A PARAMETRIC ANALYSIS OF AC INTERFERENCE CAUSED BY HIGH VOLTAGE POWER LINES ON NEIGHBORING RAILROAD TRACKS

Transcription

1 A PARAMETRIC ANALYSIS OF AC INTERFERENCE CAUSED BY HIGH VOLTAGE POWER LINES ON NEIGHBORING RAILROAD TRACKS Yexu Li and Farid Paul Dawalibi Safe Engineering Services & technologies ltd Viel, Montreal, PQ, Canada, H3M 1G4 Tel.: (514) Fax: (514) Web: I. ABSTRACT AC interference from high voltage power lines can impair the proper operation of signaling and protection systems of a railway during both normal operation and fault conditions, if the level of the current is too high and if there is a rail unbalance condition. A simplified analysis can lead to either significant unnecessary expense due to overdesign or to damage due to insufficient mitigation measures. This paper describes the mechanisms of electromagnetic interference between power lines and neighboring railroad tracks and presents a simple parametric analysis of a railroad subjected to AC interference from nearby high voltage power lines. Under steady-state conditions, the effects of some critical parameters are discussed. These parameters include: length of parallelism, distance between the power line and the railroad tracks, load current magnitude, soil resistivity and ballast resistivity, as well as the presence of shield wires and the existence of a track unbalance condition. For a single-phase-to-ground fault, three representative soil structure types are used to illustrate the ground potential rise transfer mechanism to the rail located near the energized structure. This paper provides useful information for estimating AC interference levels between power lines and nearby railroads before a detailed study is performed. Furthermore, this paper reveals some intriguing new (or unpublished) results related to inductive interference caused by three-phase power line operating under normal (steady-state) unbalanced load conditions. Keywords: Inductive interference, conductive interference, capacitive interference, railways, ballast, rail-to-ground voltage, rail-to-rail voltage, insulator joint. II. INTRODUCTION Electromagnetic interference caused by electric power lines sharing the same corridor with railways can compromise the safe operation of the signal and protection systems of the railways under both load and fault

2 conditions. Furthermore, it can result in electrical shock hazards for people touching or standing nearby the rail track [3, 5]. Most railroad and power engineers have a complete understanding of the situation whereby a transmission line introduces energy into a railroad system by inductive coupling, capacitive coupling and conduction [1]. Unfortunately, when a professional faces a real problem and must estimate the interference level before deciding to carry out a detailed study, it can be very difficult. Many factors must be considered to evaluate the AC power interference level on a rail track and to develop appropriate mitigation. The body of this paper is divided into two parts. The first part describes different computer models that were built and the large number of computer simulations carried out to determine the influence of some critical parameters that influence AC interference on a railroad from nearby power lines under steady-state conditions. These parameters include: length of parallelism, distance between the power line and the railroad tracks, load current magnitude, soil resistivity and ballast resistivity, as well as the presence of shield wires and the existence of a track unbalance condition. In the second part of this study, the mechanisms and methodology of AC interference under fault conditions are discussed. In particular, different types of soil structures are examined to illustrate how the ground potential rise is transferred as a function of distance to the energized structure. The future research work will include a detailed parametric analysis to determine the influence of other factors, such as neutral (ground, shield or static wire) ground resistances, fault current levels, length of transmission line feeder, etc., under fault conditions. III. STEADY-STATE CONDITIONS 3.1 Mechanisms of Interference Railway tracks and circuits can exhibit significant mutual inductive coupling with parallel power circuits due to the long length of exposure. Also, capacitive coupling has to be considered when the ballast is dry, because the track is poorly grounded. Inductive and capacitive (crosstalk) contributions result in unintended production of voltages at the terminals of the track circuit. This level of interference increases with decreasing separation and angle between the railways and the power lines, with increasing soil resistivity, as well as with increasing current magnitude and frequency of the power line. To limit these voltages, both inductive coupling and capacitive coupling contributions have to be addressed. The impedance of the victim circuits determines which coupling dominates. For low

