Control method and system of traction converter

Abstract

The invention relates to the technical field of traction converter control, in particular to a control method and system for a traction converter, and the method comprises the steps: constructing a d-axis equivalent circuit and a q-axis equivalent circuit of a motor, constructing a current prediction model, and obtaining a d-axis output current and a q-axis output current through the operation of the current prediction model; determining a d-axis reference current and a q-axis reference current corresponding to the maximum torque current ratio control mode or the weak magnetic control mode according to the rotating speed of the motor; and constructing an overall cost function, substituting the d-axis output current, the q-axis output current, the d-axis reference current and the q-axis reference current into the overall cost function, selecting an optimal overall cost function, and controlling on and off of an inverter switching device in the traction converter according to the optimal overall cost function. According to the invention, the problem that the performance and stability of the long-stator linear synchronous motor become poor due to a long cable and double-end power supply under the condition of double-end power supply is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of traction converter control, in particular to a control method and system of a traction converter. BACKGROUND

[0002] With the continuous development of the transportation system, high-speed maglev trains break through the speed limit of traditional wheel-rail trains and are widely used in the field of rail transportation.

[0003] To realize high-speed maglev traction power supply, a traction converter is arranged in a ground traction substation, and a long-stator linear synchronous motor is arranged in a high-speed maglev train, the motor stator side of which is arranged on the line, and the traction converter outputs three-phase variable-frequency variable-voltage alternating current to feed power to the motor stator side through a long cable, and single-end or double-end power supply modes are adopted according to the train running speed.

[0004] However, in the current power supply system, the long cable and double-end power supply mode may cause voltage loss and current fluctuation, which makes the performance and stability of the motor worse, and how to control the traction converter to cope with these influences has not yet been proposed an effective solution. SUMMARY

[0005] Embodiments of the present application provide a control method and system of a traction converter to at least solve the problem that in related technologies, long cable and double-end power supply cause the performance and stability of the long-stator linear synchronous motor to be worse in the double-end power supply case.

[0006] In a first aspect, embodiments of the present application provide a control method of a traction converter, comprising: a circuit construction step of constructing a d-axis equivalent circuit and a q-axis equivalent circuit of a motor, generating a voltage equation of the motor based on the d-axis equivalent circuit and the q-axis equivalent circuit; an output current calculation step of discretizing the voltage equation, constructing a current prediction model, and obtaining d-axis output current and q-axis output current through operation of the current prediction model; a reference current calculation step of selecting a maximum torque current ratio control mode or a field weakening control mode according to the speed of the motor, and determining the corresponding d-axis reference current and q-axis reference current in the maximum torque current ratio control mode or the field weakening control mode; The voltage obtaining step, constructing a total cost function, substituting the d-axis output current, the q-axis output current, the d-axis reference current and the q-axis reference current into the total cost function to obtain a total cost function calculation value, rolling updating the total cost function calculation values corresponding to multiple time points, and selecting the minimum value among the multiple total cost function calculation values; wherein the total cost function is the sum of a current tracking cost function, an average switching frequency limitation cost function, a line loss compensation cost function and a midpoint voltage deviation adjustment cost function; The converter control step generates corresponding pulse signals according to the numerical values of the current tracking cost function, the average switching frequency limitation cost function, the line loss compensation cost function and the midpoint voltage deviation adjustment cost function corresponding to the minimum value among the multiple total cost function calculation values, and controls the on-off of the inverter switching devices in the traction converter.

[0007] In some embodiments, the current prediction model is: ; ; wherein, is a first predicted current of the d-axis at the next time point, is a second predicted current of the d-axis at the next time point, is an equivalent resistance, is a sampling period of the controller, is an inductance of the d-axis equivalent circuit, is a first voltage of the d-axis at the current time point, is an angular velocity of the long-stator linear synchronous motor, is a first predicted current of the q-axis at the current time point, is a first predicted current of the q-axis at the next time point, is a second predicted current of the q-axis at the next time point, is an inductance of the q-axis equivalent circuit, is a first voltage of the q-axis at the current time point, is a first predicted current of the d-axis at the current time point, is a rotor flux, and k is a time point in a discrete time domain.

[0008] In some embodiments, the output current calculation step further comprises: based on the current prediction model, performing time delay compensation to obtain compensated d-axis first actual predicted current, compensated d-axis second actual predicted current, compensated q-axis first actual predicted current and compensated q-axis second actual predicted current; calculating an average of the compensated d-axis first actual predicted current and the compensated d-axis second actual predicted current to obtain the d-axis output current; calculating an average of the compensated q-axis first actual predicted current and the compensated q-axis second actual predicted current to obtain the q-axis output current.

[0009] In some embodiments, the reference current calculation step further comprises: constructing a coordinate system with the d-axis output current as the abscissa and the q-axis output current as the ordinate, constructing the current maximum constraint condition and the voltage limit constraint condition of the motor, and generating a reference current trajectory in the coordinate system according to the current maximum constraint condition and the voltage limit constraint condition; wherein the reference current trajectory comprises a first segment of current trajectory and a second segment of current trajectory, the first segment of current trajectory being a maximum torque current ratio curve from the origin to the intersection point of the current maximum constraint condition and the voltage limit constraint condition of the motor on the coordinate system when the first rated angular velocity is reached, and the second segment of current trajectory being a current limit curve from the intersection point of the current maximum constraint condition and the voltage limit constraint condition of the motor on the coordinate system when the first rated angular velocity is reached to the intersection point of the current maximum constraint condition and the voltage limit constraint condition of the motor on the coordinate system when the second rated angular velocity is reached; determining whether to run along the first segment of current trajectory or along the second segment of current trajectory according to the position of the speed of the motor in the reference current trajectory; determining to run the maximum torque current ratio control mode when running along the first segment of current trajectory; determining to run the field weakening control mode when running along the second segment of current trajectory.

[0010] In some embodiments, the d-axis reference current corresponding to the maximum torque current ratio control mode is: wherein, is the stator current of the motor, is the inductance of the d-axis equivalent circuit, is the inductance of the q-axis equivalent circuit, is the rotor flux; the q-axis current reference corresponding to the maximum torque current ratio control mode is:

[0011] In some embodiments, the calculation formula of the d-axis reference current corresponding to the field weakening control mode is:​​​​ ; wherein is the inductance of the d-axis equivalent circuit, is the inductance of the q-axis equivalent circuit, is the rotor flux, is the stator reference current of the motor, obtained according to a speed outer loop proportional-integral controller, is the angular speed of the motor, is the maximum stator voltage.

[0012] In some embodiments, the calculation formula of the current tracking cost function is: ; wherein, is the weight coefficient of the current tracking cost function, k is the time in the discrete time domain, is the d-axis output current at k+2, is the q-axis output current at k+2, is the d-axis reference current at k+2, is the q-axis reference current at k+2.

