Energy recovery control system for variable frequency drive devices directly powered by high-voltage power grid

Abstract

This invention discloses an energy recovery control system for a variable frequency drive device directly powered by a high-voltage power grid, relating to the fields of electrical drive and energy recovery control technology. It collects three-phase voltage and current data from the power grid, extracts steady-state voltage and current phasors using a sliding time window method, and updates the power grid impedance model in real time using a complex impedance calculation method. Combined with motor output power, speed changes, and torque calculations, it dynamically adjusts load operating parameters. By accurately calculating regenerated energy and adjusting the feedback path according to voltage fluctuations and frequency deviations in the power grid, it monitors the feedback current fluctuation rate and power factor to optimize the feedback path conduction ratio in real time. When the current is unstable or the power factor is abnormal, it automatically switches to a suboptimal path to ensure continuous and stable system operation. This invention effectively improves the energy recovery efficiency and grid compatibility of high-voltage power grid variable frequency drive devices and is suitable for energy recovery applications in high-power motor drive systems.

Description

Technical Field

[0001] This invention relates to the field of electrical drive and energy recovery control technology, specifically to an energy recovery control system for a frequency converter driven device directly powered by a high-voltage power grid. Background Technology

[0002] With the development of industrial automation and high-voltage frequency conversion technology, frequency conversion drive systems directly powered by high-voltage power grids are widely used in continuous heavy-load operation scenarios such as high-power pump stations, mine ventilation fans, and cooling tower fans. In these applications, the regenerative energy generated during motor operation cannot be ignored, especially during motor braking, system deceleration, or load reversal. A large amount of energy needs to be fed back to the grid or consumed in a timely and effective manner to avoid serious consequences such as voltage rise leading to system instability, equipment overheating, or even shutdown.

[0003] However, current energy recovery control strategies for frequency converters directly powered by high-voltage power grids generally suffer from the following problems: First, due to the high voltage levels of the power grid itself (6kV~10kV), conventional low-voltage energy feedback devices cannot be directly connected and require voltage matching through additional intermediate transformers, resulting in high system complexity, slow response, and significant transformer losses. Second, existing energy recovery control relies heavily on preset thresholds for judgment, failing to consider dynamic changes in load conditions and real-time fluctuations in the impedance of the energy feedback path, leading to low energy recovery efficiency and even potential energy oscillations and backflow, damaging grid stability. Third, conventional solutions lack dynamic correction mechanisms for the power factor and current harmonics of the feedback path, making them highly susceptible to harmonic pollution from the upstream power grid and malfunctions in the compensation system.

[0004] Especially in typical scenarios such as remote mountain pumped storage power stations, the variable frequency drive system has a short working cycle, large load impact, and frequent dynamic fluctuations in grid impedance. If the moment of braking energy formation cannot be accurately identified and the conduction strategy of the energy feedback path cannot be adjusted in real time, it will lead to a sudden increase in current and enhanced voltage disturbance during the feedback process. In severe cases, it may even cause the high-voltage frequency converter protection to trip, affecting the continuous operation of the system. Summary of the Invention

[0005] The purpose of this invention is to provide an energy recovery control system for a variable frequency drive device based on direct power supply from a high-voltage power grid, in order to address the shortcomings in the prior art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an energy recovery control system for a frequency converter drive device based on direct power supply from a high-voltage power grid, comprising:

[0007] The data acquisition module collects the real-time voltage U_g, current I_g, and frequency f_g of the high-voltage power grid connected to the frequency converter drive device, establishes the high-voltage side power grid impedance model Z_g, and obtains the current load operating status parameters on the motor side.

[0008] The status recognition module determines whether the current state is in the regenerative energy formation zone based on the obtained grid impedance model Z_g and load operating status parameters. If yes, it enters the energy determination module; otherwise, it maintains the original control mode.

[0009] The energy determination module estimates the energy characteristics of the regenerated energy formation section, calculates the instantaneous regenerated energy E_r, and determines whether E_r is greater than the set minimum feedback threshold E_min. If it is, the module proceeds to the path selection module; otherwise, the feedback control is terminated.

[0010] The path selection module selects the optimal feedback path R_opt based on E_r and Z_g, choosing the path with the lowest equivalent impedance as the energy conduction channel.

[0011] The conduction control module dynamically controls the selected path R_opt and synchronously adjusts the inverter's PWM output strategy to inject energy into the high-voltage grid with minimal fluctuations.

