Wide voltage adaptive frequency conversion speed regulation control system and method

Abstract

The application relates to the fields of power electronic power conversion equipment and AC-DC conversion control technology, and discloses a wide-voltage self-adaptive frequency conversion speed control system and method, which comprises a main power conversion module, a state sensing sampling module and a self-adaptive modulation control module; the self-adaptive modulation control module acquires a DC bus instantaneous voltage and calculates transient compensation gain, dynamically reconstructs effective vector action time, implements proportional contraction clamping when the sum of vector time exceeds a switching period to keep the vector time ratio constant, generates a pulse width modulation driving pulse output to a three-phase inverter bridge, and realizes transient integral volt-second balance of a single switching period under wide-range fluctuation of an input power source, blocks low-frequency pulsation transmission on the DC side, maintains the stability of a synthesized vector space phase angle under an under-voltage over-modulation working condition, reduces phase current distortion and suppresses torque collapse.

Description

Technical Field

[0001] This invention belongs to the field of power electronic energy conversion equipment and AC-DC conversion control technology, and particularly relates to a wide voltage adaptive frequency conversion speed regulation control system and method. Background Technology

[0002] In power electronic power conversion equipment and motor drive systems, the AC-DC-AC main power circuit used for mutual conversion between AC and DC is usually composed of a rectifier filter unit, a DC bus energy storage capacitor and a three-phase inverter bridge. Space vector pulse width modulation technology calculates and allocates the effective vector and zero vector action time within the switching carrier period according to the target reference voltage vector, and then controls the operation of the switching transistors in the inverter bridge to synthesize the required AC drive voltage.

[0003] When the power supply voltage fluctuates within a wide range, the rectified DC bus voltage will experience transient drops and low-frequency pulsations. Existing modulation methods typically calculate the vector action time based on the average bus voltage after low-pass filtering, making it difficult to respond promptly to instantaneous changes in the bus voltage within a single switching carrier cycle. When the DC bus voltage experiences a transient drop, the inverter output voltage vector amplitude decreases accordingly, easily leading to insufficient volt-second integration of the single-cycle phase voltage and causing a decrease in motor torque. If compensation is achieved by increasing the modulation ratio, the total action time of the two effective vectors can easily exceed the switching carrier cycle. Conventional overmodulation methods usually directly truncate the action time exceeding the cycle range, which alters the time ratio between the original effective vectors, causing a shift in the spatial phase angle of the synthesized voltage vector and increasing phase current harmonics. For example, the common... Chinese invention patent application CN102570873A discloses a dead-time compensation method for voltage space vector pulse width modulation technology. Based on the relationship between phase current polarity and voltage sector, the dead-time of the space vector action time is adjusted by increasing or decreasing the preset dead-time. This method assumes that the DC bus voltage is constant and the switching transistor has ideal volt-ampere characteristics. It only applies a fixed dead-time offset in the time domain. Under wide-range fluctuations in the power supply, the instantaneous drop and low-frequency pulsation of the DC bus voltage will cause the dead-time volt-second loss of a single switching cycle to change nonlinearly. Furthermore, it ignores the dynamic difference between the on-saturation voltage drop of the switching device and the forward voltage drop of the freewheeling diode, resulting in undercompensation or overcompensation distortion in the voltage transient jump and current zero-crossing regions. This makes it impossible to guarantee the output voltage waveform quality under wide voltage fluctuations.

[0004] Therefore, under the condition of wide fluctuations in power supply voltage, how to adjust the vector action time within the carrier period in a timely manner according to the instantaneous bus voltage, maintain the time ratio of the two effective vectors under the undervoltage overmodulation state, keep the spatial phase angle of the synthesized vector stable, and at the same time reduce the impact of bus voltage fluctuations on dead zone compensation, has become the technical problem to be solved by this invention. Summary of the Invention

[0005] To address the problems in the background art, the technical solution of the present invention is as follows: A wide voltage adaptive variable frequency speed control system, comprising:

[0006] The main power conversion module includes a rectifier and filter unit, a DC bus energy storage unit, and a three-phase inverter bridge unit connected in sequence to the controlled motor load. The rectifier and filter unit is connected to the AC power supply.

