A control method and device of a permanent magnet generator

Abstract

This application relates to a control method and apparatus for a permanent magnet generator. The method, without requiring a bidirectional DC / DC converter, achieves current limiting for energy storage battery charging by limiting the torque of the permanent magnet generator. The method includes: acquiring a DC voltage command and sampled values, calculating the difference, and obtaining the motor torque via a PI regulator; comparing the motor torque with a preset torque limit to determine the torque command; inputting the torque command into a calibrated torque current meter, outputting a torque current command and an excitation current command; calculating the first and second voltage commands for the dq axis via a PI regulator based on the difference between the sampled values ​​of the torque current and excitation current and the command values; and generating a power module drive signal through space voltage vector modulation. The advantages of this method are that it simplifies the system hardware structure, avoids the use of a bidirectional DC / DC converter, thereby reducing cost and complexity, and achieves efficient charging current control through torque current limiting, improving system reliability and energy efficiency.

Description

Technical Field

[0001] This invention relates to the field of permanent magnet generator technology, and specifically to a control method and device for a permanent magnet generator. Background Technology

[0002] In independent power systems for mobile platforms such as special vehicles, permanent magnet synchronous generators are widely used as the core power generation unit due to their high power density and efficiency. These systems typically require charging energy storage batteries to ensure a continuous and stable power supply to the load. Traditional system architectures rely on adding a bidirectional DC / DC converter between the generator and the energy storage battery. This converter performs voltage conversion and regulation, thereby achieving precise control of the battery charging current. While this design is relatively effective in current management, it has constituted a widely used technical solution in this field for a long time.

[0003] However, the approach using a bidirectional DC / DC converter has significant limitations. This additional power electronic device not only increases the overall system size, weight, and manufacturing cost, but is also a significant drawback for space-constrained and weight-sensitive special vehicles. Furthermore, the switching process between charging and discharging states in this architecture involves a certain delay, typically up to 20 milliseconds, which affects the system's dynamic response performance to some extent. Therefore, a key direction for improving existing technologies is to eliminate the bidirectional DC / DC converter while effectively limiting the charging current, thereby simplifying the system structure, reducing size and weight, and improving economic efficiency.

[0004] Faced with these challenges, the industry urgently needs an innovative control strategy that can directly and precisely control the generator itself to limit the charging current of the energy storage battery without relying on additional hardware converters. This solution requires deeply exploring the control potential of permanent magnet generators and using software algorithm optimization to replace some hardware functions, thereby achieving system miniaturization, weight reduction, and performance optimization. This is of significant practical importance for advancing the development of this technology. Summary of the Invention

[0005] To address the existing technical problems, this application provides a control method and device for a permanent magnet generator.

[0006] In a first aspect, embodiments of this application provide a control method for a permanent magnet generator. This method, without the need for a bidirectional DC / DC converter, achieves current limiting for energy storage battery charging by limiting the torque of the permanent magnet generator, and includes the following steps: The DC voltage command and DC voltage sample value are obtained, the difference between the DC voltage command and the DC voltage sample value is calculated, and the motor torque of the permanent magnet generator is calculated through the PI regulator. The motor torque is compared with the torque limit. If the motor torque is greater than the torque limit, the value of the torque command is determined to be equal to the torque limit; otherwise, the value of the torque command is equal to the motor torque. Input the torque command into the calibrated torque current meter, and output the torque current command and excitation current command through the torque current meter; Obtain the torque current sample value, calculate the difference between the torque current sample value and the torque current command, and calculate the first voltage command of the dq axis through the PI regulator; The excitation current sample value is obtained, the difference between the excitation current sample value and the excitation current command is calculated, and the second voltage command of the dq axis is calculated through the PI regulator. The first voltage command and the second voltage command of the dq axis are modulated by space voltage vector to generate a power module drive signal, which controls the permanent magnet generator to charge the energy storage battery.

[0007] Optionally, the torque limit is determined based on the current state of the energy storage battery, which includes the battery's state of charge and / or temperature.

[0008] Optionally, the parameters of the PI regulator can be set by offline calibration based on the rated power, rated speed of the permanent magnet generator and the rated charging current of the energy storage battery, or by dynamic adjustment based on the absolute value of the DC voltage difference.

[0009] Optionally, the calibration process of the torque current meter includes: based on the motor parameters of the permanent magnet generator, obtaining the torque current command and excitation current command corresponding to different torque commands through simulation or experiment, and establishing a mapping table, wherein the motor parameters include stator resistance, d-axis inductance, q-axis inductance and permanent magnet flux linkage.

[0010] Optionally, when acquiring DC voltage sample values, torque current sample values, and excitation current sample values, the sampled signals are filtered, including first-order low-pass filtering or moving average filtering.

[0011] Optionally, the space voltage vector modulation includes: determining the sector where the voltage vector is located based on the first voltage command and the second voltage command of the dq axis, calculating the duration of the effective voltage vector in the sector, and generating a corresponding pulse width modulation signal as a power module drive signal.

[0012] Secondly, embodiments of this application provide a control device for a permanent magnet generator, including: a voltage processing module configured to acquire a DC voltage command and a DC voltage sample value, calculate the difference between the two, and output the motor torque of the permanent magnet generator through a PI regulator; The torque limiting module is configured to compare the motor torque with the torque limit and output a torque command. When the motor torque is greater than the torque limit, the torque command is equal to the torque limit; otherwise, it is equal to the motor torque. The current command generation module has a built-in calibrated torque current meter and is configured to convert torque commands into torque current commands and excitation current commands. The first current regulation module is configured to acquire the torque current sample value, calculate the difference between the torque current sample value and the torque current command, and output the first voltage command of the dq axis through the PI regulator; The second current regulation module is configured to acquire the excitation current sample value, calculate the difference between the excitation current sample value and the excitation current command, and output the second voltage command of the dq axis through the PI regulator; The modulation drive module is configured to generate a power module drive signal by performing space voltage vector modulation on the first voltage command and the second voltage command of the dq axis to control the permanent magnet generator to charge the energy storage battery.

[0013] Optionally, it also includes a torque limit determination module, which determines the torque limit based on the current state of charge and / or temperature of the energy storage battery, and sends the torque limit to the torque limiting module.

