Intelligent low-voltage dual-shaft servo driving system

Abstract

The application relates to an intelligent low-voltage double-shaft servo driving system, and belongs to the technical field of servo control. The system comprises a double-shaft servo driver, an azimuth motor, an elevation motor, an azimuth angle sensor and an elevation angle sensor, wherein the double-shaft servo driver is composed of a control board and a driving board. The control board receives a turning instruction and generates a PWM control signal based on an improved magnetic field orientation control algorithm; the driving board adopts a power circuit with double MOS transistors in parallel to amplify signals and independently drive the double-shaft motor. The azimuth and elevation angle sensors detect positions in real time and feed back, and the control board adjusts the PWM signal through current loop, speed loop and position loop closed loop. The control board also performs double-shaft collaborative energy optimization management, including real-time power budget allocation, brake timing peak shifting and capacity energy buffering. The system effectively improves the driving reliability and energy efficiency under low-voltage conditions.

Description

Technical Field

[0001] This application relates to the field of servo control technology, specifically to an intelligent low-voltage dual-axis servo drive system. Background Technology

[0002] Traditional servo drive systems, used in launcher turning control applications, typically employ high-voltage power supply schemes to achieve sufficient power density and dynamic response performance. However, this approach suffers from drawbacks such as large size and weight, and poor power supply compatibility. Under low-voltage (20-32VDC) power supply conditions, existing technologies struggle to address the energy management challenges of dual-axis collaborative operation: when the azimuth and pitch axes rotate simultaneously or are subjected to transient loads (such as recoil), peak current superposition effects occur, leading to a sudden drop in bus voltage or even system instability. Furthermore, the ratio of conduction losses to switching losses in power devices changes under low-voltage conditions. Traditional current sharing control strategies struggle to ensure current balance in dual MOSFET parallel applications, while the modulation linearity of the field-oriented control algorithm decreases under conditions of large voltage fluctuations, affecting position control accuracy. Summary of the Invention

[0003] To solve or at least partially solve the above-mentioned technical problems, embodiments of this application provide an intelligent low-voltage dual-axis servo drive system.

[0004] This application provides an intelligent low-voltage dual-axis servo drive system, including a dual-axis servo driver, an azimuth motor, a pitch motor, an azimuth angle sensor, and a pitch angle sensor. The dual-axis servo driver includes a control board and a drive board. The system is characterized by:

[0005] The control board is configured to receive switching commands via a CAN interface and generate PWM control signals based on an improved field-oriented control algorithm.

[0006] The driver board is configured to receive the PWM control signal and amplify the PWM control signal using a power circuit with dual MOS transistors in parallel, so as to independently drive the azimuth motor and the pitch motor.

[0007] The azimuth angle sensor is configured to detect the actual position of the azimuth axis and feed the azimuth position signal back to the control board;

[0008] The pitch angle sensor is configured to detect the actual position of the pitch axis and feed the pitch position signal back to the control board.

[0009] The control board is also configured to adjust the PWM control signal based on the azimuth position signal and the pitch position signal through closed-loop control of current loop, speed loop and position loop;

[0010] The control board is also configured to perform dual-axis coordinated energy optimization management, including:

[0011] The power budget management module is configured to monitor bus voltage and total current in real time, and allocate instantaneous power budgets for the azimuth axis and the pitch axis, and limit the speed or acceleration of non-critical axes based on motion priority;

[0012] The braking timing management module is configured to predict the braking time points of the azimuth motor and the pitch motor, and actively stagger the braking timing to avoid the superposition of energy feedback peaks.

[0013] The capacitive energy storage management module is configured to use the bus capacitor as a temporary energy storage buffer, control the capacitor discharge to support the system when the voltage drops sharply, and prioritize the capacitor to absorb part of the feedback energy when the voltage rises sharply.

[0014] Optionally, the driver board is further configured to perform current sharing control of dual MOSFETs in parallel, including:

[0015] The current sampling module is configured to monitor the phase current of the azimuth motor and the pitch motor in real time through a sampling resistor and an isolation amplifier, and output a current feedback signal.

[0016] The current sharing adjustment module is configured to receive the current feedback signal, compare the current difference of the parallel branch of the dual MOSFETs, and adjust the pulse width of the PWM control signal based on the comparison result to balance the current distribution of the parallel branch of the dual MOSFETs.

[0017] The thermal management coordination module is configured to combine the output of the current equalization adjustment module to disperse heat through the heat dissipation path of the drive board, thereby avoiding local overheating.

[0018] Optionally, the flow equalization module is further configured to perform temperature-compensated flow equalization control, including:

[0019] The parameter adjustment module is configured to monitor the junction temperature change of the parallel branch of the dual MOS transistors in real time, and adjust the current comparison threshold of the current sharing adjustment module based on the junction temperature change.

[0020] The load response module is configured to adjust the pulse width adjustment step size of the PWM control signal according to the load change rate when the azimuth motor or the pitch motor is subjected to transient load.

[0021] The calculation optimization module is configured to calculate the real-time power loss of the parallel branch of the dual MOS transistors using an improved loss model, and optimize the current sharing parameters based on the power loss results.

[0022] Optionally, the calculation optimization module is further configured to perform real-time power loss calculation based on real-time data, including:

[0023] The data acquisition submodule is configured to acquire the current value, junction temperature, and switching frequency of the parallel branch of the dual MOS transistors in real time;

[0024] The loss calculation submodule is configured to use an improved loss model to calculate real-time power loss in combination with the current value, junction temperature, and switching frequency. The improved loss model includes a conduction loss term and a switching loss term. The conduction loss term is calculated based on junction temperature-dependent on-resistance, and the switching loss term is calculated based on switching frequency and switching energy.

[0025] The optimization submodule is configured to adjust the current comparison threshold and PWM adjustment step size of the current sharing regulation module based on the real-time power loss.

[0026] Optionally, the loss calculation submodule is further configured to perform loss model optimization based on transient response and thermal management, including:

[0027] The transient compensation submodule is configured to monitor the transient current change rate of the azimuth motor and the pitch motor, and add a transient compensation term to the improved loss model to correct the calculation deviation of switching losses.

[0028] The thermal network submodule is configured to construct a dynamic thermal resistance model of the parallel branch of the dual MOS transistors, calculate the thermal gradient based on the real-time junction temperature and ambient temperature, and feed it back to the temperature dependence parameter of the conduction loss term.

[0029] The update submodule is configured to adjust the weight coefficients of the loss model based on the transient compensation term and the thermal gradient results to optimize the calculation accuracy of real-time power loss.

[0030] Optionally, the braking timing management module is further configured to perform braking timing optimization based on load prediction, including:

[0031] The load prediction submodule is configured to analyze the motion commands and historical load data of the azimuth motor and the pitch motor in real time to predict the occurrence time and intensity of transient load events.

[0032] The staggered braking submodule is configured to calculate the optimal braking time interval for the azimuth axis and the pitch axis based on predicted transient load events, and adjust the braking timing.

[0033] The effect evaluation submodule is configured to monitor bus voltage fluctuations and energy feedback peaks during the braking process and feed them back to the load prediction submodule to optimize the prediction model.

[0034] Optionally, the peak-shifting submodule is further configured to perform braking timing optimization based on multi-parameter fusion, including:

[0035] The interval calculation unit is configured to calculate the optimal braking time interval between the azimuth axis and the pitch axis by integrating transient load intensity, bus voltage margin and system thermal state parameters.

[0036] The timing adjustment unit is configured to adjust the braking timing based on the calculated optimal braking time interval using a priority allocation strategy.

[0037] The real-time optimization unit is configured to monitor the interval adjustment effect during braking and provide real-time feedback of the optimized interval calculation parameters.

[0038] Optionally, the real-time optimization unit is further configured to perform optimization based on multi-metric feedback, including:

[0039] The effect monitoring subunit is configured to monitor the interval adjustment effect in real time during braking, including bus voltage stability, energy feedback smoothness, and system response time.