[0012] The feedback monitoring module monitors the feedback current I_r and power factor PF_r during the feedback process. When the fluctuation rate of I_r exceeds the set value ΔI_max, or PF_r is lower than the set lower limit PF_min, it automatically adjusts the conduction ratio of R_opt or switches to the suboptimal path R_sec.

[0013] The model update module updates the grid impedance model Z_g and load operating status parameters in real time based on the feedback data during the feedback process.

[0014] Preferably, the high-voltage side grid impedance model Z_g is established, including:

[0015] Collect the three-phase voltage signal and three-phase current signal of the high-voltage power supply side bus, and denot them as U_A, U_B, U_C and I_A, I_B, I_C, respectively;

[0016] The voltage and current signals are reconstructed in the time domain by synchronous sampling to obtain the phasor representation under steady-state operation.

[0017] Based on voltage and current phasors, the complex impedance calculation method is used to obtain the complex impedance value of each corresponding grid.

[0018] Based on the three-phase complex impedance results, a symmetrical component impedance model of the power grid is constructed, forming the equivalent power grid impedance model Z_g on the high-voltage side.

[0019] Preferably, the step of determining whether the current state is in the regenerative energy formation zone based on the obtained grid impedance model Z_g and load operating state parameters includes:

[0020] The mechanical output power P_m of the motor side is compared with the instantaneous active power P_g of the power grid in real time. If P_m is less than 0 and the power grid power P_g is positive, it is initially determined that the energy is in a reverse flow state.

[0021] Calculate the load speed change rate Δn_m and the electromagnetic torque change rate ΔT_m per unit time, and determine whether their product is negative;

[0022] When P_m, Δn_m, and ΔT_m meet the preset braking threshold conditions, and the equivalent grid impedance Z_g on the high-voltage side is less than the upper limit of the energy feedback impedance Z_th, it is confirmed that the region is in the regenerative energy formation zone.

[0023] Preferably, the energy characteristics of the regenerative energy formation section are estimated, and the instantaneous regenerative energy E_r is calculated, including:

[0024] Obtain the mechanical output power P_m and duration Δt of the motor side in the regenerative energy formation section, and preliminarily calculate the regenerative energy E_r1 based on the power-time product relationship;

[0025] The high-voltage power grid frequency f_g and equivalent power grid impedance Z_g within the section are collected synchronously to construct a power grid absorption capacity model based on the power grid frequency change rate df_g / dt.

[0026] By combining the parameters P_m, Z_g and df_g / dt, E_r1 is corrected to the final instantaneous regeneration energy E_r;

[0027] The power grid absorption capacity model establishes a response coefficient k_r based on actual operating data, and calculates the final result using the method E_r = E_r1×k_r.

[0028] Preferably, the step of selecting the optimal feedback path R_opt based on E_r and Z_g includes:

[0029] Obtain the current equivalent on-resistance values ​​R_1, R_2, and R_3 for all available energy feedback paths, and calculate the conduction loss based on the structure of each path.

[0030] Calculate the energy conduction efficiency η_i of E_r in each feedback path;

[0031] Compare the energy conduction efficiency of all paths and select the path with the largest η_i as the current optimal feedback path R_opt.

[0032] Preferably, the dynamic conduction control of the selected path R_opt and the synchronous adjustment of the inverter PWM output strategy include:

[0033] Based on the thermal capacitance and conduction threshold parameters of each power switching device in R_opt, set the conduction trigger current threshold and the maximum duty cycle range;

[0034] After E_r reaches the trigger condition, the power devices in the control feedback path are turned on sequentially according to the principle of minimum impact to establish a complete energy channel;

[0035] Real-time monitoring of the feedback current change rate, and using the change rate as feedback input, dynamically adjusts the inverter's pulse width modulation carrier frequency and dead time;

[0036] Based on the instantaneous voltage waveform and frequency deviation of the high-voltage side grid, the phase of the pulse width modulation reference signal is adjusted to keep the phase of the injected current in phase with the grid voltage.

[0037] Preferably, the automatic adjustment of the conduction ratio of R_opt or the switching to the suboptimal path R_sec includes:

[0038] The rate of change ΔI_r of the feedback current I_r per unit time is calculated in real time and compared with the preset fluctuation threshold ΔI_max. If ΔI_r is greater than ΔI_max, the feedback current is determined to be unstable.