[0014] Optionally, the PI regulators in the voltage processing module, the first current regulation module, and the second current regulation module are all equipped with parameter adjustment units. The parameter adjustment units are configured to dynamically adjust the proportional coefficient and integral coefficient of the PI regulator based on the rated parameters of the permanent magnet generator or the real-time voltage / current difference.

[0015] Optionally, it also includes a signal filtering module, configured to filter the DC voltage sample value, torque current sample value and excitation current sample value, including first-order low-pass filtering or moving average filtering, and send the filtered signals to the voltage processing module, the first current regulation module and the second current regulation module respectively.

[0016] This application relates to a control method for a permanent magnet generator. The core of this method lies in directly limiting the generator torque to control the charging current of the energy storage battery without using a bidirectional DC / DC converter, through a software control strategy. The method first acquires the DC voltage command and actual sampled values, and calculates the initial motor torque using a PI regulator. Then, this torque is compared in real time with a preset torque limit to determine the final torque command, ensuring that the output torque never exceeds a safe threshold. Next, a pre-calibrated torque-current meter maps the torque command to precise torque current and excitation current commands. These two current commands are then subjected to closed-loop PI control to generate corresponding dq-axis voltage commands. Finally, space vector modulation technology is used to synthesize a drive signal to directly control the permanent magnet generator to charge the battery. This method effectively eliminates the need for a bidirectional DC / DC converter by optimizing the control algorithm, significantly reducing system size, weight, and manufacturing costs, while maintaining the current limiting capability during charging, thus improving system integration and economy. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the background art, the accompanying drawings used in the embodiments of this application or the background art will be described below.

[0018] Figure 1 A diagram illustrating the architecture of a DC power supply system in a conventional special vehicle, as provided in an embodiment of this application, is shown.

[0019] Figure 2 A diagram illustrating the DC power supply system architecture that omits the bidirectional DC / DC converter provided in an embodiment of this application is shown.

[0020] Figure 3 The system architecture diagram of the energy storage battery charging current limiting control strategy provided in the embodiments of this application is shown.

[0021] Figure 4 A flowchart illustrating the torque limit calculation method provided in an embodiment of this application is shown.

[0022] Figure 5 A flowchart illustrating the method for calculating the charging current limit of an energy storage battery according to an embodiment of this application is shown.

[0023] Figure 6 A flowchart illustrating the method for calculating the charging power limit of an energy storage battery according to an embodiment of this application is shown.

[0024] Figure 7 A flowchart illustrating the method for calculating motor power limit by looking up a table, as provided in an embodiment of this application, is shown.

[0025] Figure 8A flowchart illustrating the method for calculating motor torque limits provided in an embodiment of this application is shown.

[0026] Figure 9 A schematic diagram of the current-limited charging curve of the energy storage battery provided in the embodiment of this application is shown. Detailed Implementation

[0027] The embodiments of the present invention describe the provided methods, apparatus, and electronic devices through flowcharts and / or block diagrams.

[0028] It should be understood that each block of a flowchart and / or block diagram, as well as combinations of blocks in a flowchart and / or block diagram, can be implemented by computer-readable program instructions. These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus. These computer-readable program instructions, when executed by a computer or other programmable data processing apparatus, create means for implementing the functions / operations specified in the blocks of the flowchart and / or block diagram.

[0029] These computer-readable program instructions may also be stored in a computer-readable storage medium that enables a computer or other programmable data processing device to function in a particular manner. In this way, the instructions stored in the computer-readable storage medium produce an instruction apparatus product that includes the functions / operations specified in the blocks of a flowchart and / or block diagram.

[0030] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus or other device to produce a computer-implemented process, such that the instructions that execute on the computer or other programmable data processing apparatus provide a process for implementing the functions / operations specified in the blocks of the flowchart and / or block diagram.

[0031] The embodiments of the present invention will now be described with reference to the accompanying drawings.

[0032] Figure 1The diagram illustrates the architecture of a conventional DC power supply system for special-purpose vehicles. As shown, to achieve safe and efficient charging of energy storage batteries, a commonly used solution is the integration of a bidirectional DC / DC converter. This solution precisely regulates the charging current through the bidirectional DC / DC converter, effectively maintaining the battery temperature within a reasonable range, thereby ensuring battery stability and lifespan. However, this traditional design also has significant limitations: firstly, the addition of independent power electronic devices significantly increases the overall system size and weight, hindering vehicle space optimization and maneuverability improvement; secondly, the switching of energy flow direction requires a certain response delay, typically on the order of 20 milliseconds, affecting the system's dynamic response speed. For special-purpose vehicle applications with compact spaces and a strong emphasis on maneuverability and reliability, these drawbacks make this solution less than optimal.

[0033] To address the aforementioned problems, this application provides a control method for a permanent magnet generator, such as... Figure 2 As shown, this method, without the need for a bidirectional DC / DC converter, achieves current limiting for energy storage battery charging by limiting the torque of the permanent magnet generator. The specific implementation process is as follows: Figure 3 As shown, it includes the following steps: Get DC voltage command U dcref and DC voltage sampling value U dcfdb Calculate DC voltage command U dcref With DC voltage sampling value U dcfdb The difference is used to calculate the motor torque T of the permanent magnet generator through a PI regulator. cmd ; Motor torque T cmd With torque limit T limit Comparison, if the motor torque T cmd Greater than the torque limit T limit Then determine the torque command T ref The value is equal to the torque limit T. limit Otherwise, torque command T ref The value is equal to the motor torque T cmd ; Torque command T ref Input a calibrated torque current meter, and output a torque current command i through the torque current meter. qref and excitation current command i dref ; Obtain the torque current sampling value i qfdb Calculate the torque current sampling value i qfdb With torque current command i qref The difference is calculated using a PI controller to obtain the first voltage command u for the dq axis. d ; Obtain the excitation current sampling value i dfdb Calculate the excitation current sampling value i dfdb With excitation current command i dref The difference is calculated using a PI controller to obtain the second voltage command u for the dq axis. q ; The first voltage command for the dq axis is u. d With the second voltage command u of the dq axis q After space voltage vector modulation, a power module drive signal is generated to control the permanent magnet generator to charge the energy storage battery.