[0013] In some embodiments, the calculation of the average switching frequency limitation cost function is: .

[0014] wherein, is the weight coefficient of the average switching frequency limitation cost function, k is the time in the discrete time domain, is the switching sequence at k+2, is the switching sequence at k+1, is the energy loss at k+2, is the energy loss at k+2, x is the three-phase of the long-stator linear synchronous motor, including the a-phase, the b-phase and the c-phase.

[0015] In some embodiments, the calculation formula of the line loss compensation cost function is: ; wherein, is the weight coefficient of the line loss compensation cost function, m is the first factor of the adjustment non-linear influence of the line impedance parameter cost function, n is the second factor of the adjustment non-linear influence of the line impedance parameter cost function, s is the distance from the train to the traction substation, and f is the stator frequency.

[0016] In a second aspect, the embodiments of the present application provide a control system of a traction converter, comprising: The circuit construction module is configured to construct a d-axis equivalent circuit and a q-axis equivalent circuit of the motor, and generate a voltage equation of the motor based on the d-axis equivalent circuit and the q-axis equivalent circuit; The output current calculation module is configured to perform discretization processing on the voltage equation, construct a current prediction model, and obtain a d-axis output current and a q-axis output current through calculation of the current prediction model. The reference current calculation module is configured to select a maximum torque current ratio control mode or a field weakening control mode according to a speed of the motor, and determine a corresponding d-axis reference current and a q-axis reference current in the maximum torque current ratio control mode or the field weakening control mode. The voltage acquisition module is configured to construct a total cost function, substitute the d-axis output current, the q-axis output current, the d-axis reference current and the q-axis reference current into the total cost function to obtain a total cost function calculation value, rollingly update the total cost function calculation values corresponding to a plurality of time points, and select a minimum value in the plurality of total cost function calculation values, wherein the total cost function is a sum of a current tracking cost function, an average switching frequency limitation cost function, a line loss compensation cost function and a midpoint voltage deviation adjustment cost function. The converter control module is configured to generate corresponding pulse signals according to numerical values of the current tracking cost function, the average switching frequency limitation cost function, the line loss compensation cost function and the midpoint voltage deviation adjustment cost function corresponding to the minimum value in the plurality of total cost function calculation values, and control on-off of inverter switching devices in the traction converter.

[0017] Compared with the related art, the control method and system of the traction converter provided by the embodiments of the present application comprehensively consider long cable and double-end power supply, form optimal switching states of the switching devices in the inverter, and further control input voltage and input current of the traction converter, thereby solving the problem of poor performance and stability of the long-stator linear synchronous motor caused by long cable and double-end power supply in the double-end power supply case.

[0018] Details of one or more embodiments of the present application are presented in the following drawings and description to make other features, objects and advantages of the present application more clear and easy to understand. BRIEF DESCRIPTION OF DRAWINGS

[0019] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The schematic embodiments of the present application and the description thereof are used to explain the present application, and do not constitute improper limitation on the present application. In the drawings: Figure 1 is a structural block diagram of a high-speed maglev traction power supply system according to the related art; Figure 2is a two-level cascaded topology according to the related art; Figure 3 is a diode clamped three-level topology according to the related art; Figure 4 is a three-level active neutral point clamped topology according to the related art; Figure 5 is a segmented case in a hybrid modulation strategy according to the related art; Figure 6 is a model predictive control flowchart according to the related art; Figure 7 is a value function based switch optimization flowchart according to the related art; Figure 8 is an equivalent circuit diagram of a high-speed maglev train in a double-ended parallel power supply mode; Figure 9 is a d-axis equivalent circuit diagram of a high-speed maglev train in a double-ended parallel power supply mode; Figure 10 is a q-axis equivalent circuit diagram of a high-speed maglev train in a double-ended parallel power supply mode; Figure 11 is a reference trajectory diagram of a control method of a traction converter according to an embodiment of the present application; Figure 12 is a control framework diagram of a long-stator linear synchronous motor of a high-speed maglev train according to an embodiment of the present application; Figure 13 is a flowchart of a control method of a traction converter according to an embodiment of the present application; Figure 14 is a structural block diagram of a control system of a traction converter according to an embodiment of the present application.

[0020] In the figure: 1401, circuit construction module; 1402, output current calculation module; 1403, reference current calculation module; 1404, voltage acquisition module; 1405, converter control module. DETAILED DESCRIPTION

[0021] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application is described and explained below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0022] It is apparent that the drawings in the following description merely show some examples or embodiments of the present application, and the present application can be applied to other similar situations without creative labor by those skilled in the art based on these drawings. In addition, it can be understood that although the efforts made in the development process can be complex and lengthy, some modifications, such as design, manufacture or production, etc. based on the technical content disclosed in the present application, are only routine technical means for those skilled in the art related to the content disclosed in the present application, and should not be understood as insufficient disclosure of the content disclosed in the present application.

[0023] Reference to "an embodiment" in this application means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily mutually exclusive of one another. It is explicitly contemplated that embodiments described herein can be combined with other embodiments in a non- conflicting manner.

[0024] Unless otherwise defined, technical terms or scientific terms used in the present application shall have the ordinary meaning understood by one of ordinary skill in the art to which the present application pertains. The terms "a", "an", "one", "the", and similar terms in the present application do not denote a limitation on quantity but rather denote the presence of at least one of the referenced item. The terms "including", "comprising", "having" and their variants in the present application are intended to cover non-exclusive inclusions; for example, processes, methods, systems, products, or devices that comprise a list of steps or modules (units) are not limited to the listed steps or units but can also include other steps or units not listed, or can also include other steps or units inherent in the processes, methods, products, or devices. The terms "connected", "connected", "coupled" and similar terms in the present application are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The term "multiple" in the present application means two or more. The term "and / or" describes the association relationship between the associated objects, which means that there can be three relationships, for example, "A and / or B" can mean that A exists alone, A and B exist together, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects. The terms "first", "second", "third", etc. in the present application are merely to distinguish similar objects, and do not represent a specific order for the objects.

[0025] In the development process of rail transit, traditional wheel-rail trains have always faced technical bottlenecks such as frictional resistance and wheel-rail wear due to physical contact between the wheel and the track, which has made it difficult to overcome the speed limit of 400 km / h. High-speed maglev has broken through the speed limit of traditional wheel-rail trains, significantly improving the running speed and stability of trains by realizing non-contact transmission between trains, tracks, and power supply systems. The progress of high-speed maglev technology has had a profound impact on the transportation system.

