Image Processing Reference
In-Depth Information
Tunneling router
Tunneling router
Field device A
Field device B
Fieldbus segment
Backbone
Fieldbus segment
FIGURE .
Tunneling router.
For the backbone, IP (Internet Protocol) has found widespread acceptance. Although IP networks
cannot fulfill the quality of service requirements of more demanding control applications yet since
delay cannot be fully controlled, this does not restrict their use in building automation due to the
moderate requirements present.
As a rule, however, the network layers of existing fieldbus protocols cannot be adapted for IP. his
can be overcome by “tunneling” of data packets. A tunneling router (Figure .) operates entirely
transparently to devices on the fieldbus and allows using any network as a backbone. This method
also is suited for enabling remote access via the Internet.
IT networking infrastructure has become an integral part of modern buildings. The attempt to
leverage this infrastructure for automation purposes is a natural consequence, and the shift towards
IP on the BAS backbone makes it a realistic possibility. However, it may still be advisable to deploy
separate network infrastructure for exclusive use of the BAS to avoid administrative and performance
issues. he benefit of using IP technology remains since inexpensive commercial components can be
used. Moreover, a parallel backbone infrastructure can actually be used as a cold standby for the other
network, respectively.
A project engineer thus finds a variety of applicable protocols at his disposal, as well as var-
ious options regarding network topology and function distribution. He has the choice between
using a common protocol throughout the system and only varying the network layer according to
the expected traffic load, or connecting subsystems that use specialized protocols via gateways. He
could use DDC stations to implement automation functions, or use intelligent field devices instead.
However, these decisions are not entirely independent. For instance, in a mixed protocol environ-
ment, it is advisable to colocate controller functionality in gateway devices, since both functions
require parameterization.
The requirements and conditions of each particular project ultimately determine which of the
different combinations will be ideal. If, for instance, an application domain specific bus is used, its
services and objects need to be converted for use within the associated DDC station or the BMS.
his makes integrated configuration and setup difficult. However, the hardware cost will typically be
lower than for a domain neutral system. The question whether this reduction can offset the extra
configuration effort can only be answered on a project-by-project basis.
However, the clear boundary that is established by a gateway may actually be a desirable property
since it helps in localizing sources of trouble. his is especially welcome if the systems joined by the
gateway are provided by different contractors. Therefore, it is not uncommon to see the use of two
tiered architectures with incompatible protocols in practice.
However, it is not expected to become a cost-efficient solution for field devices in the near future.
 
Search WWH ::




Custom Search