[0040] The parameter optimization subunit is configured to adjust the fusion parameters in the interval calculation unit based on the monitoring results.

[0041] Optionally, the control board is configured to perform a process of generating a PWM control signal based on an improved field-oriented control algorithm, including:

[0042] The current sampling and coordinate transformation unit is configured to acquire the motor phase current through a three-phase current sampling circuit and use an anti-interference Clarke-Park transformation to convert the phase current into direct-axis current and quadrature-axis current components in a rotating coordinate system.

[0043] The modulation compensation unit is configured to adjust the modulation ratio of the space vector modulation algorithm according to the real-time bus voltage fluctuation to ensure that the linear modulation region is maintained under low voltage conditions.

[0044] The switching sequence optimization unit is configured to optimize the switching sequence for low-voltage, high-current scenarios, reduce the number of switching operations of power devices, and balance the distribution of switching losses.

[0045] Optionally, the modulation compensation unit is further configured to perform modulation optimization based on bus voltage, including:

[0046] The voltage monitoring subunit is configured to acquire the bus voltage signal in real time and calculate the voltage fluctuation rate and available voltage margin.

[0047] The modulation ratio adjustment subunit is configured to adjust the modulation ratio of the space vector modulation algorithm according to the voltage margin, and to ensure modulation linearity by using a piecewise linear interpolation algorithm.

[0048] The overmodulation management subunit is configured to enter overmodulation mode when the voltage drops, and uses a harmonic injection strategy to extend voltage utilization.

[0049] The performance evaluation subunit is configured to monitor modulation effects, including harmonic distortion and torque ripple, and provide feedback to optimize modulation parameters.

[0050] The system provided in this application has the following beneficial effects:

[0051] The intelligent low-voltage dual-axis servo drive system provided in this application achieves efficient and reliable operation in low-voltage environments through overall structural design and functional configuration. The system employs an improved field-oriented control algorithm to generate PWM control signals, combined with closed-loop control of current, speed, and position loops, enabling precise adjustment of the azimuth and pitch axis movements. This improves the accuracy and response speed of turning, avoiding control errors caused by voltage fluctuations in traditional low-voltage servo systems. The power circuit design with dual MOSFETs in parallel reduces conduction losses and enhances driving capability, allowing the system to maintain stable output under low-voltage, high-current conditions, reducing heat generation and energy waste. Dual-axis collaborative energy optimization management optimizes energy distribution when both axes operate simultaneously through real-time power budget allocation, braking timing staggering, and capacitive energy buffering, avoiding peak current superposition and energy feedback conflicts, ensuring stable system operation under complex conditions.

[0052] Furthermore, current sharing control improves the current balance of the parallel branches of the dual MOSFETs through current sampling and regulation, preventing localized overheating and device damage. Temperature compensation and load response mechanisms enhance the system's adaptability to varying temperature environments, while loss model optimization optimizes current sharing parameters and improves efficiency by calculating power losses in real time. Braking timing prediction and peak-shifting management reduce the cumulative effect of energy feedback, and combined with multi-parameter fusion and real-time feedback, further optimize the braking process. Improvements to the field-oriented control algorithm, including modulation ratio adjustment and switching sequence optimization, ensure linear modulation and low harmonic distortion under low-voltage fluctuations, improving control accuracy and system stability. Overall, the system exhibits higher reliability, energy efficiency, and environmental adaptability in low-voltage biaxial applications. Attached Figure Description

[0053] Figure 1 A schematic diagram of an intelligent low-voltage dual-axis servo drive system provided in this application embodiment;

[0054] Figure 2 This is a schematic diagram of a control board structure provided in an embodiment of this application. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this application clearer, specific embodiments of this application will be described in further detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely for explaining this application and not for limiting it. It should also be noted that, for ease of description, only the parts relevant to this application are shown in the drawings, not all of them. Before discussing exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe operations (or steps) as sequential processes, many of these operations can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but may also have additional steps not included in the drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0056] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0057] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0058] See Figure 1 and Figure 2 A smart low-voltage dual-axis servo drive system includes a dual-axis servo driver, an azimuth motor, a pitch motor, an azimuth angle sensor, and a pitch angle sensor. The dual-axis servo driver includes a control board and a drive board.

[0059] The control board is configured to receive torsion commands via a CAN interface and generate PWM control signals based on an improved field-oriented control algorithm.

[0060] The driver board is configured to receive PWM control signals and amplify the PWM control signals using a power circuit with dual MOSFETs in parallel to independently drive the azimuth and pitch motors.

[0061] The azimuth angle sensor is configured to detect the actual position of the azimuth axis and feed the azimuth position signal back to the control board;

[0062] The pitch angle sensor is configured to detect the actual position of the pitch axis and feed the pitch position signal back to the control board;

[0063] The control board is also configured to adjust the PWM control signal based on the azimuth and pitch position signals through closed-loop control of the current loop, speed loop, and position loop.

[0064] The control board is also configured to perform dual-axis coordinated energy optimization management, including:

[0065] The power budget management module is configured to monitor bus voltage and total current in real time and allocate instantaneous power budgets for the azimuth and pitch axes, limiting the speed or acceleration of non-critical axes based on motion priority.

[0066] The braking timing management module is configured to predict the braking time points of the azimuth motor and pitch motor, and actively stagger the braking timing to avoid the superposition of energy feedback peaks.

[0067] The capacitive energy storage management module is configured to use the bus capacitor as a temporary energy storage buffer, control the capacitor discharge to support the system when the voltage drops sharply, and prioritize the capacitor to absorb part of the feedback energy when the voltage rises sharply.

[0068] The specific implementation of the intelligent low-voltage dual-axis servo drive system relates to the system's operation in a low-voltage environment. The system achieves independent control of the azimuth and pitch axes through a dual-axis servo drive, which includes a control board and a drive board. The control board receives rotation commands from the host computer via a CAN interface, and these commands contain position or speed information. The control board processes the commands based on an improved field-oriented control algorithm. This improved algorithm converts the three-phase current into direct-axis and quadrature-axis components in a rotating coordinate system through coordinate transformation, thereby generating a PWM control signal. The generation process of the PWM control signal considers the synergistic effect of the current loop, speed loop, and position loop. The current loop samples the motor phase current in real time and provides feedback for adjustment; the speed loop calculates the actual speed using an angle sensor; and the position loop compares the commanded position with the actual position to reduce errors. This closed-loop control method ensures rotation accuracy and avoids control deviations caused by voltage fluctuations.

[0069] After receiving the PWM control signal, the driver board amplifies the signal using a power circuit with dual MOSFETs in parallel. This parallel design reduces the current stress on individual devices, and current sharing control ensures uniform current distribution. The power circuit then drives the azimuth and pitch motors with the amplified signal. The azimuth and pitch motors can be permanent magnet brushless motors, which are simple in structure and highly efficient. During motor operation, the azimuth and pitch angle sensors continuously monitor the axis position. The azimuth angle sensor feeds back the azimuth position signal to the control board, and the pitch angle sensor feeds back the pitch position signal. These feedback signals are used to update the closed-loop control circuit, ensuring a fast and stable system response.

[0070] In addition, a bleeder circuit can be integrated into the driver board to manage energy feedback during motor braking. When the motor decelerates or brakes, energy is fed back to the bus. The bleeder circuit monitors the bus voltage; if the voltage exceeds a set threshold, the energy bleeder function is activated, dissipating excess energy through the braking resistor to prevent voltage overshoot from damaging components. The operation of the bleeder circuit is coordinated with the system power state to ensure stable bus voltage under low-voltage conditions.