[0026] In addition, maglev technology optimizes energy efficiency and maintenance efficiency by reducing friction and wear, and effectively reduces operating costs. With the gradual maturity of related technologies, maglev trains not only have broad application potential in short-distance and urban rail transit, but also provide new solutions for long-distance high-speed transportation, driving the modernization of global transportation systems.

[0027] In the high-speed maglev traction drive system, the traction converter is a core key device, which is placed in the ground traction substation. Compared with the wheel-rail train traction converter, the high-speed maglev traction converter has significant differences in topology and control.

[0028] The wheel-rail train traction converter is installed at the bottom of the train, which converts single-phase alternating current into three-phase variable-frequency variable-voltage alternating current to drive the rotating motor at close range. The output capacity of such converters is usually within 7 MVA.

[0029] The typical traction power supply system of high-speed maglev is shown in Figure 1 The traction converter is placed in the ground substation. Converter A-1 and converter A-2 are installed in substation A, and converter B-1 and converter B-2 are installed in substation B. The output capacity of the traction converter is usually required to be more than 20 MVA.

[0030] High-speed maglev trains use long-stator linear synchronous motors, which are divided into left and right long-stator segmented windings. The long-stator segmented winding is the basic unit of the motor stator, and the stator side of the motor is fixed to the line. When the train passes, the stator winding is magnetized and interacts with the train suspension and propulsion system to drive the train.

[0031] The substation is responsible for converting grid power into three-phase variable-frequency variable-voltage alternating current required by the maglev train to power the long-stator segmented winding. The output three-phase variable-frequency variable-voltage alternating current of the traction converter in the substation is fed to the stator side of the motor through a 30-50 km feeder cable.

[0032] The feeder cable is the power transmission channel from the substation to the long-stator segmented winding, which needs to overcome the resistance and inductance of the long cable, including , 、 and , the inductance includes 、 、 and .

[0033] Figure 1 The capacitance in the formula (1) is an element for filtering and compensation, which is used to offset the inductance of long cable and long stator segmented winding, optimize the power factor of power transmission, and make the power more efficient transmission to the stator side. The capacitance includes 、 、 and .

[0034] In order to reduce energy loss, the long stator line is divided into independent stator segments due to the long line. Only when the train runs to a certain stator segment area, the stator segment will be connected to the power supply and powered through the substation.

[0035] The high-speed maglev typical traction power supply system adopts segmented power supply mode. When the train speed is low, it runs in single-end power supply mode, that is, only one end of the line converter is used to power the stator segment, for example, only converter A-1 or converter A-2 is needed. When the train runs at high speed, it will adopt double-end power supply mode, that is, two sets of converters are connected in parallel to power the stator segment, for example, converter A-1 and converter B-1 are used in cooperation, and converter A-2 and converter B-2 are used in cooperation. The double-end parallel power supply can improve the capacity, reliability and efficiency of the system.

[0036] At present, the Japanese Yamanashi line and the Shanghai maglev line are relatively mature high-speed maglev lines. As shown in Figure 2 , the inverter side of the traction converter system of the Japanese Yamanashi line is a two-level cascade topology. At low speed, the main circuit is switched to the working mode of two half-bridges in parallel, and at high speed, the main circuit is switched to the working mode of four H-bridges in series, and is output through a transformer. As shown in Figure 3 , the inverter side of the traction converter system of the Shanghai maglev line is a diode clamped three-level NPC (Neutral Point Clamped, NPC) topology, which also has an output transformer and needs to switch the high-low speed main circuit. At low speed, the converter system works in direct mode, and two inverters are used in parallel. At high speed, the converter system works in transformer mode, and two inverters are used in series. Although the NPC topology has advantages in reducing voltage stress and harmonic content compared with the two-level cascade topology, the uneven loss of the inner and outer side switching devices limits the maximum output capacity of the converter, and further restricts the highest running speed of the train. Therefore, in order to solve this problem and meet the demand of 600km / h high-speed maglev traction, the topology needs to be improved.

[0037] The inverter side of my country's independently developed 600 km / h high-speed maglev traction converter adopts a three-level active neutral point clamped (ANPC) topology, such as... Figure 4 As shown. Compared to the traditional NPC topology, the ANPC converter not only inherits the advantages of the NPC topology in reducing voltage stress and harmonic content, but also greatly enriches the diversity of commutation methods by introducing more zero-level switching states and flexible switching state combinations. By optimizing the commutation strategy, the ANPC converter can effectively balance the loss distribution of each switching device, thereby achieving a higher output capacity while maintaining the same power rating as the NPC converter. This improvement gives the ANPC converter significant advantages in improving system performance and expanding application areas.

[0038] In the traction inverter system of high-speed maglev trains, excessively high switching frequencies increase switching losses and are limited by heat dissipation. Therefore, the switching frequency is typically controlled within a low range of several hundred hertz. Since the output frequency of the traction inverter varies considerably, the carrier ratio also changes accordingly. Therefore, as the inverter output frequency increases, the carrier ratio will inevitably decrease gradually due to the limitation of the switching frequency.

[0039] When the inverter operates at a low frequency, the carrier ratio can still remain relatively large, and the control effect of asynchronous modulation and synchronous modulation is not significantly different. However, as the inverter output frequency increases, the carrier ratio continuously decreases. If asynchronous modulation continues to be used, it will lead to asymmetry in the positive and negative half-cycles of the output waveform, resulting in asymmetry in the three-phase voltage waveform, especially noticeable at low carrier ratios. Therefore, when the carrier ratio is low, synchronous modulation, which ensures the symmetry of the three-phase voltage waveform, is typically used. Synchronous modulation, by maintaining a constant carrier ratio, ensures that the three-phase voltage waveform output by the inverter remains symmetrical at lower switching frequencies and effectively reduces low-order harmonic content. However, if a single carrier ratio synchronous modulation strategy is used for a long period, the highest switching frequency may not be fully utilized, especially at lower switching frequencies.

[0040] To address this issue, existing technologies employ a hybrid modulation strategy, combining asynchronous and synchronous modulation across the entire speed range, such as... Figure 5 As shown in Table 1, by flexibly switching the modulation method according to the changes in output frequency, the symmetry and stability of the three-phase voltage can be maintained in the low-frequency band, while the switching frequency can be fully utilized in the high-frequency band, thus ensuring that the system maintains excellent performance in all frequency ranges. The configuration of the hybrid carrier segment is shown in Table 1.

