Building and optimizing an expert system for ATP design

Automatic train protection (ATP) system is a safety critical application for railway signaling with high logic complexity, and it is typically embedded and real-time demanded. As the computational complexity of ATP system, traditional software development method cannot accommodate effectively. In this paper, we propose a novel model-based development (MBD) scheme for ATP system. We argue that high quality software development in ATP system should not only rely on function design, but also on safety verification. A model-based development life cycle and verification approach are described. Furthermore, we present a case study, which is using the tool of SCADE suite. The approach can shorten the development period, enhance system safety, and meet challenges in safety critical software development.

The next few years will see distributed real-time computer systems playing an important role in control systems of high-dependability applications, such as rail transportation. In these applications a failure in the temporal domain can be as critical as a failure in the value domain. In rail transportation, train control system has become more complex and the methods to ensure its correctness of have been considered outmoded. The safety of train control system is becoming increasingly important as computers pervade them on which human life depends and the failure to meet time deadline can have serious or even fatal consequences. This paper proposes a formal modeling and verification approach for real-time system. In the proposed method the real-time system is modeled by timed automata network (TAN) and verified by model checking which explores the state space to determine whether the system satisfies a given specification. The case study of ATP (automatic train protection) shows how the method can assist in designing more efficient and reliable real-time systems. Firstly, automatic train protection (ATP) system and timed automata network (TAN) model is proposed; secondly, the state transitions and multi-tasks ATP onboard model was modeled with TAN model, and then the time sequences of each task are expressed in UML sequence diagrams. Finally, the timing characteristics was verified to meet the requirement by UPPAAL model checker. A major conclusion of the survey is that formal methods, while still immature in some respects, can be used successfully to model and verify real-time systems.

Train Interface Unit (TIU) is the key execution unit of on board Automatic Train Protection (ATP) system. ATP system achieves the control of the train and information exchange with the train via TIU. Developing a TIU which fits for different train platform and is in compliance with the safety integrity level (SIL) 4 can greatly facilitate the system installation and lower operation and maintenance cost. This thesis has designed a TIU for a series of standard train platform. Based on the system design, four key technologies of TIU, include relay and Multifunction Vehicle Bus (MVB) interface, safety communication, emergency brake (EB) interface and service brake (SB) interface, are in-depth in anatomist. Through the thorough analysis and evaluation of the safety and real-time performance, we can get the conclusion that the TIU sub-system meets the requirements of the practical use.

Requirements-based testing is one of the most commonly used ways to ensure the correctness of software, especially for embedded control software in safety-critical domains such as spacecraft and railway systems. Many industrial standards such as the DO-333 and EN50128 also request rigorous requirements-based software testing. To test embedded control software effectively and efficiently, generating high-quality test cases automatically is extremely important. However, existing methods for generating test cases from requirements require intensive manual efforts and expertise. To address this problem, we proposed an automatic requirements-based software testing method for embedded control software. To obtain automatic test case generation and precise test oracles derivation, requirements specification should be precise and readable for the industrial practitioners. Therefore, we use the light-weight domain-specific formal description language, CASDL (Casco Accurate Specification Description Language) for the industrial practitioners to define software requirements into formal specifications at the first step. Based on the formal specification, we propose an algorithm to automatically generate test inputs that satisfy the MC/DC criteria suggested by typical industrial standards and precise test oracles can be derived by "running" the specification with such test inputs. To this end, we proposed an algorithm for simulating the formal specification to generate the test oracles, i.e., the expected outputs corresponding to the test inputs. To facilitate the application of this method in the industry, we have built a tool that can automatically perform the overall testing process. To validate and evaluate its effectiveness in real industrial projects, we have applied it in testing a real Automatic Train Protection (ATP) system provided by our industrial partner, the Casco Signal Co., Ltd (one of the largest railway control system companies in China). In the case study on ATP requirements, our approach generated test cases for 129 requirement items following MC/DC criteria and caught 40 inconsistencies between Casco's requirements and implementation.

Railways are an energy-efficient mode of transportation. Even so, efforts are being made to attain even further energy savings. The latest train-car model of the Shinkansen bullet train, while achieving enriched functionality and improved performance, has attained a 49% energy reduction compared with the 0-series Shinkansen train-cars that debuted at the start of Shinkansen's operation. Also, in the underlying train control system, technologies that will lead to energy reduction have been developed. In order to achieve further energy savings and higher-performance train control, we are developing a new train control system known as Automatic Train Protection and Block System (sometimes referred to ATP-Block system here). This new control system produces a high energy-saving effect compared with conventional signal systems based on ground equipment. We believe that realization of low-cost, highly functional energy-saving Railways will bring back users of automobiles and airlines to Railways, contributing to global energy saving and helping to prevent environmental destruction.

RFM module is widely used for the classification of customer value. The classification results based on big data are often used as portrait of customer loyalty and customer value evaluation. In this paper, RFM model is introduced innovatively to analyse the fault of on board ATP (Automatic Train Protection) system. Through the fault value classification, it can effectively predict the impact of a certain type of fault on operation. For one thing, the operation departments and equipment providers can pre-process the corresponding failures according to the evaluation results, so as to reduce operational losses. For another thing, it can be used as the basis for dealing with big data of ATP faults.

