DC voltage control for MMC‐based railway power supply integrated with renewable generation
2020; Institution of Engineering and Technology; Volume: 14; Issue: 18 Linguagem: Inglês
10.1049/iet-rpg.2020.0132
ISSN1752-1424
AutoresPeiliang Sun, Kang Li, Yongfei Li, Li Zhang,
Tópico(s)Thermal Analysis in Power Transmission
ResumoIET Renewable Power GenerationVolume 14, Issue 18 p. 3679-3689 Special Issue: Energy and Rail Transportation Integrated DevelopmentFree Access DC voltage control for MMC-based railway power supply integrated with renewable generation Peiliang Sun, Peiliang Sun orcid.org/0000-0001-8061-224X School of Electronic and Electrical Engineering, University of Leeds, Leeds, LS29 JT UKSearch for more papers by this authorKang Li, Corresponding Author Kang Li K.Li1@leeds.ac.uk School of Electronic and Electrical Engineering, University of Leeds, Leeds, LS29 JT UKSearch for more papers by this authorYongfei Li, Yongfei Li orcid.org/0000-0001-6301-3798 School of Electronic and Electrical Engineering, University of Leeds, Leeds, LS29 JT UK School of Electrical Engineering and Automation, Harbin Institute of Technology, Harbin, People's Republic of ChinaSearch for more papers by this authorLi Zhang, Li Zhang School of Electronic and Electrical Engineering, University of Leeds, Leeds, LS29 JT UKSearch for more papers by this author Peiliang Sun, Peiliang Sun orcid.org/0000-0001-8061-224X School of Electronic and Electrical Engineering, University of Leeds, Leeds, LS29 JT UKSearch for more papers by this authorKang Li, Corresponding Author Kang Li K.Li1@leeds.ac.uk School of Electronic and Electrical Engineering, University of Leeds, Leeds, LS29 JT UKSearch for more papers by this authorYongfei Li, Yongfei Li orcid.org/0000-0001-6301-3798 School of Electronic and Electrical Engineering, University of Leeds, Leeds, LS29 JT UK School of Electrical Engineering and Automation, Harbin Institute of Technology, Harbin, People's Republic of ChinaSearch for more papers by this authorLi Zhang, Li Zhang School of Electronic and Electrical Engineering, University of Leeds, Leeds, LS29 JT UKSearch for more papers by this author First published: 16 February 2021 https://doi.org/10.1049/iet-rpg.2020.0132AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Abstract The full controllable electronic device based railway feeder station offers better power quality and more flexible configurations than conventional transformer based stations. This study investigates a modular multilevel converter (MMC)-based static frequency converter station with renewable energy access. Wind power generation is coupled into the station via DC link of the back to back converter. The dynamic single-phase traction load and intermittent renewable generation bring double frequency oscillation and large deviation problems to the DC link voltage. Special design considerations and control schemes are proposed for the MMC to stabilise DC link voltage by controlling the total number of total inserted modules. The proposed control scheme resolves the voltage oscillation issue caused by single-phase load and reduces the DC link voltage deviation under 10 MW step change. A series of device-based simulations validate the control scheme which realises a reliable coupling interface for connecting the renewable generation to the DC bus. Nomenclature MMC G grid side converter MMC L traction line side converter P G grid side power P L traction load power P W wind power generation ω angular velocity of the grid T sc control system sampling rate p grid instantaneous power measured at grid side p mmc instantaneous power measured at MMC DC link p pmsg instantaneous power converted by wind generator p train instantaneous power consumed by traction load ω e M wind generator electrical angular speed i d M , i q M d- and q-axis current of the wind generator R s M wind generator stator winding resistance ψ m wind generator rotor flux linkage L m M wind generator synchronous inductance a average capacitor voltage factor i g a , i g b , i g c three-phase current at the grid side of MMC u g a , u g b , u g c three-phase voltage at the grid side of MMC u C i capacitor voltage of the ith submodule of MMC C sub MMC submodule capacitor v d M the d-axis generator terminal voltage v q M the q-axis generator terminal voltage L br MMC branch inductor R br equivalent resistance of one MMC branch i j u current in the MMC phase j upper branch, j = a , b , c , e , f i j l current in the MMC phase j lower branch, j = a , b , c , e , f i dc MMC DC link current u