[0034] The first step of this method is to obtain the DC voltage command U. dcref and the actual DC voltage sampling value U dcfdb The difference between the two is calculated, and this difference is input into the PI controller to calculate the initial motor torque T of the permanent magnet generator. cmd The DC voltage command U here dcref This represents the target stable DC bus voltage that the system expects to maintain, while the sampled value U dcfdb This refers to the real-time output voltage of the power generation system. A PI regulator is a controller that combines proportional and integral actions, capable of quickly and smoothly eliminating system deviations based on the magnitude and accumulation of errors. Torque commands are generated through PI regulation in the outer voltage loop. The principle behind this is that the generator's output torque is directly related to its output power, which in turn is closely related to the stability of the DC bus voltage. The core purpose of this step is to establish an external voltage closed-loop control, enabling the generator's output to automatically track the voltage command, maintain the stability of the DC bus voltage, and provide a stable foundation for subsequent current limiting control. The beneficial effect of this design is that it achieves the system's basic voltage regulation function, ensuring that the generator side can autonomously maintain a stable DC voltage platform without connecting a bidirectional DC / DC converter, providing a reliable energy source for battery charging.

[0035] After calculating the initial motor torque T cmd Then, a key step in this method is to compare it with a preset torque limit T. limit Compare them. If T cmd Greater than T limit Then the final torque command T ref Will be restricted to T limit Conversely, T ref equal to T cmd This torque limit T limitIt's not a fixed value, but a dynamic value calculated based on parameters such as the charging current limit of the energy storage battery, the current DC voltage, the generator speed, and the controller efficiency. Its fundamental purpose is to limit the generator's output power within a safe range for the battery. This comparison and limitation constitutes the core of the entire charging current limiting strategy. The principle is that the electromagnetic torque output by the permanent magnet generator is proportional to the current it generates (and thus the output power). By limiting the upper limit of the torque command, the maximum power that the generator can output is limited at the source, thereby indirectly limiting the charging current to the energy storage battery and preventing excessive battery temperature rise or damage due to overcharging. The beneficial effect of this step is that it achieves software current limiting without additional hardware (such as bidirectional DC / DC converters). It transforms the complex problem of charging current control into a simple and effective clamping operation of the generator torque command, greatly simplifying the system architecture.

[0036] Determine the final torque command T ref The next step is to input this data into a pre-calibrated torque-current meter. This meter stores the optimal torque-current command i corresponding to achieving optimal control performance (such as maximum torque-current ratio, field weakening control, etc.) under different operating conditions (mainly different speeds and torques). qref and excitation current command i dref By using a table lookup method, the value related to the current T can be obtained quickly and accurately. ref The current command value is matched to the rotational speed. This step utilizes the vector control theory of permanent magnet synchronous motors to decouple torque control into independent control of the d-axis (excitation component) and q-axis (torque component) currents. The lookup table method is an efficient strategy for achieving nonlinear mapping, based on offline precise measurements and optimized calculations of generator characteristics. The advantage of this step is that it provides an optimal current reference for motor control, ensuring efficient and smooth generator operation under a given torque command, while also providing a precise command target for subsequent current closed-loop control. Optionally, this mapping relationship can also be implemented through an online calculation model, but the lookup table method is generally more advantageous in terms of real-time performance.

[0037] Current command (i) was obtained qref and i dref After that, the system enters the current closed-loop control stage. This method collects the actual q-axis current (torque current) sampling value i of the generator. qfdb and d-axis current (excitation current) sampling value i dfdb These two current error signals are then compared with their respective command values ​​to generate current error signals. These two current error signals are then fed into their respective PI controllers, which dynamically calculate the d-axis voltage command u to be applied to the generator based on the error. d and q-axis voltage command uq This forces the actual current to quickly and accurately track the commanded current. This step is the inner loop control in a typical dual-loop control system, and its response speed is much faster than the external voltage or speed loop. Its principle is to precisely achieve the torque (or power) control target through direct and rapid current control, and to suppress the impact of disturbances such as load surges on the system. Its beneficial effect lies in achieving high-precision, high-dynamic performance control of the generator current, ensuring that the torque command can be executed quickly and accurately, which is a key guarantee for the stable and reliable operation of the entire control system.

[0038] Finally, the calculated d-axis voltage command u d and q-axis voltage command u q After being processed by a space voltage vector modulation algorithm, the voltage is converted into a drive signal to control the switching action of the power module (usually a three-phase inverter bridge). SVPWM is an optimized pulse width modulation technique that synthesizes the desired voltage space vector on the motor stator windings by controlling the turn-on and turn-off sequence and timing of the power switching devices in the inverter. Its principle is to approximate an ideal circular rotating magnetic field through the synthesis of voltage vectors, thereby enabling the motor to obtain smoother torque and higher voltage utilization. The drive signal generated after SVPWM modulation directly controls the power module to generate the required three-phase AC voltage, ultimately driving the permanent magnet generator to operate in the set state, realizing the charging control of the energy storage battery. The beneficial effect of this step is that it efficiently and accurately converts the voltage command calculated by the digital controller into actual power conversion, completing the final link from control signal to physical energy transfer. Furthermore, SVPWM technology itself helps reduce harmonic losses and torque ripple, improving the overall system efficiency and stability.

[0039] In summary, this method constructs a complete software control chain through a series of interconnected steps, from voltage stabilization, torque current limiting, current optimization command generation, high-precision current tracking to final power drive. Its core innovation lies in successfully replacing the indispensable bidirectional DC / DC converter hardware unit in traditional solutions through a sophisticated control algorithm, particularly the dynamic limiting strategy for torque commands. Ultimately, it achieves the control objective of ensuring safe and reliable charging of the energy storage battery while simplifying the system hardware structure and reducing size, weight, and cost.

