Light Rail Transit System Energy Flow Analysis for the Case of Addis Ababa City: For the Application of Regenerative Energy and Energy Storage

: Significant amounts of energy can be saved by installing energy storage on an electrified transit system allowing energy from braking to be captured. The amount of energy saved is dependent on the amount of energy transferred during braking which relies on the drive cycle and the vehicle parameters. Additional energy is saved through a lower overall energy requirement of train, which results in a reduction in traction supply losses. The overall benefit can be determined by analyzing the energy flow through components in an electrified transit system and hence determining the change in energy dissipations when energy storage is installed. In this paper, electrified transit system energy flows are analyzed for the implementation of energy storage system on board on Addis Ababa light rail transit. The methodology used assesses energy flows in the traction system, establishing where energy is dissipated. The analysis is performed for a specified drive cycle. Finally, the analysis showed that 36 % of traction energy could be saved through the implementation of energy storage on the Addis Ababa light rail transit.


Tractive Resistance
Energy is used to counteract frictional forces in all traction applications. Davis [6] defines the traction resistance as a quadratic function of vehicle speed.
If a vehicle velocity is treated as time, ( ) can be defined as a resistive force in terms of time ( ).

The Driveline
The light rail vehicle transforms electrical energy from the traction supply grid to kinetic energy. The conversion can be divided into several phases; Power electronic converter, electrical conduction, mechanical transmission and traction motor, all of these conversion processes have loss mechanisms as shown in fig. 3.

Train Electrical Conductors
Electric energy is transported by electric conductors from the pantograph to the electronic power converter, which dissipates heat energy.
The cable lengths in the train are usually short and the resistance levels are low, which can be considered as insignificant transmission losses inside the vehicle.

Power Conditioner
In order to achieve the necessary torque, the supply of electrical power is controlled by means of power electronic converters. Many new train use induction motors powered by inverter. Inverters transform a DC supply to the VVVF (three-phase variable voltage frequency) supply. Modern inverter consists of six switches (IGBTs with anti-parallel diodes) as shown in fig. 4. The switches are working to provide the appropriate frequency and voltage for a three-phase AC output. The diode conduction, _ losses ca be described as Where ( ) is the diode junction voltage and ( ) is the diode current, when the diode conducts this equals the line motor current.
( ) can be written as a function of the current as shown below, The values for and can be derived [7] from manufacturer's datasheets [8]. For a Siemens BSM75GB120 these values are = 1.25V and = 7.71 Ω. The IGBT conduction losses can be described as Where, ( ) is the IGBT junction voltage and ( ) is the IGBT forward current, which equals the motor current, when the IGBT is conducting, varies is a function of the current.
Where is 1.83V and is 17.33 Ω [8]. The switching losses, and are a function of the IGBT forward current. Fig. 5 shows the switching losses for a Siemens device, BSM75GB120. The energy loss over one cycle is the sum of the energy losses from each switching operation. The switching losses vary over the sinewave. To simplify the calculation, the RMS current is used to find the switch on and switch off losses from the graph. The switching power, ℎ switching can be described as.
The inverter loss energy is dissipated as heat, and therefore there is a cooling requirement. For the purpose of analyzing energy flows, the gate drive control circuit and cooling energy requirements are considered as auxiliary loads.

The Induction Motor
Induction motor loss mechanisms include ohmic losses, iron losses and frictional and windage losses [9]. The frictional and windage losses can be considered as part of the train's frictional loss. When the coefficients of the Davis equation are determined, the motor frictional and windage losses are included in the tractive resistance analysis. The ohmic losses that occur in the stator and rotor and are dependent on the stator and rotor resistances, and and stator and rotor RMS currents, and .
_ ℎ There are three iron losses, , mechanisms, Hysteresis loss ℎ , eddy current loss, and anomalous loss, . Iron losses can be simplified and related to the magnetization current. An induction motor can be represented as a quasi-steady state equivalent circuit, Fig.6, using resistances to represent the three loss mechanisms, , and , representing the stator copper loss, the rotor copper loss and the iron loss respectively. The equivalent circuit can be evaluated to determine the motor losses.