e , u f traction line side MMC terminal voltage of phases e , f u j u voltage of all inserted submodules in MMC phase j upper branches, j = a , b , c , e , f u j l voltage of all inserted submodules in MMC phase j lower branches, j = a , b , c , e , f u dc MMC DC link voltage V dc nominal MMC DC link voltage N number of submodules in one MMC phase u C ave average capacitor voltage of MMC u o output voltage of the traction line side MMC i o output current of the traction line side MMC i d , i q the d and q axis current of the grid side MMC m j u modulation index of MMC phase j upper branch, j = a , b , c , e , f m j l modulation index of MMC phase j upper branch, j = a , b , c , e , f u j com common mode voltage of MMC phase j, j = a , b , c , e , f i j com common mode current of MMC phase j, j = a , b , c , e , f m j com common mode modulation index of MMC phase j, j = a , b , c , e , f u j dif differential mode voltage of MMC phase j, j = a , b , c , e , f i j dif differential mode current of MMC phase j, j = a , b , c , e , f m j dif differential mode modulation index of MMC phase j, j = a , b , c , e , f Δ u dc ref DC link voltage stabilisation control reference u j cir ref MMC phase j circulating current control reference, j = a , b , c u j osc ref traction side converter oscillation current suppression control reference in MMC phase j, j = e , f ω r resonant angular velocity of the resonant controller φ traction load angle Δ u C j u i , Δ u C j l i the ith capacitor voltage balance control references for upper and lower branch of MMC phase j 1 Introduction The landscape change of the energy sector for embracing significant renewable penetration has made the electrified railway an environmentally more friendly means of transportation than on-road vehicles and air transport. The AC power supply scheme is the most consolidated technology which enjoys popularity in current high-speed railway applications, and the 25 kV single-phase AC power supply scheme is widely adopted in the traction power supply system in many countries such as Japan, China, UK and France etc. However, most of the existing 25 kV AC supply feeder stations are realised by conventional transformers which lead to several power quality issues such as voltage unbalance, low-power factor and harmonics issues. Various compensation schemes and balanced transformers have been researched and implemented in railway systems to improve the power quality [1]. So far, the power electronic technologies have become mature for high-voltage and high-power applications with the advancement of high-power semiconductor switches. Full power electronic converter based railway traction feeder station have been researched and used [2]. These power electronic converter based substations are also named as 'static frequency converter (SFC)' in the community and which are back to back conversion systems. The power quality issues brought by transformers can be solved in the full converter based substation. Moreover, the SFC station enables effective feeding of regenerative power back to the grid. The modular multilevel converter (MMC) topology has been intensively researched in recent years and has shown its advantages in high-voltage direct current (HVDC) transmission system and high-power conversion applications [3-6]. The successful applications of power electronic converter not only improve the power quality but also facilitate a more flexible power supply system for the future electrical railway system (ERS). In the past decade, some pioneering work has been carried out to verify the feasibility of integrating renewable energy source (RES) and energy storage system (ESS) into ERS. The European Union initiated the MERLIN project [7] to achieve a more sustainable and optimised energy usage in European electric mainline railway system in 2012. Japanese researchers demonstrated the potential of using photovoltaic (PV) panels on platform roofs and railway premises to introduce the solar power into the ERS in 2013 [8]. Zero-emission operation has been achieved in a local 'Hiraizumi Station' with solar energy generation and lithium-ion batteries [9]. To improve the energy efficiency, the East Japan Railway Company also studied economic benefit of the DC railway regenerative energy utilisation by ESS [10]. In [11], PV generation with hybrid ESS is connected to the cophase traction power system for coordinating regenerative braking energy and local renewable energy, and energy management strategy is optimised to achieve the lowest daily cost. Şengör