An energy management method for a hybrid energy storage system connected to a traction power supply system

Abstract

The present invention discloses an energy management strategy for a hybrid energy storage system connected to a traction power supply system. An energy density type energy storage medium - lithium battery and a power density type energy storage medium - supercapacitor are selected to construct a hybrid energy storage type RPC system, and the system principle and control strategy are studied. In order to improve the utilization rate of regenerative braking energy and extend the cycle service life of the energy storage medium, an energy management strategy for the hybrid energy storage type system based on VMD is proposed. In order to more accurately obtain the high - and low - frequency power in the energy storage power, the IMF corresponding to the minimum value of the cross - correlation coefficient of the IMFs after VMD decomposition is used as the threshold, and the high - and low - frequency energy storage power is reconstructed to obtain the target power value of the energy storage medium and realize the charge and discharge of the energy storage medium through controlling the bidirectional DC / DC. Hardware - in - the - loop test experiments are carried out on the StarSim power electronics small - step real - time simulator of Shanghai Yuankuan Energy (Modeling Tech) for the proposed control strategy, and long - working - condition verification and analysis are carried out on MATLAB / Simulink in combination with the measured data of a traction substation on the Lanxin Line.

Description

Technical Field

[0001] The present invention belongs to the field of energy management strategies for electrified railways, and particularly relates to an energy management method for a hybrid energy storage system connected to a traction power supply system. Background Art

[0002] As of the end of 2021, the national railway operating mileage reached 150,000 kilometers, and the electrification rate reached 73.3%. In terms of comprehensive energy consumption, it increased by 5.7% compared with 2020. With the increase in the operating mileage of electrified railways in China, railway energy consumption will continue to increase. When the train is in the braking state, regenerative braking energy will be generated. In some special sections, this part of the energy can reach 10%-30% of the traction energy. If it is directly fed back to the traction substation, not only will the energy not be effectively utilized, but also due to the large amount of negative sequence and harmonic content in it, it will aggravate the negative sequence voltage unbalance degree and reduce the power quality of the power grid. To solve the negative sequence problem of electrified railways, the railway power conditioner (energy storage type railway power conditioner, RPC) proposed by scholars has received extensive attention due to its excellent comprehensive compensation performance, port extensibility and other advantages. With the proposal of the concept of "carbon neutrality", it is of great significance to reduce transportation energy consumption and carbon dioxide emission intensity and promote the low-carbon operation of railways. To solve the problem of regenerative braking energy utilization, some scholars have proposed adding an energy storage system to the intermediate DC side of the RPC to form an energy storage type railway power conditioner to realize the two-way power flow between the power supply arms. Among them, the RPC is used to improve the negative sequence problem, and the energy storage system stores and releases the regenerative braking energy to realize the peak shaving and valley filling of the load power.

[0003] However, most of the current railway energy storage type railway power conditioner schemes use a single energy storage medium. Due to the large total amount of regenerative braking energy, it is difficult for a single energy storage to realize the utilization of regenerative braking energy with large power and large energy. The hybrid energy storage system composed of these two energy storage media with power density type and energy density type has become a research hotspot at home and abroad. The energy management strategy of the hybrid energy storage system is responsible for coordinating the energy flow between the hybrid energy storage system and the main system. The allocation of power commands is a key issue in the energy management control of the hybrid energy storage system. A reasonable power allocation strategy can not only improve the energy utilization rate but also extend the service life of the energy storage medium.

[0004] In the microgrid system, some scholars have obtained the charge and discharge power commands of each medium of the hybrid energy storage system by using the adaptive first-order low-pass filter algorithm, the spectrum analysis algorithm, the wavelet packet decomposition algorithm, and the ensemble empirical mode decomposition algorithm respectively. In the railway system, some scholars have used the first-order low-pass filter algorithm for the internal power distribution of the hybrid energy storage system to realize the reasonable utilization of regenerative braking energy. There are also those who use the discrete Fourier transform for power distribution. To sum up, algorithms for obtaining the power commands of the hybrid energy storage system through low-pass filtering, spectrum analysis, wavelet packet decomposition, VMD, etc. are more widely used in the microgrid. In the hybrid energy storage system of electrified railways, the energy management and power distribution strategies still need to be further studied.

[0005] Based on the above problems, this patent proposes an energy management method for a hybrid energy storage type RPC system based on variational mode decomposition (VMD) for the energy management strategy problem of the hybrid energy storage system accessing the traction power supply system in electrified railways. Summary of the Invention

[0006] In order to solve the technical problems existing in the background art, the purpose of the present invention is to provide an energy management method for a hybrid energy storage system accessing the traction power supply system.