[0041] Table 1 Hybrid Carrier Segment Configuration Table

[0042] The use of hybrid modulation strategies can effectively improve the output waveform quality of inverters across the entire speed domain. However, during system operation, different modulation strategies are required for different carrier ratios, inevitably leading to switching issues between asynchronous and synchronous modulation, as well as between different synchronous modulation modes. During switching between these modulation modes, abrupt changes in voltage phase at the switching moment can cause discontinuities in the fundamental voltage, resulting in voltage or current abrupt changes and surges. These abrupt changes can cause severe vibrations in the motor under hybrid modulation modes, affecting the stable operation of the system. Ensuring smooth switching of the motor under hybrid modulation modes and avoiding the shocks and vibrations caused by switching is crucial for improving the stability and operational quality of the motor system. However, achieving this goal faces certain difficulties because precise switching between different modulation modes is required. This not only necessitates real-time monitoring of voltage and current changes but also requires complex algorithms to determine the optimal switching timing to ensure a smooth system transition. The high algorithmic complexity during the switching process increases the implementation difficulty and computational burden of the control system.

[0043] Model Predictive Control (MPC) is an advanced control strategy based on mathematical models. It achieves precise control of dynamic systems by optimizing the control input online within each control cycle. The basic idea of ​​MPC is to use the system's mathematical model to predict its behavior over a future period and then solve for the optimal control input based on the prediction, thereby optimizing system performance. MPC is not only applicable to linear systems but also effectively handles nonlinear systems, showing significant advantages, especially in complex systems involving multivariable coupling, such as... Figure 6 As shown.

[0044] In MPC applications, it is often necessary to predict important variables such as current and voltage based on the dynamic behavior of the system. To accurately predict the system's current behavior, a series of predictive models are typically used in practical applications. These models estimate the current value at future times by processing data from the current moment. By selecting and combining multiple predictive models, MPC can flexibly switch modulation methods according to changes in the output frequency within different control cycles, thereby optimizing system performance.

[0045] Specifically, in the current prediction process of MPC, the predicted current value , It is unknown, but is generally considered to be true when the sampling period is small: .

[0046] in, The reference current at time k+1 is Components of the coordinate axis, Reference current at k+1 time Components of the coordinate axis, Predicted current at k+1 time Components of the coordinate axis, Predicted current at k+1 time Components of the coordinate axis.

[0047] The reference signal transmission process will cause errors due to delay, and the Lagrange extrapolation method is used to calculate the reference current at the next time according to the reference current at the current time and the reference current at the previous time, and the calculation formula is .

[0048] Wherein, Reference signal at the next time, Reference current at the current time, Reference current at the previous time.

[0049] Further rolling optimization makes the actual current as close as possible to the given current. In each sampling period, first, the actual current and the given current value are obtained, and the current is predicted using the prediction model for 27 basic voltage vectors. Then, based on the loss prediction model and the loss prediction value function, the corresponding switching state is selected, and the target function is used for rolling optimization, so as to determine the switching state with the optimal dynamic characteristics. On the basis of completing the model prediction, a cost function J is constructed, which comprehensively considers the working condition requirements and the overall system performance. The value function J aims to optimize the overall operation performance of the system by balancing various control objectives, such as working condition adaptability, system energy efficiency, and loss minimization, and the specific process is as shown in Figure 7 The online optimization of the ANPC traction converter low-loss control scheme is carried out, and the corresponding formula is: . Wherein, Total value function, Current tracking error cost function, Midpoint potential offset cost function, Switching frequency cost function, Switching tube loss cost function.

[0050] Specifically, the calculation formula of the current tracking error cost function is: .

[0051] Wherein, Optimization target weight coefficient of current tracking error cost function, Components of the coordinate axis of the predicted current at k+1 time, Components of the coordinate axis of the reference current at k+1 time. Components of the coordinate axis of the reference current at k+1 time. Components of the coordinate axis of the reference current at k+1 time.

[0052] Specifically, the calculation formula of the midpoint potential offset cost function is: .

[0053] Wherein, is the optimization target weight coefficient of the midpoint potential offset cost function, is the midpoint voltage.

[0054] Specifically, the calculation formula of the switching frequency cost function is: .

[0055] Wherein, is the optimization target weight coefficient of the switching frequency cost function, is the switching state.

[0056] Specifically, the calculation formula of the switching tube loss cost function is: .

[0057] Wherein, is the optimization target weight coefficient of the switching tube loss cost function, is the switching tube loss, wherein is the number of switching tubes in one phase.

[0058] The technology of driving motor based on model predictive control has certain limitations in practical application. First, the existing control method is limited in considering factors, and cannot fully optimize the smooth driving in the full speed range, which may cause performance fluctuations at high speed, affecting the stability and efficiency of the motor. Especially in high speed or heavy load conditions, the dynamic response of the motor may not be optimal, thereby affecting the overall performance.

[0059] Second, the existing MPC control method is mainly aimed at conventional motor applications, and lacks design for high-speed maglev long-stator linear motors. This type of motor has complex magnetic field characteristics and high dynamic response requirements, and the existing control scheme may not be accurate enough when dealing with rapid changes in high-speed maglev systems, making it difficult to achieve ideal control effect.

[0060] In addition, the existing technology does not fully consider key factors in high-speed maglev scenarios, such as long cables and double-ended power supply systems. Long cables may cause voltage loss, signal delay and other problems, while double-ended power supply systems may cause current fluctuations, which all affect the performance and control stability of the motor. Therefore, the existing MPC control method cannot fully meet the special application requirements, limiting its application effect in high-speed maglev long-stator linear motors.

[0061] To solve the above problems, the present application provides a control method and system for traction converter, as shown in Figure 13 The control method comprises: The circuit construction step S101 constructs the d-axis equivalent circuit and the q-axis equivalent circuit of the long-stator linear synchronous motor in the double-ended power supply mode, and generates the voltage equation of the long-stator linear synchronous motor based on the d-axis equivalent circuit and the q-axis equivalent circuit.

[0062] By constructing the d-axis and q-axis equivalent circuits and generating the voltage equation, the electrical characteristics of the long-stator linear synchronous motor in the double-ended power supply mode can be accurately mathematically described, providing an accurate model basis for subsequent control algorithm design and improving the pertinence and accuracy of control.

[0063] The output current calculation step S102 discretizes the voltage equation, constructs a current prediction model, and obtains the d-axis output current and the q-axis output current through current prediction model operation.

[0064] The reference current calculation step S103 selects the maximum torque current ratio control mode or the field weakening control mode according to the speed of the long-stator linear synchronous motor, and determines the corresponding d-axis reference current and q-axis reference current in the maximum torque current ratio control mode or the field weakening control mode.

[0065] According to the motor speed, the maximum torque current ratio control mode or the field weakening control mode is automatically selected, which can make the motor maintain good performance under different operating conditions. The maximum torque current ratio control mode is used at low speed to fully utilize the torque output capability of the motor, and the field weakening control mode is switched to at high speed to expand the speed range of the motor, thereby improving the operating efficiency and reliability of the motor.

