IFZONE PROJECT: IMPROVING CIRCULATION IN NEUTRAL SECTIONS. ADIF. Railway Technology Centre. Severo Ochoa, , Malaga, Spain.

Transcription

1 IFZONE PROJECT: IMPROVING CIRCULATION IN NEUTRAL SECTIONS José Conrado Martínez Acevedo 1) ( ), Carlos Tobajas Guerra 2) ( ), Luis Fernando Almenara 1) ( ), Jorge Iglesias Díaz 2), Juan Cabello Álvarez 2) ( Antonio Berrios Villalba 1) 1) ADIF. Railway Technology Centre. Severo Ochoa, , Malaga, Spain. 2) ADIF. Titan , Madrid, Spain. Txomin Nieva ( ), Mikel Aiarzaguena ( TRAINELEC. Industrial Park Katategi - Plot 3 bis , IRURA-Gipuzkoa, Spain. ABSTRACT Electrical phase separation Neutral Sections are installations that are necessary in High-Speed lines as they prevent electrical phase unbalance in the three phase grids that power them. The existence of neutral sections may entail some occasional operating problems and therefore an attempt is made to minimise how many of them there are and improve operation. In this sense, this paper proposes new technical approaches directed at this reduction and presenting and adopting new techniques that improve the operation of the trains when they run through them. All of these approaches are subject to the IFZONE Investigation Project. Although both aspects can be considered general, the paper is based on the experience and experimentation of the Administrador de Infraestructuras Ferroviarias (ADIF - Railway Infrastructure Administrator) in the neutral sections of the Spanish High-Speed network. 1. INTRODUCTION In the electrification of High-Speed railways (therefore, with single-phase alternating current), the power supply to the traction electrical substations is carried out via different electrical phases so that the balance produced on the three-phase transport lines is as low as possible 1. A standard substation with these characteristics is schematised in Figure 1. There are generally not more than two voltage transformers, which may be fed from the equal or different electrical phases. In the first situation, phase rotation would be carried out by substation while in the second it would be carried out by transformers. In any event, and according to the adopted rotation, the sections of catenary powered from each transformer will be in that electrical phase and therefore the transition between sections of a different phase will have to be foreseen in order not to connect them when the train passes. These sections are commonly called Electrical phase separation neutral sections (hereinafter they will only be called neutral sections). As can be deduced, in the case of adopting a rotation by substation, having a neutral section in it is not necessary, but rather only between collateral substations (Figure 1). If rotation is carried out by a transformer, a neutral section in the substation and between collateral substations will be required. It should be pointed out that the voltages that can be applied between the terminals of the primary winding of the transformers are the line-to-line voltages that can be obtained combining the three phase voltages of the transport three-phase grid. On their part, the voltages in the secondary winding are obtained applying the transformer transformation ratio to the primary voltages. As to be expected, the time rates of the secondary voltages do not vary as they are calculated multiplying the primary voltages by a scale. 1 The special characteristics of networks powered at 15 kv/1.67 Hz have not been considered.

2 Figure 1. Single-phase alternating current substation: a) Voltage transformers powered from the electrical phases themselves; b) Voltage transformers powered from four different electrical phases. (Source: ADIF). 2. TECHNICAL CHARACTERISTICS Considering the case of the Spanish High-Speed network, a neutral section is built with two insulated overlaps between which the de-energised catenary is installed (Figure 2). According to the indicated diagram, catenary 1 is powered from the electrical phase 2 (voltage V1) while catenary 2 is powered from phase 2 (voltage 2). The de-energised catenary is installed between the two (catenary 3). The train that is coming along catenary 1 enters into contact with catenary 3 via the first sectioning. Therefore, when the pantograph has lost contact with catenary 1, it only rubs catenary 3 until it reaches the second sectioning, when it starts to come into contact with catenary 3. It should be pointed out that the train enters the neutral section moving by mechanical inertia as its traction switch has been disconnected. It will reconnect when it leaves to be able to be powered from the new stretch again. V1 V2 Figure 2. Mechanical layout of a neutral section of catenary. (Source: ADIF). Logically, on passing from catenary 1 to catenary 3, the pantograph will connect to voltage V1. The same will happen with the V2 voltage on passing on the sectioning of catenary 2. The rest of the time,