3 impedance circuits (track circuits), inductive coupling is dominant. For high impedance circuits (communication circuits), capacitive coupling is dominant [7]. 3.2 Methodology Interference levels due to inductive and capacitive coupling must be studied under load and fault conditions. For maximum emergency load conditions, accounting for the maximum expected load unbalance, the rail-to-ground and rail-to-rail voltages are computed throughout the entire length of interest of the railways. The interference level should not exceed 25 V rail-to-ground, 5 V across insulating joints and 5 V rail-to-rail [6]. 3.3 Computer Models The objective of any study of AC interference is to identify the source, the victim and the energy transfer path to facilitate a cost-effective mitigation. Computer simulations and field measurements are two commonly used methods. In this study, a large number of computer simulations are carried out to determine the influence of some critical parameters that influence the inductive and capacitive coupling under steady-state conditions [4, 8]. Figures 1a and 1b show a cross-sectional view of the system modeled in this study. Two typical transmission line configurations are studied, namely, a horizontal configuration and a vertical configuration. A railway runs parallel to the transmission line for a total length of 2 km. A 1.5 m rail-to-rail distance is used for all the computer models. Varying one parameter at a time through a range of values creates a series of simulations that demonstrate the effect of this parameter on the interference level. 3.4 Results and Discussion Induced voltages on the track are proportional to the load current. Also, the induced voltages increase with increasing parallel length. This can be proven with no effort. Figure 2 shows how maximum induced rail-to-ground, maximum induced rail current (the one closer to the transmission line) and maximum induced rail-to-rail voltage per unit length change as a function of the distance between the rail and the center line of the transmission line, for a horizontal configuration. To ensure normal rail system operations and personnel safety, rail-to-ground voltages and rail-to-rail voltages are limited to 25 V and 5 V, respectively, in accordance with ANSI/IEEE Standard 8 safety criteria and the AREMA 21 Blue Book [6]. These limits can be reached under certain conditions in practice. In this case, mitigation is required [2, 9-1]. It is interesting to note that the rail-to-rail voltage reaches a

4 minimum at a separation distance of about 2m, before rising slightly again at larger separations and then vanishing at large distances. Figure 3 reports the same interference quantities as Figure 2, but for a vertical transmission line configuration. One interesting conclusion can be drawn from Figure 3. The static wires increase the unbalance of the phase wires and causes a larger induced emf in the rail track under steady-state conditions for this configuration. However, it should be emphasized that this conclusion applies only for some load conditions. Furthermore, under fault conditions, a static wire will always reduce the AC interference levels on the tracks. It is important to identify if existing railroad signal equipment is compatible with the rail-to-rail voltage resulting from normal rail system unbalances. The rail-to-rail voltage is typically in the range of a few percent of the induced rail-to-ground voltage, for nominally balanced rail conditions, but it can be considerably higher for an unbalanced condition. Unbalanced conditions may include a shorted insulated joint, a broken rail, an earthed lead-in conductor, or a degraded resistance joint. Figures 4-6 illustrate the results of unbalanced condition simulations. Each graph shows the rail-to-ground voltages for a pair of tracks along the railroad. From them, the corresponding rail-to-rail voltages can be determined. As can be seen, unbalanced conditions create higher rail-to-rail voltages, though they produce lower rail-to-ground voltages in some cases. Figure 4 shows the results corresponding to a broken rail unbalanced condition. The two curves represent the railto-ground voltages of the two rail tracks (one is under normal conditions, and the other is broken down at the center of the block), respectively. It clearly indicates that although the rail-to-ground voltage of the broken rail decreases simply because of a shorter parallelism, the rail-to-rail voltage increases. It is therefore logical to conclude that a shorted insulated joint will exhibit an increase of both rail-to-ground and rail-to-rail voltages. Figure 5 shows what happens when one of the tracks has a grounded conductor, while the other track is in its normal condition. The grounded conductor has a 1 ohms ground resistance that is located at the end of the track. One can conclude that this grounded conductor lowers the rail-to-ground potential at the grounded end, at the expense of the other end. Furthermore, it raises the rail-to-rail voltages along the railroad tracks. Figure 6 shows the rail-to-rail voltages caused by a degraded joint. A pair of insulated joints is modeled at the center of each track block. The good insulator joint resistance is 1 ohms, while the degraded one is 85 ohms. The unbalance of this joint-pair is relatively small so that the maximum rail-to-rail voltage (.6 V) may not result in