[0071] The control board and drive board are coupled through internal connection circuits to achieve data exchange and power transmission. The control board also performs a self-test function, checking the voltage status and motor connection status at the initial power-on stage. The self-test process includes checking the control voltage, drive voltage, motor winding continuity, and sensor status. If an abnormality is detected, fault information is reported to ensure safe system startup.

[0072] Dual-axis coordinated energy optimization management is one of the core functions of the system. The power budget management module monitors the bus voltage and total current in real time, and allocates instantaneous power budgets to the azimuth and pitch axes according to motion priority. For example, during turning, if the total current approaches the limit, the speed or acceleration of non-critical axes is limited to avoid peak current accumulation. The braking timing management module predicts the braking time points of the azimuth and pitch motors, and actively staggers the braking sequence based on motion curves and load changes to reduce the risk of energy feedback peak accumulation. The capacitive energy storage management module uses the bus capacitor as a temporary energy storage buffer. During voltage drops, it controls the capacitor discharge to support system power supply, and during voltage surges, the capacitor preferentially absorbs part of the feedback energy to smooth voltage fluctuations.

[0073] The entire system emphasizes real-time performance and adaptability. The control board continuously runs the field-oriented control algorithm, adjusting the PWM signal to match load changes. The driver board ensures efficient power output, sensor feedback provides the basis for closed-loop control, and energy management optimizes system energy efficiency. The system effectively handles energy distribution issues in dual-axis collaborative operation, reduces the impact of voltage fluctuations, improves motion smoothness, and enhances overall performance through self-checking and optimization management. The implementation method reflects the design considerations of a low-voltage servo system, such as reducing losses, adapting to a wide voltage range, and handling transient loads, thereby meeting the needs of demanding application scenarios.

[0074] In some implementations, the driver board is also configured to perform current sharing control of the dual MOSFETs in parallel, including:

[0075] The current sampling module is configured to monitor the phase current of the azimuth motor and pitch motor in real time through sampling resistors and isolation amplifiers, and output current feedback signals.

[0076] The current sharing adjustment module is configured to receive the current feedback signal, compare the current difference of the parallel branch of the dual MOSFETs, and adjust the pulse width of the PWM control signal based on the comparison result to balance the current distribution of the parallel branch of the dual MOSFETs.

[0077] The thermal management coordination module is configured to combine with the output of the current sharing adjustment module to disperse heat through the heat dissipation path of the driver board, thereby avoiding local overheating.

[0078] In the intelligent low-voltage dual-axis servo drive system, the drive board executes a current-sharing control process using dual parallel MOSFETs to ensure stable operation of the azimuth and pitch motors under low-voltage, high-current conditions. The current sampling module monitors the motor phase current in real time through sampling resistors and isolation amplifiers. The sampling resistors are connected in series with the motor phase lines, and the isolation amplifier converts the current signal into a differential voltage signal and outputs a current feedback signal, which is transmitted to the current-sharing adjustment module for processing. The monitoring process is performed at high frequency, matched to the PWM switching cycle, to capture transient current changes and avoid errors caused by sampling delays.

[0079] After receiving the current feedback signal, the current sharing adjustment module compares the current differences in the parallel branches of the dual MOSFETs and calculates the current imbalance. Based on the comparison results, the module adjusts the pulse width of the PWM control signal, slightly reducing the conduction time of the high-current branch and slightly increasing the conduction time of the low-current branch by fine-tuning the duty cycle, thereby achieving current balance. The adjustment process is performed smoothly to avoid system oscillations caused by sudden changes and ensure control stability. The current sharing adjustment module also considers the impact of temperature changes on current distribution, updating adjustment parameters in real time to adapt to different operating conditions.

[0080] The thermal management coordination module, in conjunction with the current sharing adjustment module, distributes heat through the heat dissipation path of the driver board. The driver board is mounted on the system baseplate, and heat is conducted to the external heat dissipation surface via thermal pads. The heat dissipation path is designed for uniform distribution to avoid hotspot formation. The module monitors the temperature rise of the power devices; when a localized temperature increase is detected, it assists in adjusting the current sharing parameters or temporarily reducing the load to coordinate heat dissipation. The thermal management process is synchronized with the current sharing control, ensuring the reliability of the system during long-term operation.

[0081] Throughout the current sharing control process, current sampling provides an accurate data foundation, current sharing adjustment achieves fine current balance, and thermal management enhances system robustness. This method effectively prevents device overheating or damage caused by current imbalance, improving the efficiency and lifespan of the dual MOSFET parallel structure. The system maintains stable current output under low-voltage conditions, reducing energy loss and adapting to load changes, ensuring consistent and reliable drive performance for both azimuth and pitch axes.

[0082] In some implementations, the flow sharing adjustment module is also configured to perform temperature-compensated flow sharing control, including:

[0083] The parameter adjustment module is configured to monitor the junction temperature change of the parallel branch of the dual MOS transistors in real time and adjust the current comparison threshold of the current sharing adjustment module based on the junction temperature change.

[0084] The load response module is configured to adjust the pulse width of the PWM control signal according to the load change rate when the azimuth motor or pitch motor is subjected to transient load.

[0085] The calculation optimization module is configured to calculate the real-time power loss of the parallel branch of the dual MOS transistors using an improved loss model, and optimize the current sharing parameters based on the power loss results.

[0086] In the intelligent low-voltage dual-axis servo drive system, the current sharing adjustment module performs a temperature-compensated current sharing control process, improving system stability under varying temperature environments through real-time monitoring and adjustment. The parameter adjustment module integrates a temperature sensor, such as a thermistor, installed near the parallel branch of the dual MOSFETs to directly detect junction temperature changes. Sensor data is sampled at high frequency, and the junction temperature value is input to the processing unit after analog-to-digital conversion. The module adjusts the current comparison threshold based on junction temperature changes. When the junction temperature rises, the threshold is appropriately lowered to improve current sharing sensitivity and avoid current unevenness caused by temperature drift. The adjustment process is performed smoothly, avoiding abrupt changes that could affect system response and ensuring that the threshold change matches the temperature curve.

[0087] The load response module monitors load changes of the azimuth and pitch motors, acquires phase current data in real time through a current sampling circuit, and calculates the load change rate. The load change rate is expressed as the differential value of the current, using the formula: ,in This is the current value. The value represents time and reflects the intensity of the transient load. The module adjusts the step size of the PWM control signal pulse width based on the rate of change. When the rate of change is large, the step size is increased for a faster response; when the rate of change is small, the step size is decreased to maintain stability. This correction process works in conjunction with current sharing regulation to ensure that current balance is not affected when subjected to transient loads such as recoil forces.

[0088] The calculation optimization module uses an improved loss model to calculate the real-time power loss of the parallel branch of two MOSFETs. The model includes conduction loss and switching loss terms. The conduction loss is calculated based on the junction temperature-dependent on-resistance, as shown in the formula: ,in For branch current, For junction temperature The on-resistance is given by [formula missing], which increases with increasing temperature. Switching losses are calculated based on the switching frequency and switching energy, using the following formula: [formula missing]. ,in For switching frequency, For switching energy. The module collects current, junction temperature, and switching frequency data in real time, and substitutes them into the model to calculate the total power loss. Furthermore, based on the loss results, the current sharing parameters are optimized, such as adjusting the current comparison threshold or PWM pulse width, to ensure that the current sharing control is more conservative when the power loss is high, thus avoiding overheating.

[0089] The entire temperature compensation process emphasizes real-time performance. The parameter adjustment module handles the impact of temperature, the load response module responds to load changes, and the calculation optimization module optimizes loss management. The system uses software algorithms to achieve data exchange between modules. Temperature monitoring provides environmental input, load change rate detection captures transient events, and loss calculation provides an energy benchmark, ensuring consistent performance of current sharing control under different operating conditions. This method reduces current deviations caused by temperature changes or load fluctuations, improves the reliability and efficiency of the dual MOSFET parallel structure, and reduces energy waste through optimization. The implementation method focuses on process control, requiring no complex external intervention, and achieves robust operation under low-voltage environments.

