Trains equipped with automatic train operation (ATO) systems are operated between stations according to the speed commands they receive from balises. These commands define a particular speed profile and running time, with associated energy usage (consumption). The design of speed profiles usually takes into account running times and comfort criteria, but not energy consumption criteria. In this article, a computer-aided procedure for the selection of optimal speed profiles, including energy consumption, which does not have an effect on running times, is presented. To this end, the equations and algorithms that define the train motion and ATO control have been modelled and implemented in a very detailed simulator. This simulator includes four independent modules (ATO, motor, train dynamics, and energy consumption), an automatic generator of every possible profile and a graphical assistant for the selection of speed commands in accordance with decision theory techniques. The results have been compared with measured data in order to adjust and validate the simulator. The implementation of this new procedure in the Madrid underground has led to a 13 per cent of energy saving. As a result, the decision has been taken to redesign all the ATO speed profiles on this underground.

This paper presents experimental results of running resistance tests. Running resistance is determined for conventional passenger trains, freight trains, and the X2 high-speed train. The influence of variables such as speed, number of axles, number of coaches, axle load, track type, and train length is studied. The running resistance is expressed in a general form by a second degree polynomial. The three terms in the polynomial are functions of these variables. The magnitude of the first term is speed independent and varies with number of axles, axle load, and type of track. The second term varies with speed and train length. No influence of axle load is distinguished. The third term is related to the air drag and varies with the speed squared and train configuration. It can be divided into two parts. One part is constant and depends upon the front and rear of the train, and another part increases approximately linearly with train length.

This paper presents a simulation-based model for manual driving strategies that will minimize energy consumption for highspeed trains. Specific characteristics of both high-speed lines (HSLs) and manual driving strategies are considered in order to obtain achievable designs that can be tested on commercial services. The proposed design model calculates a list of efficient high-level commands to be systematically executed by the driver on an HSL along the trip. The design is based on a detailed simulation model of the train's motion (taking into account track and train characteristics and operational constraints), combined with a genetic algorithm to select the best driving. Continuous control solution by mathematical optimization is avoided, as it is not an appropriate reference for manual driving in HSL. The validation of the simulation model is focused on running resistance, tractive/braking efficiencies, and consumption of auxiliary equipment, and shows differences between real measurements and simulated results which are lower than 2% both in run time and energy consumption. Finally, a real case is presented in which the proposed model was used to design efficient driving strategies that were subsequently implemented on commercial services along the Spanish HSL Madrid-Barcelona in both directions, measuring average energy savings of 23 and 18%, respectively, when the efficient driving strategies were compared with measured standard manual driving. The future scope will be the application of this model to online recalculation of driving commands. (Non-member) received the Master's degree in electrical engineering in 1990, the Licentiate of Engineering (tekn. Lic.) degree in 1995, and the Ph.D. (tekn. Dr.) degree in 2001 from the Royal Institute of Technology (KTH), Stockholm, Sweden. Currently, he is a Visiting Researcher with Pontificia Comillas University, Madrid, Spain. His research interests include simulation in train operations, running resistance, and physical modeling and analysis of emissions of pollutants in train operations.

Running resistance of ore trains consisting of Uad-type wagons is determined from full-scale measurements on Malmbanan. Tests are also run in curves with the Uad equipped with three piece bogies where the axles are non-steerable and an ore wagon equipped with bogies allowing the axles to better align themselves on straight track and more radially in curves, thus making them steerable. Influence of speed, axle load, curve radii, and train length is studied and quantified. The running resistance is parameterized and expressed in a general way so that it can be calculated for any Swedish ore train consisting of Uad-type wagons. The study shows that the increase in running resistance is linear due to the increasing axle load on tangent track and train length. The increase in resistance due to curves is significant and increases as the curve radius decreases. If the axles align themselves radially, the curve resistance reduces by 40 per cent, compared with the Uad. The results show which parameters in a running resistance formula should be paid extra attention when constructing a train model for simulation purposes. A comparison is made between ore trains and ordinary Swedish loco-hauled freight trains. The energy consumption of an ore train is not much affected if the operational speed increases from 50 to 60 km/h. Also, a reduced aerodynamic drag has a very little effect on the consumption due to the low operational speed. In this article, a review of the study is made with conclusions.

This article proposes a simple method to determine train running resistance. The resistance is determined by calculating the change in kinetic and potential energy of a coasting train between successive measurement positions. The strength of this method is that the measuring equipment needed is kept at a minimum and it is not limited to a track having a constant grade, thus making this method suitable, in particular, for long freight trains running in mountain areas. An error analysis is performed for this method and the probable error sources are discussed.