This work focuses on the construction of an expert system shell for airborne equipment design. Traditional expert systems are constructed using a single monolithic software program for a specific application. Since the airborne equipment design is critical & complex, it demands the use of expert system technology. Since present aircrafts take in different equipments for different purposes, it is not feasible to think in terms of independently developed monolithic expert system program for each equipment. To overcome these problems, we have designed and developed a complex competent system i.e., a generic component based expert system for airborne equipment design. The main advantage of the component-based approach is that, the knowledge base is not hard coded and these components can be coupled with any domain-specific knowledge bases. The expert system with its expertise knowledge is made to guide the designer by providing design guidelines & testing procedures for the desired equipment. The expert system also audits the design and provides guidelines to modify & to improve the design.

Expert-system technology has emerged from beneath the umbrella of Artificial Intelligence and into the mainstream of engineering problem solving. There already are several notable examples of the application of this state-of-the-art technology in the petroleum industry. Clearly, expert systems that run efficiently on standard personal computers are particularly of value. A new expert-system generator that overcomes the speed, memory and mass-storage limitations of XT/AT-class personal computers is presented in this paper. The shell is optimized for the specific characteristics of the target user, knowledge expert and computer. Expert systems developed using this shell currently are being used by our technical personnel for drilling fluid and related applications.

In Memoriam: Dr. Clive L. Dym (1942–2016)

This tutorial E-book is based on the long-standing experience of teaching railway operations and signalling at TU Braunschweig and other universities and higher vocational training institutions in different parts of the world. It explains the fundamental principles all railway signalling systems have in common. This is done in a generic way that does not focus on specific national solutions. It covers basic elements and terms, principles for safe train separation, interlocking principles, automatic train protection, and level crossing protection. It also contains a glossary with definitions of relevant terms.

Although the use of expert systems (ESs) as a problem-solving tool is relatively new in the transportation profession, it has been recognized that a well-designed ES can contribute in areas where knowledge domain requires an integration of expertise from various disciplines or is very complex and difficult to quantify. To minimize the effort involved in developing essential ES mechanisms, researchers often prefer to use ES shells which provide a standard inference procedure, knowledge representation, and a user-friendly design environment. However, commercial ES shells tend to be designed for only certain classes of problems, and are not as general and flexible as programming languages. More than 150 commercial ES shells are available in the microcomputer market. Most ES shells are ususally claimed by their vendors to be usable for almost any conceivable applications, but such claims have often proved to be extravagant. Hence, the selection of an efficient yet cost-effective commercial shell is one of the most critical tasks in ES development. In reviewing the the vital task of ES shell selection, this paper presents our experience in evaluating more than 30 ES shells for developing an ES for signal design. The proposed ES includes signal settings, phasing plans, controllers, and detector locations.

A Second-Generation Expert System for Computer-Aided Control System Design

Formalizing a subset of ERTMS/ETCS specifications for verification purposes

A major environmental concern of today’s scientists is the inefficient exploitation of natural resources. The land is the ultimate source of wealth and the foundation on which civilization is constructed. Inappropriate land use, leads to destruction of the land resource, poverty and other social problems, and even to the destruction of civilization. To avoid such phenomena, land evaluation is employed, for rational land use planning and appropriate and sustainable use of natural and human resources (Rossiter, 1994). The management of land use is an interdisciplinary activity that relies on large amounts of information from different sources. Land evaluators need to collect information from soil surveyors, climatologists and census takers on land resource. They also need the expert knowledge of soil scientists, agronomists and economists on land use. In addition, land evaluators must select and apply the most appropriate analytical methods to evaluate land qualities and to combine these into overall physical and/or economic suitability. This evaluation is then calibrated against expert judgement and related experience. Finally they must present the results of the evaluation with reports and maps. This output has to be dynamic, considering the continuous refinement of the whole land evaluation process. The above characteristics of land evaluation denote that the management of such a process definitely requires the support of computer systems, especially expert systems, remote sensing and image processing systems, and geographical information systems (GIS). Such systems exist, but they are usually stand-alone units, hence human intervention (land evaluators) for the flow of information from one system towards the other is indispensable. Therefore, integrated systems are highly desirable. The latest research and development trends in this area progressively encompass Artificial Intelligence (AI) techniques to a greater extent, in order to achieve an optimal performance in the analysis of the vast geographical data. Expert systems were included early on, in an effort to model the domain knowledge of land evaluation from experts. Now, such systems introduce fuzzy logic to cope with uncertainty within the data sources and the inference procedure. Machine learning techniques are also included to model the land evaluation procedures when expert knowledge is insufficient or even absent. In general, there exists an amount of both symbolic and non-symbolic AI techniques, which scientists are keen on combining and integrating with traditional land information systems. This chapter is structured as follows. An overview of three of the most used AI techniques in land evaluation problems is given. Following that, the next section introduces ISLE (Tsoumakas and Vlahavas, 1999), an Intelligent System for Land Evaluation that is designed as a framework for the integration of AI techniques with a geographical information system. The final section discusses conclusions and future trends in this field.

A topology-dependent expert system for DC design of a self-biased single transistor stage was developed and tested. The test case system was implemented in Turbo Prolog and was run on an XT-compatible personal computer. The rule-based system includes four groups of rules: heuristic, general network, topology-dependent network, and default rules. It is shown that the expert system mimics the human design procedure and can handle mathematically ill-defined design problems. A main inherent feature of the proposed expert system is its ability to produce a design solution for any given combination of non-conflicting input specifications. >