[0090] The system effectively suppresses temperature drift and transient interference, maintains current balance, and enhances overall durability. The process design takes into account the specific requirements of low-voltage servo systems, such as wide temperature range adaptability and dynamic load response, and simplifies implementation complexity through modular processing.

[0091] In some implementations, the computational optimization module is also configured to perform real-time power loss calculations based on real-time data, including:

[0092] The data acquisition submodule is configured to acquire the current value, junction temperature and switching frequency of the parallel branch of the dual MOSFETs in real time;

[0093] The loss calculation submodule is configured to use an improved loss model to calculate real-time power loss by combining current value, junction temperature, and switching frequency. The improved loss model includes a conduction loss term and a switching loss term. The conduction loss term is calculated based on junction temperature-dependent on-resistance, and the switching loss term is calculated based on switching frequency and switching energy.

[0094] The optimization submodule is configured to adjust the current comparison threshold and PWM adjustment step size of the current sharing regulation module based on real-time power loss.

[0095] In the intelligent low-voltage dual-axis servo drive system, the calculation and optimization module performs real-time power loss calculation based on real-time data, improving system energy efficiency through data acquisition, loss calculation, and parameter optimization. The data acquisition submodule operates at high frequency, acquiring the current value, junction temperature, and switching frequency of the parallel branch of the dual MOSFETs in real time. The current value is acquired through a sampling resistor and an isolation amplifier. The sampling resistor is connected in series in the motor phase line, and the isolation amplifier converts the current signal into a differential voltage and outputs a digital value. The sampling frequency matches the PWM switching cycle to ensure real-time data accuracy. The junction temperature is detected by a temperature sensor integrated near the MOSFET, such as a thermistor. The sensor data is converted from analog to digital and input to the processing unit. The junction temperature value is updated every millisecond, reflecting the thermal state of the device. The switching frequency is directly read from the PWM generation circuit of the control board, and the frequency value is adjusted according to the load based on the timer setting of the improved field-oriented control algorithm. The acquisition process is continuous, and the data is transmitted to the loss calculation submodule via the internal bus, providing input for the loss model.

[0096] The loss calculation submodule uses an improved loss model to calculate real-time power loss, which includes conduction loss and switching loss terms. The conduction loss term is calculated based on the junction temperature-dependent on-resistance, as shown in the formula: ,in This is the branch current value. For junction temperature The on-resistance is determined by the temperature curve, which increases with temperature and is obtained from a pre-stored temperature-resistance curve table. Switching losses are calculated based on the switching frequency and switching energy, using the following formula: ,in For switching frequency, and These represent the switching energy for turning on and off, respectively, and are functions of junction temperature. The values ​​are extracted from MOSFET datasheets or experimental data and updated with temperature changes. Total power loss is the sum of the two terms: The calculation process is executed once per PWM cycle, and the model parameters are adjusted in real time to ensure accurate loss estimation under low voltage and high current conditions.

[0097] The optimization submodule adjusts the current comparison threshold and PWM adjustment step size of the current sharing regulation module based on real-time power loss results. After receiving the loss value, the module compares it with the preset threshold. If the loss is high, the current comparison threshold is lowered to improve current sharing sensitivity and avoid overcurrent; simultaneously, the PWM adjustment step size is increased to speed up the response. The adjustment process is smooth, using a gradient descent algorithm to fine-tune parameters and prevent oscillations caused by sudden changes. The optimization submodule also interacts with the thermal management coordination module, coordinating adjustments to the heat dissipation strategy when the loss indicates an overheating risk. The entire process operates in a closed loop, with loss calculation providing the data foundation and optimization actions ensuring system stability under varying operating conditions.

[0098] The parameter adjustments of the optimization submodules are synchronized with the system status in real time. This method improves the accuracy of power loss estimation, makes current sharing control more adaptable to temperature changes and load fluctuations, reduces energy waste, and enhances the system's reliability in low-voltage environments.

[0099] In some implementations, the loss calculation submodule is also configured to perform loss model optimization based on transient response and thermal management, including:

[0100] The transient compensation submodule is configured to monitor the transient current change rate of the azimuth and pitch motors and add a transient compensation term to the improved loss model to correct the calculation deviation of switching losses.

[0101] The thermal network submodule is configured to construct a dynamic thermal resistance model for the parallel branch of dual MOS transistors, calculate the thermal gradient based on the real-time junction temperature and ambient temperature, and feed it back to the temperature dependence parameter of the conduction loss term.

[0102] The update submodule is configured to adjust the weighting coefficients of the loss model based on the transient compensation term and thermal gradient results to optimize the calculation accuracy of real-time power loss.

[0103] In the intelligent low-voltage dual-axis servo drive system, the calculation and optimization module performs a loss model optimization process based on transient response and dynamic thermal management. It improves the accuracy of power loss calculations through transient compensation, thermal network modeling, and updates. The transient compensation submodule is integrated into the control loop, monitoring the transient current change rate of the azimuth and pitch motors in real time. The monitoring process utilizes the high-frequency sampling capability of the current sampling circuit to collect phase current data and calculate the current differential value, as shown in the formula: ,in This is the current value. For time, this value reflects the severity of load changes. When a transient event is detected, such as a current spike caused by recoil, the submodule adds a transient compensation term to the improved loss model. The compensation term takes the form of... ,in This is the transient compensation coefficient, determined based on experimental data. The compensation term corrects for calculation errors in switching losses, as transient changes can lead to inaccurate estimations of switching energy. Adding this term updates the total power loss formula to... This allows for a more accurate reflection of actual losses.

[0104] The thermal network submodule constructs a dynamic thermal resistance model for the parallel branch of two MOSFETs to handle the impact of junction temperature variations on conduction losses. The module is based on real-time junction temperature and ambient temperature data; the junction temperature is obtained from a temperature sensor, and the ambient temperature is read from the system environmental monitoring unit. The thermal resistance model includes the junction-to-case thermal resistance. Thermal resistance from shell to environment The model calculates the thermal gradient using the following formula: ,in For the junction temperature, For ambient temperature, For real-time power loss, The thermal time constant is used. The thermal gradient result is fed back to the temperature-dependent parameter of the conduction loss term. This makes the on-resistance change with temperature more accurate, reducing estimation errors caused by thermal delay.

[0105] The update submodule adjusts the weighting coefficients of the loss model based on the transient compensation term and thermal gradient results. The module analyzes the compensation term and thermal gradient data in real time and uses an iterative algorithm to optimize the weights, such as adjusting the contribution ratio of conduction loss and switching loss to the total loss. The optimization process is based on monitored voltage stability and thermal state; if the transient compensation term is large, the weight of switching loss is increased; if the thermal gradient indicates overheating risk, the weight of conduction loss is decreased. After adjustment, the loss model is more adapted to the operating conditions, and the calculation accuracy is improved. The entire optimization process runs in a closed loop, executed once per PWM cycle to ensure real-time performance.

[0106] The transient compensation submodule's high-frequency monitoring ensures rapid response to load changes, the thermal network submodule's thermal resistance model improves temperature adaptability, and the weight adjustment of the update submodule enhances model robustness. This approach reduces loss calculation errors caused by transient events and thermal dynamics, improving the system's energy efficiency management under low voltage and high current conditions.

[0107] In some implementations, the braking timing management module is also configured to perform load prediction-based braking timing optimization, including:

[0108] The load prediction submodule is configured to analyze the motion commands and historical load data of the azimuth and pitch motors in real time to predict the occurrence time and intensity of transient load events.

[0109] The staggered braking submodule is configured to calculate the optimal braking time interval for the azimuth and pitch axes based on predicted transient load events and adjust the braking timing.

[0110] The effect evaluation submodule is configured to monitor bus voltage fluctuations and energy feedback peaks during the braking process and feed them back to the load prediction submodule to optimize the prediction model.