[0066] The voltage acquisition step S104 constructs a total cost function, substitutes the d-axis output current, the q-axis output current, the d-axis reference current and the q-axis reference current into the total cost function, obtains the total cost function calculation value, rolls the total cost function calculation values corresponding to multiple time points, and selects the minimum value among the multiple total cost function calculation values.

[0067] The total cost function is the sum of the current tracking cost function, the average switching frequency limitation cost function, the line loss compensation cost function and the midpoint voltage deviation adjustment cost function.

[0068] The converter control step S105 generates corresponding pulse signals according to the numerical value of the current tracking cost function, the numerical value of the average switching frequency limitation cost function, the numerical value of the line loss compensation cost function and the numerical value of the midpoint voltage deviation adjustment cost function corresponding to the minimum value among the multiple total cost function calculation values, and controls the on-off of the inverter switching devices in the traction converter.

[0069] By constructing the overall cost function and optimizing the inverter switch state based on it, the key factors affecting the system performance are comprehensively considered, which can ensure the current tracking accuracy, reduce the switching frequency to reduce the switching loss, compensate the line loss to improve the system efficiency, and thus improve the comprehensive performance of the traction inverter control system of the high-speed maglev train.

[0070] The equivalent circuit of the high-speed maglev train in the double-end parallel power supply mode in the three-phase coordinate system is shown in Figure 8 . 、 、 and 、 、 respectively represent the output voltage of the two-end AC converter, 、 、 and 、 、 respectively represent the output current of the two-end converter, 、 、 is the end voltage of the long-stator linear motor, 、 、 is the three-phase current of the long-stator linear motor, 、 is the resistance of the two-side feeding cable, 、 is the inductance of the two-side feeding cable.

[0071] By Clarke transformation, the equivalent circuit of the long-stator linear synchronous motor in the d-q axis coordinate system is derived, as shown in Figure 9 and Figure 10 .

[0072] Combined with Figures 8-10 , the voltage equation of the long-stator linear synchronous motor is constructed as follows: .

[0073] wherein, is the pole pitch of the long-stator linear synchronous motor, is the speed of the train, , is the rotor magnetic flux, is the excitation current, is the mutual inductance between the rotor and the stator, and p is the number of motor pole pairs.

[0074] In some embodiments, the current prediction model is: , .

[0075] wherein, is the first predicted current of the d-axis at the next time, is the second predicted current of the d-axis at the next time, is the equivalent resistance, is the sampling period of the controller, is the inductance of the equivalent circuit of the d-axis, is the first voltage of the d-axis at the current time, is the angular velocity of the long-stator linear synchronous motor, is the first predicted current of the q-axis at the current time, is the first predicted current of the q-axis at the next time, is the second predicted current of the q-axis at the next time, is the inductance of the equivalent circuit of the q-axis, is the first voltage of the q-axis at the current time, is the first predicted current of the d-axis at the current time, is the rotor flux, and k is the time in the discrete time domain.

[0076] The current prediction model comprehensively considers various electrical characteristic parameters of the motor, such as the equivalent resistance, inductance, voltage, angular velocity, rotor flux, and the mutual relationship therebetween, and can accurately reflect the dynamic change rule of the d-axis current of the long-stator linear synchronous motor in the running process. Compared with some simplified models, the current prediction model can more truly simulate the electromagnetic process inside the motor and provides strong support for accurately predicting the current.

[0077] Accurate current prediction is the basis of many advanced control algorithms. By using the model to predict the current, reliable input data can be provided for subsequent control algorithms, control errors and instability phenomena caused by inaccurate current prediction can be reduced, the reliability and stability of the entire traction inverter control algorithm can be improved, and the safe and stable operation of the high-speed maglev train can be ensured.

[0078] In some embodiments, the output current calculation step further comprises: Based on the current prediction model, through time delay compensation, the compensated first actual predicted current of the d-axis, the compensated second actual predicted current of the d-axis, the compensated first actual predicted current of the q-axis, and the compensated second actual predicted current of the q-axis are obtained.

[0079] Specifically, the compensated first actual predicted current of the d-axis and the compensated first actual predicted current of the q-axis are: .

[0080] The average of the compensated d-axis first actual predicted current and the compensated d-axis second actual predicted current is calculated to obtain a d-axis output current.

[0081] The average of the compensated q-axis first actual predicted current and the compensated q-axis second actual predicted current is calculated to obtain a q-axis output current.

[0082] The two-step prediction and compensation mechanism fully considers various interference factors existing in the actual system. Compared with the method relying only on a single prediction model, it can more effectively eliminate prediction errors. By compensating the predicted current, the compensated d-axis and q-axis first actual predicted currents can more accurately reflect the current values required by the actual operation of the motor, providing a more reliable basis for subsequent motor control, and helping to improve the precision and stability of motor control.

[0083] Since the two sets of converters for double-ended power supply need to provide electric energy with equal amplitude and consistent phase, and the maglev long-stator linear motor is equipped with one set of linear motor on each side of the left and right stators, only the analysis of and is needed, which are representative, and have similar analysis methods. In order to obtain the predicted value of the dq current, the voltage equation of the long-stator linear synchronous motor is discretized. In this process, only the main impedance parameters are retained, and the line impedance parameters will be calculated in the subsequent cost function. Given the short control period of the controller, the first-order Euler method is used for discretization to ensure high calculation accuracy. Thus, the d-axis first predicted current , the d-axis second predicted current , the q-axis first predicted current , and the q-axis second predicted current at the next moment can be obtained as follows: , .

[0084] Considering the sampling and calculation delay of the control algorithm, two-step prediction is usually used for time delay compensation to improve the prediction accuracy of the model. Therefore, the compensated d-axis first actual predicted current , the compensated d-axis second actual predicted current , the compensated q-axis first actual predicted current , and the compensated q-axis second actual predicted current are as follows: , .

[0085] Further, according to the compensated actual predicted current value, the d-axis output current and the q-axis output current are obtained, and the calculation formula is: .

[0086] When the d-axis output current or the q-axis output current at a certain moment is calculated, only the d-axis first actual predicted current and the d-axis second actual predicted current read at the corresponding moment are substituted.

[0087] The application determines the optimal control input by optimizing the cost function. The cost function can be divided into four main parts: current tracking error, average switching frequency limitation, midpoint voltage offset adjustment, and line loss compensation.

[0088] In some embodiments, the calculation formula of the current tracking cost function is: .

[0089] wherein, is the weight coefficient of the current tracking cost function, k is the moment in the discrete time domain, is the d-axis output current at the moment k+2, is the q-axis output current at the moment k+2, is the d-axis reference current at the moment k+2, is the q-axis reference current at the moment k+2.

[0090] By minimizing the current tracking cost function, accurate control of the d-axis and q-axis currents is achieved. It can quickly respond to changes in the reference current, reduce steady-state and dynamic errors, make the motor current closer to the ideal waveform, and improve the torque control accuracy and system dynamic performance.

