Intelligent transport systems development. Vadim Shmal

INTEGRATED INTELLIGENT TRANSPORTATION PROCESS MANAGEMENT SYSTEMS

      Currently, the development of various automated control systems in railway transport is increasingly taking place in the direction of their intellectualization. As a rule, intelligent railway systems are created to control individual processes.

      World experience shows that the greatest effect is achieved when developing and implementing an integrated interconnected complex of intelligent systems. In this case, a unified information support is created, the mutual influence of managed processes is taken into account.

      General integration principles

      An illustrative example of the need to create an integrated complex of intelligent systems is the existing network (TCC) and regional (RTCC) automated dispatch control centers. There are dozens of automated workstations (AWSs) in various areas of organization of the transportation process, maintenance and repair of infrastructure and rolling stock devices, as well as security. Each AWS as a human-machine system performs a specific target function. However, a full-fledged interconnection of these functions can be carried out only with the integrated construction of a complex of intelligent dispatch systems. In principle, we can talk about a unified intelligent system in automated dispatch control centers. Let’s consider this provision in relation to regional (road) control centers – RTCC.

      In each RTCC, a hierarchical dispatching structure solves tasks of three main types:

      1) ensuring loading in accordance with the daily and current loading plans;

      2) ensuring the passage of trains (including those performing local work) in accordance with the traffic schedule, the formation plan and the plan for the transfer of wagons along internal and external joints with unconditional compliance with traffic safety;

      3) performing various kinds of special transportation and tasks.

      There are obvious direct relationships in the work of various dispatchers when implementing these types of tasks. Close relationships also occur when solving tasks of various types, so delays in the passage of trains (task type 2) may entail non-fulfillment of tasks for tasks of types 1 and 3. Untimely completion of a special task (task type 3, for example, the promotion of a train with oversized cargo) may cause disruption of the transfer of trains and wagons at the joints (task type 3) and loading plan (task type 1), etc.

      Therefore, synchronous integrated intellectualization of the AWSs of the entire control unit of the RTCC is advisable.

      The main provisions are defined, the implementation of which is a necessary condition for the intellectualization of management processes in regional dispatch centers. These include:

      ■ the use of principles for the development of automated process control systems (TP ACS);

      ■ ensuring efficiency in solving various types of tasks and resolving emerging conflict situations;

      taking into account market conditions in the work of control centers;

      ■ saving all kinds of resources.

      When building management processes, it is necessary that the developed algorithms for solving specific tasks (for RTCC dispatchers these are operational tasks) make it possible to obtain rational, and if possible, optimal solutions. For this condition, it is necessary to have a sufficient amount of information about the processes, take into account the influence of various factors, including disturbing influences, as well as constantly monitor the situation on the basis of special feedback subsystems.

      It is these requirements that are taken into account when building TP ACS as closed control systems with feedback.

      Each dispatcher constantly accumulates experience, which is used when making decisions. Therefore, when developing intelligent systems, it is important to use the principle of their self-learning.

      At the present stage of development of intelligent RTCC systems, the control solutions developed should be used in the «adviser» mode. With the accumulation of experience in the operation of such systems, the refinement of the complex of factors and algorithms taken into account, the transition to the automatic mode of their operation will be carried out.

      The dispatcher’s work proceeds in the constant adoption of operational decisions. The degree of efficiency depends on the needs and capabilities of forecasting specific situations.

      The need for an operational forecast can extend over a very long period. Let’s imagine the situation in a RTCC, the scope of which includes a large seaport, and the cargo comes from loading stations located at distances of several thousand kilometers. Linking the approach of wagons with the approach of ships, especially taking into account weather conditions, requires a forecast of the operational situation for 10—15 days ahead.

      A multi-day forecast is also required to solve the problem of organizing the turnover of locomotives and locomotive crews. At the same time, a forecast for 20—30 minutes may be sufficient for the train dispatcher to solve a specific conflict situation of train traffic on the section.

      Therefore, for each task performed in the RTCC, the developer of an intelligent management system determines the required forecast period and the real possibilities of obtaining it based on relevant information, including those available in existing databases (APOMS-2, etc.).

      In the classical formulation, the well-known problem of the distribution of empty wagons is considered as a transport problem of linear or dynamic programming with cost optimization at a minimum of wagon-kilometers. If the «just in time» condition is met, it is necessary to take into account the additional condition of dynamics in terms of the time of receipt and the time of «consumption» (feeding for loading) of empty wagons. Developers of intelligent systems should take this into account.

      In market conditions, guaranteed delivery of goods is one of the main tasks of JSC «Russian Railways». This is the most important indicator of the quality of the company’s products, its competitiveness.

      It is required to develop an intelligent system with the objective function of minimizing fines paid by JSC «Russian Railways» due to late delivery of goods. The methodology for solving this problem within the framework of the RTCC should be based on the ranking of wagons arriving at the railway (in the region) with varying degrees of delay in relation to the delivery dates, determining regulatory measures to accelerate the promotion of such wagons, taking into account the degree of their delay, developing proactive measures for wagons with possible violations of the delivery time of goods.

      Some of the stated provisions have already found practical implementation.

      The Oktyabrskaya railway experience

      On the Oktyabrskaya mainline, a specialized dispatch center for a high-speed traffic polygon (SRTCC) has been created, in which the general principles and basic provisions of the integrated intelligent transportation process management system are implemented. Particular importance is attached to ensuring traffic safety.

      The approved technical means, including innovative solutions for monitoring the position of mobile objects (trains) based on the GLONASS system, were used in the creation of the SRTCC. As an information base, the systems of AWS train dispatcher, AWS of the energy dispatcher, automated system for analysis of planning and execution of intermissions, Warning systems – v.2 and automated rolling stock control system, monitoring the condition of rolling stock), сomplex automated system of accounting, control of elimination of failures of technical means and analysis of their reliability, «GID -URAL» (traffic schedule) are used) and others. Information exchange is organized with specialized infrastructure systems – paths, automation and telemechanics, etc.

      The system-wide part of the hardware package is implemented within the Oracle Windows Cluster Server DBMS. At the same time, regional – level application complexes are implemented within the framework of the high-speed traffic control center, and the network level – in the form of operational