[0007] To solve the technical problems, the technical solution of the present invention is as follows:

[0008] An energy management method for a hybrid energy storage system accessing the traction power supply system, the method comprising:

[0009] S1. Determine the peak shaving threshold P tra and the valley filling threshold P reg of the traction power supply system, and determine the maximum charging value P bat_cmax , the maximum discharge value P bat_dmax of the lithium battery, the maximum charging value P sc_cmax , the maximum discharge value P sc_dmax of the super capacitor, the change range SOC bat_min , SOC bat_max of the state of charge of the lithium battery, and the change range SOC sc_min , SOC sc_max of the state of charge of the super capacitor;

[0010] S2. Detect the effective values U L , U R of the voltages on the left and right power supply arms of the traction side in the traction power supply system and the effective values I L , I R , and detect the state of charge SOC bat , SOC sc of the super capacitor and the lithium battery in the hybrid energy storage system;

[0011] S3. Calculate the active power P of the left and right power supply arms on the traction side based on the effective values of the voltages and currents of the left and right power supply arms on the traction side. L and P R . Obtain the total active power P of the load based on the active powers of the left and right power supply arms on the traction side. Z ; Obtain the maximum charging value P of the total active power hybrid energy storage system based on the maximum charging and discharging values of the lithium battery, the maximum charging and discharging values of the super capacitor. es_cmax and the maximum discharging value P es_dmax ;

[0012] S4. Based on the obtained peak shaving threshold and valley filling threshold, the change range of the state of charge (SOC) of the lithium battery and the super capacitor, the state of charge of the super capacitor and the lithium battery, the total active power of the load, the maximum charging and discharging values of the total active power hybrid energy storage system, construct the charge and discharge strategy of the hybrid energy storage system, and determine the target power value of the energy storage system according to the charge and discharge strategy of the hybrid energy storage system.

[0013] S5. Decompose using the VMD decomposition algorithm to obtain K IMFs and calculate the cross-correlation coefficient, and reconstruct the target power value of the lithium battery according to the cross-correlation coefficient and the target power value of the super capacitor That is, the energy management strategy for the hybrid energy storage system connected to the traction power supply system is realized.

[0014] Furthermore, in step S3, the active powers P of the left and right power supply arms on the traction side L and P R are calculated by the formulas: P L =U L I L ; P R =U R I R ; The total active power P of the load Z is calculated by the formula: P Z =P L +P R ; The maximum charging value P of the total active power hybrid energy storage system es_cmax is calculated by the formula: P es_cmax =P bat_cmax +P sc_cmax ; The maximum discharging value P es_dmax is calculated by the formula: P es_dmax =P bat_dmax +P sc_dmax .

[0015] Furthermore, in step S4, the constructed charge and discharge strategy of the hybrid energy storage system is divided into three modes:

[0016] (1) Valley filling energy storage mode

[0017] When P Z ≤P reg and SOC bat or SOC sc is less than their respective maximum state of charge values, the energy storage system absorbs regenerative braking energy, which can be specifically divided into the following two cases. When P Z <P es_cmax , the energy absorbed by the energy storage system is |P es_cmax |, and the remaining energy is sent back to the power grid; when P reg >P Z ≥P es_cmax , the energy absorbed by the energy storage system is |P Z |, and no energy is sent back to the power grid; therefore, the target power value of the energy storage system in this mode is:

[0018]

[0019] (2) Standby mode

[0020] When P reg <P Z <P tra or SOC bat 、SOC sc is not within the normal operating range, the energy storage system is in standby mode, and its target power value is:

[0021]

[0022] (3) Peak shaving and energy releasing mode

[0023] When P Z ≥P tra and SOC bat or SOC sc is greater than their respective minimum state of charge values, the energy storage system releases energy to reduce the peak load and the load demand of the traction transformer, and the remaining energy is provided by the power grid; it is specifically divided into the following two cases. When P tra ≤P Z <P tra +P es_dmax , the energy released by the energy storage system is |P tra -P Z |, when P Z ≥P tra +P es_dmax , the energy storage system releases the maximum energy P es_dmax ; therefore, the target power value of the energy storage system in the peak shaving and energy releasing mode is:

[0024]

[0025] Furthermore, in step S5, the VMD decomposition algorithm is used to decompose Specifically:

[0026] Using the VMD decomposition algorithm, i.e., the constrained variational model, to decompose As shown in Equation (4):

[0027]

[0028] In the formula: {u k} is the total active power of the load The set of IMFs obtained by VMD decomposition; {ω k} is the set of IMF center frequencies; K is the number of IMFs; δ(t) is the impulse function;

[0029] By introducing the Lagrange multiplier function to solve the optimal solution of the above variational problem, its expression is:

[0030]

[0031] In the formula, λ is the Lagrange multiplication operator; a is the quadratic penalty factor;

[0032] The alternating multiplier algorithm is used to solve Equation (4), and by iteratively updating u k , ω k and λ, the IMFs are obtained, and then the modal components and their center frequencies are obtained through Fourier transform as follows:

[0033]

[0034] In the formula are respectively The sequences after Fourier transform of λ(ω), u i (ω); Is the Wiener filter of the current residual; Is the current IMF center frequency;

[0035] The update formula of the Lagrange multiplication operator is:

[0036]

[0037] In the iterative solution process, each IMF and its center frequency are continuously updated until the iterative stop condition is met and the entire loop ends. The iterative stop condition is:

[0038]

[0039] Wherein, e is a given determination accuracy, and e > 0;

[0040] That is, K IMFs are obtained.

[0041] Further, in step S5, the specific calculation of the cross-correlation coefficient is as follows:

[0042] Determine the cross-correlation coefficient R(IMF k , IMF k+1 ) as the cross-correlation coefficient between the connected IMF k and IMF k+1 :

[0043]

[0044] Wherein, N is the number of sampling points; are respectively the standard deviation and mean value of the k-th IMF; are respectively the standard deviation and mean value of the (k + 1)-th IMF. Select the IMF corresponding to the minimum cross-correlation coefficient k as the boundary, and reconstruct the IMF 1-k into the low-frequency power P of the energy storage system esL , and compare it with the maximum value of the lithium battery charge and discharge to obtain the target power value of the lithium battery as:

[0045]

[0046] The target power of the supercapacitor is:

[0047]

[0048] Compared with the prior art, the advantages of the present invention are as follows:

[0049] This method selects the energy density type energy storage medium - lithium battery and the power density type energy storage medium - supercapacitor to construct a hybrid energy storage type RPC system, and studies the system principle and control strategy. In order to improve the utilization rate of regenerative braking energy and extend the cycle service life of the energy storage medium, an energy management strategy for the hybrid energy storage type system based on VMD is proposed. In order to more accurately obtain the high and low frequency powers in the energy storage power, according to the IMF corresponding to the minimum cross-correlation coefficient after VMD decomposition as the threshold, the high and low frequency energy storage powers are reconstructed, and the target power values of the energy storage medium are obtained and the charge and discharge of the energy storage medium are realized by controlling the bidirectional DC / DC. The proposed control strategy is tested in hardware-in-the-loop on the StarSim power electronics small-step real-time simulator of Shanghai YuanKuan Energy (Modeling Tech), and long-term condition verification and analysis are carried out on MATLAB / Simulink combined with the measured data of a traction substation on the Lanzhou-Xinjiang line. Description of the Drawings

[0050] Figure 1 and the structure diagram of the hybrid energy storage type RPC system;

[0051] Figure 2 and the control strategy of the hybrid energy storage type RPC system;

[0052] Figure 3 and the energy management strategy of the hybrid energy storage type RPC system;

[0053] Figure 4 and the energy modes of the hybrid energy storage system;

[0054] Figure 5 and the diagram of the relationship between energy modes;

[0055] Figure 6 and the diagram of the StarSim hardware-in-the-loop experimental platform;

[0056] Figure 7 and the intermediate DC side voltage;

[0057] Figure 8 and the waveform diagrams of the voltage and current of the RPC converter;

[0058] Figure 9 and the waveform diagrams of the load powers of the left and right power supply arms;

[0059] Figure 10 and the target power value curve of the hybrid energy storage system;

[0060] Figure 11 and the IMF spectrogram;

[0061] Figure 12 and the target power value curves of the lithium battery and the super capacitor;

[0062] Figure 13 and the state of charge diagram of the energy storage medium;

[0063] Figure 14 and the comparison diagrams of the powers of the left and right power supply arms before and after compensation;

[0064] Figure 15 and the comparison of the compensation powers of the traction substation;

[0065] Figure 16 and the comparison of the negative sequence current diagrams before and after compensation. Specific implementation manners

[0066] The following describes the specific implementation manners of the present invention in combination with embodiments:

[0067] It should be noted that the structures, ratios, sizes, etc. shown in this specification are only used to cooperate with the content disclosed in the specification for those familiar with this technology to understand and read, and are not used to limit the implementation conditions of the present invention. Any modification of the structure, change of the proportional relationship, or adjustment of the size, without affecting the efficacy that the present invention can produce and the purpose that can be achieved, should still fall within the scope covered by the technical content disclosed in the present invention.