et al. [12] also presented a mixed-integer programming model to minimise the daily operational cost of a similar system where the model considers the dynamics of train load, pricing scheme, stochastic nature of the state of energy and uncertain PV generation. Zhu et al. [13] used passivity-based stability criterion to assess the stability of a PV plant tied into medium voltage DC railway electrification system, and proposed a virtual impedance control scheme. Railway power supply integrated with RESs and hybrid ESSs relies heavily on the ICT technologies to operate reliably, effectively and efficiently. Therefore, researchers have considered the next generation of ERS to be a smart railway system, where information and communication technologies are used to improve the overall controllability. Eduardo et al. [7] discussed key features in a smart railway system, including 'smart train operation', 'smart operation of railway power supply', and 'smart interaction with other power systems'. Aguado et al. [14] proposed a methodology for optimal operation planning of railway energy systems considering the uncertainties associated with RES through scenario tree approach. Novak et al. [15] presented a hierarchical coordination scheme for substation energy flow control and the individual traction control, and used a stationary ESS to minimise energy consumption. Through interactions with other power systems, the ERS is transformed from a passive energy consumer into a proactive system which has the capability to respond to various grid demands. Sun et al. [16] evaluated the impact of battery-based railway transportation on power grid operation, and demonstrated that the mobile battery storage can relieve the transmission congestion while reducing the operation costs. Most of the research works mainly focus on RES and ESS installation capacity and power flow optimisation. The accurate system dynamics in medium voltage AC railway supply system with RES integration has not been fully investigated. Renewable energy such as wind and solar energy are intermittent and cannot be accurately predicted. Similarly, the traction load has similar characteristics due to rail condition, carrying weight, weather, and other factors. Unlike the conventional transformer based station, the power electronic converter based station needs to be controlled in real time to transfer the energy among the grid, RES and the train. The intermittent renewable energy generation and rapid changing train load will impose more stringent requirements on the converter performance. The instantaneous power consumed by single-phase traction load introduces double frequency oscillation into the supply system [17]. The existence of sudden load change, intermittent renewable generation and oscillating single-phase power challenge the stable operation of the hybrid railway power supply system. In this paper, a full power MMC solution is proposed for railway traction power supply integrated with RES. The aforementioned problems of single-phase oscillation and high-power change are addressed by utilising floating capacitor in the MMC. The remainder of the paper is organised as follows. Section 2 discusses the configurations of integrating RES with substation. Section 3 details the modelling and control of wind generation system. MMC modelling and control details are introduced in Section 4. Simulation results of several scenarios are evaluated in Section 5 and Section 6 concludes the paper. 2 Railway traction power supply integrated with renewable generations 2.1 Topologies Six different configurations with RES integrated into railway traction supply system are illustrated in Fig. 1. Configurations presented in Figs. 1a and b do not need any modification in the original stations, RES is directly connected to the power grid or traction overhead line. These two configurations are typical in conventional smart grid and their design and control are studied in [18]. Fig. 1Open in figure viewerPowerPoint Different connection schemes for integrating RES into railway supply system (a) RES interfaced with high-voltage distribution grid, (b) RES interfaced with medium voltage 25 kV overhead line, (c) RES interfaced with power grid compensator, (d) RES interfaced with railway power conditioner, (e) RES interfaced with cophase supply conditioner, (f) RES interfaced with static converter based station Figs. 1c and d represent the scenarios where the RES is connected to the grid side power quality compensator or railway power conditioner so that the DC/AC