame to the master 60, unlike EtherCat. The transfer frame transmitted from each master 60 is transferred to the signal 71 and the point 72 which are slaves, and ends. Therefore, the transfer frame is duplicated in each data transfer device 10 and transferred, so that a network multicast for each railway line can be realized. In the present embodiment, T-LAN # 1 and T-LAN # 2 are alternately assigned to the network in units of 10 ms.

  FIG. 29 shows a configuration example of the entire time table 235 in the present embodiment. FIG. 29 shows a data transfer system configuration on the network for a period of 20 ms. Here, as described above, it is shown that T-LAN # 1 and T-LAN # 2 are alternately configured every 10 ms. ing.

  FIG. 30 shows a configuration example of the individual time table 236 based on the overall time table 235 of FIG. 29 for the data transfer devices TM_D, TM_G, and TM_A. That is, in FIG. 30, in order to configure the T-LAN # 1 having Cont_α exemplified in FIG. 28 as the master 60, between the ports of TM_D 2-4, 2-5, 2 and 2 −6 is connected, and TM_G is connected between ports 3-1 via TM_E and TM_F. Also, in order to configure T-LAN # 2 with Cont_β illustrated in FIG. 28 as the master 60, TM_D ports 3-5 and 3-6 are connected during the period of time 010 to 019, and TM_E, TM_F The TM_G is connected to the port 3-2, and the TM_A is connected to the port 1-2 via the TM_A.

  According to the embodiment of the present invention, the data transfer device 10 and the master 60 can construct a designated route at a designated time according to the schedule distributed from the schedule server 20. Since each data transfer device 10 and the node transfer the transfer frame in units of control frames that do not perform buffering, it is possible to transfer data with guaranteed real-time properties.

  In addition, according to the present invention, the data transfer device 10 can realize multicast data transfer in which real-time performance is guaranteed by performing one-input multiple-output data transfer.

  Further, according to the present invention, multicast communication having different forms can be realized periodically by switching the path of the transfer frame by the data transfer apparatus 10 at each time.

1 Data Transfer System 10 Data Transfer Device 11 NIC
13 Memory 15 GPS Receiver / 1PPS Generator 17 PLL Frequency Synthesizer 17 19 Microcomputer 20 Schedule Server 21 Network Interface Card 22 CPU
23 Memory 24 Keyboard 25 Mouse 26 Input / output controller 27 Monitor controller 60 Master 61 Network interface card 62 CPU 63 Memory 68 Network interface card 70 Slave 80 Control LAN
192 Individual time table reception module 193 Time base transfer switching module 232 Initial information setting module
234 Initial information table 234 233 Resource allocation module 235 Overall time table 236 Individual time table 237 Distribution time schedule 238 Time table distribution module 239 NIC I / F driver 682 GPS receiver / 1PPS generator 684 PLL frequency synthesizer 684
686 Microcomputer 688 Memory 6862 Individual time table reception module 6863 Time base transfer switching module

Claims (10)