[0091] Considering the switching loss problem of the inverter, the switching selection is optimized by limiting the amplitude of switching state change to reduce switching loss. In some embodiments, the calculation of the average switching frequency limitation cost function is: .

[0092] wherein, is the weight coefficient of the average switching frequency limitation cost function, k is the moment in the discrete time domain, is the switching sequence at the moment k+2, is the switching sequence at the moment k+1, is the energy loss of the i-th switching tube at the moment k+2, is the energy loss of the i-th switching tube at the moment k+2, x is the three phases of the long-stator linear synchronous motor, including phase a, phase b, and phase c.

[0093] The energy loss can be calculated.

[0094] By balancing the current tracking performance and switching loss, the average switching frequency limit cost function enables the system to maintain high efficiency under different operating conditions. At light load, the switching frequency can be appropriately reduced to reduce loss, and at heavy load, sufficient current control accuracy is ensured to achieve global optimization of system efficiency.

[0095] Since the ANPC topology needs to ensure that the midpoint voltage remains stable throughout the operation process, in some embodiments, the cost function of the midpoint voltage offset is: .

[0096] wherein, is the weight coefficient of the cost function of the midpoint voltage offset, is the midpoint voltage.

[0097] During the operation of the maglev train, the parameters of the feeding cable will change with the distance of the train operation. The longer the distance between the train and the traction substation, the larger the cable parameters. The parameter change of the line resistance can be expressed as: .

[0098] wherein, s is the distance of the train to the traction substation, and f is the stator frequency.

[0099] The parameters of the cable line resistance that change with the distance of the train operation need to be considered. In some embodiments, according to the parameters of the line resistance, the calculation formula of the line loss compensation cost function is: .

[0100] wherein, is the weight coefficient of the line loss compensation cost function, m is the first factor of the adjustment nonlinearity influence of the line impedance parameter cost function, n is the second factor of the adjustment nonlinearity influence of the line impedance parameter cost function, s is the distance of the train to the traction substation, and f is the stator frequency.

[0101] By dynamically adjusting the control strategy when the line impedance changes, the line loss compensation cost function effectively reduces the power loss on the power supply line. This improves energy utilization efficiency, reduces energy waste, and reduces the operating cost of the high-speed maglev train.

[0102] , , and can be set to a numerical value according to actual needs.

[0103] The overall cost function is: .

[0104] By enumerating the cost function values under different voltage vectors, the voltage vector corresponding to the lowest overall cost function is obtained. The voltage vector is sent to the ANPC, and then through the long cable to the long stator linear synchronous motor of the train, to realize information transmission and control.

[0105] Specifically, a plurality of basic voltage vectors are obtained, which are existing data in the art and refer to the voltage state set that the converter can output, representing the state of the switching devices of the inverter of the converter. Under different basic voltage vectors, the d-axis output current, the q-axis output current, the d-axis reference current and the q-axis reference current are obtained, which are input into the overall cost function to generate different overall cost function values corresponding to different basic voltage vectors. A minimum value is selected from these values, and the basic voltage vector corresponding to the minimum value is the optimal voltage vector. The switching sequence of the switching devices of the inverter of the ANPC converter is controlled by sending the optimal voltage vector to the ANPC converter in the form of a pulse signal, and then a unique input current is determined. The input current is transmitted to the long stator linear synchronous motor of the train through the long cable, forming an optimal input voltage, and further realizing energy saving optimization.

[0106] In some embodiments, the reference current calculation step further comprises: A coordinate system is constructed with the d-axis output current as the horizontal coordinate and the q-axis output current as the vertical coordinate. The current maximum constraint condition and the voltage limit constraint condition of the motor are constructed, and the reference current trajectory is generated in the coordinate system according to the current maximum constraint condition and the voltage limit constraint condition.

[0107] The reference current trajectory includes a first current trajectory and a second current trajectory. The first current trajectory is a maximum torque current ratio curve from the origin to the intersection point of the current maximum constraint condition and the voltage limit constraint condition of the motor on the coordinate system when the first rated angular velocity is reached. The second current trajectory is a current limit curve from the intersection point of the current maximum constraint condition and the voltage limit constraint condition of the motor on the coordinate system when the first rated angular velocity is reached to the intersection point of the current maximum constraint condition and the voltage limit constraint condition of the motor on the coordinate system when the second rated angular velocity is reached.

[0108] According to the position of the speed of the motor in the reference current trajectory, it is determined whether to run along the first current trajectory or the second current trajectory.

[0109] When running along the first current trajectory, it is determined to run in the maximum torque current ratio control mode.

[0110] When running along the second current trajectory, it is determined to run in the field weakening control mode.

[0111] By constructing the current maximum constraint condition and the voltage limit constraint condition, the reference current trajectory is generated, which can reasonably divide the operation area of the motor. The control mode to be adopted by the motor under different speed and load conditions is determined, which provides clear guidance for stable operation and efficient control of the motor, and helps to improve the overall performance of the motor control system.

[0112] According to the motor speed, the control mode is automatically judged and switched, realizing the optimized switching of the control mode. The problem of performance degradation or unstable operation of the motor caused by improper selection of the control mode is avoided. The maximum torque current ratio control mode is adopted at low speed to improve the load carrying capacity of the motor, and the field weakening control mode is switched to at high speed to meet the demand of high speed operation of the train, so that the motor can maintain good performance in the whole operation range, and the operation efficiency and reliability of the high speed maglev train are improved.

[0113] In some embodiments, the corresponding d-axis reference current of the maximum torque current ratio control mode is:

[0114] wherein, is the stator current of the long-stator linear synchronous motor, is the inductance of the d-axis equivalent circuit, is the inductance of the q-axis equivalent circuit, is the rotor flux.

[0115] The corresponding q-axis current reference of the maximum torque current ratio control mode is:

[0116] Accurate reference current calculation makes the motor torque output more stable, reduces torque ripple and speed fluctuation. This helps to improve the smoothness and ride comfort of the train operation, while reducing the wear and vibration of mechanical parts, prolonging the service life of the equipment, and enhancing the reliability and stability of the whole traction system.

[0117] In some embodiments, the calculation formula of the corresponding d-axis reference current in the field weakening control mode is:

[0118] wherein is the inductance of the d-axis equivalent circuit, is the inductance of the q-axis equivalent circuit, is the rotor flux, is the stator reference current of the long-stator linear synchronous motor, which is obtained according to the speed outer loop proportional integral controller, ​​​​​The angular velocity of the long-stator linear synchronous motor, The maximum stator voltage.

[0119] At high speed, the increase of back electromotive force is effectively compensated by accurately calculating the d-axis reference current, ensuring that the stator voltage does not exceed the maximum output capability of the inverter. This avoids current distortion and system instability caused by voltage over-limiting, ensuring control accuracy and reliability at high speed.