[0068] At the same time, the terms such as "upper", "lower", "left", "right", "middle", and "one" cited in this specification are also only for the sake of clarity in narration, rather than used to limit the scope of implementation of the present invention. The change or adjustment of their relative relationship, without substantial change in the technical content, should also be regarded as the scope of implementation of the present invention.

[0069] Embodiment 1:

[0070] The present invention selects an energy density type energy storage medium - lithium battery and a power density type energy storage medium - super capacitor to construct a hybrid energy storage type RPC system, and studies the system principle and control strategy. In order to improve the utilization rate of regenerative braking energy and extend the cycle service life of the energy storage medium, an energy management strategy for the hybrid energy storage type system based on VMD is proposed. In order to more accurately obtain the high and low frequency powers in the energy storage power, the IMF corresponding to the minimum cross-correlation coefficient of the IMFs after VMD decomposition is used as the threshold, and the high and low frequency energy storage powers are reconstructed to obtain the target power values of the energy storage medium, and the charging and discharging of the energy storage medium are realized by controlling the bidirectional DC / DC. The proposed control strategy is tested in a hardware-in-the-loop test experiment on the StarSim power electronics small-step real-time simulator of Shanghai YuanKuan Energy (Modeling Tech), and long-term working condition verification and analysis are carried out on MATLAB / Simulink combined with the measured data of a traction substation on the Lan-Xin line.

[0071] 1 Hybrid energy storage type RPC system

[0072] 1.1 System structure

[0073] The structure diagram of the hybrid energy storage type RPC system is as Figure 1 shown.

[0074] Figure 1 In it, since the measured data used in this article is from a traction substation on the Lan-Xin line, the grid level is 330 kV. P g is the grid output power; P tL , P tR are the active powers of the left and right power supply arms output by the traction substation respectively, and P cL , P cR are the active powers compensated by the RPC for the left and right power supply arms respectively, and P bat , P scThey are the output powers of the lithium battery and the super capacitor respectively.

[0075] Figure 1 The hybrid energy storage type RPC system consists of an RPC and a hybrid energy storage system. The hybrid energy storage type RPC system is connected to the left and right power supply arms through two step-down transformers. The converters VSC L 、VSC R in the RPC are connected through an intermediate DC capacitor to achieve energy flow. The hybrid energy storage system consists of a bidirectional DC / DC converter, a lithium battery and a super capacitor. The lithium battery and the super capacitor are connected to the intermediate DC capacitor through their respective bidirectional DC / DC converters. An energy management strategy is adopted to control the charge and discharge of the energy storage medium to realize the utilization of regenerative braking energy.

[0076] 1.2 System working principle

[0077] The energy relationship diagram between the system structures is as shown in Figure 1 Define P es as the total output power of the hybrid energy storage system, P L 、P R as the active powers of the loads on the left and right power supply arms respectively, and P Z as the total active power of the load, which is equal to the sum of the active powers of the loads on the left and right power supply arms. If the direction of the arrow in Figure 1 is defined as the positive direction of power flow, then the power relationship between the various parts of the system is:

[0078]

[0079] The active powers P′ L 、P′ R of the left and right power supply arms after being compensated by the hybrid energy storage type RPC system are respectively:

[0080]

[0081] Then the active powers P′ L 、P′ R compensated by the RPC for the left and right power supply arms are respectively:

[0082]

[0083] 2. Control strategy of the hybrid energy storage type RPC system

[0084] The control strategy of the hybrid energy storage type RPC system consists of two parts: the control of the hybrid energy storage system and the control of the RPC. The control schematic diagram is as shown in Figure 2 shown.

[0085] The control of the hybrid energy storage system consists of four parts: energy management strategy, VMD power distribution strategy, current closed-loop control, and PWM modulation. The energy management strategy and VMD power distribution strategy are elaborated in detail in Section 3.

[0086] Total active power P of the load Z The target power values of the lithium battery and the supercapacitor are obtained through the energy management strategy and VMD power distribution strategy and are respectively divided by the lithium battery voltage U bat and the supercapacitor voltage U sc to obtain the target current values and Finally, the actual value I of the inductor current in the current loop feedback bat (I sc ) and The difference is PI-regulated and then pulse waves are generated through PWM to control their respective bidirectional DC / DC converters, thereby realizing the charge and discharge control of the energy storage device.

[0087] The RPC control strategy consists of three parts: voltage outer loop, current inner loop, and PWM modulation.

[0088] To ensure the normal operation of the RPC and the smooth connection of the hybrid energy storage system, the voltage in the intermediate DC link needs to be kept stable. Therefore, through the voltage outer loop control, the difference between the target value of the intermediate DC voltage and the actual voltage value U dc is PI-regulated to achieve the stability of the DC side voltage.

