WO2006008522A1

WO2006008522A1 - Ip addressing

Info

Publication number: WO2006008522A1
Application number: PCT/GB2005/002844
Authority: WO
Inventors: Aled Hughes; Ray Last
Original assignee: Norgren Limited
Priority date: 2004-07-21
Filing date: 2005-07-21
Publication date: 2006-01-26
Also published as: GB0416243D0

Abstract

An Ethernet addressing means for an industrial network comprising at least one controller means (2) and at least one control device node (3, 4, 5, 6) arranged on the network (7), the addressing means (11) comprising at least one user-operate e switch (8) on the control device node (3, 4, 5, 6), the settings of the or each switch (8) being used by the addressing means (11) to form a node-name, wherein an address assigning means (9) assigns an IP address to the control device node (3, 4, 5, 6), a registration means (10) registers the node-name with the IP address, and the controller means (2) then communicates with the control device node (3, 4, 5, 6) using the node-name, which is resolved into the IP address by the registration means (10). When a control device node (3, 4, 5, 6) is replaced, it can be programmed, using the switches (8), with the node-name of the old device.

Description

ADDRESSING IN INDUSTRIAL CONTROL NETWORK USING NODE NAME RESOLUTION

This invention relates to assigning Internet Protocol (IP) addresses to a device on a network and in particular to assigning IP addresses to devices on an industrial network.

In industry it is common for there to be many systems that perform manufacturing processes or the like and that are provided with instructions remotely by a central controller system. For example, each system may be pneumatically operated and therefore controlled by a set of valves that are mounted onto a valve island. The valve island comprises a plurality of solenoid-actuated control valves that control the pneumatic signals required to operate the various functions of the associated system. The valve island therefore receives electrical control signals from the central controller system to determine which control valves it should actuate to perform the manufacturing process. The central controller system, typically a Programmable Logic Controller (PLC), may control many valve islands.

While in the past the central controller system may have been connected directly to each valve island it controls, it is now more common for the valve islands and the controller system to be arranged on a network, such as a Fieldbus network. In a Fieldbus network each device has a specific address that is used to allow the controller system to communicate with it. The Fieldbus address of each device is usually set by DIP switches on the device itself. Thus, arranging the devices on a network is more flexible, as each device can be controlled by sending addressed control information over the network.

On an Ethernet network each device requires a unique IP address that the controller system uses to communicate with it. There are several methods in common usage for assigning these unique IP addresses, such as Dynamic Host Configuration Protocol (DHCP) , BOOTP or Static addressing (where a user sets the address) . DHCP uses a server to allocate IP addresses from a range, which is determined by the user. Devices using DHCP must lease an IP address for a set period of time, which is user defined, and then renew the lease at the expiry time. This means that IP addresses that are not being used can be allocated to a different device. By using DHCP the user need not know what the IP address is, as operation of the network is transparent to the user, allowing devices to be plugged in and be operational almost instantaneously with no configuration issues.

Conventional Ethernet networks using DHCP operate effectively, as each client (with its automatically DHCP assigned IP address) can send instructions, such as requests for web pages, to web servers having substantially fixed IP addresses (or an associated URL name that is resolved by a DNS server). In industrial networks the controller system is equivalent to the client in that it sends control instructions to the devices, which are therefore equivalent to servers. However, in industrial networks it is the devices or servers that are automatically assigned IP addresses. As can be appreciated, in industrial Ethernet networks it is a more complex problem to ensure that the controller system controls the correct device at the correct IP address.

If the user wishes to replace a device in an industrial network with a new one, the new device could be programmed with the IP address of the replaced device. It is impractical to use DIP switches or the like, as used in Fieldbus networks, to set the IP address due to the number of digits required (four eight-bit numbers) . Further, as each number in the IP address has a particular relevance they cannot be set arbitrarily. Alternatively, the new device could be set using a PC and a crossover cable although this does not simplify the operation of an Ethernet based industrial network.