[0040] The core point of this invention is through T limit For T cmd This limitation is used to limit the charging current of energy storage batteries, such as... Figure 4 As shown, T limitThe calculation process is as follows: This flowchart clearly shows the entire reverse calculation process of the motor torque limit during the charging of the energy storage battery. Its starting point is not the motor itself, but rather the safety boundary of the energy storage battery. Specifically, it is necessary to first determine the real-time maximum safe charging current limit i based on the charging and discharging characteristic curve of the energy storage battery (this curve comprehensively reflects the influence of factors such as the battery's state of charge and temperature on the acceptable charging current). cha Next, the system multiplies this current limit by the current DC bus voltage to calculate the maximum allowable charging power limit P on the battery side. cha However, this power transfer from the generator to the battery involves controller losses. Therefore, the next crucial step is to consult a pre-calibrated generator controller efficiency table to determine the battery-side power P. cha Based on the corresponding efficiency value, the maximum electromagnetic power limit P that the generator side actually needs to output is calculated. gen The final step in the process is to set the motor power limit P. gen Combined with the current real-time speed of the permanent magnet generator, through the physical formula T limit = P gen The final motor torque limit T is calculated using / ω (where ω is the mechanical angular velocity). limit This reverse derivation process, from the battery's safe current to the generator's torque command, ensures that the generator's output capacity is constrained at the source. This allows the entire system to naturally guarantee that the charging current will never exceed the battery's safe tolerance range without relying on additional hardware.

[0041] like Figure 5 As shown, this represents the charging current limit i for the energy storage battery. cha The calculation flowchart shows how to calculate the charging current limit i of the energy storage battery based on its charge / discharge characteristic curves, temperature, and state of charge (SOC). cha .like Figure 6 As shown, based on the sampled DC voltage U dcfdb Calculate the charging power limit P of the energy storage battery cha .like Figure 7 As shown, based on the motor speed n fdb and the sampled DC voltage U dcfdb The motor power limit P is obtained by looking up the efficiency table of the calibrated generator controller. gen Finally, as Figure 8 As shown, according to the motor power limit P gen and motor speed n fdb Calculate the motor torque limit T limit .

[0042] like Figure 9As shown, the voltage and charging current curves of the energy storage battery during current-limited charging process can be seen. limit For T cmd The limitations effectively restrict the charging current of energy storage batteries.

[0043] In some embodiments, the torque limit is optionally determined based on the current state of the energy storage battery, which includes the battery's state of charge and / or temperature. The torque limit is not determined using a fixed threshold, but rather dynamically correlated with the real-time state of the energy storage battery, where key state parameters include the battery's state of charge and temperature. The state of charge reflects the battery's current remaining charge level, while temperature directly affects the rate and safety of the internal chemical reactions within the battery. Specifically, the control system continuously monitors these two parameters and dynamically queries or calculates the maximum charging current limit that the battery can safely accept under the current operating conditions, based on a pre-established battery characteristic model—typically embedded in the controller in the form of a data table or functional relationship. This current limit is then combined with the system DC voltage to convert it into a charging power limit. Taking into account the generator speed and system efficiency, the corresponding generator torque limit is finally derived in reverse. The beneficial effects of this dynamic limit strategy are significant. It enables the charging process to intelligently adapt to the actual health condition of the battery. For example, when the battery temperature is high or the state of charge is close to saturation, it automatically reduces the torque (and power) input, thereby effectively preventing overcharging and overheating, and greatly improving the safety and reliability of the system. At the same time, this strategy optimizes the charging curve, which helps to extend the life of the energy storage battery. From a system perspective, it achieves fully software-based intelligent energy management without relying on additional hardware sensors or complex circuits, further consolidating the core advantage of simplifying the hardware architecture through control algorithms.

[0044] In some embodiments, the parameters of the PI controller can be optionally set by offline calibration based on the rated power, rated speed of the permanent magnet generator, and rated charging current of the energy storage battery, or by dynamic adjustment based on the absolute value of the DC voltage difference. The parameter setting of the PI controller employs a flexible and engineered strategy, which can be achieved through two complementary methods: offline calibration based on key system parameters, or dynamic adjustment based on the system's operating state. Offline calibration is fundamental, relying on core system parameters such as the rated power, rated speed of the permanent magnet generator, and rated charging current of the energy storage battery. Through pre-modeling, simulation, or physical platform testing, a set of optimal PI parameters is determined to ensure stable and rapid response of the system under typical operating conditions. The principle behind this method is that these rated parameters define the main dynamic characteristics of the system, and the parameters calibrated accordingly provide a robust initial reference for the controller. Optionally, to further improve the system's adaptability and control quality under a wide range of operating conditions, a dynamic adjustment mechanism can also be introduced. For example, the PI parameters are fine-tuned based on the absolute value of the difference between the DC voltage command and the sampled value: when the voltage deviation is large, the control action is appropriately enhanced to accelerate the response speed; when the system is close to steady state and the deviation is small, the control action is weakened to avoid overshoot and oscillation. This essentially simulates a nonlinear control characteristic, giving the system both speed and stability. The benefits of this parameter setting method are multifaceted. Offline calibration ensures that the control system has basically reliable control performance from the outset, reducing the complexity of on-site debugging. The dynamic adjustment mechanism, on the other hand, gives the controller adaptive capability, enabling it to cope with non-ideal operating conditions such as sudden load changes and speed fluctuations. This effectively suppresses the risk of slow response or instability that may occur in traditional fixed-parameter PI regulators when facing large-scale dynamic processes, thereby improving the stability of the DC bus voltage and the robustness of the system across the entire operating range. The combination of these two methods ensures that the control method can still achieve high-precision and high-reliability charging control even with the hardware simplification of eliminating the bidirectional DC / DC converter.

[0045] In some embodiments, optionally, the calibration process of the torque current meter includes: based on the motor parameters of the permanent magnet generator, obtaining the torque current command and excitation current command corresponding to different torque commands through simulation or experiment, and establishing a mapping table, wherein the motor parameters include stator resistance, d-axis inductance, q-axis inductance, and permanent magnet flux linkage. The calibration of the torque current meter is a crucial offline preparation process, the core of which lies in establishing a precise mapping relationship between the torque command and the optimal current command. The implementation of this process first relies on the core motor parameters of the permanent magnet generator, including stator resistance, d-axis inductance, q-axis inductance, and permanent magnet flux linkage. The stator resistance determines the copper loss of the windings, the d-axis and q-axis inductances characterize the characteristics of the motor's magnetic circuit, and the permanent magnet flux linkage reflects the rotor magnetic field strength. Based on these parameters, high-precision simulation can be performed through finite element analysis or mathematical models, or systematic testing of the physical motor can be conducted on an experimental platform to obtain the optimal torque current command i corresponding to achieving specific optimization objectives (such as maximum torque-to-current ratio, unity power factor control, or field weakening speed extension) under different speed and torque command combinations. qref and excitation current command i dref Subsequently, these discrete, optimized data points are organized and stored into a multi-dimensional lookup table, which constitutes the calibrated torque and current table used internally by the controller. Essentially, this process transforms the complex electromagnetic relationships of the motor and the nonlinear control optimization problem into a static data model that can be quickly looked up by the real-time system through prior calculations and experiments. The benefits of this offline calibration method are significant. It transforms the complex real-time calculation task of solving for the optimal current command into a simple table lookup operation, greatly reducing the computational burden on the controller and ensuring the system's rapid response capability. By injecting pre-optimized current commands, it can be ensured that the permanent magnet generator operates in the high-efficiency or optimal performance region throughout its entire operating range, thereby significantly improving the overall energy efficiency and control accuracy of the system. Furthermore, this mapping table fully considers the parameter characteristics of the motor itself, making the control strategy highly targeted to the motor, enhancing the stability and reliability of the system, and laying a precise benchmark for subsequent current closed-loop control.