[0089] According to the voltages u L , u R of the left and right power supply arms, the currents i L , i R and the compensation powers P cL , P cR of the left and right power supply arms, the compensation currents i Lc , i Rc of the hybrid energy storage type RPC system on the left and right power supply arms are calculated, and their expressions are:

[0090]

[0091] In the formula, i' L , i' R are the currents of the left and right power supply arms after compensation respectively; U L , U R are the effective values of the voltages of the left and right power supply arms respectively.

[0092] Finally, current inner loop control is performed on the compensation current and the outer loop stable current, and pulse waves are generated through PWM to control the left and right converters of the RPC.

[0093] 3. Energy Management Strategy of Hybrid Energy Storage Type RPC System

[0094] The flow and interaction of the energy between the energy storage system and the traction power supply system depend on the energy management strategy, and the specific idea is as Figure 3 shown. The energy management strategy consists of two parts: the charge and discharge strategy of the hybrid energy storage system and the power distribution strategy. The charge and discharge strategy of the hybrid energy storage system is responsible for reasonably distributing the energy between the hybrid energy storage system and the traction power supply system to achieve the function of peak shaving and valley filling; the power distribution strategy realizes the power distribution between the lithium battery and the super capacitor in the hybrid energy storage system.

[0095] 3.1 Energy Management Modes

[0096] Define the power value of the traction working condition of the system as positive, and the power value of the braking working condition of the system as negative; the power value absorbed by the energy storage medium is positive, and the released power value is negative. For the convenience of expression, let P tra and P reg be the peak shaving threshold and valley filling threshold of the system respectively; P es_cmax and P es_dmax are the maximum charge and discharge values of the energy storage system respectively, and P es_cmax = -P es_dmax ; SOC bat is the state of charge of the lithium battery; SOC bat_min and SOC bat_max are the minimum and maximum values of the state of charge of the lithium battery respectively; SOC sc is the state of charge of the super capacitor; SOC sc_min and SOC sc_max are the minimum and maximum values of the state of charge of the super capacitor respectively; P bat_cmax and P bat_dmax are the maximum charge and discharge values of the lithium battery respectively, and P bat_cmax = -P bat_dmax ; P sc_cmax and P sc_dmax are the maximum charge and discharge values of the super capacitor respectively, and P sc_cmax = -P sc_dmax .

[0097] Figure 4 In, according to the load power conditions of the left and right power supply arms, the charge and discharge strategy of the energy storage system is divided into three energy modes: peak shaving and energy releasing mode, valley filling and energy storage mode, and standby mode. The analysis of each mode is as follows:

[0098] (1) Valley filling and energy storage mode

[0099] When P Z ≤ P reg and SOC bat or SOC scWhen it is less than their respective state of charge maximum values, the energy storage system absorbs regenerative braking energy, which can be specifically divided into the following two cases. When P Z <P es_cmax , the energy absorbed by the energy storage system is |P es_cmax |, and the remaining energy is sent back to the power grid; when P reg >P Z ≥P es_cmax , the energy absorbed by the energy storage system is |P Z |, and no energy is sent back to the power grid. Therefore, the target power value of the energy storage system in this mode is:

[0100]

[0101] (2) Standby mode

[0102] When P reg <P Z <P tra or SOC bat 、SOC sc is not within the normal working range, the energy storage system is in standby mode, and its target power value is:

[0103]

[0104] (3) Peak shaving and energy release mode

[0105] When P Z ≥P tra and SOC bat or SOC sc is greater than their respective state of charge minimum values, the energy storage system releases energy to reduce the load peak and reduce the load demand of the traction transformer, and the remaining energy is provided by the power grid. It is subdivided into the following two cases. When P tra ≤P Z <P tra +P es_dmax , the energy released by the energy storage system is |P tra -P Z |. When P Z ≥P tra +P es_dmax , the energy storage system releases the maximum energy P es_dmax . Therefore, the target power value of the energy storage system in the peak shaving and energy release mode is:

[0106]

[0107] Determine the energy working mode of the hybrid energy storage system according to the total active power P Z of the load and the state of charge of the energy storage medium. The energy mode relationship Figure 5 is shown as:

[0108] 3.2 Hybrid Energy Storage Power Allocation Strategy Based on VMD

[0109] VMD can effectively avoid modal aliasing and achieve the separation of components with similar frequencies. To obtain accurate high and low power of energy storage and extend the life of energy storage media, a hybrid energy storage power allocation strategy based on VMD is adopted.