3 the voltage of catenary 3 will vary depending on the difference of the electrical phases of the voltages of the ends and on the impedance of the de-energised section. It can be seen that if the time rates of the voltages of the collateral voltage transformers are consecutive (60º phase difference), the electric voltage between the ends of the neutral section is roughly the nominal voltage of the secondary of the transformers. If the rates are not consecutive (phase difference of 120º), the voltage between ends will be roughly 3 times the nominal voltage of the secondary. 3. OPERATING ASPECTS A neutral section may be announced with enough time for the train to be able to carry out the opening operation of the traction switch 2. As seen above, the opening requirement is due to the pantograph energising the de-energised catenary for a few seconds using the end sectionings, an action that may involve the electrical bypassing of both routes if the train has two pantographs that connect catenaries 1 and 2 via catenary 3 when passing over the sectionings. Furthermore, in short neutral sections, it has been seen that a very high speeds the arc extinction might not have occurred when the pantograph is already in the second sectioning. In any event, the greater the length of the deenergised section, the fewer problems will occur although the loss of speed the train suffers will generally be higher. The need for installing neutral sections in high-speed lines may entail the following problems: 1. Breakdowns on the catenary if a train enters with the traction switch closed. In the event of circulating under ATP this type of incidents will not occur. 2. Trains that stop in the section de-energised (for different reasons) and which cannot start up again on their own. Normally a train stopping affects the regularity of the rest of the trains circulating behind it, as some minutes are required for energising the de-energised section and being able to start to run again. 3. Generation of fatigue in the train s traction equipment due to continuous openings and closings of the circuit. It should be pointed out here that the train s lost speed when passing through the neutral section 3 is approximately 9 km/h with a total time passed since traction of roughly 22 s and having run roughly 1600 m. Even so, it has been seen that the existence of neutral sections on the line does not affect the total route time and the total amount of electric energy consumed. Even so, and as to be expected, the ideal situation for operation would be to have as few neutral sections as possible. 4. REDUCTION OF THE AMOUNT OF NEUTRAL SECTIONS As in the current situation it is not possible to get rid of the neutral section as an installation that prevents the unbalance of the network s phases, the objective would be reduce the amount of them. As seen above, they may be situated between collateral substations or in the electric substation itself if the phase rotation is carried out by the voltage transformer. It is important to highlight that the neutral section located in the substation is more critical than the one between collateral substations as in the case of an incorrect connection between adjacent routes, the short circuit produced would be more harmful for the installation. The main objective is therefore to do away this neutral section in substations. 2 Automatic process if circulating under ATP. 3 Data obtained after the tests carried out by ADIF on Madrid-Barcelona High-Speed Line and Madrid- Seville High-Speed Line (a mean circulation speed of 270 km/h is considered).

4 Electrical phase rotation in the substation or in the transformers, and therefore the amount of neutral sections that will finally exist is the outcome of the forecast of the demand of electric power from the supply grid. In this sense, it is easy to see that the unbalance entered in the supply point increases the higher the required power. Therefore, a rotation at transformer level would be associated to a demanding traffic scenario (high power and low frequency) that would possibly affect the balance of the network s phases in the event of only adopting a rotation on the substation level. It should be pointed out that in Spain the requirement imposed by the technical operator of the transport system 4, is for unbalance to not exceed 0.7% for a duration of minutes and 1% for a duration of seconds. If the most restrictive traffic scenario were not to be reached until a period of time has passed, with other prior scenarios existing, the temporary cancelling of the substation s neutral section may be viable from a technical point of view. In this situation, it would be useful to have an analysis to be able to assess the degree of unbalance in the network according to the frequency of trains that circulate on the line. Furthermore, and considering that the unbalance increases the lower the phase difference between adjacent transformers, it would also be useful to have a specific analysis for each of the phase difference configurations. All of these considerations should be born in mind from the moment the installation is designed, as a requirement would be for the three electrical phases of the grid to be laid out in the substation (consult diagram in Figure 1). In this way, a possible reconnection would be simpler from a construction point of view. As to be expected, all of these actions should be coordinated at the right time with the technical operator of the transport network. 5. INTRODUCTION OF A STATIC SWITCHING SYSTEM Currently some administrators 5 including ADIF 6 are starting to try static switching systems to allow the training making the transition between electrical phases directly, without being affected in any way and with the aim of improving the operating capacity of the exploitation. These type of systems use switches on each side of the neutral section, which are represented by semiconductor equipment that allow carrying out a switching in a very short space of time. A very important characteristic of this type of switching is that the exact position of the train should be born in mind. If the neutral section schematised in Figure 2 is considered, the operating principle of the system would be as follows: The train that is coming along catenary 1 is detected by a detection system that informs the switching system of the next entry of the train into the neutral section. At this moment, a close command is established to switch 1 so that the de-energised section is powered at voltage V1. The train does not receive the open command from the traction disconnector at any time. Again, the detection system should detect that the train is totally situated inside the deenergised section. When this condition occurs, an open command is issued on switch 1 and a close command on switch 2 so that the de-energised section is powered at voltage V2. This process is short enough for the train not to detect a lack of voltage in the catenary (which 4 Red Eléctrica de España (REE Spanish Electricity Grid). 5 It should be pointed out that Japanese high-speed lines already have these type of systems in service (for some years), although they are continuing to experiment and optimise the initial development. 6 The experimental system designed by ADIF together with other companies is called SCZN (Sistema de Conmutación de Zona Neutra Neutral Section Switching System) and is subject to a Research Project financed by the Ministry of Science and Innovation (IFZONE Project).

