The expert system for automatic control of train traffic belongs to the class of information systems of automation and tele mechanics. The model, which reflects the informational links between the main goal and the characteristics of the automatic train control system, is informational. It can be used to conduct a qualitative analysis and synthesis of the system.
In order to build an information model of the system, it is necessary, first of all, to formulate the main target tasks of the system and to determine the amount of information that needs to be processed when solving the main target tasks .
The main task of the expert system for automatic control of train movement is automatic control  behind the running train to the train in front. In this case, the necessary throughput must be ensured with complete traffic safety . To do this, in general, you need to have the following information in the system: characteristics of the movement of the ahead running train : actual speed , location , braking performance , length ; characteristics of movement behind the running train B2; actual speed v2, location , braking performance ; characteristics of the path of movement: track profile i, track plan R; own operational and technical characteristics of the Q system: performance and errors.
Some of these characteristics are a priori known, have the nature of parameters for automatic control and can be stored in the system memory even before the start of the automatic traffic control process (breaking characteristics, train lengths, track profiles and plans, system characteristics) [4, 5, 6]. Some of the characteristics are a priori unknown, depend on the movement, have a variable character for automatic control of movement and must be entered into the control device of the expert system during operation (speed and location of trains).
The control device of the automatic train control system processes the information , , , received from the controlled objects (trains B1 and B2), taking into account the information stored in the memory of the system and issues the train B2 an order for the programmed speed v2pr and some additional information that makes it easier for the driver driving train B2 (distance between B1 and B2, actual train speed , profile i and track plan R).
Thus, the information model of the expert system for automatic control of train traffic can be represented in the form shown in Fig. 1.
In this information model, train means any train behind it, and train means any train . in front.
The direction, nature and volume of information flows indicated on the model are necessary and sufficient for the most complex and high-quality of the known cases of automatic regulation of train traffic, therefore this model is generalized. Information models of other cases (systems) of automatic control of train traffic are special cases of the given generalizations of the model and are obtained from the latter by deleting certain connections (information flows) or characteristics (nature or volume of information). Therefore, the analysis performed on the generalized information model is a generalized analysis of the automatic control of train traffic.
Fig.1. Information model of the expert system for automatic control
of train traffic
The generalized model can be reduced to a form that is more convenient for analysis, as well as its concepts can be expanded. To do this, let us focus on the fact that the main task of the automatic control system for train traffic is to automatically control the movement behind the B2 train going on to the B1 train in front, ensuring the required throughput with complete traffic safety. From this it can be seen that the main task is to automatically regulate the movement behind the running train B2. The ahead train B1 acts as a hindrance to the movement behind the train B2, disturbing the movement of the train B2. This interference requires a corresponding change in the characteristics of movement ( , ) behind the running train , when changing the characteristics ( , ) of the train in front.