[0039] The current power factor PF_r is detected synchronously, and it is determined whether it is lower than the set lower limit PF_min. If it is lower than PF_min, it is considered that the feedback deviates from the grid phase.

[0040] Based on the judgment results of ΔI_r and PF_r, the duty cycle distribution of each power device in the current feedback path R_opt is dynamically adjusted;

[0041] If I_r and PF_r still do not return to the set range after adjustment, then the suboptimal path R_sec will be activated.

[0042] Preferably, the step of updating the grid impedance model Z_g and load operating state parameters in real time based on feedback data includes:

[0043] The instantaneous values ​​of three-phase voltage and three-phase current during the energy feedback cycle are collected, and the voltage and current phasors under steady-state conditions are extracted using the sliding time window method.

[0044] Using the updated phasor data, the complex impedance value of each corresponding grid complex impedance is recalculated using the complex impedance calculation method, and a new Z_g is constructed using the symmetrical component method.

[0045] The output voltage, active power, and speed changes of the motor are calculated synchronously, and the load operating status parameters are dynamically updated in conjunction with the torque calculation model.

[0046] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0047] 1. This invention, by updating the impedance model Z_g of the high-voltage power grid and the operating state parameters of the motor load in real time, can accurately reflect the dynamic changes of the power grid and load, ensuring that the current and voltage phases are always matched during the energy feedback process, thus avoiding system failures caused by power grid instability or load fluctuations. Through this technology, this invention achieves the ability to adaptively adjust the high-voltage power grid in real time, effectively improving the stability and efficiency of the feedback process and avoiding the risk of power grid quality being affected by fluctuations in feedback current or a decrease in power factor.

[0048] 2. This invention utilizes the sliding time window method to extract steady-state operating conditions of voltage and current signals, calculate complex impedance, and apply the symmetrical component method. This enables the grid impedance model Z_g to possess higher dynamic adaptability, allowing it to adjust in real time to adapt to changes in grid load. The dynamic updating of load operating state parameters further ensures the accuracy of load conditions, making feedback path selection and energy regulation more precise, thereby improving the overall energy feedback system's response speed, accuracy, and safety. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0050] Figure 1 This is a flowchart of the system modules of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] For examples, please refer to Figure 1 As shown in this embodiment, the energy recovery control system for the frequency converter drive device based on direct power supply from a high-voltage power grid includes:

[0053] The data acquisition module collects the real-time voltage U_g, current I_g, and frequency f_g of the high-voltage power grid connected to the variable frequency drive device, establishes the high-voltage side power grid impedance model Z_g (i.e., the high-voltage side equivalent power grid impedance), and obtains the current load operating status parameters on the motor side.

[0054] The three-phase voltage and three-phase current signals of the high-voltage power supply bus are collected and denoted as U_A, U_B, U_C and I_A, I_B, I_C, respectively. The voltage and current signals are sampled using voltage transformers and current transformers installed on the high-voltage input bus side, with a sampling frequency of not less than 10kHz.

[0055] The acquired voltage and current signals are synchronously sampled. A high-precision A / D converter is used to synchronously sample the three-phase signals, and a dual-channel phase-locked loop method is used to reconstruct the time domain of each signal, filtering out high-frequency harmonics and transient fluctuations, and extracting the steady-state power frequency phasor representations of voltage and current. The phasor representations include amplitude and phase information.

[0056] Based on the obtained voltage and current phasors, the complex impedance value for each phase of the power grid is calculated using a complex impedance calculation method. Specifically, the calculation involves ratioing the voltage phasor to the corresponding current phasor for each phase and multiplying the ratio by the exponential form of the phase angle to recover the complex form of the impedance. For example, the complex impedance Z_A of phase A is calculated as follows: Where θ_A is the phase difference angle between U_A and I_A, and j is the imaginary unit; the calculation methods for the complex impedances Z_B and Z_C of phase B and phase C are the same.

[0057] Based on the calculation results of Z_A, Z_B, and Z_C, the grid impedance is decomposed using the symmetrical component method to construct positive-sequence, negative-sequence, and zero-sequence impedance components. The equivalent grid impedance model Z_g on the high-voltage side is defined as the scalar value of the positive-sequence impedance Z_1 at the operating frequency f_g, used to characterize the grid's complex impedance response capability to energy feedback flow. If asymmetrical operation of the grid is detected, the negative-sequence impedance Z_2 is further compensated to correct the model's accuracy.