5 would involve opening the disconnector) and there is therefore no loss of traction. The train leaves the neutral section fed by electrical phase 2. In a given point of the exit of the neutral section, the detection system identifies the total passing of the train and starts to normalise the neutral section (opening of switch 2), and is ready for the next train to pass. It should be pointed out that the detection system may be represented by a track circuit specially designed for this function (as is the case of the Japanese system), or by a series of electromagnetic pedals that detect the position of the train wheel (system used by ADIF). Figure 4 represents the schematic form of the shunting sequence of the switching system when the train passes through the neutral section. Figure 4. Shunting sequence of the switching system. (Source: IZONE Project). As indicated above, the switching system should be designed to act in a given time that is determined by the characteristics of the material that circulates on the High-Speed line. Thus, if a detection system based on electromagnetic pedals is used (the case of ADIF), the main characteristic that determines the system s reaction time is the distance between the first wheel of the train and the nearest pantograph that it may carry in service.

6 The length of the de-energised catenary is also a fundamental parameter for analysing the viability of installing this type of systems. Specifically, it has been able to conclude that if this length is lower than established for an interoperable neutral section (402 m), installing switches will not be viable. The reason lies in considering the circulation in double composition of a 200 m train. In this case, if the distance existing between the train s first wheel and the farthest away pantograph from the second composition the train may carry in service (a small auxiliary distance derived from the coupling of the two trains should also be considered), it is concluded that this distance is the one that determines the maximum length of the de-energised section to be used. Therefore, the greater the de-energised section the simpler it will be to install a system with these characteristics 7. Considering the definitive situation of the pedal and the maximum train circulation speed, the time in which the system should react for carrying out the switching can be calculated. It has been concluded that in the case of ADIF the minimum reaction time should be around 0.3 s. 6. INTRODUCTION OF AN INDEPENDENT PROTECTION SYSTEM Another possible improvement for the operating capacity of the existing neutral sections on the High- Speed line is the introduction of a system ADIF has called SPZN (Sistema de Protección de Zona Neutra Neutral Section Protection System). It is an independent system to the neutral section control and is designed to prevent the train from entering the neutral section with the traction disconnector closed (a situation that could occur in the case of circulating without ATP, for example, due to a breakdown of the latter). The operating principle is based on measuring current using a sensor located in an area near to the rubbing area of the two catenaries in the sectioning. In normal operating conditions, circulation of current along this area should not exist, and if it is detected the sensor would emit a trigger command to the substation s switches. Analysing the type of sensor to be used is important to develop an easy to maintain and reliable system. 7. CONCLUSIONS The connection of voltage transformers in the same traction electric-power substation should be carried out between the same electrical phases so that the catenary neutral section in front of the substation can be eliminated. The adjacent transformers belonging to different traction substations are connected with a phase difference of 60º achieving in this way that the voltage existing between the ends of the catenary neutral section are roughly the same as the nominal voltage in the secondary of the transformers. An important improvement in the existing neutral sections would be represented by the introduction of a switching system (SCZN) that allows the train to make a transition between electrical phases without being affected in any whatsoever. Another system to be introduced would be the SPZN (protection system) preventing accidental connection between different electrical phases. 7 Thus, for example, in the Japanese railway network, a neutral section may be 1000 m long.