According to a first aspect of the invention we provide an Ethernet addressing means for an industrial network comprising at least one controller means and at least one control device node arranged on the network, the addressing means comprising at least one user-operable switch on the control device node, the settings of the or each switch being used by the addressing means to form a node-name, wherein an address assigning means assigns an IP address to the control device node, a registration means registers the node-name with the IP address, and the controller means then communicates with the control device node using the node-name, which is resolved into the IP address by the registration means .

As the controller means uses the user programmed node-name to communicate with the control device node, the IP address of the control device node is irrelevant. Thus, when a control device node is replaced, it can be programmed, using the switches, with the node-name of the old device. The new device will then obtain an IP address from the address assigning means and this will be registered against the node-name in the registration means.

Preferably the or each control device node controls the operation of a valve island that comprises a plurality of solenoid operated valves. Preferably the controller means comprises a programmable logic controller (PLC).

Preferably, the addressing means comprises three user operable switches on the control device node. Preferably the or each switch is decimal and thus has ten different positions. The node-name may be the values of the or each switch. Preferably the node-name is formed from the model name of the device having the switch values appended thereto.

The address assigning means may use a DHCP server to assign the IP address. Alternatively, the address assigning means may use Bootstrap Protocol (BOOTP) to assign the IP address. Alternatively, the address assigning means may use a static IP address, set by the user. Alternatively, the address assigning means may use Address Resolution Protocol (ARP) or Reverse Address Resolution Protocol (RARP) to assign the IP address.

Preferably, the address assigning means attempts to obtain the IP address using DHCP, Bootstrap and Static IP methods in turn until an IP address is obtained. It will be appreciated that the address assigning means may attempt to use the above methods in any given order.

Preferably, the registration means is a Domain Name Server (DNS) , which registers the node-name as the Ethernet host name with the IP address from the address assigning means. Alternatively, the registration means may comprise means to broadcast or register the node-name as a NetBIOS network name, thereby associating it with the IP address from the address assigning means.

Preferably, the addressing means ensures that no other device on the network has the same node-name before beginning communication over the network.

According to a second aspect of the invention we provide a method of addressing a control device node on an industrial Ethernet network including at least one control device node having at least one user- operable switch, a controller means, an addressing means, an address assigning means and a registration means, the method using the addressing means to perform the steps of; generating a node-name for the control device node using the values on the or each user-operable switch; obtaining an IP address from the address assigning means; registering the node-name and the IP address on the registration means such that the node-name is used by the controller means to send instructions to the control device node.

When the controller means wishes to instruct a control device node it uses its node-name, which is resolved into its associated IP address by the registration means. Thus, as the user has means to input a node-name, which is then registered, the IP address that is used becomes irrelevant.

Preferably the controller means comprises a programmable logic controller (PLC) .

Preferably, the control device node has three user operable switches. Preferably the or each switch is decimal and thus has ten different positions.

The node-name may be the values of the or each switch. Preferably, the node-name is generated by appending the values of the or each switch to a model name of the control device node. The model name may be factory set.

The address assigning means may use a DHCP server to assign the initial address. Alternatively, the address assigning means may use Bootstrap

Protocol (BOOTP) to assign the initial address. Alternatively, the address assigning means may use a static IP address, set by the user, as the initial address. Preferably, the address assigning means attempts to obtain the initial address using DHCP, Bootstrap and Static IP methods in turn until an IP address is obtained. It will be appreciated that the address assigning means may attempt to use the above methods in any given order.

Preferably, the registration means is a Domain Name Server (DNS), which registers the node-name as the Ethernet host name with the IP address from the address assigning means. Alternatively, the registration means may comprise means to broadcast the node-name as a NetBIOS network name, thereby associating it with the IP address from the address assigning means .

There now follows by way of example only a detailed description of the present invention with reference to the accompanying drawings in which:

Figure 1 shows a diagram of an industrial Ethernet network;

Figure 2 shows a flow chart illustrating the first and second aspects of the invention; and

Figure 3 shows a flow chart illustrating an embodiment of the invention.