[0111] In an intelligent low-voltage dual-axis servo drive system, the braking timing management module performs a load prediction-based braking timing optimization process. This improves the system's energy management efficiency during braking by predicting transient load events, calculating the optimal braking time interval, and monitoring feedback. The load prediction submodule is integrated into the control board's software algorithm, analyzing motion commands and historical load data from the azimuth and pitch motors in real time. Motion commands are received from the CAN interface and include position or speed adjustment information. Historical load data is stored in the control board's memory, recording events such as recoil force occurrence time and intensity. The analysis process uses a sliding window algorithm to process recent data sequences and identify load patterns, such as inferring the probability of transient load occurrences based on acceleration change trends. The prediction results output as estimates of the time and intensity of load events, guiding braking timing adjustments.

[0112] The peak-shifting submodule calculates the optimal braking time interval between the azimuth and pitch axes based on load prediction results. The calculation process considers load intensity, system priority, and energy feedback risk; the interval value is determined using a weighting function, as shown in the formula. ,in Indicates the predicted load intensity. Indicates motion priority (e.g., azimuth axis turning takes priority over pitch axis turning). This represents the energy feedback risk assessment factor. (Function) Based on experimental data fitting, the interval values ​​are ensured to adapt to the fluctuations of low-pressure scenarios. After calculation, the module adjusts the braking timing, for example, delaying the braking initiation time of the pitch axis so that the azimuth axis completes braking first, avoiding the superposition of energy feedback peaks. The adjustment process is carried out smoothly, with gradual changes in timing to prevent system oscillations.

[0113] The performance evaluation submodule monitors bus voltage fluctuations and energy feedback peaks during braking to optimize the prediction model. Monitoring utilizes a voltage sampling circuit to collect bus voltage data in real time, calculates fluctuation amplitude and frequency, and records the number of triggers and duration of the discharge circuit to assess the stability of energy feedback. Monitoring data is fed back to the load prediction submodule to update the historical database and adjust prediction algorithm parameters, such as correcting the sliding window size or weighting coefficients. The feedback loop runs continuously to ensure the prediction model improves as operating conditions change.

[0114] This method reduces voltage spikes or energy waste caused by braking conflicts, improves the reliability of dual-axis collaborative operation, and makes the system more adaptable to load changes in low-voltage environments. The synergy between braking timing adjustment and energy management functions enhances overall energy efficiency, while a feedback mechanism prevents accumulated errors.

[0115] In some implementations, the peak-shifting submodule is also configured to perform braking timing optimization based on multi-parameter fusion, including:

[0116] The interval calculation unit is configured to calculate the optimal braking time interval between the azimuth and pitch axes by integrating transient load intensity, bus voltage margin and system thermal state parameters.

[0117] The timing adjustment unit is configured to adjust the braking timing based on the calculated optimal braking time interval using a priority allocation strategy.

[0118] The real-time optimization unit is configured to monitor the interval adjustment effect during braking and provide real-time feedback of the optimized interval calculation parameters.

[0119] In the intelligent low-voltage dual-axis servo drive system, the peak-shifting submodule performs a braking timing optimization process based on multi-parameter fusion. This process improves the accuracy of braking energy management by comprehensively calculating braking intervals, adjusting timing, and optimizing real-time feedback. The interval calculation unit collects transient load intensity, bus voltage margin, and system thermal state parameters in real time. Load intensity is obtained from the output of the load prediction submodule, representing the torque magnitude of transient events; voltage margin is monitored by the bus voltage sampling circuit, calculating the ratio of the current voltage to the rated voltage; and thermal state parameters are read from temperature sensors, reflecting the system's heat dissipation conditions. These parameters are processed using a weighted fusion algorithm, as shown in the formula: ,in Indicates the optimal braking time interval. Based on the basic interval threshold, , and These represent load strength, voltage margin, and thermal state factor, respectively. , , These are weighting coefficients. The calculation process considers the coupling relationship between parameters; for example, when the voltage margin is low, the coefficient is increased. The weighting is prioritized to ensure voltage stability. The interval value is updated once per braking cycle to ensure adaptation to changes in operating conditions.

[0120] The timing adjustment unit adjusts the braking timing based on the calculated optimal braking time interval using a priority allocation strategy. The unit first analyzes the motion states of the azimuth and pitch axes, determining the current turning phase through position commands. If the azimuth axis is in a critical positioning segment, it is assigned higher priority. During adjustment, lower-priority axes are braked first, while the braking timing of higher-priority axes is delayed, with the delay time following a specific pattern. The adjustment process employs a gradual acceleration algorithm to avoid mechanical shocks caused by abrupt timing changes. Simultaneously, the unit monitoring system responds to changes in motor speed, fine-tuning the interval value to compensate for external disturbances.

[0121] The real-time optimization unit monitors the interval adjustment effect during braking, including bus voltage fluctuation amplitude, energy feedback peak frequency, and system response delay. Monitoring data is acquired through voltage sampling and discharge circuit trigger recording. The unit uses a recursive algorithm to analyze data deviations; if excessive voltage fluctuations or increased response delays are detected, the weighting coefficients in the interval calculation unit are adjusted. , , The optimization process iterates using a sliding window approach, preserving recent data trends to ensure that parameter updates are both timely and avoid oscillations.

[0122] The entire optimization process achieves closed-loop control through software algorithms. The interval calculation unit provides theoretical interval values, the timing adjustment unit executes actual actions, and the real-time optimization unit corrects model parameters. This method significantly improves the adaptability of braking timing under dynamic loads and reduces energy feedback superposition or voltage instability problems caused by timing conflicts. The implementation process emphasizes the rationality of parameter fusion and the smoothness of adjustment, enabling the azimuth and pitch axes to work in coordination during the braking phase, thus enhancing the system's reliability in low-voltage scenarios.

[0123] In some implementations, the real-time optimization unit is also configured to perform optimization based on multi-metric feedback, including:

[0124] The effect monitoring subunit is configured to monitor the interval adjustment effect in real time during braking, including bus voltage stability, energy feedback smoothness, and system response time.

[0125] The parameter optimization subunit is configured to adjust the fusion parameters in the interval calculation unit based on the monitoring results.

[0126] In the intelligent low-voltage dual-axis servo drive system, the real-time optimization unit executes an optimization process based on multi-index feedback. It improves the accuracy of braking timing management by monitoring braking effects and adjusting parameters from multiple dimensions. The effect monitoring subunit simultaneously tracks several key indicators during braking, including bus voltage stability, energy feedback smoothness, and system response time. Bus voltage stability is analyzed by sampling the bus voltage signal at high frequency to calculate the voltage fluctuation rate, i.e., the amplitude and frequency of voltage changes per unit time. Energy feedback smoothness is evaluated by recording the number of triggers and duration of the discharge circuit to determine whether the energy feedback is stable. System response time is timed from the issuance of the braking command to the actual entry of the system into a stable braking state, reflecting the timeliness of timing adjustments. Monitoring data is uploaded to the processing unit in real time to form a comprehensive evaluation result.

[0127] The parameter optimization subunit adjusts the fusion parameters in the interval calculation unit based on monitoring results. This unit establishes a parameter adjustment mapping relationship: when the bus voltage fluctuation rate is high, the weight of the voltage margin-related parameter β is adjusted first to enhance voltage stability control; when the energy feedback smoothness is poor, the weight of the load intensity parameter α is adjusted to optimize load distribution; when the system response time is delayed, the weight of the thermal state parameter γ is adjusted synchronously to improve system dynamic performance. The parameter adjustment adopts a gradual strategy, avoiding system oscillations through small-step iterations, while considering the coupling effects between parameters to ensure a smooth and reliable adjustment process.

