Layered structure of DCS communication network - Database & Sql Blog Articles

Brand AVX TPSE226M035R0125 Low impedance tantalum capacitor AVX 22

Electronic scale crystal oscillator 3.2*2.5mm 3225 16M (16.000MHZ) 12PF 10PPM 20PPM 30PPM

SMD aluminum electrolytic capacitor

Spot Remarks 2017+

1. Advantages and Disadvantages of Hierarchical Structure After extensive analysis, based on openness, reliability and real-time transmission, the communication network of each dcs system adopts hierarchical structure without exception, process acquisition data upload, control data downlink Different levels of communication networks can adopt different topologies and different communication protocols according to actual needs. The mutual cooperation of each level constitutes the overall communication system. From the current point of view, each DCS system generally has a workshop-level fieldbus communication network or a remote communication network, a control network between electronic devices, and an operation management network in the control room. As shown in Figure 1, a typical system is configured according to this configuration. ABB's PROCONTROLP, Siemens' TELEPERMXP, Hitachi's HLACS5000, Bailey's Svmphony and Honeywell's TDC3000. To simplify the system hierarchy, some systems combine control networks and operations management networks into one network, such as Westinghouse's OVATION, FOXBORO's I/A, and MCS's MAXl000+. It should be noted that FIG. 1 is only a schematic diagram, and each layer network topology may be any of a ring type, a bus type, and a star type.

In the hierarchical structure of the DCS communication system, the field bus layer is used to connect the field smart meter, the remote bus is used to connect the remote I/O cabinet, and the control network layer is mainly used for communication between the controllers and the process data server. The management network layer is used for communication between the operator station, the engineering station and the process data server. As for the whole plant information management layer (MIS), it is mainly used for production management, optimization scheduling and personnel management of the whole plant. Since the MIS network generally does not adopt redundant configuration, and the relative wiring is complicated and the reliability is relatively low, the power plant automation expert proposes to add a real-time monitoring information network (SIs) between the operation network and the MIS network, which is dedicated to hanging. Connect to the plant-level public system, each unit DCS system and load distribution station, etc., for real-time monitoring and optimization of multi-units. In addition, there are internal communication networks in each automation cabinet. Different systems have different systems, but generally adopt a deterministic master-slave communication network or token communication network, which will not be discussed here.
Considering that most of the data only needs to be transmitted in the underlying network, only a few need to transmit across the network through the upper layer network. Therefore, after the entire communication system is reasonably layered, on the one hand, the network traffic of each layer can be reasonably isolated, which is beneficial to Improve the real-time performance of each layer of communication, on the other hand, it is also conducive to the reliability of the network. Layered can effectively isolate network communication faults at various layers. For example, the fault of the operation management network will not affect the normal operation of the control network. Another advantage of using a layered structure is that DCS vendors can select a reasonable network topology and communication protocol according to their needs, and minimize the cost on the basis of real-time and reliability of the entire communication system, thereby improving the overall system. Competitiveness, such as operating the network layer in Figure 1, effectively increases the openness of the system by adopting the popular standard industrial Ethernet structure and TCP/IP communication protocol. At the same time, because Ethernet is very cheap, the overall cost is effectively reduced. . It should be noted that in this layered system, the reliability of connecting communication devices between different networks becomes a key factor, and high-quality standard gateways are used when possible.
2. Further improve communication efficiency through segmentation Due to the low efficiency of Ethernet under heavy load, special attention should be paid to effectively reducing the communication load of Ethernet, especially when the control layer uses the Ethernet structure. Compared with the control network, although the real-time requirements of the operation layer network are much lower (generally, due to human visual effects, the refresh rate of the operation screen data should not be higher than 1 second, otherwise the operator is prone to fatigue), but Ethernet Due to the inherent shortcomings of transmission uncertainty, it is always beneficial to effectively reduce the communication load. Three methods can be used for this. First, higher speed Ethernet, such as 100Mbps Fast Ethernet; second, switched Ethernet The network replaces the shared Ethernet to avoid the collision domain, thus improving the transmission efficiency. The typical representative is MCS's MAX1000+ system; the third is to use reasonable segmentation technology. All three methods have their own characteristics, which are reflected in the existing DCS. Since the first two methods are familiar to everyone, they will not be described here. The third case is briefly discussed here.
Since the hierarchical network system shown in Figure] generally adopts the configuration of dual server and multi-operator station, the operation management network can be artificially divided into two relatively independent Ethernet networks, and two networks are connected through the bridge. The servers are respectively connected to two Ethernet networks, and the operator stations are equally divided on two networks. For example, at present, a typical unit DCS is equipped with five operator stations, so it can be distributed according to 2 and 3, as shown in Figure 2. Show. In this way, under normal circumstances, each server only serves the corresponding operator station, thereby reducing the network transmission load and server load, and can significantly improve the response speed of the system. If an abnormal situation occurs, such as a server, then the two Ethernets through the bridge become a whole. At this time, the working server serves all the operator stations at the same time. Although the response speed is slightly slower, it does not affect at all. Normal use. This method not only makes full use of two servers, but also can significantly change the system response speed when only one bridge is added, so it is a "low input, high return" solution. According to the author's understanding, currently SYMPHONY of Bailey Company, PROCONTROI of ABB Company. Both P and Hitachi’s HIACS5000+ have adopted this measure.

After adopting the above scheme, the communication efficiency and reliability of the operation management network have been greatly improved. In fact, similar segmentation measures are also used in the control network layer. The most typical ones are the Hitachi HIACS3000/5000 system and the FOXBORO I/ASeries system. The schematic diagram is shown in Figure 3. At this time, the control layer network is divided into systems according to the system. Combustion systems, soda systems and water supply systems. Through the division, the communication load of each control network segment is effectively reduced, the real-time communication is greatly improved, and the reliability is also significantly improved. However, it should be noted that in this way, it is very important to control the division of the network segment and the reasonable distribution of the control cabinet. When possible, placing the control cabinet with frequent data exchange on the same control network segment can greatly improve the communication. real-time. For example, according to the relevant information provided by Hitachi, in the Hitachi HIACS3000/5000 system, the data transmission period in the cabinet is 100 ms, and the data exchange of the control cabinet on the same control network takes 200 ms. If it is necessary to exchange data between the control cabinets on different control networks. It takes 400ms. The potential advantage of this approach is that the physical dispersion of the control cabinet can be better achieved, because the control network segment can be flexibly configured as needed, which is better reflected in FOXBORO's DCS practical application system.