[0046] In some embodiments, optionally, when acquiring DC voltage, torque current, and excitation current sample values, the sampled signals are filtered. This filtering process includes first-order low-pass filtering or moving average filtering. Filtering the DC voltage, torque current, and excitation current sample values ​​is a crucial step in ensuring the stable and reliable operation of the control system. This process is mainly accomplished by introducing a digital filtering algorithm after the analog-to-digital converter and before the controller uses these sample values. Commonly used methods include first-order low-pass filtering or moving average filtering. First-order low-pass filtering allows low-frequency signals (representing real physical quantity changes) to pass through smoothly while significantly attenuating high-frequency components (usually noise interference). Its filtering strength is determined by a parameter called the time constant. Moving average filtering, on the other hand, performs an arithmetic average of data from multiple consecutive sampling times, using the averaging effect to smooth random fluctuations. These filtering processes essentially seek the optimal balance between real-time performance and accuracy, aiming to retain useful information such as DC voltage and current that reflect the true state of the system, while suppressing high-frequency glitches introduced by electromagnetic interference, sensor noise, or switching device operation. The beneficial effects of this measure are critical. The filtered sampling signal is smoother and purer, more accurately reflecting the system's operating state. This provides reliable feedback for subsequent PI regulators (voltage loop, current loop), effectively preventing controller malfunctions caused by signal noise, such as high-frequency jitter or oscillation in the PI regulator output. This significantly improves the stability and control accuracy of the entire control system, especially the DC bus voltage, and enhances the system's anti-interference capability. In the aforementioned control architecture that does not rely on hardware DC / DC converters, this software robustness is particularly important. It improves system reliability from the signal source, ensuring that the software-based torque current limiting strategy can be executed accurately and smoothly.

[0047] In some embodiments, optionally, space voltage vector modulation includes: determining the sector where the voltage vector is located based on the first voltage command and the second voltage command of the dq axis, calculating the duration of the effective voltage vector in the sector, and generating a corresponding pulse width modulation signal as a power module drive signal. In some embodiments, the implementation process of space voltage vector modulation is a key step in converting control commands into actual drive signals. This process first uses the dq axis voltage command (u) calculated by the current loop PI regulator. d and u qThe algorithm transforms the vector to a stationary two-phase α-β coordinate system to determine the location of a target voltage space vector. Then, it determines which predefined sector this vector occupies within the complex space vector diagram, forming the basis for subsequent synthesis actions. Next, based on the volt-second balance principle, it calculates the required duration of action for the two adjacent basic non-zero voltage vectors and the zero vector within that sector, ensuring that their average effect is equivalent to the target voltage vector within a fixed switching cycle. Finally, based on these calculated durations, a specific switching sequence is used to generate six pulse-width modulation (PWM) signals for the switching devices, such as insulated-gate bipolar transistors (IGBTs), in the final control power module. The benefits of this modulation strategy are significant. Compared to traditional sinusoidal pulse-width modulation, space voltage vector modulation (SVC) can more fully utilize the DC bus voltage, improving voltage utilization and enabling the motor to output higher torque and power under the same DC voltage conditions. Simultaneously, by optimizing the switching sequence, this method effectively reduces harmonic distortion of the output current and torque ripple in the motor, resulting in smoother and quieter motor operation and reduced additional losses caused by harmonics. Furthermore, its algorithm structure is well-organized, facilitating programming implementation by digital signal processors, ensuring real-time performance and reliability of control, and providing an efficient and precise execution end for the entire hardware-free DC / DC converter control system.

[0048] This application provides a control device for a permanent magnet generator, including: a voltage processing module configured to acquire a DC voltage command and a DC voltage sample value, calculate the difference between the two, and output the motor torque of the permanent magnet generator through a PI regulator; The torque limiting module is configured to compare the motor torque with the torque limit and output a torque command. When the motor torque is greater than the torque limit, the torque command is equal to the torque limit; otherwise, it is equal to the motor torque. The current command generation module has a built-in calibrated torque current meter and is configured to convert torque commands into torque current commands and excitation current commands. The first current regulation module is configured to acquire the torque current sample value, calculate the difference between the torque current sample value and the torque current command, and output the first voltage command of the dq axis through the PI regulator; The second current regulation module is configured to acquire the excitation current sample value, calculate the difference between the excitation current sample value and the excitation current command, and output the second voltage command of the dq axis through the PI regulator; The modulation drive module is configured to generate a power module drive signal by performing space voltage vector modulation on the first voltage command and the second voltage command of the dq axis to control the permanent magnet generator to charge the energy storage battery.