[0110] 3.2.1 Variational Mode Decomposition Method

[0111] VMD is an adaptive signal processing algorithm that can decompose non-stationary and non-linear signals into IMFs with different center frequencies. The optimal solution of the variational model can be searched iteratively, and its constrained variational model is shown in Equation (8):

[0112]

[0113] where: {u k} is the total active power of the load the set of IMFs obtained by VMD decomposition; {ω k} is the set of IMF center frequencies; K is the number of IMFs; δ(t) is the impulse function.

[0114] The optimal solution of the above variational problem is solved by introducing the Lagrange multiplier function, and its expression is:[[]]

[0115]

[0116] where λ is the Lagrange multiplication operator; α is the quadratic penalty factor.

[0117] The alternating multiplier algorithm is used to solve Equation (8). By iteratively updating u k , ω k and λ, the IMFs are obtained, and then the modal components and their center frequencies are obtained through Fourier transform as follows:[[]]

[0118]

[0119] where are respectively the sequences of λ(ω) and u i (ω) after Fourier transform; is the Wiener filter of the current remaining amount; is the current IMF center frequency. The update formula of the Lagrange multiplication operator is:[[]]

[0120]

[0121] During the iterative solution process, each IMF and its center frequency are continuously updated until the entire loop ends after meeting the iterative stop condition, which is:

[0122]

[0123] where e is the given determination accuracy, and e > 0.

[0124] 3.2.2 Determination of the target power of the energy storage medium

[0125] After VMD decomposition, K IMF components with frequencies from low to high are obtained. To more accurately reconstruct the high- and low-frequency powers, according to the cross-correlation coefficient, which reflects the correlation between IMFs, the greater the cross-correlation coefficient, the greater the degree of association between variables.

[0126] Define R(IMF k , IMF k+1 ) as the cross-correlation coefficient between the adjacent IMF k and IMF k+1 :

[0127]

[0128] where N is the number of sampling points; are the mean and standard deviation of the k-th IMF respectively; are the mean and standard deviation of the (k + 1)-th IMF respectively.

[0129] Select the IMF k corresponding to the minimum cross-correlation coefficient as the boundary, and reconstruct the IMF 1-k into the low-frequency power P esL of the energy storage system, and compare it with the maximum charge and discharge value of the lithium battery to obtain the target power value of the lithium battery as:

[0130]

[0131] Then the target power of the supercapacitor is:

[0132]

[0133] Example 2:

[0134] To verify the effectiveness of the control strategy and energy management strategy of the hybrid energy storage type RPC system, hardware-in-the-loop test experiments are verified through the StarSim power electronics small-step real-time simulator of Shanghai Yuankuan Energy (Modeling Tech); in MATLAB / Simulink, the measured power of a certain traction substation's power supply arm on the Lanxin Line is used as simulation data for long-term measured data verification.

[0135] 4.1 StarSim Real-time Verification

[0136] On the StarSim experimental platform shown in Figure 6 Set the working condition as P L = 5MW, P H = 5MW, The intermediate DC-link voltage is obtained as shown in Figure 7 The intermediate DC-link voltage is stable near the preset value of 3200V.

[0137] The voltages u sL and u sR of the left and right converters and the currents i sL and i sR waveform diagrams are as shown in Figure 8 Under the set working conditions, the left and right converters work normally and the energy flow is realized.

[0138] 4.2 MATLAB Simulation Verification

[0139] In MATLAB / Simulink, simulation verification is carried out according to the measured power data of the power supply arm (sampling frequency is 0.25 s / time). The waveforms of the powers P L and P R of the left and right power supply arms and the total active power P z of the load are as shown in Figure 9 .

[0140] The target power of the hybrid energy storage system is obtained through the charge and discharge strategy of the hybrid energy storage system, and the waveform is as Figure 10 .

[0141] VMD needs to set the number of IMFs K. The value of K has a great influence on the decomposition effect of VMD. According to the central frequencies under different K values in Table 3, when K = 7, the central frequencies differ too much and there is an under-decomposition problem; when K = 9, similar frequencies appear and over-decomposition occurs. Therefore, K = 8 is taken, and the IMF spectrogram is obtained according to the decomposition results.

[0142] According to the VMD decomposition results, the cross-correlation coefficients between adjacent IMFs are calculated, and the results are shown in Table 1.

[0143] Table 1 Cross-correlation Coefficients of IMFs

[0144]

[0145] According to Table 1, R(IMF6, IMF7) is the smallest, and the correlation between IMF6 and IMF7 is the smallest. Therefore, IMF 1-6 is selected to be reconstructed into the low-frequency power, and the target power value of the lithium battery is obtained after amplitude limiting The remaining power is high-frequency power, and the target power value of the supercapacitor can be obtained after amplitude limiting. The target power waveform is as Figure 12 , the lithium battery undertakes the power with large energy, and the supercapacitor undertakes the power with frequent fluctuations and small energy.