converter required in Figs. 1a and b are cancelled. Both compensator and conditioner are responsible for addressing the power quality issues due to traction load. The power generated by RES will be used to balance the grid current or compensate power flow between two supply arms. RES rating is limited by the conditioner capacity which is usually a portion of the feeder station. Ma et al. [19] analysed a multi-port railway power conditioner of Fig. 1d. Configurations in Figs. 1e and f integrate the renewable generation into the station. Cophase connection with RES complicates the system design but provides more flexible control for traction supply because the number of neutral zone can be reduced or eliminated. Liu et al. [11] investigated the energy management for this connection scheme. Topologies presented in Figs. 1a–e can be used regardless of the feeder station configuration. Due to the distinctive benefits of SFC-based substation [20], configurations illustrated in Fig. 1f is investigated in this paper, where RES are connected to the medium DC voltage link of the back to back converter. Railway feeder stations usually have power rating from 10 to 80 MW, so this topology has the capacity to accept more renewable generations than configurations illustrated in Figs. 1c–e. 2.2 Proposed static converter station integrated with wind power generation Fig. 2 presents the equivalent diagram of the proposed supply system. MMC back to back converter is used to convert three-phase voltage power into a single phase. As illustrated in Fig. 3, the MMC on the grid side is denoted as MM CG and MMC connected to the traction overhead line is denoted as MM CL . The wind turbine generator (WTG) system brings power P W into the DC link, and the power flow of MM CG denoted as P G can be controlled in bi-direction depending on the DC link voltage. P L denotes traction side power and is delivered by MM CL . Fig. 2Open in figure viewerPowerPoint Configuration of a wind power connection into static converter based railway substation Fig. 3Open in figure viewerPowerPoint Power flow illustration for the hybrid railway supply system The range of power flowing into or out of the grid ( P G ) is determined by wind power, train load and regenerative efficiency. Generally, regeneration power will be less than 80% of the maximum traction power, so we can define the power rating of MM CG as: max ( P G + ) ≥ P L max min ( P G − ) ≤ − 0.8 P L max + P W max , (1) where P G + denotes power flowing from the grid to DC link via MM CG , P G − denotes power from DC link back to the grid via MM CG , P L max and P W max denote the maximum wind power generation and traction load power, respectively. However, if extra ESS is included in this system, then more wind power can be accepted without an increase MM CG 's power rating. Different power flow patterns and the symbol definition are illustrated in Fig. 3. Based on the energy conservation principle, the following equation (ignoring the energy inside inductors) can be derived: ∫ t 1 t 2 P W + P G − P L d t = ∑ C sub u C i 2 ( t 2 ) − u C i 2 ( t 1 ) 2 , (2) where C sub represents each capacitor's capacitance in the system and u C i ( t ) is the corresponding voltage at time t. The fast and stable operation of wind power delivery and traction power supply depends on a stable DC link voltage. If the DC link voltage has large deviations or large magnitude oscillations, the system performance will be compromised and the device connected to the DC-bus may behave unexpectedly [21]. As mentioned in Section 1, P L and P W by nature have adverse effects on DC voltage stabilisation. Thus the MM CG must robustly control DC link voltage by changing P G and internal submodule states. Ideally, all capacitor voltages vary around their nominal values, but it is hard to achieve during transient states. For example, if P L changes from 0 to maximum in a short time in acceleration mode (or reversely in braking mode), the capacitor voltage will inevitably vary. It takes time to restore u C ave back to nominal its value and during that time DC link voltage will be affected. Section 4 addresses this problem in details. It is preferable to build the wind energy conversion system (WECS) near the location of SFC station to reduce transmission losses. We assume the WECS are a set of WTGs and transmit the power through HVDC line into SFC station. There are numerous research works conducted on different topologies and control schemes for connecting wind farm and DC transmission [22]. The detailed wind farm control is beyond the scope of this paper, and therefore only one equivalent generator unit is considered for simplicity. 