[0049] The data execution process of this device begins with the voltage processing module, which continuously acquires the DC voltage command signal and the actual DC voltage sample value output by the generator. It calculates the difference between the two and sends it to a PI regulator (a controller combining proportional and integral operations for quickly and smoothly eliminating system deviations) for processing, thereby generating an initial motor torque signal. This torque signal is then sent to the torque limiting module, where it is compared in real-time with a torque limit dynamically calculated based on the safety status of the energy storage battery. The comparator executes a clamping logic: if the motor torque exceeds the limit, it outputs a torque command equal to that limit; otherwise, it directly outputs the original motor torque value. The core purpose of this is to limit the generator's output power at the source, ensuring that the charging current does not exceed the battery's safety threshold. Next, the current command generation module begins operation. It has a pre-stored offline calibrated torque-current table. This module receives the torque command and maps it to a set of optimal torque-current and excitation current commands through a lookup table. This set of commands aims to enable the motor to operate efficiently under the current conditions. Subsequently, the process enters the current closed-loop regulation stage: the first current regulation module receives the torque current command and its sampled feedback value, performs a PI calculation on the difference between the two, and outputs a voltage command for controlling the q-axis (torque component); simultaneously, the second current regulation module performs a similar operation on the excitation current command and feedback, generating a d-axis (excitation component) voltage command. Finally, the modulation drive module synthesizes these two dq-axis voltage commands and uses space vector modulation technology (an optimized pulse width modulation strategy that can efficiently utilize DC voltage and reduce harmonics) to generate a precise pulse signal to drive the switching devices in the power module, ultimately controlling the permanent magnet generator to charge the energy storage battery in a controlled and safe manner. The entire process is interconnected, achieving reliable charging management with simplified hardware through a pure electric control strategy.

[0050] In some embodiments, optionally, a torque limit determination module is also included, which determines the torque limit based on the current state of charge and / or temperature of the energy storage battery, and sends the torque limit to the torque limiting module. The system further includes a dedicated torque limit determination module, the core function of which is to dynamically calculate and set the torque limit based on the real-time operating state of the energy storage battery. This process begins with continuous monitoring of key parameters of the energy storage battery, particularly its state of charge (i.e., remaining charge level) and temperature. This real-time data is input into a pre-stored battery characteristic model within the module, which is typically based on the battery's chemical characteristics and safety boundaries, mapping the maximum acceptable charging current under the current state. Subsequently, the module combines this current limit with the system DC voltage to calculate the maximum allowable charging power limit on the battery side. Then, considering factors such as generator speed and system efficiency, a series of calculations (e.g., combining...) are performed... Figure 7(The process of looking up the motor power limit shown in the table) ultimately leads to the reverse derivation of the corresponding generator torque limit T. limit This value is then sent to the torque limiting module as a basis for comparison and clamping. The beneficial effect of this design is that it elevates charging safety control from a passive response to an active prevention. By dynamically adjusting the torque limit based on the battery's state of charge and temperature—two key health indicators—the system can intelligently adapt to the battery's actual needs: for example, automatically reducing charging power when the battery temperature is too high or the charge is close to saturation, thereby fundamentally preventing overcharging and overheating risks, significantly improving system safety and reliability, and helping to extend battery life. The introduction of this module enables the entire control strategy to achieve fully software-based, adaptive, and high-level energy management by simplifying the hardware by eliminating the need for a bidirectional DC / DC converter.

[0051] In some embodiments, optionally, the PI regulators in the voltage processing module, the first current regulation module, and the second current regulation module are all equipped with parameter adjustment units. These parameter adjustment units are configured to dynamically adjust the proportional and integral coefficients of the PI regulators based on the rated parameters of the permanent magnet generator or the real-time voltage / current difference. The PI regulators used in the voltage processing module, the first current regulation module, and the second current regulation module are all equipped with parameter adjustment units. The core function of these units is to optimize the dynamic performance of the controller. The implementation process embodies a hierarchical strategy: First, the system performs offline calibration based on key parameters such as the rated power and rated speed of the permanent magnet generator, setting a set of basic proportional coefficients (Kp) and integral coefficients (Ki) for each PI regulator. This set of parameters ensures that the system has basic stability and response speed under typical design conditions. Based on this, the parameter adjustment unit also has online adaptive capability, that is, it dynamically fine-tunes these coefficients according to the difference between the real-time acquired voltage or current command and the feedback value. For example, when the system starts up or a sudden load change causes a large voltage or current deviation, the unit will appropriately increase the proportional gain to speed up the response, and may adjust the integral gain to prevent integral saturation. Conversely, when the system is close to steady state and the deviation is small, it tends to decrease the proportional gain and optimize the integral action to pursue higher steady-state accuracy and suppress overshoot. This dynamic-static combined parameter tuning strategy brings significant benefits. Offline calibration provides a reliable control reference for the system, greatly reducing the complexity of on-site debugging. Online dynamic adjustment gives the controller strong adaptive capabilities, enabling it to easily cope with complex nonlinear factors and a wide range of operating conditions, such as generator speed fluctuations or load steps, in actual operation. This not only effectively improves the system's response speed and control accuracy, but more importantly, it enhances the system's robustness, ensuring that the entire charging control system can still maintain high stability and reliability even with the simplified architecture that eliminates the bidirectional DC / DC converter, achieving the design goal of simplifying hardware without sacrificing performance.

[0052] In some embodiments, the system optionally includes a signal filtering module configured to filter the DC voltage sample value, torque current sample value, and excitation current sample value, including first-order low-pass filtering or moving average filtering, and send the filtered signals to the voltage processing module, the first current regulation module, and the second current regulation module, respectively. The system also integrates a dedicated signal filtering module, which plays a crucial role in the data acquisition path. Its implementation involves immediately applying a digital filtering algorithm to the obtained raw digital signal after sampling and analog-to-digital conversion of the DC voltage, torque current, and excitation current. Common methods include first-order low-pass filtering or moving average filtering. First-order low-pass filtering allows low-frequency signals representing real physical changes to pass through smoothly by setting a cutoff frequency, while effectively attenuating high-frequency components caused by electromagnetic interference and switching noise. Moving average filtering, on the other hand, performs an arithmetic average of data from multiple consecutive sampling points, using the averaging effect to smooth out random fluctuations in the signal. The filtered clean signal is sent to the voltage processing module and the two current regulation modules as feedback input for their PI controllers to perform accurate calculations. The beneficial effect of this design is that it significantly improves the reliability and stability of the control system from the signal source. Noise mixed in the raw sampled signal can easily cause the PI controller to malfunction, resulting in high-frequency jitter in the output command and affecting the stable operation of the system. The filtering module acts as a "purifier," filtering out these harmful interference components and providing accurate and smooth feedback values ​​for subsequent control loops. This not only ensures the control accuracy of the voltage outer loop and current inner loop, making the DC bus voltage more stable, but also enhances the overall anti-interference capability of the system. For this simplified system architecture that relies on software algorithms for precise control, this measure provides a crucial foundation for system robustness.