[0146] The state of charge of the lithium battery and the supercapacitor is as Figure 13 shown;

[0147] The power waveforms of the left and right power supply arms before and after compensation are as Figure 14 shown. Through the energy flow of the hybrid energy storage type RPC system, the load imbalance degree of the left and right power supply arms is reduced.

[0148] According to Figures 12 - 14 , it can be obtained that from 0 to 99 s, for most load conditions, P z ≥5 MW, the system is in the peak shaving and energy releasing mode, the hybrid energy storage system releases energy, the SOC of the lithium battery bat is in a decreasing state, and the SOC of the supercapacitor sc fluctuates and responds to the charge and discharge of the target power demand; from 104 s to 215 s, the system is in the standby mode, the energy storage system does not work, and the RPC works to transfer the power of the left and right power supply arms; from 215 s to 300 s, P z ≤0 MW, the system is in the valley filling and energy storage mode, the hybrid energy storage system stores energy, and the SOC of the lithium battery bat is in an increasing state.

[0149] P T , P' T are the required power values before and after compensation of the traction substation respectively. According to Figure 15 , it can be obtained that in the peak shaving and energy releasing mode, the system realizes the release of regenerative braking energy, effectively reducing the compensation power of the traction substation; in the valley filling and energy storage mode, the system realizes the recovery of regenerative braking energy, regulating the railway energy consumption.

[0150] The comparison of the negative sequence currents I ﹣ , I' ﹣ before and after compensation is as Figure 16 , the negative sequence current drops significantly compared with that without compensation, and the maximum value of the negative sequence current drops from 21.21 A to 8.33 A.

[0151] Conclusion:

[0152] The hybrid energy storage type RPC system for electrified railways is studied, and the following conclusions are obtained for its system control strategy, energy management strategy and power distribution strategy:

[0153] 1) Analyzed the energy flow relationship of the hybrid energy storage type RPC system, and gave the system control strategy, including the RPC control strategy and the hybrid energy storage system control strategy, to realize the storage and release of regenerative braking energy, reduce the train energy consumption demand, and promote the low-carbon development of electrified railways;

[0154] 2) Obtained the energy management strategy of the hybrid energy storage type RPC system. The charging and discharging of the hybrid energy storage system are divided into three modes: peak shaving and energy releasing, valley filling and energy storage, and standby, to realize the energy flow between the energy storage system and the traction power supply system;

[0155] 3) Aiming at the internal power distribution of the hybrid energy storage system, proposed a VMD-based hybrid energy storage power distribution algorithm. Determine the high-frequency and low-frequency boundaries according to the IMF cross-correlation coefficient, effectively realize the high-frequency and low-frequency distribution of the energy storage power, which is beneficial to improving the utilization rate of regenerative braking energy and prolonging the life of the energy storage medium.

[0156] The above has made a detailed description of the preferred implementation mode of the present invention. However, the present invention is not limited to the above implementation mode. Within the knowledge scope of those skilled in the art, various changes can be made without departing from the purpose of the present invention.

[0157] Many other changes and modifications can be made without departing from the concept and scope of the present invention. It should be understood that the present invention is not limited to a specific implementation mode, and the scope of the present invention is defined by the appended claims.