An industrial network 1 is illustrated in Figure 1 and comprises a controller means 2 and four control device nodes 3, 4, 5, 6 connected to an Ethernet network represented by a network cable 7. Each of the control device nodes 3, 4, 5, 6 has three user operable switches 8 that form part of an addressing means 11. The switches 8 are decimal rotary switches that allow a three-figure number to be input in the range 0-999. The network 7 also includes an address assigning means 9 and a registration means 10.

Each control device node 3, 4, 5, 6 may be further connected to a valve island (not shown) or the like that controls automated machinery, for example. The controller means 2 comprises a Programmable Logic

Controller (PLC) that communicates instructions to the control device nodes 3, 4, 5, 6 via the network 7. The address assigning means 9 comprises a DHCP server that is in communication with the address means 11 of each control device node 3, 4, 5, 6. Further, the registration means 10 comprises a Domain Name Server (DNS) .

The first and second aspects of the invention are illustrated by a flow chart 12 in Figure 2. The flow chart 12 shows how the addressing means 11 of a control device node 3, 4, 5, 6 obtains an IP address for communicating on the Ethernet network 7 to receive instructions from the controller means 2.

The first stage 13 represents the starting up of one of the nodes 3, 4, 5, 6 when it is first introduced onto the network 7, either when the network is first set-up or as a replacement for a previous node. Prior to the first stage 13, the user should set the switches 8 to either predetermined values in accordance with the control program running on the controller means 2, or to correspond to the node being replaced.

In the second stage 14, the address means 11 generates a node-name using the values of switches 8. The node-name may be the model name of the device having the switch values appended thereto, for example. The node-name allows the control device node 3, 4, 5, 6 to be identified on the network 7. Stage 15 represents the addressing means 11 obtaining an IP address by a user-selected method. In the example network shown in Figure 1, the address means 11 contacts the DHCP server 9 which "leases" the control device node 3, 4, 5, 6 an IP address in accordance with the Dynamic Host Configuration Protocol. It will be appreciated that the address assigning means 9 may alternatively use a static IP address stored in the control device node and thus the address assigning means 9 may form part of the addressing means.

The addressing means 11 then registers the node-name and the IP address obtained from the address assigning means 9 with the registration means 10, which in the example network shown in Figure 1 , is a DNS server. The DNS server therefore provides an association between the user-set node-name (registered as the Ethernet host name) and the DHCP set IP address. Although in this example a DNS server is used to associate the node-name and IP address, any suitable registration method maybe employed depending on the protocols supported by the network. For example, the registration means may comprise means to broadcast the node-name as a NetBIOS network name, which is registered locally at each device on the network, thereby associating it with the IP address.

The controller means 2 is programmed to refer to each control device node 3, 4, 5, 6 by its node-name (Ethernet host name) which is resolved by the DNS server 10 into the IP address of the control device node 3, 4, 5, 6. Thus, the IP address of each control device node 3, 4, 5, 6 is irrelevant as the node-name can be set simply by adjustment of the switches 8, obviating the need for reprogramming of the PLC 2 or programming of the addressing means 11 using an additional computer. Figure 3 illustrates the operation of an embodiment of the addressing means 11 from stage 15 onwards as shown in Figure 2. The addressing means 11 in this example has five modes of operation 20, 21, 22, 23, 24 depending on the value of the switches 8. The invention is illustrated in the first mode 20, while the remaining modes allow the addressing means 11 to be flexible to accommodate the preferences of the end-user.