5 problems under this unbalance condition. However, if the unbalance increases, a high rail-to-rail voltage can occur and may eventually cause interference on the signaling equipment. The following are some results that show how the induction changes with other parameters. The maximum induced rail-to-ground voltage and the maximum induced track current increase with increasing soil resistivity (Table 1) under unbalanced steady-state or fault conditions, but the rail-to-rail voltage changes are small. This table is based on the following system configuration. - Horizontal three-phase transmission line configuration (with 1 ka on one phase and ka on the other two phases), no static wires - Track block length: 2 km (1.24 miles) - Separation distance between the T/L and the rail track: 2 m - Track type: 6 lb - Ballast resistivity: 1, ohm-m The above results also apply for this three-phase horizontal transmission line with a significant, moderate or slight unbalance in load currents. However, a detailed analysis of various load unbalance levels on the phase that is the closest to the rails revealed some unexpected and intriguing results that are new (or are still unpublished). A series of computer models were developed in which the two furthest phases from the rails carry 1 ka load current while the remaining phase load current was varied from.5 to 1 ka, for a uniform soil resistivity ranging from 1 to 1, ohm-m. The computation results show that: o The rail-to-ground voltage decreased rapidly from a maximum of about 194 V (1, ohm-m soil) or 48 V (1 ohm-m soil) at.5 ka to a minimum of about 9.6 V at.97 ka for a 1, ohm-m soil and to below 11.5 V at about.9 to.92 ka for the 1 ohm-m soil. o Remarkably, the interference levels started to increase again from the minimum level reached to about 16 V (1, ohm-m) or about 15.4 V (1 ohm-m) for a 1 ka load current in the phase closest to the rail (i.e., a balanced load condition). Furthermore, under this particular load condition, it was found that significant variations of soil resistivity have practically no impact on the interference levels. This negligible impact is also observed for a load current of.95 ka on the phase closest to the rails (the interference levels along the rails are quite close for all soil resistivity values ranging from 1 to 1, ohm-m) o For all other load currents, the interference levels behave as expected, i.e., they decrease with a decrease of the soil resistivity.

6 These phenomena are attributed to a cancellation effect of the transmission line geometrical unbalance with respect to the rails by the load unbalance introduced by the phase closest to the rail. However, although this fact explains very well the presence of a minimum in the interference levels, it does not provide a satisfactory explanation for the relative insensitivity of the interference levels to soil resistivity values at two distinct load unbalance conditions. Clearly, a more detailed analysis of the phenomena is needed to better understand what is happening exactly in such conditions. Obviously, an unbalance condition on another phase may provide quite different results. This investigation will be carried out in the near future. The induced rail-to-ground voltages increase with increasing ballast resistivity. However, it has a small influence on the rail-to-rail voltage under steady-state conditions (Table 2). The other parameters for this case are: - Horizontal three-phase transmission line configuration with balanced load currents, no neutral wires - Track block length: 2 km (1.24 miles) - The separation distance between the T/L and the rail track: 2 m - Track type: 6 lb - Soil resistivity: 1 ohm-m Various rail tracks were modeled. Table 3 shows that the track type has a very small effect on the level of the induction. This table applies to the same system configuration as given above except that the ballast resistivity is kept constant at 1, ohm-m. IV. FAULT CONDITIONS 3.5 Mechanisms of Interference During a single-phase-to-ground fault on the power line, induced potentials in a rail track can be very high, due to the intense magnetic field caused by the large current that may flow in the faulted phase wire. The levels of induced rail-to-ground and rail-to-rail voltages mainly depend on fault current levels, type of the static wires, ground resistances of the static wires, separation distances between the power line and the rail tracks, rail exposure lengths, as well as soil structure type and soil resistivity values. Under fault conditions, conductive coupling can be a major concern, especially at the locations where railroads are close to the transmission line structures. These structures may inject large amount of currents into the soil during power line fault conditions. Such structures include power line towers and substation and power plant grounding systems. This coupling can result in large rail-to-ground voltages, especially when the ballast is dry. This is because