[0128] The optimization process forms a closed-loop control, with effect monitoring providing real-time feedback and parameter optimization enabling dynamic adjustments. The monitoring algorithm uses a sliding window to manage data, retaining recent historical records to identify trends, while the optimization algorithm ensures the system's adaptability under complex low-pressure conditions by weighted balancing of the priorities of different indicators. This approach significantly improves the intelligence level of braking timing management, making the coordination of the azimuth and pitch axes more precise and reliable during the braking phase.

[0129] In some implementations, the control board is configured to perform a process of generating PWM control signals based on an improved field-oriented control algorithm, including:

[0130] The current sampling and coordinate transformation unit is configured to acquire the motor phase current through a three-phase current sampling circuit and use an anti-interference Clarke-Park transformation to convert the phase current into direct-axis current and quadrature-axis current components in a rotating coordinate system.

[0131] The modulation compensation unit is configured to adjust the modulation ratio of the space vector modulation algorithm according to the real-time bus voltage fluctuation to ensure that the linear modulation region is maintained under low voltage conditions.

[0132] The switching sequence optimization unit is configured to optimize the switching sequence for low-voltage, high-current scenarios, reduce the number of switching operations of power devices, and balance the distribution of switching losses.

[0133] In the intelligent low-voltage dual-axis servo drive system, the control board executes a process of generating PWM control signals based on an improved field-oriented control algorithm. Precise control is achieved through current sampling and coordinate transformation, modulation compensation, and switching sequence optimization. The current sampling and coordinate transformation unit acquires the motor phase currents through a three-phase current sampling circuit. This circuit includes sampling resistors and isolation amplifiers. The sampling resistors are connected in series with the motor phase lines, and the isolation amplifier converts the current signal into a differential voltage and outputs it to the processing unit. The sampling process is performed at high frequency, matching the PWM switching frequency to ensure real-time data transmission. After acquiring the phase currents, the unit uses an anti-interference Clarke-Park transformation to convert the currents in the three-phase stationary coordinate system into direct-axis and quadrature-axis current components in the rotating coordinate system. The Clarke transformation converts the three-phase currents into two-phase stationary coordinate system currents, and the Park transformation further projects the two-phase currents onto the rotating coordinate system. Anti-interference algorithms, such as filters and time deviation compensation, are incorporated during the transformation process to reduce the impact of switching noise and sampling delay. The transformed direct-axis and quadrature-axis currents are used for the core calculations of the field-oriented control algorithm.

[0134] The modulation compensation unit adjusts the modulation ratio of the space vector modulation algorithm based on real-time bus voltage fluctuations. The unit monitors the bus voltage signal, acquires the voltage value through a voltage sampling circuit, and calculates the voltage fluctuation rate and available voltage margin. Based on the voltage margin, the modulation compensation unit adjusts the modulation ratio, using a piecewise linear interpolation algorithm to divide the modulation ratio into multiple intervals, adjusting linearly within each interval to ensure a linear modulation region is maintained under low-voltage conditions. When the voltage drops, the modulation compensation unit intelligently enters overmodulation mode, employing a harmonic injection strategy to expand voltage utilization, for example, injecting the third harmonic to compensate for insufficient voltage. The adjustment process is performed smoothly, avoiding output torque pulsations caused by sudden modulation changes.

[0135] The switching sequence optimization unit optimizes the switching sequence for low-voltage, high-current scenarios, reducing the number of switching operations on power devices and balancing the distribution of switching losses. The unit analyzes the action sequence of the fundamental voltage vector and redesigns the switching sequence, such as adjusting the distribution position of the zero vector to reduce switching actions within a single PWM cycle. Simultaneously, the unit adjusts the switching frequency based on heatsink temperature and output power, optimizing efficiency while ensuring heat dissipation. The optimized switching sequence works in conjunction with the modulation compensation unit to ensure uniform distribution of switching losses under low-voltage fluctuations and prevent localized overheating.

[0136] The entire PWM control signal generation process operates in a closed loop. Current sampling provides the feedback basis, coordinate transformation enables magnetic field orientation, modulation compensation adapts to voltage fluctuations, and switching sequence optimization improves efficiency. The system reduces hardware complexity through algorithm integration, utilizes existing circuit resources for current sampling, and implements modulation compensation and switching sequence optimization in software. This approach improves the quality control of the PWM signal in low-voltage scenarios, making the drive of the azimuth and pitch motors more precise and reliable.

[0137] The improved field-oriented control algorithm achieves decoupled control through coordinate transformation, thereby improving torque response speed;

[0138] Modulation compensation ensures modulation linearity under voltage fluctuations and avoids nonlinear distortion;

[0139] Optimized switching sequences reduce switching losses and extend device lifespan. The system maintains high performance in low-voltage environments, meeting demanding application requirements.

[0140] In some implementations, the modulation compensation unit is also configured to perform modulation optimization based on bus voltage, including:

[0141] The voltage monitoring subunit is configured to acquire the bus voltage signal in real time and calculate the voltage fluctuation rate and available voltage margin.

[0142] The modulation ratio adjustment subunit is configured to adjust the modulation ratio of the space vector modulation algorithm according to the voltage margin, and a piecewise linear interpolation algorithm is used to ensure modulation linearity.

[0143] The overmodulation management subunit is configured to enter overmodulation mode when the voltage drops, and uses a harmonic injection strategy to extend voltage utilization.

[0144] The performance evaluation subunit is configured to monitor modulation effects, including harmonic distortion and torque ripple, and provide feedback to optimize modulation parameters.

[0145] In the intelligent low-voltage dual-axis servo drive system, the modulation compensation unit performs a modulation optimization process based on the bus voltage. By real-time monitoring of voltage fluctuations, adjusting the modulation ratio, managing overmodulation, and evaluating modulation effects, it improves the system's control accuracy and stability under low-voltage conditions. The voltage monitoring subunit is integrated into the signal processing circuit of the control board, acquiring the bus voltage signal in real time. The acquisition process utilizes a voltage sampling circuit, which includes voltage divider resistors and operational amplifiers, to convert the high-voltage bus signal into a processable low-voltage signal and output digital values ​​through an analog-to-digital converter. The sampling frequency is set relatively high to match the PWM switching cycle, ensuring real-time data transmission. After acquisition, the voltage monitoring subunit calculates the voltage fluctuation rate, i.e., the amplitude and frequency of voltage change per unit time, and simultaneously calculates the available voltage margin, expressed as the ratio of the current voltage to the rated voltage. The calculation is based on a sliding window algorithm, processing recent voltage sequences to identify fluctuation trends and provide a basis for subsequent adjustments.

[0146] The modulation ratio adjustment subunit adjusts the modulation ratio of the space vector modulation algorithm based on voltage margin. The adjustment process employs a piecewise linear interpolation algorithm, dividing the modulation ratio into multiple intervals, such as high margin, medium margin, and low margin regions. Within each interval, the modulation ratio changes linearly with the voltage margin. When the voltage margin is high, the modulation ratio maintains a standard value to ensure output linearity; when the voltage margin decreases, the modulation ratio is appropriately reduced to avoid entering the nonlinear region and causing distortion. The interpolation algorithm provides a smooth transition, preventing torque ripple caused by abrupt changes. Adjustment is performed in real time, updated once per PWM cycle to ensure rapid response. The unit also collaborates with a field-oriented control algorithm to fine-tune the modulation parameters according to voltage changes, maintaining system stability.

[0147] The overmodulation management subunit intelligently enters overmodulation mode upon detecting a voltage drop. The unit monitors the voltage margin, and triggers overmodulation when the margin falls below a set threshold. The overmodulation mode employs a harmonic injection strategy, such as injecting a third harmonic component, to compensate for insufficient voltage and expand voltage utilization. The injection process is achieved by modifying the space vector modulation waveform; the harmonic components are superimposed on the fundamental frequency to enhance output capability. The unit simultaneously manages the entry and exit of overmodulation, gradually de-modulating when the voltage recovers to avoid sudden output changes. The management strategy considers system load conditions; for example, it triggers overmodulation earlier under heavy load to ensure consistent performance.