Mechanical Transmission
The final stage of the drive is the mechanical transmission. The transmission consists of a gear box, usually made of up a driver gear on the motor shaft and a gear on the axle. Losses occur when one gear drives another [11]. The losses can be related to the coefficient of Where ℎ is the mechanical power transmitted, 1 is the angle of approach for the driver gear and 1 is the angle of recess for the driver gear. Frictional losses of the mechanical transmission, including bearing losses are considered as part of the frictional losses of the train. If frictional coefficients are determined using rundown tests, the effects of friction within the drive line are considered.

Auxiliary loads
Auxiliary loads are additional loads which do not produce tractive power. These include ventilation systems, braking systems, doors, control and monitoring systems, driver support facilities, battery charging and passenger comfort functions [12]. Ventilation is crucial to remove excess heat from traction and auxiliary equipment including power converters, motors and control equipment. The ventilation requirements are dependent on the size of the equipment and the heat energy produced, i.e. the losses. Most electric vehicles contain pneumatic systems to operate braking systems, doors and other equipment. A compressor is required to maintain the pneumatic system's pressure; energy is consumed by the compressor [13]. Control and monitoring systems play a vital role in the operation of electric traction vehicles. Control and monitoring systems control traction functions, drives, braking and passenger systems.
Driver support facilities include communication systems (e.g. radios), instrumentation, screen demisters, horns and lighting. Passenger comfort functions include lighting, heating, air conditioning, passenger information systems and CCTV. Auxiliary loads are supplied through auxiliary converters.

Traction Power Supply System
In electrified traction systems, vehicles are powered by electricity which is supplied from a local distribution network through a traction supply system. For the purpose of analyzing the energy flows of an electrified transit system, energy flows in the electricity supply system are considered from the point of connection to the local distribution system. The traction electrical supply system includes the traction substations, the conductor system and a contact system. For DC systems, substations are located every few kilometers along the system (depending on the systems utilization and the voltage level). Substations consist of a transformer and a rectifier. Typically, 12pulse rectifiers are used to reduce harmonic distortion on the local distribution network. Fig.7 shows a traction substation layout.

Fig.7 DC Traction Substation Layout [14]
Transformer losses are divided into two categories, copper losses in the windings and core losses, due to hysteresis and eddy currents losses [15]. These loss mechanisms are similar to those described for induction motors.
Where 1 2 and 2 represent the resistance of the primary winding, and wye and delta windings of the secondary respectively.
The four diodes that are conducting each conduct the rated current. The power loss in the rectifier can be described as, Where is the substation output current. The overall loss of the substation is determined by adding the transformer losses to the rectifier losses and can be described as a quadratic function of the substation current.
The power loss can be related to the substation power, This is integrated to calculate the energy dissipated.

Traction Supply System
This can be generalized for any system and described as the sum of copper losses: Where is the current and is the resistances of sections of the supply system. To consider the energy loss, the power loss is integrated over time. The currents and resistances of each section can be described as functions of time, ( ) and (t) respectively.

Case Study: Addis Ababa light rail transit system
The analysis described in this research paper can be applied to the city of Addis Ababa light rail system to determine the distribution of energy dissipation. Energy consumption of the light rail transit must be analyzed over a complete drive cycle as shown in Fig 9. The drive cycle covers one kilometer.

Fig.8 Speed profile
The Power dissipated overcoming frictional forces is calculated using equation (25). The case study is based on Addis Ababa light rail transit and so the frictional parameters of the light rail can be approximated to those of electric motor train of Addis Ababa.

Mechanical Transmission Losses
Gear box losses can be determined by using equation (26), To determine this, the mechanical power, ℎ is required. The mechanical power is the power required to overcome friction and accelerate the vehicle, it is assumed that the vehicle travels on a level track, and hence no energy is required to climb a gradient.
Friction is calculated in the previous section as shown in section 3.1. The power to accelerate the tram can be calculated from equations of motion.

= * (28)
Where m is the equivalent mass, which is the sum of the vehicle mass and the equivalent mass of the vehicle rotational parts. The mechanical power for the complete drive cycle is shown in Fig.11.