[0120] The high-speed magnetic levitation long-stator linear synchronous motor (LS-LSM) has a wide speed range and requires field weakening control. In the high-speed region, the linear motor has maximum voltage and current constraints, which are represented as:

[0121] where, is the maximum stator current, is the maximum stator voltage.

[0122] Further, the voltage limit constraint condition is derived to ensure that the motor's voltage output does not exceed the maximum output voltage of the inverter.

[0123] The specific form is:

[0124] According to the maximum current constraint and voltage limit constraint conditions of the linear motor, the current limit circle and voltage limit ellipse of the linear motor are generated in the dq plane, as shown in Figure 11 . In combination with Figure 11 , the red reference current trajectory moves along the OABC curve. In the linear speed range of the LS-LSM, the OA segment represents the current trajectory, which follows the maximum torque per ampere (MTPA) curve to determine the reference current. When the LS-LSM speed reaches the rated angular velocity and the output power reaches the maximum, the current trajectory moves along the AB segment of the current limit circle and increases the demagnetizing current component, thereby achieving further field weakening speed expansion. In this process, the field weakening control strategy is used to determine the reference current value. When the LS-LSM speed reaches , the motor enters the deep field weakening region, and the reference current will move along the BC segment trajectory, following the maximum torque voltage ratio curve for control.

[0125] Specifically, the maximum torque current ratio control method is as follows: When the LS-LSM is in the linear speed range, in order to maximize the traction output of the LS-LSM, the current loss can be minimized while maintaining the same torque, thereby improving the energy efficiency of the system.​​​

[0126] The train traction force equation is constructed: .

[0127] Based on the train traction force equation and the maximum current constraint of the linear motor, a Lagrange auxiliary function is constructed: .

[0128] Wherein, F, is only used as a function for construction, and has no actual physical meaning, is a weight coefficient that can be defined artificially.

[0129] In the Lagrange auxiliary function, the partial derivatives of , and are solved respectively, and the following is obtained: .

[0130] Further solving obtains the d-axis current reference value and the q-axis current reference value.

[0131] The d-axis current reference value is: .

[0132] The q-axis current reference value is: .

[0133] Specifically, the field-weakening control method is as follows: Since the model predictive current control algorithm has no voltage feedback, the feedforward method is considered to calculate the current reference value of the PMSM in the field-weakening state. When entering the field-weakening state, the stator voltage of the motor reaches the maximum amplitude, and the reference current will be at the edge position of the voltage limit ellipse.

[0134] Substitute the maximum current constraint of the motor into the voltage limit constraint condition, and replace with , and the following is obtained: .

[0135] Further solving obtains the field-weakening reference current, which is expressed as: .

[0136] Wherein, is the stator reference current of the long-stator linear synchronous motor, which is obtained according to the speed outer loop proportional-integral controller.

[0137] Based on the above derivation, the control logic block diagram of the model predictive control algorithm in the full speed domain can be obtained, as shown in Figure 12The traction converter is confirmed based on the actual position of the train. Through coordinate transformation, the equivalent circuit of the long-stator linear synchronous motor of the train is converted from a three-phase coordinate system to a d-q axis coordinate system, facilitating data calculation and control. First, current model prediction is performed, and a time delay compensation method is used to generate d-axis output current and q-axis output current. Through the speed of the long-stator linear synchronous motor of the train, it is determined that the current control mode is a maximum torque current ratio control mode or a field weakening control mode, and further d-axis reference current and q-axis reference current are generated. Through the d-axis output current, the q-axis output current, the d-axis reference current and the q-axis reference current, an overall cost function is constructed, different voltage vectors are substituted into the overall cost function, and the calculation values of the overall cost function are calculated. Through comparison, the minimum overall cost function calculation value is obtained, and the corresponding voltage vector is the optimal solution, representing the sequence of the switching devices in the inverter in the traction converter. The voltage vector is sent to the ANPC. The unique input current is determined by the sequence of the ANPC switching devices, transmitted to the long-stator linear synchronous motor of the train through the long cable, and the optimal input voltage is formed, thereby realizing the effect of train energy saving optimization.

[0138] It should be noted that the steps shown in the above flow or the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0139] As Figure 14 shown, the embodiment provides a control system of a traction converter, comprising; The circuit construction module 1401 is configured to construct the d-axis equivalent circuit and the q-axis equivalent circuit of the motor, and generate the voltage equation of the motor based on the d-axis equivalent circuit and the q-axis equivalent circuit.

[0140] The output current calculation module 1402 is configured to discretize the voltage equation, construct a current prediction model, and obtain d-axis output current and q-axis output current through current prediction model operation.

[0141] The reference current calculation module 1403 is configured to select a maximum torque current ratio control mode or a field weakening control mode according to the speed of the motor, and determine the corresponding d-axis reference current and q-axis reference current in the maximum torque current ratio control mode or the field weakening control mode.

[0142] The voltage acquisition module 1404 is configured to construct an overall cost function, substitute the d-axis output current, the q-axis output current, the d-axis reference current and the q-axis reference current into the overall cost function, obtain the overall cost function calculation value, roll the overall cost function calculation value corresponding to multiple time points, and select the minimum value in the multiple overall cost function calculation values.

[0143] wherein the overall cost function is a sum of the current tracking cost function, the average switching frequency limitation cost function, the line loss compensation cost function and the midpoint voltage offset adjustment cost function.

[0144] The converter control module 1405 is configured to generate the corresponding pulse signals according to the values of the current tracking cost function, the average switching frequency limitation cost function, the line loss compensation cost function and the midpoint voltage offset adjustment cost function corresponding to the minimum value in the plurality of overall cost function calculation values, and control the on-off of the inverter switching devices in the traction converter.

[0145] The technical features of the above-described embodiments can be combined in any manner. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described, but it should be considered that any combination of the technical features is within the scope of the present disclosure as long as the combination does not result in contradictions.

[0146] The above-described embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present application, and these are within the scope of protection of the present application. Therefore, the scope of protection of the patent of the present application should be subject to the appended claims.