7 the soil potential rises due to the currents discharged at the structures. Railways located near faulted structures, however, will remain at a relatively low potential, since the dry ballast has a high resistance and offers poor grounding. The magnitude of the conductive coupling is strongly influenced by the soil structure. It decreases with increasing distance away from the faulted structure, but the rate of decrease depends upon the soil structure. 4.1 Methodology Usually (but not always), the worst-case fault location for a given railway section is the power system structure (substation, tower, etc.) closest to that section of the railway. In order to protect a railway system (rail tracks, signaling and relay systems, etc.) adequately, it is necessary to simulate a fault on each tower throughout the railway s length of exposure to the power lines, in order to determine the worst-case interference levels everywhere [4, 8]. For each fault, the following quantities are computed: 1. Induced rail-to-earth and rail-to-rail voltages throughout the railway length of exposure 2. Injected currents at the faulted power structures (used for the conductive interference study) 3. Distribution of the fault currents between the static wires and the faulted structures The envelopes of the maximum values of these quantities are determined and the worst-case locations are found [8]. For the conductive interference, because soil characteristics and currents injected into the faulted structures can vary from one location to another, it is important to model representative towers throughout the right-of-way in order to determine worst-case conductive interference effects on a railway system. The soil structures along the right-of-way have to be determined from soil resistivity measurements. The foundations of the tower and other grounds or metallic structures attached to it are then modeled in the corresponding soil structure. The current calculated in the inductive interference study is injected into the earth at each structure. The earth and rail potentials are then computed as a function of distance from the structure up to the rail tracks. Finally, the total interference level is determined as the vector sum of the inductive, capacitive and conductive coupling components. 4.2 Results and Discussions Figures 7 shows the transferred ground potential rise as a function of distance from the energized structure (a 3 m long rod) buried in various soil structures. Three soil structures were selected:

8 Soil structure 1: Uniform soil: 1 ohm-m. Soil structure 2: Top layer resistivity / Bottom layer resistivity: 1/1 ohm-m. Top layer thickness: 2 m. Soil structure 3: Top layer resistivity / Bottom layer resistivity: 1/1 ohm-m. Top layer thickness: 2 m. A current of 1 ka is injected into the structure. Figure 7 demonstrates that the soil structure has a considerable impact on the potential transferred to a track that is some distance away from the structure. Soil resistivity is a key element to determine the conductive interference level. However, the fault current level, neutral or static wire ground resistances are also critical parameters to determine accurate AC interference levels under fault conditions. A detailed study on this subject will be carried out in the near future. V. CONCLUSION The following important conclusions can be drawn from the preceding discussions: AC interference between power lines and railways can be divided into three key categories: inductive, capacitive and conductive coupling. This interference could jeopardize the proper operation of the signal and protection systems of railways. Furthermore, it can produce unsafe touch and step voltages. Interference levels can be estimated based on the right-of-way network system before a detailed study is performed. Soil resistivity, length of parallelism, distances between the power line and railroad tracks, load current magnitudes, the existence of a track unbalance condition and the presence of static wires, etc., have important effects on the induced emf and on rail-to-rail voltages during normal steady-state conditions. In contrast, ballast resistivity and track type have relatively small impact on the induced interference. During fault conditions, the total electromagnetic interference on a rail track must include conductive coupling. The soil structure is a key element to determine the transferred potentials to the rail track locations. However, fault current levels, the type of static wires and the ground resistances of the towers will have an important impact on the total interference. Several interesting and intriguing results were uncovered under unbalanced load (steady-state) conditions of a horizontal three-phase transmission line. First, a minimum interference level was achieved for a specific unbalance load condition on the phase closest to the rail. Furthermore, the interference levels exhibited an