[0058] While constructing the Z_g model, the current load operating status parameters on the motor side are read in real time through the sampling interface built into the inverter body. These parameters include mechanical output power P_m, speed n_m, and electromagnetic torque T_m. The power P_m is calculated based on the product of the active components of the output current and voltage. The speed n_m is measured in real time by the encoder, and the torque T_m is calculated through current closed-loop control.

[0059] The status identification module determines whether the current state is in the regenerative energy formation zone based on the obtained grid impedance model Z_g and load operating status parameters. If yes, it enters the energy determination module; otherwise, it maintains the original control mode.

[0060] The mechanical output power P_m at the motor output terminal and the instantaneous active power P_g on the high-voltage grid side are obtained to characterize the direction of energy transfer between the load and the grid. The mechanical output power P_m is obtained by converting the motor current, voltage, and power factor; the instantaneous active power P_g of the grid is calculated by instantaneously multiplying and summing the three-phase voltage and three-phase current signals. If P_m is detected to be less than zero, it indicates that the load is no longer absorbing energy and is showing reverse power output. At the same time, P_g is positive, meaning that the grid is still in the state of absorbing energy. Therefore, it is preliminarily determined that there is a trend of reverse energy transfer from the load to the grid, constituting an energy reverse flow state.

[0061] To further verify whether the motor is in a mechanical deceleration or regenerative braking process, the rate of change of motor speed n_m (Δn_m) and the rate of change of electromagnetic torque T_m (ΔT_m) per unit time are calculated. Δn_m is obtained by dividing the speed difference between two consecutive sampling periods by the sampling period time T, and ΔT_m is calculated in the same way. Multiplying Δn_m and ΔT_m results in a negative value, indicating that at the current moment there is a situation of "speed decreasing while torque is positive" or "speed increasing while torque is negative," suggesting that the motor is in the energy braking conversion phase.

[0062] Based on a comprehensive analysis of the energy flow direction and dynamic load characteristics, determine whether the following two conditions are met simultaneously:

[0063] First, the power and dynamic parameters meet the regenerative braking threshold judgment criteria, that is, P_m is less than zero, and Δn_m multiplied by ΔT_m is less than zero;

[0064] Second, the equivalent grid impedance Z_g of the high-voltage side currently established is less than the preset upper limit of the energy feedback impedance Z_th. The upper limit of impedance Z_th is the maximum complex impedance value that can be allowed to ensure stable grid connection of the feedback path. It is determined based on historical operating data and grid connection specifications, and its specific value is set in the range of 1 to 5 ohms according to different equipment levels.

[0065] When both of the above conditions are met, it is determined that the system has entered the regenerative energy formation phase. At this time, the control process will switch to the energy determination module to perform subsequent instantaneous energy calculation and feedback path selection. If either condition is not met, the current original control mode will be maintained, and the energy feedback path will be blocked to ensure system stability and safe operation of the power grid.

[0066] The energy determination module estimates the energy characteristics of the regenerated energy formation section, calculates the instantaneous regenerated energy E_r, and determines whether E_r is greater than the set minimum feedback threshold E_min. If it is, the module proceeds to the path selection module; otherwise, the feedback control is terminated.

[0067] Obtain the mechanical output power P_m of the motor side and the corresponding duration Δt within the current regenerative energy formation section. The mechanical output power P_m is calculated from the output current, voltage, and power factor measured on the inverter side, and the unit is kilowatts; the duration Δt is determined by the time interval during which the energy flow determination result remains continuously valid, and the unit is seconds. Based on the product relationship between power and time, the regenerative energy within this section is initially calculated, denoted as the first-stage regenerative energy E_r1, and the calculation formula is: , where a negative value of P_m indicates that energy is fed back from the load side to the grid side.

[0068] To improve the dynamic accuracy of energy estimation, the actual operating frequency f_g of the high-voltage power grid in the current section and the equivalent high-voltage side grid impedance Z_g at the corresponding time are collected simultaneously. The frequency difference between two sampling periods is divided by the time interval to calculate the grid frequency change rate df_g / dt. The frequency change rate df_g / dt reflects the trend of the grid's absorption capacity change due to load fluctuations during regeneration. Combining f_g, df_g / dt, and Z_g, a grid absorption capacity model is constructed to correct the initial energy value E_r1.