[0053] Example

[0054] This application provides a control method for a permanent magnet generator. The core of this method lies in its independence from hardware (bidirectional DC / DC converter) and instead utilizes a sophisticated software control strategy to directly limit the output torque of the permanent magnet generator, thereby effectively limiting the charging current of the energy storage battery. The specific implementation of this method includes a series of logically rigorous steps, which will be described in detail below with reference to optional embodiments.

[0055] First, the steps of acquiring the DC voltage command and DC voltage sample value are performed. The DC voltage command represents the target value of the stable DC bus voltage that the system expects to maintain, which is usually set by the upper-level energy management system according to system requirements. The DC voltage sample value is obtained in real time by acquiring the actual DC bus voltage at the generator output terminal through a high-precision voltage sensor. The difference between the command value and the sample value is calculated, and this difference signal is input to a PI regulator. The PI regulator is a controller that combines proportional and integral actions; the proportional element provides fast response, while the integral element is used to eliminate steady-state errors. Based on the input voltage deviation signal, the regulator calculates the initial value of the permanent magnet generator electromagnetic torque required to maintain DC voltage stability, denoted as the motor torque Tcmd. In one embodiment, to ensure the stability of the system under different operating conditions, the parameters of the PI regulator (proportional coefficient Kp and integral coefficient Ki) can be calculated and calibrated offline based on key system parameters such as the rated power, rated speed of the permanent magnet generator, and rated charging current of the energy storage battery, for example, by pre-determining a set of robust parameters using the Ziegler-Nichols rule or other engineering tuning methods. Optionally, to further improve dynamic performance, the parameters of the PI regulator can also be dynamically fine-tuned based on the absolute value of the DC voltage difference. For example, when the voltage deviation is large, the proportional coefficient can be appropriately increased to speed up the response, while when the system is close to steady state, the integral action is emphasized to improve control accuracy. This step constitutes the outer loop voltage control, the beneficial effect of which is to establish the system's basic voltage regulation capability and provide a stable DC voltage platform for subsequent current limiting control.

[0056] Next, the step of comparing the motor torque with the torque limit is performed. The calculated motor torque T cmd It will be sent to a comparison logic unit and compared with a preset torque limit T. limit Real-time comparison is performed. This torque limit T limit It is not fixed, but dynamically determined based on the real-time state of the energy storage battery. In a key optional embodiment, the torque limit T limit The determination process is a reverse calculation process: First, based on the current state of charge and temperature of the energy storage battery, the maximum safe charging current I allowed by the battery in this state is determined by querying its characteristic curve or the built-in battery model. bat_max For example, when the battery temperature is high or the state of charge is close to its upper limit, I bat_max This will be adjusted accordingly. Then, this current limit is multiplied by the current DC bus voltage to obtain the maximum allowable charging power limit P on the battery side. cha Finally, considering the generator speed and system efficiency (which can be obtained by looking up a table), P... cha Converted to the power that needs to be limited on the generator side, and then according to the formula T limit= P / ω (where ω is the mechanical angular velocity) to calculate the real-time torque limit T. limit The comparison logic is: if T cmd >T limit The final output torque command T ref Clamped as T limit Otherwise, T ref = T cmd The beneficial effect of this step is that the core current limiting strategy is realized. By limiting the mechanical power input at the source, it indirectly and effectively limits the final charging current, ensuring the charging safety of the battery, while avoiding the size, weight and cost associated with using a bidirectional DC / DC converter.

[0057] Next, the step of inputting the torque command into the calibrated torque-current meter is performed. The determined torque command T... ref The current is fed into a pre-calibrated torque-current table (or current mapping table). This table stores the optimal torque-current command I required to achieve specific optimization objectives (such as maximum torque-current ratio MTPA, or field weakening control) at different operating points (mainly under different speeds and torque commands). qref and excitation current command I dref The calibration of this torque-current meter is itself an important offline process. In one alternative embodiment, the calibration process is based on the precise motor parameters of the permanent magnet generator, including stator resistance, d-axis inductance, q-axis inductance, and permanent magnet flux linkage. Electromagnetic field simulation is performed using finite element analysis software, or the actual motor is tested on an experimental bench, collecting voltage and current data under different torque and speed commands. Then, through calculation or optimization algorithms, the optimal IL that meets the target performance is determined. d I q The data is combined and ultimately organized into a multidimensional lookup table, which is then embedded into the controller. The benefit of this step is that it transforms the complex problem of calculating the optimal real-time current into an efficient lookup operation, reducing the computational burden on the controller and ensuring that the motor always operates within its efficient or optimal performance range.

[0058] Subsequently, the system executes the steps of acquiring the torque current sample value, calculating its difference from the command, and calculating the first voltage command for the dq axis using a PI regulator. The system acquires the sample value I of the actual q-axis current (i.e., the torque current component) output by the permanent magnet generator through a current sensor (such as a Hall sensor). qfdb Calculate the sampled value and the command value I output by the torque current meter. qref The difference between the two values ​​is calculated, and this current error signal is sent to a dedicated PI controller (also known as a q-axis current controller). This PI controller dynamically calculates and outputs the q-axis voltage command U based on the error. qThe purpose is to force the actual torque current to track the command value quickly and accurately. Similarly, the steps of acquiring the excitation current sample value, calculating its difference from the command, and calculating the second voltage command of the dq axis using a PI regulator are executed in parallel. The sample value I of the d-axis current (i.e., the excitation current component) is acquired. dfdb , and instruction value I dref After comparison, a d-axis voltage command U is generated through another PI regulator (d-axis current regulator). d These two steps together constitute the inner loop control of the current. In an optional embodiment, to improve control accuracy and anti-interference capability, the current sample value I... qfdb and I dfdb Before being fed into the PI regulator, the generator current can be digitally filtered. This filtering can be a first-order low-pass filter, designed to attenuate high-frequency noise caused by power device switching; or a moving average filter, which smooths random fluctuations by averaging multiple sampling points. The filter parameters must be set to balance noise suppression with ensuring system response speed. The inner current loop provides highly dynamic and precise control of the generator current, crucial for accurate execution of torque commands. Its fast response characteristics also help suppress disturbances caused by load changes.