9 almost complete insensitivity to the soil resistivity variations for two distinct unbalance load conditions on the phase that is the closest to the rails (namely.95 and 1. ka, while the other two phases were at 1. ka). VI. ACKNOWLEDGEMENTS The authors would like to thank Mr. Robert D. Southey, Manager of Applied R&D and Dr. Jinxi Ma, Manager of Analytical R&D at Safe Engineering Services & technologies ltd., for their insightful comments and constructive criticism. VII. REFERENCES [1] F. P. Dawalibi, J. Ma, and Y. Li, Mechanisms of Electromagnetic Interference between Electrical Networks and Neighboring Metallic Utilities, APC, Chicago, April [2] R. D. Southey and F. P. Dawalibi, Computer Modelling of AC Interference Problems for the Most Cost-Effective Solutions, CORROSION 98, Paper No [3] A. H. E. Manders, G. A. Hofkens, and H. Schoenmarkers, Inductive Interference of the Signal and Protection System of the Netherlands Railways by High Voltage Overhead Lines Running Parallel with the Railways, International Conference on Large High Voltage Electric Systems, CIGRE [4] Y. Li, F. P.Dawalibi, J. Ma, and R. D. Southey, Integrated Analysis Software for Electromagnetic Interference between Power Lines and Neighboring Utilities, The International Conference on Electrical Engineering, ICEE 21, Xi an, China, July 22-26, 21. [5] D. C. Carpenter and R. J. Hill, Railroad Track Electrical Impedance and Adjacent Track Crosstalk Modeling Using the Finite- Element Method of Electromagnetic Systems Analysis, IEEE Transactions on Vehicular Technology, Vol. 42, No. 4, November [6] Marvin J. Frazier, Eilis M. Logan and Brian S. Cramer, Blue Book on Inductive Coordination Task Force Progress Report, AREMA 21 Annual Conference & Exposition, Chicago, September 9-13, 21. [7] Brian S. Cramer, AC Power Interference in Railroad Systems, AREMA 21 Annual Conference & Exposition, Chicago, September 9-13, 21. [8] Y. Li, F. P. Dawalibi and R. D. Southey, Automatic Analysis of Electromagnetic Interference between Power Lines and Neighboring Railways, AREMA 21 Annual Conference & Exposition, Chicago, September 9-13, 21.

10 [9] M. J. Frazier, Utility Corridor Design: Transmission Lines, Railroad, and Pipelines, EPRI EL-4147, Volume 1, Project 192-2, Final Report, July [1] A. Taflove and K. R. Umashankar, Mutual Design of Overhead Transmission Lines and Railroad Communication and Signal Systems, EPRI EL-331, Volume 1, Project 192-2, Final Report, October VIII. BIOGRAPHIES Ms. Yexu Li received the B.Sc. degree in Geophysics from Beijing University and the M.Sc. degree in Seismology from the Chinese Academy of Sciences in 1986 and 1989, respectively. She received the M.Sc. degree in Applied Geophysics from Ecole Polytechnique of the University of Montreal in 1996 and the Graduate Diploma in Computer Sciences from Concordia University in From 1995 to 1998, she worked as a Geophysicist with SIAL Geosicences Inc. in Montreal, and was involved in geophysical EM survey design, data acquisition and processing as well as interpretation. She joined Safe Engineering Services & technologies ltd. in Montreal in March 1998 as a scientific researcher and software developer. She is presently working on AC interference studies and software development. Ms. Li has coauthored more than ten papers on electromagnetic interference analysis and geophysics. Dr. Farid P. Dawalibi (M'72, SM'82) was born in Lebanon in November He received a Bachelor of Engineering degree from St. Joseph's University, affiliated with the University of Lyon, and the M.Sc. and Ph.D. degrees from Ecole Polytechnique of the University of Montreal. From 1971 to 1976, he worked as a consulting engineer with the Shawinigan Engineering Company, in Montreal. He worked on numerous projects involving power system analysis and design, railway electrification studies and specialized computer software code development. In 1976, he joined Montel-Sprecher & Schuh, a manufacturer of high voltage equipment in Montreal, as Manager of Technical Services and was involved in power system design, equipment selection and testing for systems ranging from a few to several hundred kv. In 1979, he founded Safe Engineering Services & technologies, a company specializing in soil effects on power networks. Since then he has been responsible for the engineering activities of the company including the development of computer software related to power system applications. He is the author of more than one hundred and sixty papers on power system grounding, lightning, inductive interference and electromagnetic field analysis. He has written several research reports for CEA and EPRI. Dr. Dawalibi is a corresponding member of various IEEE Committee Working Groups, and a senior member of the IEEE Power Engineering Society and the Canadian Society for Electrical Engineering. He is a registered Engineer in the Province of Quebec.