[0148] The performance evaluation subunit monitors the modulation effect, including harmonic distortion and torque ripple. Harmonic distortion is obtained by analyzing the motor current waveform using Fast Fourier Transform (FFT) to calculate the total harmonic distortion rate and assess modulation quality. Torque ripple is indirectly inferred from speed loop fluctuations, reflecting output smoothness. Monitoring data is fed back to the modulation compensation unit in real time for parameter optimization. For example, if harmonic distortion is high, the interpolation parameters of the modulation ratio adjustment subunit are adjusted; if torque ripple is significant, the trigger threshold of the over-modulation management subunit is corrected. The feedback loop operates in a closed loop to ensure that modulation optimization adapts to changes in operating conditions.

[0149] Throughout the optimization process, voltage monitoring provides the input reference, modulation ratio adjustment maintains linearity, overmodulation management expands the dynamic range, and performance evaluation ensures long-term reliability. The system integrates each subunit through software algorithms, utilizing existing hardware resources such as voltage sampling circuits and control processors to reduce implementation complexity. This approach improves the adaptability of the modulation stage under low-voltage fluctuations, reduces control deviations caused by voltage changes, and enhances the drive consistency of the azimuth and pitch motors.

[0150] In an intelligent low-voltage dual-axis servo drive system, the parameter optimization unit, as an enhancement module of the improved field-oriented control algorithm, enhances the system's stability under low-voltage fluctuations by adjusting control loop parameters in real time. The parameter optimization unit monitors system state variables, including bus voltage, output current, motor speed, and ambient temperature. This data is acquired from a sensor network, such as voltage sampling circuits, current isolation amplifiers, and temperature sensors. The monitoring process is based on the PWM switching frequency to ensure real-time data acquisition.

[0151] The parameter optimization unit comprises four core submodules: a state monitoring submodule, a parameter mapping submodule, a real-time adjustment submodule, and an iterative learning submodule. The state monitoring submodule is responsible for collecting system operation data and processing recent data sequences using a sliding window algorithm to identify trend changes. For example, when a rapid drop in bus voltage from its rated value is detected, the state monitoring submodule will flag the event and calculate the volatility. The parameter mapping submodule establishes a correspondence between state variables and control parameters based on fuzzy logic, converting monitoring data into parameter adjustment commands. The mapping relationship is obtained through training with experimental data and stored in a lookup table structure.

[0152] The submodule's execution parameters are dynamically optimized in real time, using the gradient descent method to fine-tune the PID gains of the current loop, speed loop, and position loop. When the voltage fluctuation rate increases, the submodule adjusts according to the formula... ,in For the new gain value, This is the original gain value. For adaptive coefficients, This refers to voltage fluctuation rate. The adjustment process is carried out smoothly to avoid system oscillations caused by sudden changes. The iterative learning submodule optimizes the mapping relationship through historical performance data and updates the lookup table content using recursive least squares, enabling the system to self-optimize over time.

[0153] After the current sampling and coordinate transformation unit outputs the direct-axis and quadrature-axis currents, the parameter optimization unit adjusts the integral gain of the current loop in real time to ensure that the current tracking accuracy does not decrease when the voltage fluctuates. When the modulation compensation unit adjusts the modulation ratio, the parameter optimization unit simultaneously optimizes the speed loop parameters to maintain consistent dynamic response.

[0154] The parameter optimization unit adapts control parameters to varying operating conditions, resolving the issue of fixed parameter mismatch in low-voltage environments. The system remains stable under voltage fluctuations, responds quickly to sudden load changes, and maintains accuracy under temperature variations. The unit's learning mechanism also extends algorithm lifespan and reduces maintenance requirements. This approach enhances system reliability in harsh low-voltage scenarios, providing an intelligent control foundation for dual-axis servo drives.

[0155] In the intelligent low-voltage dual-axis servo drive system, the parameter optimization unit enhances robustness through a hardware-based co-optimization layer. Building upon existing state monitoring, the parameter optimization unit adds real-time tracking of power device junction temperature and electromagnetic compatibility (EMC) indicators. Junction temperature data is acquired from a temperature sensor embedded near the MOSFET on the driver board, with the sampling frequency synchronized with the PWM cycle to ensure real-time data accuracy. EMC adaptability is assessed by monitoring power line noise and signal integrity, utilizing existing current and voltage sampling circuits to analyze the noise spectrum and identify interference events. Monitoring data is linked to algorithm parameter adjustments. When the junction temperature rises, the parameter optimization unit reduces the gain of the current loop to decrease heat accumulation caused by switching losses; when increased EMC noise is detected, the parameter learning process is paused to prevent erroneous data from contaminating the model.

[0156] The hardware co-optimization layer employs a priority scheduling mechanism to handle parameter conflicts during multi-axis coordination. The layer includes arbitration logic that allocates parameter adjustment permissions based on motion state. For example, when the azimuth axis is in the precise positioning stage, its velocity loop parameters are optimized first, while the pitch axis parameters remain conservatively set to prevent mutual interference between the two axes. The arbitration result is transmitted to the control loops of each axis via the control board's bus, ensuring synchronized adjustment actions. The optimization layer also coordinates with the bleeder circuit and heat dissipation path. When parameter adjustments indicate an overheating risk, the bleeder circuit is triggered to activate in advance, or the cooling fan speed is adjusted to provide protection.

[0157] The learning mechanism of the parameter optimization unit enhances robustness by introducing outlier filtering and redundancy checks to improve the reliability of iterative learning. When storing historical data, records from extreme conditions, such as high-temperature or high-noise events, are marked. The learning algorithm weights these data to reduce their impact on the model. The verification process uses checksum algorithms to verify data integrity and prevent parameter drift caused by transmission errors. The entire optimization layer runs as a background task with lower priority than real-time control, but it is updated frequently to ensure timely response.

[0158] The hardware co-optimization layer enables the parameter optimization unit to adapt more comprehensively to low-pressure environments. The system maintains control stability under thermal fluctuations, preserves parameter accuracy under EMC interference, and avoids conflicts during multi-axis coordination. This improves overall reliability and extends system lifespan.

[0159] In intelligent low-voltage dual-axis servo drive systems, a dynamic thermal balancing control unit can also be included. This unit optimizes heat dissipation management by integrating load prediction and real-time thermal monitoring. As an extension module of the control board software, the dynamic thermal balancing control unit obtains future load trend data for the azimuth and pitch axes from the load prediction submodule, including motion commands and transient events such as recoil. Simultaneously, the unit collects heat distribution data in real time through a multi-point temperature sensor network on the drive board, generating a heat map. A fuzzy logic algorithm is used to calculate the thermal risk index. When an increase in load on a certain axis is predicted, the heat dissipation strategy is adjusted in advance, such as activating auxiliary cooling measures or dynamically allocating power consumption. The thermal balancing strategy works in conjunction with motion priority to ensure uniform heat load distribution during dual-axis collaborative operation, avoiding localized overheating. The unit operates at high frequency, matched to the PWM cycle, ensuring timely response. This unit can significantly reduce the probability of localized overheating, improve the lifespan of power devices, and reduce the risk of thermal failure. Through predictive control and optimized heat dissipation energy consumption, the system remains stable during long-term high-load operation and adapts to rapid changes in low-voltage scenarios.