Claims

1. An energy management method for a hybrid energy storage system connected to a traction power supply system, characterized in that, The method includes: S1. Determine the peak shaving threshold \(P\) of the traction power supply system tra and the valley filling threshold \(P\) reg , and determine the maximum charging value \(P\) of the lithium battery in the hybrid energy storage system bat_cmax , the maximum discharging value \(P\) bat_dmax , the maximum charging value \(P\) of the super capacitor sc_cmax , the maximum discharging value \(P\) sc_dmax , the change range of the state of charge (SOC) of the lithium battery \(SOC\) bat_min , \(SOC\) bat_max and the change range of the state of charge (SOC) of the super capacitor \(SOC\) sc_min , \(SOC\) sc_max ; S2. Detect the effective values of the voltages U L and U R of the left and right power supply arms on the traction side in the traction power supply system, and the effective values of the currents I L and I R . Detect the state of charge SOC bat and SOC sc of the supercapacitor and the lithium battery in the hybrid energy storage system; S3. Calculate the active power P of the left and right power supply arms on the traction side based on the effective values of the voltage and current of the left and right power supply arms on the traction side L , P R . Obtain the total active power P of the load based on the active power of the left and right power supply arms on the traction side Z ; Obtain the maximum charging power P of the total active power hybrid energy storage system based on the maximum charging value and discharging value of the lithium battery, and the maximum charging value and discharging value of the super capacitor es_cmax , and the maximum discharging power P es_dmax ; S4. Based on the obtained peak shaving threshold and valley filling threshold, the change range of the state of charge (SOC) of the lithium battery and the supercapacitor, the state of charge of the supercapacitor and the lithium battery, the total active power of the load, the maximum charging value and the maximum discharging value of the total active hybrid energy storage system, construct the charge and discharge strategy of the hybrid energy storage system, and determine the target power value of the energy storage system according to the charge and discharge strategy of the hybrid energy storage system In step S4, the charge-discharge strategy of the hybrid energy storage system is divided into three modes: (1) Valley filling energy storage mode When P Z ≤P reg and SOC bat or SOC sc is less than their respective state of charge maximum values, the energy storage system absorbs the regenerative braking energy, which can be specifically divided into the following two cases. When P Z <P es_cmax , the energy absorbed by the energy storage system is |P es_cmax |, and the remaining energy is sent back to the power grid; when P reg >P Z ≥P es_cmax , the energy absorbed by the energy storage system is |P Z |, and no energy is sent back to the power grid; therefore, the target power value P*es of the energy storage system in this mode is: (2) Standby mode When P reg <P Z <P tra or SOC bat 、SOC sc is not within the normal operating range, the energy storage system is in standby mode, and its target power value P*es is: (3) Peak shaving and energy releasing mode When P Z ≥ P tra and SOC bat or SOC sc is greater than their respective state of charge minimum values, the energy storage system releases energy, reduces the peak load, and decreases the traction transformer load demand. The remaining energy is provided by the power grid. Specifically, it is divided into the following two cases. When P tra ≤ P Z < P tra + P es_dmax , the energy released by the energy storage system is |P tra - P Z |. When P Z ≥ P tra + P es_dmax , the energy storage system releases the maximum energy P es_dmax . Therefore, the target power value P*es of the energy storage system in the peak shaving and energy release mode is: S5. Decompose using the VMD decomposition algorithm Obtain K IMFs and calculate the cross-correlation coefficient, and reconstruct the target power value of the lithium battery according to the cross-correlation coefficient and the target power value of the supercapacitor That is, the energy management strategy for the hybrid energy storage system to access the traction power supply system is realized; In step S5, the specific calculation of the cross-correlation coefficient is as follows: Determine the cross-correlation coefficient R(IMF k , IMF k+1 ) as the cross-correlation coefficient between the connected IMF k and IMF k+1 : Where N is the number of sampling points; are the standard deviation and mean of the k-th IMF respectively; are the standard deviation and mean of the (k + 1)-th IMF respectively. Select the IMF corresponding to the minimum cross-correlation coefficient k as the boundary to reconstruct the IMF 1-k into the low-frequency power P of the energy storage system esL , and compare it with the maximum value of the lithium battery charge and discharge to obtain the target power value of the lithium battery as: The target power of the supercapacitor is: In step S3, the active power P of the left and right power supply arms on the traction side L 、P R The calculation formula is: P L =U L I L ;P R =U R I R ;i L 、i R are the currents of the left and right power supply arms after compensation respectively; U L 、U R are the effective values of the voltages of the left and right power supply arms respectively; The total active power P of the load Z The calculation formula is: P Z =P L +P R ;The maximum charging value P of the total active power hybrid energy storage system es_cmax The calculation formula is: P es_cmax =P bat_cmax +P sc_cmax ;The maximum discharging value P es_dmax The calculation formula is: P es_dmax =P bat_dmax +P sc_dmax ; In step S5, the VMD decomposition algorithm is used for decomposition Specifically: Using the VMD decomposition algorithm, i.e., constrained variational model decomposition As shown in Equation (4): Where: {u k} is the total active power of the load IMF set obtained by VMD decomposition; {ω k} is the IMF center frequency set; K is the number of IMFs; δ(t) is the impulse function; The optimal solution of the above variational problem is solved by introducing the Lagrange multiplier function, and its expression is: In the formula, λ is the Lagrange multiplication operator; a is the quadratic penalty factor; The alternating multiplier algorithm is used to solve Equation (4), and u is updated iteratively k , ω k and λ to obtain the IMF, and then the modal components and their central frequencies are obtained through Fourier transform as follows: where are respectively λ(ω), u i the sequences after Fourier transform of is the Wiener filter of the current remaining amount; is the current IMF center frequency; The update formula of the Lagrange multiplication operator is: In the iterative solution process, each IMF and its center frequency are continuously updated until the iteration stop condition is met and the entire loop ends. The iteration stop condition is: In the formula, e is the given determination accuracy, e > 0; That is, K IMFs are obtained.

Images