[0069] The aforementioned grid dynamic parameters are incorporated into the correction algorithm to construct a response coefficient k_r for regenerative energy adjustment. This response coefficient k_r is obtained through a regression model established using historical operating data and the transient response behavior of the grid; its value is inversely proportional to Z_g and directly proportional to df_g / dt. When the grid impedance Z_g is small, it indicates strong absorption capacity, and k_r is close to 1; when the frequency changes drastically, it indicates a decrease in grid connection stability, and k_r decreases. The final calculation method for the regenerative energy E_r is as follows: The value of k_r ranges from 0.6 to 1.2.

[0070] The corrected instantaneous regenerated energy E_r is compared with the set minimum feedback threshold E_min. The minimum feedback threshold E_min is the critical energy lower limit required to overcome path conduction energy consumption and switch triggering threshold for high-voltage side energy feedback operation. Its value is set during system design based on path conduction resistance, control triggering energy consumption, and grid voltage level, and is typically between 500 joules and 2000 joules. If the calculated E_r is greater than E_min, an energy feedback path selection operation is performed; otherwise, the current feedback process is terminated, the original control mode is maintained, and weak energy is prevented from causing reverse disturbances to grid stability.

[0071] The path selection module selects the optimal feedback path R_opt based on E_r and Z_g, choosing the path with the lowest equivalent impedance as the energy conduction channel.

[0072] Obtain the equivalent on-resistance values ​​of all available energy feedback paths under the current operating conditions. Assume there are three alternative energy feedback paths: Path 1, Path 2, and Path 3, with corresponding equivalent on-resistance values ​​denoted as R_1, R_2, and R_3, respectively. The equivalent on-resistance includes the sum of the main feedback power devices (such as bidirectional thyristors and IGBTs), filter inductors, resistors, and line connection impedances in the path. Specifically, the equivalent impedance value at the current moment is calculated by modeling the technical parameters of each device (on-state voltage drop, current rating, thermal resistance coefficient) and the path topology, combined with the actual operating current. The unit is ohms.

[0073] Based on the known instantaneous regenerative energy E_r and the equivalent grid impedance Z_g on the high-voltage side, the energy conduction efficiency η_i is calculated for each feedback path. This conduction efficiency is used to evaluate the proportion of energy that can actually be transferred from the load side into the grid through each path. The specific calculation method is as follows: Where R_i is the equivalent conduction impedance of path i, and Z_g is the equivalent grid impedance on the high-voltage side. When R_i is small, the energy transmission loss is small and the conduction efficiency η_i is high; conversely, the efficiency decreases.

[0074] The energy conduction efficiencies η_1, η_2, and η_3 corresponding to all paths are compared, and the one with the highest efficiency is selected as the current optimal energy feedback path R_opt. This path will be prioritized for conduction in subsequent steps, and the relevant power switching devices will be driven by the control device to maximize the injection of E_r into the high-voltage grid, reduce feedback losses, and improve energy utilization.

[0075] To ensure the dynamic adaptability of the path selection process, the response time, current fluctuation level, and operational stability level of the selected path R_opt under the current Z_g condition will be recorded synchronously during each path evaluation process to build a path response database for path selection optimization and switching criterion adjustment in subsequent control cycles.

[0076] The conduction control module dynamically controls the selected path R_opt and synchronously adjusts the inverter's PWM output strategy to inject energy into the high-voltage grid with minimal fluctuations.

[0077] Based on the thermal capacitance and conduction threshold parameters of each power switching device in the selected feedback path R_opt, the conduction trigger current threshold I_th and the maximum allowable duty cycle D_max are set. The thermal capacitance value is the energy that the switching device can withstand per unit time, determined based on its packaging structure, material thermal conductivity, and cooling method, and is expressed in joules per degree Celsius. The conduction threshold parameter is the minimum turn-on current of the device within its safe conduction range, expressed in amperes. Based on the thermal capacitance and maximum conduction capability of each device, the conduction initiation condition I_th for the current operating cycle, and the maximum duty cycle D_max corresponding to avoid thermal breakdown, are calculated comprehensively, typically ranging from 0.4 to 0.9.

[0078] Upon determining that the instantaneous regenerative energy E_r exceeds the minimum feedback threshold E_min, the controller activates the energy feedback mode, sequentially turning on the power devices in R_opt according to the minimum impact principle. The minimum impact principle prioritizes the turning on of devices with low equivalent impedance and short response time, and sets a turn-on delay Δt between adjacent devices to prevent electromagnetic disturbances caused by sudden current surges. The power device turn-on sequence is based on the device switching losses, while simultaneously meeting the device withstand voltage and rated current requirements to ensure that the energy channel establishes a complete closed path without exceeding design limits.