Claims

1. A control method of a traction converter, characterized by, The method comprises the following steps: a circuit construction step of constructing a d-axis equivalent circuit and a q-axis equivalent circuit of the motor, and generating a voltage equation of the motor based on the d-axis equivalent circuit and the q-axis equivalent circuit; an output current calculation step of discretizing the voltage equation, constructing a current prediction model, and obtaining a d-axis output current and a q-axis output current through calculation of the current prediction model; a reference current calculation step of selecting a maximum torque current ratio control mode or a field weakening control mode according to a speed of the motor, and determining a corresponding d-axis reference current and a q-axis reference current in the maximum torque current ratio control mode or the field weakening control mode; a voltage acquisition step of constructing a total cost function, substituting the d-axis output current, the q-axis output current, the d-axis reference current and the q-axis reference current into the total cost function to obtain a total cost function calculation value, rolling updating the total cost function calculation values corresponding to a plurality of time points, and selecting a minimum value in the plurality of total cost function calculation values; wherein the total cost function is a sum of a current tracking cost function, an average switching frequency limitation cost function, a line loss compensation cost function and a midpoint voltage deviation adjustment cost function; a converter control step of generating corresponding pulse signals according to a value of the current tracking cost function, a value of the average switching frequency limitation cost function, a value of the line loss compensation cost function and a value of the midpoint voltage deviation adjustment cost function corresponding to the minimum value in the plurality of total cost function calculation values, and controlling on-off of an inverter switching device in a traction converter.

2. The control method of a traction converter according to claim 1, characterized by, The current prediction model is: ； ； wherein, is the first predicted current of the d-axis at the next time instant, is the second predicted current of the d-axis at the next time instant, is the equivalent resistance, is the sampling period of the controller, is the inductance of the equivalent circuit of the d-axis, is the first voltage of the d-axis at the current time instant, is the angular speed of the long-stator linear synchronous motor, is the first predicted current of the q-axis at the current time instant, is the first predicted current of the q-axis at the next time instant, is the second predicted current of the q-axis at the next time instant, is the inductance of the equivalent circuit of the q-axis, is the first voltage of the q-axis at the current time instant, is the first predicted current of the d-axis at the current time instant, is the rotor flux, k is the time instant in the discrete time domain.

3. The control method of a traction converter according to claim 2, characterized in that, The output current calculation step further comprises: delay compensation based on the current prediction model to obtain compensated d-axis first actual prediction current, compensated d-axis second actual prediction current, compensated q-axis first actual prediction current and compensated q-axis second actual prediction current; calculating an average value of the compensated d-axis first actual prediction current and the compensated d-axis second actual prediction current to obtain the d-axis output current; calculating an average value of the compensated q-axis first actual prediction current and the compensated q-axis second actual prediction current to obtain the q-axis output current.

4. The control method of a traction converter according to claim 1, characterized by, The reference current calculation step further comprises: constructing a coordinate system with the d-axis output current as the abscissa and the q-axis output current as the ordinate, constructing a current maximum constraint condition and a voltage limit constraint condition of the motor, and generating a reference current trajectory in the coordinate system according to the current maximum constraint condition and the voltage limit constraint condition; The reference current trajectory comprises a first segment current trajectory and a second segment current trajectory, the first segment current trajectory is a maximum torque current ratio curve from an origin to a point of intersection of a current maximum constraint condition and a voltage limit constraint condition of the motor at a first rated angular velocity on a coordinate system, and the second segment current trajectory is a current limit curve from the point of intersection of the current maximum constraint condition and the voltage limit constraint condition of the motor at the first rated angular velocity to a point of intersection of the current maximum constraint condition and the voltage limit constraint condition of the motor at a second rated angular velocity on the coordinate system; determining whether to run along the first segment current trajectory or the second segment current trajectory according to a position of the speed of the motor in the reference current trajectory; determining to run the maximum torque current ratio control mode when running along the first segment current trajectory; determining to run the field weakening control mode when running along the second segment current trajectory.

5. The control method of a traction converter according to claim 4, characterized by, the maximum torque current ratio control mode corresponding to the d-axis reference current is: ； wherein, is the stator current of the electric machine, is the inductance of the d-axis equivalent circuit, is the inductance of the q-axis equivalent circuit, is the rotor flux; the maximum torque current ratio control mode corresponding to the q-axis current reference is: 。 6. The control method of a traction converter according to claim 4, characterized by, The calculation formula of the d-axis reference current corresponding to the field weakening control mode is: ； wherein is the inductance of the d-axis equivalent circuit, is the inductance of the q-axis equivalent circuit, is the rotor flux, is the stator reference current of the electrical machine, obtained from the speed outer loop proportional-integral controller, is the angular speed of the electrical machine, is the maximum stator voltage.

7. The control method of a traction converter according to claim 1, characterized by, The calculation formula of the current tracking cost function is: ； wherein, is a weight coefficient of the current tracking cost function, k is a time in a discrete time domain, is a d-axis output current at k+2 time, is a q-axis output current at k+2 time, is a d-axis reference current at k+2 time, is a q-axis reference current at k+2 time.

8. The control method of a traction converter according to claim 1, characterized by, The calculation of the average switching frequency limitation cost function is: 。 wherein, is a weight coefficient of the average switching frequency limitation cost function, k is a time in a discrete time domain, is a switching sequence at k+2 time, is a switching sequence at k+1 time, is an energy loss at k+2 time, is an energy loss at k+2 time, x is three phases of the long-stator linear synchronous motor, including a phase, b phase and c phase.

9. The control method of a traction converter according to claim 1, characterized by, The calculation formula of the line loss compensation cost function is: ； wherein, is a weight coefficient of the line loss compensation cost function, m is a first factor of the adjustment nonlinearity influence of the line impedance parameter cost function, n is a second factor of the adjustment nonlinearity influence of the line impedance parameter cost function, s is the distance of the train to the traction substation, and f is the stator frequency.

10. A control system of a traction converter, characterized in that, comprises: a circuit construction module configured to construct a d-axis equivalent circuit and a q-axis equivalent circuit of the motor, and generate a voltage equation of the motor based on the d-axis equivalent circuit and the q-axis equivalent circuit; an output current calculation module configured to discretize the voltage equation, construct a current prediction model, and obtain d-axis output current and q-axis output current through calculation of the current prediction model; a reference current calculation module configured to select a maximum torque current ratio control mode or a field weakening control mode according to the speed of the motor, and determine d-axis reference current and q-axis reference current corresponding to the maximum torque current ratio control mode or the field weakening control mode; a voltage acquisition module configured to construct a total cost function, substitute the d-axis output current, the q-axis output current, the d-axis reference current and the q-axis reference current into the total cost function to obtain total cost function calculation values, rollingly update the total cost function calculation values corresponding to a plurality of time points, and select a minimum value in the plurality of total cost function calculation values; wherein the total cost function is a sum of a current tracking cost function, an average switching frequency limitation cost function, a line loss compensation cost function and a midpoint voltage offset adjustment cost function; a converter control module configured to generate corresponding pulse signals according to numerical values of the current tracking cost function, the average switching frequency limitation cost function, the line loss compensation cost function and the midpoint voltage offset adjustment cost function corresponding to the minimum value in the plurality of total cost function calculation values, and control on-off of inverter switching devices in a traction converter.

Images