11 Table 1. The maximum induced rail-to-ground, rail-to-rail voltages and track current change as a function of the soil resistivity under unbalanced load conditions. Soil Resistivity Rail-to-Ground Voltage (V) Induced Current Rail-to-Rail Voltage (V) (ohm-m) (the track closer to the T/L) (A) Table 2. The maximum induced rail-to-ground and rail-to-rail voltages change as a function of the ballast resistivity under normal load conditions. Ballast Resistivity. (ohm-m) Rail-to-Ground Voltage (V) (the track closer to the T/L) Induced Current (A) Rail-to-Rail Voltage (V) Table 3. The maximum induced rail-to-ground, rail-to-rail voltages and track current change with different rail track types under normal load conditions. Track Type Rail-to-Ground Voltage (V) Induced Current Rail-to-Rail Voltage (V) (lb) (the track closer to the T/L) (A)

12 Figure 1. A typical cross-section of the system modeled: (a) Horizontal configuration, (b) Vertical configuration Induced EMF (V/kA/kM) Analysis Assumptions: Load Current: 1 ka Track Block Length: 2 km Soil Resistivity: 1 ohm-m Ballast Resistivty: 1, ohm-m No Static Wire Rail-to-Ground Potential Induced Rail Current Rail-to-Rail Voltage Induced Current (A) Separation Distance (m) Figure 2. Induced EMF per unit length of rail track as a function of separation distance from the transmission line: horizontal configuration.

13 Induced EMF V/kA/kM) Analysis Assumptions: Load Current: 1 ka Track Block Length: 2 km Soil Resistivity: 1 ohm-m Ballast Resistivty: 1, ohm-m Rail-to-Ground Potential (No Neutral) Induced Rail Current (No Neutral) Rail-to-Ground Potential (Steel Neutral) Induced Rail Current (Steel Neutral) Induced Current (A) Separation Distance (m) Figure 3. Induced EMF per unit length of rail track as a function of separation distance from the transmission line: vertical configuration. Rail-to-Ground Voltage Magnitude(V) Analysis Assumptions: Load Current: 1 ka Track Separation Distance from the T/L: 1 m Ballast Resistivity: 1, ohm-m Soil Resistivity: 1 ohm-m Track Block Length: 2 km Unbroken Rail Maximum Rail-to-Rail Voltage Broken Rail Distance Along the Rail Track (M) Figure 4. Induced rail-to-ground potentials under an unbalanced condition: broken rail.

14 Rail-to-Ground Voltage Magnitude (V) Analysis Assumptions: Load Current: 1 ka Track Separation Distance from the T/L: 1 m Ballast Resistivity: 1, ohm-m Soil Resistivity: 1 ohm-m Track Block Length: 2 km Grounded Conductor: 1 ohms 7.5 V Ungrounded Rail Track Grounded Rail Track Distance Along the Rail Track (M) Figure 5. Induced rail-to-ground potentials under an unbalanced condition: an earthed lead-in conductor. Rail-to-Ground Voltage Magnitude (V) Analysis Assumptions: Load Current: 1 ka Track Separation Distance from the T/L: 1 m Ballast Resistivity: 1, ohm-m Soil Resistivity: 1 ohm-m Track Block Length: 2 km Good IJ Resistance: 1 ohms Degraded IJ Resistance: 85 ohms.6 V With Degraded IJ With Good IJ Distance Along the Rail Track (M) Figure 6. Induced rail-to-ground potentials under an unbalanced condition: a degraded resistance joint.

15 Soil Structure 1 (Uniform Soil) 15 Soil Structure 2 (Two-layer: 1/1 ohm-m, 2 m) Soil Structure 3 (Two-layer: 1/1 ohm-m, 2 m) Potential Profile Magnitude (kv) Distance from Center of Faulted Structure (m) Figure 7. Transferred ground potential rise as a function of the distance from an energized structure in different soil structures.