[0079] During the feedback process, the rate of change of the feedback current I_r, i.e., the change in current per unit time dI_r / dt, is monitored in real time. This value is used as the feedback input to dynamically adjust the carrier frequency f_c and dead time t_d in the inverter's pulse width modulation strategy. Specifically, when dI_r / dt is greater than the set upper limit ΔI_max, f_c is appropriately reduced to slow down the current rise, while the dead time t_d is increased to improve the tolerance of the drive circuit to feedback disturbances. Conversely, when the current is stable, f_c can be appropriately increased and t_d shortened to improve waveform control accuracy and response speed. The adaptive adjustment process of the above parameters is achieved through a proportional-integral-derivative (PID) closed-loop control algorithm, with a control period of no more than 10 milliseconds.

[0080] Based on the currently acquired instantaneous voltage waveform and frequency offset Δf_g of the high-voltage side power grid, the output phase φ_ref of the pulse width modulation reference signal is dynamically adjusted to ensure that the phase of the feedback current is consistent with the phase of the power grid voltage. Specifically, the three-phase voltage of the power grid is extracted using a fast Fourier transform to obtain the main frequency component, and its zero-phase angle φ_g is calculated. Then, the phase φ_ref of the PWM reference signal is adjusted to within the range of φ_g ± 5 degrees, achieving in-phase operation of the injected current and voltage.

[0081] The feedback monitoring module monitors the feedback current I_r and power factor PF_r during the feedback process. When the fluctuation rate of I_r exceeds the set value ΔI_max, or PF_r is lower than the set lower limit PF_min, it automatically adjusts the conduction ratio of R_opt or switches to the suboptimal path R_sec.

[0082] The rate of change ΔI_r of the feedback current I_r per unit time is calculated in real time and serves as an important indicator for measuring the stability of the feedback current. ΔI_r is the difference between the feedback current in the current sampling period and the current in the previous period, divided by the sampling time interval Δt, and is measured in amperes per second. A feedback current fluctuation threshold ΔI_max is set as the judgment criterion. This threshold is set based on the allowable current ramp-up rate of the power device, the safe operating current range, and the grid's immunity, and is typically ranged from 100 to 500 amperes per millisecond. When ΔI_r exceeds ΔI_max, it is determined that the current feedback current has experienced abnormal fluctuations, posing a risk of unstable transmission.

[0083] The power factor PF_r is synchronously detected during the current feedback process to evaluate the phase relationship between the feedback current and the grid voltage. PF_r is obtained by calculating the ratio of grid active power to apparent power, reflecting whether the feedback current is operating in phase with the grid voltage. If PF_r is lower than the set lower limit of power factor PF_min, which is usually set to 0.9, it indicates that the feedback current is lagging or leading, deviating from the grid voltage phase, which may cause reverse power oscillation or reduce energy feedback efficiency.

[0084] Based on the judgment results of ΔI_r and PF_r, the duty cycle of the power devices within the feedback path is optimized and adjusted. The duty cycle adjustment is based on a pulse width modulation control model, dynamically adjusting the on-time ratio of each power device. This shortens the on-time of high-fluctuation path segments and lengthens the on-time of relatively stable segments, thereby optimizing the current waveform and phase matching without changing the feedback path. This adjustment process is completed using proportional-integral control, with the feedback input being the deviation between ΔI_r and PF_r. The control period is less than 10 milliseconds to ensure fast response.

[0085] If, after the aforementioned duty cycle adjustment, the monitored ΔI_r is still greater than ΔI_max, and PF_r is still lower than PF_min, it indicates that the current feedback path R_opt can no longer meet the stable operating conditions. At this point, the controller triggers a path switching command, shutting down the main power switching devices of the current path R_opt and turning on the preset suboptimal path R_sec. The selection of R_sec is based on the previous energy conduction efficiency ranking results, ensuring that the switched path has acceptable transmission efficiency and stronger current control capability. During the switching process, the order and delay of the turn-on are controlled to avoid system impact caused by instantaneous current surges.

[0086] The model update module updates the grid impedance model Z_g and load operating status parameters in real time based on the feedback data during the feedback process.