A first communication unit that generates and transmits a transfer frame; a plurality of second communication units that receive and store the transfer frame; and the first communication unit and the second communication. A communication network that connects the first communication unit and the second communication unit via the communication network, the first communication unit and the second communication unit connected to the first communication unit and the second communication unit A data transfer system including a data transfer unit that executes data transfer between
The data transfer unit executes transmission / reception of the transfer frame among a data transfer control unit for controlling the data transfer, a storage unit for storing the transfer frame, and the first communication unit. One communication port and a second communication port for performing transmission and reception of the transfer frame between at least one second communication unit, and time synchronization with the first communication unit is made ,
The storage unit includes a first storage area that stores the transfer frame received by the first communication port, and a second storage area that stores the transfer frame transmitted and received by the second communication port. ,
When the first communication port receives the transfer frame during a predetermined period set in advance, the data transfer control unit stores the transfer frame in the first storage area, and Transferred to a storage area and transmitted from the second communication port;
Data transfer system. The data transfer system according to claim 1, wherein
The data transfer unit further includes a third communication port for performing transmission and reception of the transfer frame with at least one other second communication unit,
The storage unit further includes a third storage area for storing the transfer frame transmitted and received by the third communication port,
When the first communication port receives the transfer frame in a predetermined period, the data transfer control unit stores the transfer frame in the first storage area, and also stores the second storage area and the second storage area. Transferred to three storage areas, and transmitted from the second communication port and the third communication port,
Data transfer system. The data transfer system according to claim 1, wherein
Associating and storing a plurality of network configurations including the first communication unit and the predetermined second communication unit, and a predetermined period required to establish each network configuration The entire configuration information storage unit, and the storage content of the entire configuration information storage unit, in each network configuration, the period in which the first communication unit communicates with the corresponding second communication unit, and the In the data transfer unit, generate individual schedule information including a period during which the transfer frame is transferred between the first communication port and the second communication port, and the first communication unit and the A schedule management unit for delivering to the data transfer unit;
Data transfer system. The data transfer system according to claim 1, wherein
The data transfer control unit discards the data of the transfer frame received by the first communication port in a time zone other than the predetermined period without storing it in the storage unit;
Data transfer system. The data transfer system according to claim 3, wherein
The schedule management unit includes a data transfer path between the data transfer units included in one network configuration and a data transfer path between the data transfer units included in another network configuration in the storage contents of the overall configuration information storage unit , The contention of the data transfer path is resolved by shifting in time the period during which any one of the network configurations is established.
Data transfer system. The data transfer system according to claim 5, wherein
The schedule management unit includes a data transfer path between the data transfer units included in one network configuration and a data transfer path between the data transfer units included in another network configuration in the storage contents of the overall configuration information storage unit Are determined to be competing in the same period, and it is determined that the contention of the data transfer path cannot be resolved by shifting the period during which any of the network configurations is established, Output information indicating that the desired network configuration cannot be established,
Data transfer system. The data transfer system according to claim 3, wherein
The storage content of the overall configuration information storage unit is set for the time corresponding to the longest transfer cycle among the transfer cycles of the transfer frame required by the network configuration.
Data transfer system. A first communication unit that generates and transmits a transfer frame; a plurality of second communication units that receive and store the transfer frame; and the first communication unit and the second communication. A communication network that connects the first communication unit and the second communication unit via the communication network, and the first communication unit and the second communication unit connected to the first communication unit and the second communication unit via the communication network. A data transfer unit that executes data transfer between the data transfer unit, the data transfer unit, a data transfer control unit for controlling the data transfer, a storage unit for storing the transfer frame, and the first transfer unit A first communication port that performs transmission / reception of the transfer frame with a communication unit; and a second communication port that performs transmission / reception of the transfer frame with at least one second communication unit; The storage unit includes A control method of a data transfer system comprising: a first storage area for storing the transfer frame received by one communication port; and a second storage area for storing the transfer frame transmitted and received by the second communication port. There,
Take time synchronization between the first communication unit and the data transfer unit,
When the first communication port receives the transfer frame during a predetermined period set in advance by the data transfer control unit, the transfer frame is stored in the first storage area, and the second Transferred to a storage area and transmitted from the second communication port;
Control method for data transfer system. A method for controlling a data transfer system according to claim 8, comprising:
The data transfer unit further includes a third communication port for performing transmission and reception of the transfer frame with at least one other second communication unit,
The storage unit further includes a third storage area for storing the transfer frame transmitted and received by the third communication port,
When the first communication port receives the transfer frame during a predetermined period by the data transfer control unit, the transfer frame is stored in the first storage area, and the second storage area and the second storage area Transferred to three storage areas, and transmitted from the second communication port and the third communication port,
Control method for data transfer system. A method for controlling a data transfer system according to claim 8, comprising:
The data transfer system includes a plurality of network configurations configured to include the first communication unit and the predetermined second communication unit, and a predetermined number required to establish each network configuration. Communication with the second communication unit corresponding to the first communication unit in each network configuration from the entire configuration information storage unit that stores the period in association with each other and the stored contents of the entire configuration information storage unit And in the data transfer unit, generate individual schedule information including a period for executing transfer of the transfer frame between the first communication port and the second communication port, and A schedule management unit for delivering to the first communication unit and the data transfer unit;
Control method for data transfer system.

Images