[0160] In intelligent low-voltage dual-axis servo drive systems, the health status prediction unit enables predictive maintenance through long-term monitoring of system parameters. Integrated into the control board software, this unit serves as an enhancement module for self-testing, collecting current waveforms, temperature curves, and vibration data (via a newly added vibration sensor). Data is recorded periodically, and machine learning algorithms, such as time series analysis, are used to train a health model and calculate a health index. When a performance degradation trend is detected, such as a slow increase in the on-resistance of a MOSFET, the unit triggers an early warning and adjusts control parameters, such as the gain of an improved field-oriented control algorithm, to compensate for the performance decline. This unit works in conjunction with the maintenance interface to generate maintenance suggestions, reducing downtime. It can identify potential faults in advance, improving system reliability and reducing sudden shutdowns. Adjustments extend component lifespan, making it suitable for long-term deployment in harsh low-voltage environments and enhancing task continuity.

[0161] The above description is merely a preferred embodiment and the technical principles employed in this application. This application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions that can be made by those skilled in the art will not depart from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of this application, the scope of which is determined by the scope of the claims.

Claims

1. An intelligent low-voltage dual-axis servo drive system, comprising a dual-axis servo driver, an azimuth motor, a pitch motor, an azimuth angle sensor, and a pitch angle sensor, wherein the dual-axis servo driver comprises a control board and a drive board, characterized in that: The control board is configured to receive switching commands via a CAN interface and generate PWM control signals based on an improved field-oriented control algorithm. The driver board is configured to receive the PWM control signal and amplify the PWM control signal using a power circuit with dual MOS transistors in parallel, so as to independently drive the azimuth motor and the pitch motor. The azimuth angle sensor is configured to detect the actual position of the azimuth axis and feed the azimuth position signal back to the control board; The pitch angle sensor is configured to detect the actual position of the pitch axis and feed the pitch position signal back to the control board. The control board is also configured to adjust the PWM control signal based on the azimuth position signal and the pitch position signal through closed-loop control of current loop, speed loop and position loop; The control board is also configured to perform dual-axis coordinated energy optimization management, including: The power budget management module is configured to monitor bus voltage and total current in real time, and allocate instantaneous power budgets for the azimuth axis and the pitch axis, and limit the speed or acceleration of non-critical axes based on motion priority; The braking timing management module is configured to predict the braking time points of the azimuth motor and the pitch motor, and actively stagger the braking timing to avoid the superposition of energy feedback peaks. The capacitive energy storage management module is configured to use the bus capacitor as a temporary energy storage buffer, control the capacitor discharge to support the system when the voltage drops sharply, and allow the capacitor to absorb part of the feedback energy first when the voltage rises sharply. The driver board is also configured to perform current sharing control of the dual MOSFETs in parallel, including: The current sampling module is configured to monitor the phase current of the azimuth motor and the pitch motor in real time through a sampling resistor and an isolation amplifier, and output a current feedback signal. The current sharing adjustment module is configured to receive the current feedback signal, compare the current difference of the parallel branch of the dual MOSFETs, and adjust the pulse width of the PWM control signal based on the comparison result to balance the current distribution of the parallel branch of the dual MOSFETs. The thermal management coordination module is configured to combine the output of the current equalization adjustment module to disperse heat through the heat dissipation path of the drive board, thereby avoiding local overheating; The flow equalization module is also configured to perform temperature-compensated flow equalization control, including: The parameter adjustment module is configured to monitor the junction temperature change of the parallel branch of the dual MOS transistors in real time, and adjust the current comparison threshold of the current sharing adjustment module based on the junction temperature change. The load response module is configured to adjust the pulse width adjustment step size of the PWM control signal according to the load change rate when the azimuth motor or the pitch motor is subjected to transient load. The calculation optimization module is configured to calculate the real-time power loss of the parallel branch of the dual MOS transistors using an improved loss model, and optimize the current sharing parameters based on the power loss results.

2. The system according to claim 1, characterized in that, The computational optimization module is also configured to perform real-time power loss calculations based on real-time data, including: The data acquisition submodule is configured to acquire the current value, junction temperature, and switching frequency of the parallel branch of the dual MOS transistors in real time; The loss calculation submodule is configured to use an improved loss model to calculate real-time power loss in combination with the current value, junction temperature, and switching frequency. The improved loss model includes a conduction loss term and a switching loss term. The conduction loss term is calculated based on junction temperature-dependent on-resistance, and the switching loss term is calculated based on switching frequency and switching energy. The optimization submodule is configured to adjust the current comparison threshold and PWM adjustment step size of the current sharing regulation module based on the real-time power loss.

3. The system according to claim 2, characterized in that, The loss calculation submodule is also configured to perform loss model optimization based on transient response and thermal management, including: The transient compensation submodule is configured to monitor the transient current change rate of the azimuth motor and the pitch motor, and add a transient compensation term to the improved loss model to correct the calculation deviation of switching losses. The thermal network submodule is configured to construct a dynamic thermal resistance model of the parallel branch of the dual MOS transistors, calculate the thermal gradient based on the real-time junction temperature and ambient temperature, and feed it back to the temperature dependence parameter of the conduction loss term. The update submodule is configured to adjust the weight coefficients of the loss model based on the transient compensation term and the thermal gradient results to optimize the calculation accuracy of real-time power loss.

4. The system according to claim 1, characterized in that, The braking timing management module is also configured to perform braking timing optimization based on load prediction, including: The load prediction submodule is configured to analyze the motion commands and historical load data of the azimuth motor and the pitch motor in real time to predict the occurrence time and intensity of transient load events. The staggered braking submodule is configured to calculate the optimal braking time interval for the azimuth axis and the pitch axis based on predicted transient load events, and adjust the braking timing. The effect evaluation submodule is configured to monitor bus voltage fluctuations and energy feedback peaks during the braking process and feed them back to the load prediction submodule to optimize the prediction model.

5. The system according to claim 4, characterized in that, The peak-shifting submodule is also configured to perform braking timing optimization based on multi-parameter fusion, including: The interval calculation unit is configured to calculate the optimal braking time interval between the azimuth axis and the pitch axis by integrating transient load intensity, bus voltage margin and system thermal state parameters. The timing adjustment unit is configured to adjust the braking timing based on the calculated optimal braking time interval using a priority allocation strategy. The real-time optimization unit is configured to monitor the interval adjustment effect during braking and provide real-time feedback of the optimized interval calculation parameters.

6. The system according to claim 5, characterized in that, The real-time optimization unit is also configured to perform optimization based on multi-metric feedback, including: The effect monitoring subunit is configured to monitor the interval adjustment effect in real time during braking, including bus voltage stability, energy feedback smoothness, and system response time. The parameter optimization subunit is configured to adjust the fusion parameters in the interval calculation unit based on the monitoring results.

7. The system according to claim 1, characterized in that, The control board is configured to perform a process of generating PWM control signals based on an improved field-oriented control algorithm, including: The current sampling and coordinate transformation unit is configured to acquire the motor phase current through a three-phase current sampling circuit and use an anti-interference Clarke-Park transformation to convert the phase current into direct-axis current and quadrature-axis current components in a rotating coordinate system. The modulation compensation unit is configured to adjust the modulation ratio of the space vector modulation algorithm according to the real-time bus voltage fluctuation to ensure that the linear modulation region is maintained under low voltage conditions. The switching sequence optimization unit is configured to optimize the switching sequence for low-voltage, high-current scenarios, reduce the number of switching operations of power devices, and balance the distribution of switching losses.

8. The system according to claim 7, characterized in that, The modulation compensation unit is also configured to perform modulation optimization based on bus voltage, including: The voltage monitoring subunit is configured to acquire the bus voltage signal in real time and calculate the voltage fluctuation rate and available voltage margin. The modulation ratio adjustment subunit is configured to adjust the modulation ratio of the space vector modulation algorithm according to the voltage margin, and to ensure modulation linearity by using a piecewise linear interpolation algorithm. The overmodulation management subunit is configured to enter overmodulation mode when the voltage drops, and uses a harmonic injection strategy to extend voltage utilization. The performance evaluation subunit is configured to monitor modulation effects, including harmonic distortion and torque ripple, and provide feedback to optimize modulation parameters.

Images