[0059] Finally, the process involves generating power module drive signals by performing Space Vector Pulse Widgeting (SVPWM) on the two dq-axis voltage commands. The resulting Ud and Uq commands are first transformed from the rotating dq coordinate system to the stationary two-phase α-β coordinate system through an inverse coordinate transformation. The SVPWM modulation algorithm then determines the sector containing the target voltage vector based on the α and β voltage components. Next, based on the volt-second balance principle, the required duration of action of the two adjacent fundamental non-zero voltage vectors and the zero vector within one switching cycle is calculated to synthesize the desired voltage vector. Finally, according to a preset switching sequence, PWM drive signals are generated to control the on / off state of the six switching transistors (such as IGBTs or MOSFETs) in the three-phase full-bridge power module. These signals ultimately control the power module to generate the required three-phase AC voltage, driving the permanent magnet generator to operate, thereby achieving charging control of the energy storage battery. Compared to traditional SPWM, SVPWM technology offers advantages such as higher DC voltage utilization, lower current harmonics and torque ripple, resulting in smoother and more efficient motor operation.

[0060] In summary, this embodiment, through the organic combination of the above steps, constructs a complete permanent magnet generator control scheme that does not rely on a bidirectional DC / DC converter. This method achieves hardware function substitution through software algorithms, simplifying the system structure, reducing size, weight, and cost, while effectively ensuring the charging safety of the energy storage battery through torque current limiting. Furthermore, it utilizes multi-loop control, optimized lookup tables, and advanced modulation techniques to ensure the system has excellent dynamic performance and operating efficiency.

[0061] The above description is merely a specific implementation of the embodiments of the present invention, but the protection scope of the embodiments of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the embodiments of the present invention should be included within the protection scope of the embodiments of the present invention. Therefore, the protection scope of the embodiments of the present invention should be determined by the protection scope of the claims.

Claims

1. A control method for a permanent magnet generator, characterized in that, This method, without the need for a bidirectional DC / DC converter, limits the charging current of the energy storage battery by limiting the torque of the permanent magnet generator, and includes the following steps: The DC voltage command and DC voltage sample value are obtained, the difference between the DC voltage command and the DC voltage sample value is calculated, and the motor torque of the permanent magnet generator is calculated through a PI regulator. The motor torque is compared with the torque limit. If the motor torque is greater than the torque limit, the value of the torque command is determined to be equal to the torque limit; otherwise, the value of the torque command is equal to the motor torque. The torque command is input into the calibrated torque current meter, and the torque current command and excitation current command are output through the torque current meter. Obtain the torque current sample value, calculate the difference between the torque current sample value and the torque current command, and calculate the first voltage command of the dq axis through the PI regulator; Obtain the excitation current sampling value, calculate the difference between the excitation current sampling value and the excitation current command, and calculate the second voltage command of the dq axis through the PI regulator; The first voltage command and the second voltage command of the dq axis are subjected to space voltage vector modulation to generate a power module drive signal, which controls the permanent magnet generator to charge the energy storage battery.

2. The method according to claim 1, characterized in that, The torque limit is determined based on the current state of the energy storage battery, which includes the battery's state of charge and / or temperature.

3. The method according to claim 1 or 2, characterized in that, The parameters of the PI regulator are set in the following ways: offline calibration based on the rated power, rated speed of the permanent magnet generator and the rated charging current of the energy storage battery, or dynamic adjustment based on the absolute value of the DC voltage difference.

4. The method according to claim 1, characterized in that, The calibration process of the torque current meter includes: based on the motor parameters of the permanent magnet generator, obtaining the torque current command and excitation current command corresponding to different torque commands through simulation or experiment, and establishing a mapping table. The motor parameters include stator resistance, d-axis inductance, q-axis inductance, and permanent magnet flux linkage.

5. The method according to claim 1, characterized in that, When acquiring the DC voltage sample value, torque current sample value, and excitation current sample value, the sampled signals are filtered. The filtering process includes first-order low-pass filtering or moving average filtering.

6. The control method for a permanent magnet generator according to claim 1, characterized in that, The space voltage vector modulation includes: determining the sector where the voltage vector is located based on the first voltage command and the second voltage command of the dq axis, calculating the duration of the effective voltage vector in the sector, and generating a corresponding pulse width modulation signal as the driving signal of the power module.

7. A control device for a permanent magnet generator, used to implement the control method described in claim 1, characterized in that, include: The voltage processing module is configured to acquire DC voltage commands and DC voltage sample values, calculate the difference between the two, and output the motor torque of the permanent magnet generator through a PI regulator. The torque limiting module is configured to compare the motor torque with a torque limit and output a torque command. When the motor torque is greater than the torque limit, the torque command is equal to the torque limit; otherwise, it is equal to the motor torque. The current command generation module, which has a built-in calibrated torque current meter, is configured to convert the torque command into a torque current command and an excitation current command. The first current regulation module is configured to acquire torque current sampling values, calculate the difference between the torque current sampling values ​​and the torque current command, and output the first voltage command of the dq axis through a PI regulator. The second current regulation module is configured to acquire the excitation current sampling value, calculate the difference between the excitation current sampling value and the excitation current command, and output the dq axis second voltage command through the PI regulator; The modulation drive module is configured to generate a power module drive signal by performing space voltage vector modulation on the first voltage command and the second voltage command of the dq axis to control the permanent magnet generator to charge the energy storage battery.

8. The apparatus according to claim 7, characterized in that, It also includes a torque limit determination module, which determines the torque limit based on the current state of charge and / or temperature of the energy storage battery, and sends the torque limit to the torque limiting module.

9. The control device for a permanent magnet generator according to claim 7, characterized in that, The PI regulators in the voltage processing module, the first current regulation module, and the second current regulation module are all equipped with parameter adjustment units. The parameter adjustment units are configured to dynamically adjust the proportional coefficient and integral coefficient of the PI regulator based on the rated parameters of the permanent magnet generator or the real-time voltage / current difference.

10. The control device for a permanent magnet generator according to claim 7, characterized in that, It also includes a signal filtering module, which is configured to filter the DC voltage sample value, torque current sample value and excitation current sample value, including first-order low-pass filtering or moving average filtering, and send the filtered signals to the voltage processing module, the first current regulation module and the second current regulation module respectively.

Images