3 Wind power generator modelling and control The wind generation system is simplified into a single machine unit. In this section, a single 10 MW permanent magnet synchronous generator (PMSG) with three-phase three-level neutral point clamped (NPC) converter is used to represent the WTG system Fig. 4. The medium voltage DC side of NPC connects directly to the DC link inside the SFC station. Fig. 4Open in figure viewerPowerPoint PMSG with NPC three-level converter 3.1 Wind power synchronous generator modelling Fig. 4 illustrates the circuit diagram of a typical wind energy conversion unit using an NPC converter. Table 1 lists the parameters of the equivalent WECS. Ignoring the effects of slotting, saturation, end effect etc. the simplified PMSG dynamic equations in dq-frame can be written as (3): v d M = R s M i d M − ω e M L q M i q M + L d M d i d M d t v q M = R s M i q M + ω e M L d M i d M + L q M d i q M d t + ω e M ψ m , (3) where v d M , v q M are the machine terminal voltages in dq frame; ω e M is the electrical angular speed. Here it is assumed that the generator is a surface mount magnet generator, viz. L d M = L q M = L m M . The machine parameters are shown in Table. 1. Table 1. System parameters of wind power generation system PMSG machine rated line voltage V l M 26.4 kV rated stator frequency f s M 20 Hz number of pole pairs p 8 rated rotor flux linkage ψ m 172 Wb stator winding resistance R s M 1.6 Ω synchronous inductance L m M 65.6 mH NPC converter rated DC voltage V dc 48 kV NPC capacitors C 1 , 2 M 500 μF switching frequency f npc 2000 Hz Parameters do not necessarily represent a real system. The wind turbine model is also reduced to a controlled torque source. In this simplified system, the generator is controlled in constant speed operation with fast response to external torque input. We simulate different wind power generation by changing the value of applied torque on PMSG's shaft. 3.2 Generator controller design To achieve a fast response, a modified deadbeat control is adopted as a machine controller. Deadbeat control is a model-based control method which is equivalent to a simplified implementation of the horizon one model predictive controller. Discretise (3) using forward Euler method with controller sample time T sc , the prediction of dq axis currents can be explicitly predicted: i d M ( k + 1 ) i q M ( k + 1 ) = 1 − T sc R s M L m M T sc ω e M ( k ) − T sc ω e M ( k ) 1 − T sc R s M L m M i d M ( k ) i q M ( k ) + T sc 1 L m M 0 0 T sc 1 L m M v d M ( k ) v q M ( k ) + 0 − T sc ω e M ( k ) ψ m L m M . (4) As described in Fig. 5, reference current i q m ref is updated by PI controller and i d m ref is set to zero for maximum torque per ampere control. Fig. 5Open in figure viewerPowerPoint Controller diagram for PMSG drive Therefore, we can calculate the required terminal voltage to generate the exact reference current in the next time step by v d M ref v q M ref = R s M − 1 T sc L m M − L m M ω e M ( k ) L m M ω e M ( k ) R s M − 1 T sc L m M i d M ( k ) i q M ( k ) + 1 T sc L m M 0 0 1 T sc L m M i d M ref i q M ref + 0 ψ m ω e M ( k ) (5) Equation (5) expresses the control equations inside the deadbeat controller which calculate terminal voltage command to control i d M , i q M to track the reference signal without error. However, the sampling time T sc cannot be infinitesimal, so modelling error deteriorates its current control performance. For this reason, parallel integral controllers are added (Fig. 5) to improve the control robustness [23]. The composite voltage command signals are then transformed back to abc frame and the space vector pulse width modulation method is used to decide switching states in the NPC controller. 4 MMC-based SFC station 4.1 Modelling and design of the back to back converter 4.1.1 Single-phase model and control principle of MMC Fig. 6 shows the circuit topology of MMC-based static back to back converter. The back to back converter has three phases (a,b,c) working as a grid connection converter and two phases (e,f) connecting with railway traction overhead line. Different phases have a similar structure and each of them is composed of two sets of a series-connected half-bridge modules with branch inductor L br . The DC link voltage is measured as u dc , and a virtual reference point is used to derive circuit equations. u g a , u g b , u g c are three phase voltages with star connection, and thus i g a + i g b + i g c = 0. For simplicity, we consider discussion of the single-phase MMC model in the following explanation. Fig. 6Open in figure viewerPowerPoint Topology of the MMC back to back static converter Symbols u j u and u j l represent the voltage of all inserted modules in upper and lower branches in phase j, respectively. By Kirchhoff's law the current and voltage in phase a follow (6)–(8). u g j = − R br i j u − L br d i j u d t − u j u + 0.5 u dc (6) u g j = + R br i j l + L br d i j l d t + u j l − 0.5 u dc (7) i g j = − i j u + i j l (8) Then add and subtract (6) and (7) the following expressions of terminal voltage u g j and DC link voltage u dc are obtained: u dc = u j u + u j l + 2 R br i j u + i j l 2 + 2 L br d d t i j u + i j l 2 (9) u g j = − u j u + u j l 2 + R br − i j u + i j l 2 + L br d d t − i j u + i j l 2 (10) We control MMC dynamics by switching on and off of each submodule to decide voltage of u j u and u j l . If the number of submodules N → ∞ or the switching frequency is sufficiently high, we can use modulation indices m j u , m j l to represent the branch voltages: u j u = m j u N u C ave , u j l = m j l N u C ave , (11) where u C ave is the average capacitor voltage. Equations (8)–(10) imply that the AC terminal side value is decided by the difference value of upper and lower branch, and the DC side is determined by common values of two branches. Therefore it is easier to view MMC topology in differential and common mode model, and the differential/common mode values are defined as x j com = ( x j u + x j l ) / 2 , x j dif = ( − x j u + x j l ) / 2 , (12) where x represents modulation index, branch voltage or branch current, and the subscripts com , dif represent the common mode value and differential mode value, respectively. For example, i j dif represents the differential mode current of phase j which is defined by upper and lower branch currents i j u and i j l . Then the terminal voltage and current of one MMC phase can be rewritten using differential and common mode values: u dc = m j com 2 N u C ave + i j com 2 R br + d d t i j com 2 L br , (13) u g j = m j dif N u C ave + i j dif R br + d d t i adif L br , (14) i g j = 2 i j dif . (15) The MMC single-phase equivalent model in (13) and (14) are illustrated in Fig. 7. In brief, u g j is controlled by m j dif and u dc is controlled by m j com . In normal steady state operation, average capacitor voltage in each branch equals to 1 / N of the nominal DC link voltage V dc . Consequently, in most of applications, N modules are inserted in series to maintain DC link voltage at each instance by default, which gives m j com = 0.5. Usually, i j com has to be controlled to suppress excessive AC components, then a small sinusoidal signal will be added to m j com to achieve this objective. Fig. 7Open in figure viewerPowerPoint Simplified equivalent differential common-mode model of single MMC phase As the differential mode shows that the MMC can be controlled in the same way as conventional two-level converter, most of the previous works only stabilise DC voltage by controlling average capacitor voltages and use m j com = 0.5 to indirectly maintain the DC voltage. However, if the load change drastically, namely i dc change greatly in very short period of time, average capacitor voltage will no longer be the nominal value ( u C ave ≠ V dc / N). Modulation signal of m j com = 0.5 can not guarantee stable DC link voltage and u dc changes with average capacitor voltage synchronously. The conventional approach does not fully take advantage of the special characteristics in MMC. In fact, the total number of inserted submodules can be modified to compensate transient voltage drop. For this reason, it is necessary to deliberately control m j com such that DC link voltage has less deviation from the nominal value in transient state. 4.1.2 Parameter design of MMC back to back converter Assume right after a sudden load change, the average capacitor voltage becomes: u C ave = a V dc N , (16) where a is the average capacitor voltage factor, and we assume a ∈ [ 0.5 , 1.5 ]. By (13), ignoring the voltage across the inductor, m com has to be set to 0.5 / a in order to stabilise the DC voltage perfectly. Note that m j u , m j al ∈ [ 0 , 1 ], and therefore m a dif can vary from − 0.5 to 0.5 if and only if m j com = 0.5. Also in this case, the no load voltage output range is u g j ∈ [ − 0.5 V dc , + 0.5 V dc ]. But under the condition when the average capacitor voltage has large deviation from the nominal, differe
Referência(s)