[0087] Instantaneous values ​​of three-phase voltage and three-phase current are collected during the energy feedback cycle. Instantaneous data of three-phase voltages U_A, U_B, U_C and currents I_A, I_B, I_C of the high-voltage grid bus are acquired via a high-frequency sampling interface. Then, the voltage and current phasors under steady-state conditions are extracted using a sliding window method. Specifically, the sliding window method averages data from multiple consecutive sampling cycles to remove high-frequency interference and instantaneous fluctuations, thereby obtaining phasor representations of voltage and current representing the steady-state condition. The phasor representation is in complex form, where the amplitude and phase information of voltage and current are obtained through Fourier transform, expressed as follows: , where θ_A is the phase of phase A voltage.

[0088] Using the updated voltage and current phasor data, the complex impedance values ​​Z_A, Z_B, and Z_C for each phase of the power grid are recalculated using the complex impedance calculation method. The complex impedance Z is calculated from the ratio of voltage to current, as shown in the formula: Where θ_Ui and θ_Ii are the phase angles of voltage and current, respectively. By calculating the complex impedances Z_A, Z_B, and Z_C of each phase, the equivalent impedance model Z_g of the high-voltage side power grid is further constructed using the Symmetrical Component Method. This model consists of positive-sequence impedance Z_1, negative-sequence impedance Z_2, and zero-sequence impedance Z_0. To ensure the accuracy and dynamic adaptability of the model, the actual situation of asymmetrical operation of the power grid is taken into account.

[0089] The system synchronously calculates the output voltage, active power, and speed changes at the motor terminals, and dynamically updates the load operating parameters based on the torque estimation model. The motor output voltage is obtained through feedback control of the frequency converter, while the active power P_m is calculated using the relationship between current, voltage, and power factor. (Where V is voltage, I is current, and PF is power factor) Calculated. Speed ​​changes are acquired in real-time via an encoder, while torque T_m is calculated using the relationship between current and motor speed. (Where k is a constant and n_m is the motor speed) Calculation. Combining the relationship between the motor's power and torque, the load operating status parameters are updated to ensure that the system can reflect real-time changes in load conditions.

[0090] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. An energy recovery control system for a variable frequency drive device directly powered by a high-voltage power grid, characterized in that: include: The data acquisition module collects the real-time voltage U_g, current I_g, and frequency f_g of the high-voltage power grid connected to the frequency converter drive device, establishes the high-voltage side power grid impedance model Z_g, and obtains the current load operating status parameters on the motor side. The status identification module determines whether the current state is in the regenerative energy formation zone based on the obtained grid impedance model Z_g and load operating status parameters. If so, it enters the energy determination module. Otherwise, maintain the original control mode; The energy determination module estimates the energy characteristics of the regenerated energy formation section, calculates the instantaneous regenerated energy E_r, and determines whether E_r is greater than the set minimum feedback threshold E_min. If it is, the module proceeds to the path selection module; otherwise, the feedback control is terminated. The path selection module selects the optimal feedback path R_opt based on E_r and Z_g, choosing the path with the lowest equivalent impedance as the energy conduction channel. The conduction control module dynamically controls the selected path R_opt and synchronously adjusts the inverter's PWM output strategy to inject energy into the high-voltage grid with minimal fluctuations. The feedback monitoring module monitors the feedback current I_r and power factor PF_r during the feedback process. When the fluctuation rate of I_r exceeds the set value ΔI_max, or PF_r is lower than the set lower limit PF_min, it automatically adjusts the conduction ratio of R_opt or switches to the suboptimal path R_sec. The model update module updates the grid impedance model Z_g and load operating status parameters in real time based on the feedback data during the feedback process.

2. The energy recovery control system for a frequency converter drive device based on direct power supply from a high-voltage power grid according to claim 1, characterized in that: Establish the high-voltage side grid impedance model Z_g, including: Collect the three-phase voltage signal and three-phase current signal of the high-voltage power supply side bus, and denot them as U_A, U_B, U_C and I_A, I_B, I_C, respectively; The voltage and current signals are reconstructed in the time domain by synchronous sampling to obtain the phasor representation under steady-state operation. Based on voltage and current phasors, the complex impedance calculation method is used to obtain the complex impedance value of each corresponding grid. Based on the three-phase complex impedance results, a symmetrical component impedance model of the power grid is constructed, forming the equivalent power grid impedance model Z_g on the high-voltage side.