Fig.10 Mechanical power of a single train
The coefficients can be taken from [11] as ′ = 0.0272, 1 =0.3691 rads and 1 = 0.3045 rads. The total energy dissipated through the gears is 35.56 kJ. The energy loss profile is shown in Fig.12.

Fig.11 Power dissipated in the gears
The energy dissipated through the gears is small, and so for the purpose of analysis is added to the frictional losses.

Induction Motor
The induction motor losses can be divided into three categories; ohmic losses, iron losses, frictional and windage losses. Frictional and windage losses are considered as frictional losses of the vehicle. An induction motor can be described as an equivalent circuit as shown in Fig.6. Where Xr' and Rr' are equivalent values.
The power dissipated in ′ (1 − )/ represents the mechanical power generated by the induction motor,  When the minimum and maximum motor supply frequencies are determined, the induction motor can be solved using numerical methods. The bisection method, converges to a solution with a minimal error within a few steps. At each step, equations (32), (33) and (34) are evaluated and the torque is determined. The process is repeated using the bisection rule until the difference between the calculated torque and the required torque, the error, decreases to an acceptable level.
When the supply frequency, and therefore the supply voltage has been determined, the equivalent circuit is solved to find the power dissipated in the stator resistor, rotor resistor and core loss resistor to determine the stator copper loss, the rotor copper loss and the iron core loss respectively. Addis Ababa light rail transit uses four 130 KW motor. Table 2 shows the motor parameters. The total energy dissipated by each induction motor over the simple drive cycle is 300 kJ. This gives 1200 kJ for the two induction motors. Fig.15 shows the total motor loss profile for the vehicle.

Power Electronic Converter
The induction motor is driven by an inverter. An inverter consists of six IGBTs with antiparallel diodes. Inverter losses can be divided into two categories, conduction losses and switching losses. The conduction loss of an inverter is dependent on the motor current and whether the diode or IGBT is conducting, Where is a constant, for the Siemens devices this is calculated to be 3 x10 -4 V. The total loss for the inverter can be described as a function of current.
Taking a switching frequency of 20 kHz and a DC link Voltage of 590 V, the total energy dissipated in each inverter is calculated to be 145 kJ. The power electronics driving the motor has a total energy dissipation of 580 kJ; Fig.16 shows the power dissipation profile for the power electronics on the vehicle.

Auxiliary Load
The auxiliary load for the City Class tram is assumed constant; measurements have indicated that the average auxiliary load is 20 KW.
Over the drive cycle, 2000 kJ of energy is consumed by the auxiliary loads.

Traction Supply System Losses
The traction supply losses can be determined by solving the traction supply network.  [19]. By applying nodal voltage analysis to the system the currents in each section of conductor can be determined. The losses can be determined and summed to find the total transmission loss. Fig.18 shows the power loss dissipation profile for the traction supply system, 124. kJ of energy are dissipated in the supply system conductors.

Fig. 18 Traction Supply System Losses
The analysis of the distribution network is also used to determine the currents drawn from each substation, allowing the substation losses to be calculated. The substation losses can be determined using equation (

Braking Resistor Dissipation
During braking, kinetic energy of the vehicle is transferred through the vehicle driveline, the remaining energy is dissipated through braking resistors. The total energy dissipated during the simple drive cycle is 3900 kJ; Fig.20 shows the braking power profile.  The results show the energy dissipated for a simple a complete drive cycle. Table 3 shows the breakdown of energy consumption over this drive cycle. The breakdown only considers the vehicle losses. Based on the analysis is observed that nearly 40 % of total energy dissipates on braking resistor and this tells us that a great deal energy could be saved if energy storing devices were applied on Addis Ababa light rail system.

Conclusions
This paper has presented a detailed methodology for analyzing energy flows in a traction system for Addis Ababa light rail transit. The analysis is used to determine the energy dissipated in each component of an electrified transit system. Application of the analysis to the Addis Ababa light rail transit system revealed that 36 % of energy used in an electrified traction system is dissipated in braking. This emphasizes the potential of installing energy storage on an electric traction system. The analytical tools described form a good basis to construct a model, which could be used to simulate electrified transit systems. This could be used to determine the potential benefits of energy storage.