The first mode of operation 20 is entered by setting the switches 8 to a value between 0 and 254. The addressing means 11 uses two software flags to determine which method the address assigning means 9 will use to assign an IP address to the control device node 3, 4, 5, 6. The software flags may be user set, or may be set during manufacture. Further, the flags may be set so that the addressing means 11 attempts all supported methods to obtain an IP address, which is advantageous as the addressing means therefore requires little user-input and is thus easy to set-up. Thus, the addressing means 11 checks the status of the DHCP software flag 25. If the flag is set to "Yes" the addressing means 11 attempts to contact an address assigning DHCP server to obtain an IP address, which is represented in stage 26. If the DHCP flag is set to "No" , or if the attempt to obtain an IP address using DHCP has failed, the addressing means 11 proceeds to check the status of the BOOTP flag 27. If the BOOTP flag is set to "Yes" the addressing means 11 attempts to contact an address assigning BOOTP server 28 to obtain an IP address. Similarly, if the BOOTP flag is set to "No" , or if the attempt to obtain an IP address using BOOTP has failed, the addressing means 11 uses a Static IP address 29 as the IP address. The Static IP address 29 is pre-programmed into the address means 11 and stored in NVRAM or the like.

Once an IP address has been obtained, regardless of the method used by the address assigning means, it is passed with the node-name to be registered 31 by the registration means 10. The registration means 10 checks 32 that the registration has been successful and if so, communication 33 across the network can begin.

The controller means 2 then uses the node-name (Ethernet host name) , which is resolved into an IP address by the registration means 10, to send instructions to the control device nodes 3, 4, 5, 6. As the switches 8 are used to set the node-name used by the controller means 2, control device nodes can be easily installed and replaced.

However, if the addressing means 11 cannot register the node-name or if the node-name is found to be a duplicate, the addressing means 11 uses MAC communications 35. In this state, the IP address is set to 0.0.0.0 (which is a non-valid address) and the addressing means 11 accepts incoming updates and static address changes from an additional system or the like.

The second mode of operation 21 is entered when the switches are set to a value between 255 and 499. In this mode, the addressing means 11 uses a BOOTP address assigning means 36 to obtain an IP address. The third mode of operation 22 is entered when the switches are set to a value between 500 and 699. In this mode, the addressing means 11 uses a DHCP address assigning means 37 to obtain an IP address. The fourth mode of operation 23 is entered when the switches are set to a value between 700 and 899. In this mode, the address assigning means 9 uses a Static IP address 38. In each of the second to fourth modes of operation 21, 22, 23 the IP address is checked 34 to ensure it is not a duplicate. The controller means 2 can communicate 33 with the control device node 3, 4, 5, 6, otherwise the addressing means enters MAC communication mode 35. The fifth mode of operation 24 is entered when the switches are set to a value between 900 and 999. This mode of operation resets 39 the addressing means 11 by restoring the default factory settings from NVRAM or the like. Once the reset is complete, a message 40 is generated to inform the user that the reset has been successfully completed. The addressing means 11 then remains in this mode until the switches 8 are altered.

Claims

1. An Ethernet addressing means for an industrial network comprising at least one controller means (2) and at least one control device node (3,4,5,6) arranged on the network (7) , the addressing means (11) comprising at least one user-operable switch (8) on the control device node (3,4,5,6), the settings of the or each switch (8) being used by the addressing means (11) to form a node-name, wherein an address assigning means (9) assigns an IP address to the control device node (3,4,5,6) , a registration means (10) registers the node-name with the IP address, and the controller means (2) then communicates with the control device node (3,4,5,6) using the node-name, which is resolved into the IP address by the registration means (10) .

2. An Ethernet addressing means according to claim 1, in which the or each control device node (3,4,5,6) controls the operation of a valve island that comprises a plurality of solenoid operated valves .

3. An Ethernet addressing means according to claim 1 or claim 2, in which the controller means (2) comprises a programmable logic controller

(PLC) .

4. An Ethernet addressing means according to any preceding claim, in which the addressing means (11) comprises three user operable switches (8) on the control device node (3,4,5,6) .

5. An Ethernet addressing means according to any preceding claim, in which the or each switch (8) is decimal and thus has ten different positions.

6. An Ethernet addressing means according to any preceding claim, in which the node-name is the values of the or each switch (8) .