3. The energy recovery control system for a frequency converter drive device based on direct power supply from a high-voltage power grid according to claim 1, characterized in that: The step of determining whether the current state is in the regenerative energy formation zone based on the obtained grid impedance model Z_g and load operating state parameters includes: The mechanical output power P_m of the motor side is compared with the instantaneous active power P_g of the power grid in real time. If P_m is less than 0 and the power grid power P_g is positive, it is initially determined that the energy is in a reverse flow state. Calculate the load speed change rate Δn_m and the electromagnetic torque change rate ΔT_m per unit time, and determine whether their product is negative; When P_m, Δn_m, and ΔT_m meet the preset braking threshold conditions, and the equivalent grid impedance Z_g on the high-voltage side is less than the upper limit of the energy feedback impedance Z_th, it is confirmed that the region is in the regenerative energy formation zone.

4. The energy recovery control system for a frequency converter drive device based on direct power supply from a high-voltage power grid according to claim 1, characterized in that: The energy characteristics of the regenerative energy formation section are estimated, and the instantaneous regenerative energy E_r is calculated, including: Obtain the mechanical output power P_m and duration Δt of the motor side in the regenerative energy formation section, and preliminarily calculate the regenerative energy E_r1 based on the power-time product relationship; The high-voltage power grid frequency f_g and equivalent power grid impedance Z_g within the section are collected synchronously to construct a power grid absorption capacity model based on the power grid frequency change rate df_g / dt. By combining the parameters P_m, Z_g and df_g / dt, E_r1 is corrected to the final instantaneous regeneration energy E_r; The power grid absorption capacity model establishes a response coefficient k_r based on actual operating data, and calculates the final result using the method E_r = E_r1×k_r.

5. The energy recovery control system for a frequency converter drive device based on direct power supply from a high-voltage power grid according to claim 1, characterized in that: The step of selecting the optimal feedback path R_opt based on E_r and Z_g includes: Obtain the current equivalent on-resistance values ​​R_1, R_2, and R_3 for all available energy feedback paths, and calculate the conduction loss based on the structure of each path. Calculate the energy conduction efficiency η_i of E_r in each feedback path; Compare the energy conduction efficiency of all paths and select the path with the largest η_i as the current optimal feedback path R_opt.

6. The energy recovery control system for a frequency converter drive device based on direct power supply from a high-voltage power grid according to claim 1, characterized in that: The strategy of dynamically controlling the selected path R_opt and synchronously adjusting the inverter's PWM output includes: Based on the thermal capacitance and conduction threshold parameters of each power switching device in R_opt, set the conduction trigger current threshold and the maximum duty cycle range; After E_r reaches the trigger condition, the power devices in the control feedback path are turned on sequentially according to the principle of minimum impact to establish a complete energy channel; Real-time monitoring of the feedback current change rate, and using the change rate as feedback input, dynamically adjusts the inverter's pulse width modulation carrier frequency and dead time; Based on the instantaneous voltage waveform and frequency deviation of the high-voltage side grid, the phase of the pulse width modulation reference signal is adjusted to keep the phase of the injected current in phase with the grid voltage.

7. The energy recovery control system for a frequency converter drive device based on direct power supply from a high-voltage power grid according to claim 1, characterized in that: The automatic adjustment of the conduction ratio of R_opt or switching to the suboptimal path R_sec includes: The rate of change ΔI_r of the feedback current I_r per unit time is calculated in real time and compared with the preset fluctuation threshold ΔI_max. If ΔI_r is greater than ΔI_max, the feedback current is determined to be unstable. The current power factor PF_r is detected synchronously, and it is determined whether it is lower than the set lower limit PF_min. If it is lower than PF_min, it is considered that the feedback deviates from the grid phase. Based on the judgment results of ΔI_r and PF_r, the duty cycle distribution of each power device in the current feedback path R_opt is dynamically adjusted; If I_r and PF_r still do not return to the set range after adjustment, then the suboptimal path R_sec will be activated.

8. The energy recovery control system for a frequency converter drive device based on direct power supply from a high-voltage power grid according to claim 1, characterized in that: The step of updating the grid impedance model Z_g and load operating status parameters in real time based on feedback data includes: The instantaneous values ​​of three-phase voltage and three-phase current during the energy feedback cycle are collected, and the voltage and current phasors under steady-state conditions are extracted using the sliding time window method. Using the updated phasor data, the complex impedance value of each corresponding grid complex impedance is recalculated using the complex impedance calculation method, and a new Z_g is constructed using the symmetrical component method. The output voltage, active power, and speed changes of the motor are calculated synchronously, and the load operating status parameters are dynamically updated in conjunction with the torque calculation model.

Images