智能建筑：协议大战带来的和平 关键词： 智能建筑市场，LonWorks,BACnet,TCP/IP,IT,Niagara,Java, XML/Web Service, 能源危机，集成智能建筑。 Some business sectors have managed to get further than the automobile industry. The IT world has the de-facto Wintel (Windows plus Intel) standard in which (assuming you dont have an Apple, Unix or Linux computer) you can swap files between different systems without a problem. This standard came about by pure commercial pressure. Whether or not you agree that a few companies should impose their technology on an industry, there are plenty of examples, which show it to be vital to industry and market expansion. VCRs, electricity supply, telephones and railways are examples of innovations, which did not really begin to expand until one technical standard rose to dominance. Most markets follow the same general path. Initially, the market is fragmented as suppliers see the opportunities in different ways. Engineers design products independently according to their own experience and prejudice. As the market grows and customers require products from different companies to inter-operate, many ad-hoc schemes to link and integrate these systems are developed. Through a process of economic natural selection and customer pressure, the number of technologies is whittled down to a few. Finally, a dominant standard is imposed often by the largest supplying company in the market (as Microsoft has done in PCs) or by a powerful buyer (such as the US Department of Defense did with TCP/IP). It is not obvious at the outset, which technology will be triumphant but invariably one is. From a technical point of view the Intelligent Building market can be seen as having passed the initial stage and is now consolidating. Only five years ago there were plenty of contenders for the standard building control protocol. As well as proprietary offerings, there were many strong contenders such as Profibus, N2, Modbus, FND, and CAB. This list was whittled down until it seemed there were only two protocols with a chance: BACnet and LonWorks. Until recently, the proponents of each seemed to spend as much time attacking the other as promoting themselves. Now peace has broken out largely thanks to some new kids on the block: Niagara and XML(eXtensible Markup Language). At the recent BuilConn convention in Dallas it became apparent that there may be a place for all of these technologies. If one of them finally does become predominant then the process will be gradual but will not hinder the larger movement: the realization of the Intelligent Building. So are these protocols competitive or complementary? Well confusingly, a bit of both (whoever said life would be simple?). What became clear at BuilConn is that each has a role to play and because each is developing quickly, choice, economy and span of control is increasing rapidly for building owners, operators and occupants. LonWorks, invented, developed and promoted by the Echelon corporation can be viewed as building automation for plumbers; the output of one device is fed into the input of another using a common data pipe. This is often the LonBus but it can be power line, wireless or a number of others. So controls bought from one company can be hooked up to the machinery from another fairly easily. LonWorks does not depend on a central computer but it does require the use of devices with the Echelon Neuron chip or its equivalent. LonWorks systems can start small and grow incrementally to powerful multi-site control networks by sending the messages over LANs and WANs. If LonWorks was designed by a plumber then BACnet was designed by a committee of computer geeks. The basis of BACnet is the abstraction of the real world into a number of software objects which when assembled correctly simulate the behaviour of a real-world item like a boiler or a fan. Describing one of these devices in BACnet is a bit like describing a car by its components (take 1 engine, add 4 wheels, connect together with a trans