7. An Ethernet addressing means according to any preceding claim, in which the node-name is formed from the model name of the device having the switch values appended thereto.

8. An Ethernet addressing means according to any preceding claim, in which the address assigning means (9) uses a DHCP server to assign the IP address.

9. An Ethernet addressing means according to any of claims 1 to 7, in which the address assigning means (9) uses Bootstrap Protocol (BOOTP) to assign the IP address.

10. An Ethernet addressing means according to any of claims 1 to 7, in which the address assigning means (9) uses a static IP address, set by the user to assign the IP address.

11. An Ethernet addressing means according to any of claims 1 to 7, in which the address assigning means (9) uses Address Resolution Protocol (ARP) .

12. An Ethernet addressing means according to any of claims 1 to 7, in which the address assigning means (9) uses Reverse Address Resolution

Protocol (RARP) to assign the IP address.

13. An Ethernet addressing means according to any of claims 1 to 7, in which the address assigning means (9) attempts to obtain the IP address using DHCP, Bootstrap and Static IP methods in turn until an IP address is obtained.

14. An Ethernet addressing means according to any preceding claim, in which the registration means (10) is a Domain Name Server (DNS) , which registers the node-name as the Ethernet host name with the IP address from the address assigning means.

15. An Ethernet addressing means according to any of claims 1 to 13, in which the registration means (10) comprises means to broadcast or register the node-name as a NetBIOS network name, thereby associating it with the IP address from the address assigning means (9) .

16. An Ethernet addressing means according to any preceding claim, in which the addressing means (11) ensures that no other device on the network has the same node-name before beginning communication over the network (7) .

17. A method of addressing a control device node on an industrial Ethernet network (7) including at least one control device node (3,4,5,6) having at least one user-operable switch (8) , a controller means (2) , an addressing means (11) , an address assigning means (9) and a registration means (10) , the method using the addressing means (11) to perform the steps of; generating a node-name (14) for the control device node (3,4,5,6) using the values on the or each user-operable switch (8) ; obtaining an IP address (15) from the address assigning means (9) ; registering (16) the node-name and the IP address on the registration means (10) such that the node-name is used by the controller means (2) to send instructions to the control device node (3,4,5,6) .

18. A method of addressing a control device node according to claim 17, in which the controller means (2) comprises a programmable logic controller (PLC) .

19. A method of addressing a control device node according to claim 17 or claim 18, in which the control device node (3,4,5,6) has three user operable switches (8) .

20. A method of addressing a control device node according to any of claims 17 to 19, in which the or each switch (8) is decimal and thus has ten different positions.

21. A method of addressing a control device node according to any of claims 17 to 20, in which the node-name is the values of the or each switch (8) .

22. A method of addressing a control device node according to any of claims 17 to 21, in which the node-name is generated by appending the values of the or each switch (8) to a model name of the control device node (3,4,5,6) .

23. A method of addressing a control device node according to claim 22, in which the model name is factory set.

24. A method of addressing a control device node according to any of claims 17 to 23, in which the address assigning means (9) uses a DHCP server to assign the initial address.

25. A method of addressing a control device node according to any of claims 17 to 23, in which the address assigning means (9) uses Bootstrap Protocol (BOOTP) to assign the initial address.

26. A method of addressing a control device node according to any of claims 17 to 23, in which the address assigning means (9) uses a static IP address, set by the user, as the initial address.

27. A method of addressing a control device node according to any of claims 17 to 23, in which the address assigning means (9) attempts to obtain the initial address using DHCP, Bootstrap and Static IP methods in turn until an IP address is obtained.

28. A method of addressing a control device node according to any of claims 17 to 27, in which the registration means (10) is a Domain Name Server (DNS), which registers the node-name as the Ethernet host name with the IP address from the address assigning means (9) .

29. A method of addressing a control device node according to any of claims 17 to 27, in which the registration means (10) comprise means to broadcast the node-name as a NetBIOS network name, thereby associating it with the IP address from the address assigning means (9) .