[0007] The state sensing sampling module includes a DC bus voltage transient sampling unit located at both ends of the DC bus energy storage unit, and a phase current detection unit located at the output side of the three-phase inverter bridge unit.

[0008] The adaptive modulation control module has signal input terminals connected to the DC bus voltage transient sampling unit and the phase current detection unit, respectively, and its output terminal connected to the three-phase inverter bridge unit, including:

[0009] The bus fluctuation transient feedforward compensation unit is used to obtain the instantaneous voltage value of the DC bus. When the preset bus reference voltage value remains unchanged, it calculates the output transient voltage compensation gain coefficient according to the inverse relationship with the instantaneous voltage value of the DC bus.

[0010] The vector timing dynamic reconstruction unit is used to generate the basic effective vector duration based on the target reference voltage vector, multiply the transient voltage compensation gain coefficient by the basic effective vector duration, and output the first effective vector duration. Second effective vector action time and zero vector action time ;

[0011] The overmodulation dynamic boundary clamping unit is used to clamp the first effective vector according to the preset switching carrier period and the first effective vector's action time when the sum of the two effective vector action times exceeds the preset switching carrier period. and the second effective vector action time The sum is The ratio of the sums generates an overmodulation scaling factor to implement proportional contraction clamping, keeping the ratio of the first effective vector action time to the second effective vector action time constant and the final zero vector action time zero;

[0012] The pulse width modulation signal generation unit is used to generate a space vector pulse width modulation drive pulse based on the corrected vector action time and output it to the three-phase inverter bridge unit.

[0013] The overmodulation volt-second loss cross-cycle accumulation compensation unit has its signal input connected to the overmodulation dynamic boundary clamping unit. When triggering the overmodulation proportional contraction clamping, the overmodulation volt-second loss cross-cycle accumulation compensation unit calculates the effective vector volt-second product lost within a single cycle based on the effective vector time difference before and after clamping and the instantaneous DC bus voltage value, and stores it in the time domain. When the sum of the effective vector action times in subsequent switching carrier cycles is less than the preset switching carrier cycle, the accumulated volt-second product loss is converted into a time compensation amount and added equally to the first and second effective vector action times of the unsaturated cycle. The total effective vector action time after the addition is capped at the preset switching carrier cycle.

[0014] Preferably, the DC bus voltage transient sampling unit includes a high-frequency voltage divider sampling circuit and a linear isolation amplifier circuit, with a signal conditioning delay of less than 2μs. The DC bus voltage transient sampling unit is synchronously connected to the symmetrical triangular carrier generation unit inside the adaptive modulation control module. The analog-to-digital conversion sampling is synchronously triggered at the apex and trough of the symmetrical triangular carrier to obtain the instantaneous DC bus voltage value, and the instantaneous DC bus voltage value is input to the bus fluctuation transient feedforward compensation unit in real time.

[0015] Preferably, the adaptive modulation control module further includes: a magnetic saturation and resonance suppression unit, with the signal input terminal connected to the bus fluctuation transient feedforward compensation unit; the magnetic saturation and resonance suppression unit is used to inject an anti-phase damping voltage component into the target reference voltage vector in the stationary two-phase coordinate system when the instantaneous voltage value of the DC bus exceeds the preset overvoltage safety threshold, and to lower the direct-axis demagnetizing current setpoint in the synchronous rotating coordinate system, thereby suppressing the high-frequency switching resonance generated by the DC bus energy storage unit and the parasitic inductance of the main circuit, and reducing the transient magnetic flux density of the stator core of the controlled motor load.

[0016] Preferably, the pulse width modulation signal generation unit outputs a seven-segment symmetrical space vector pulse sequence; when the sum of the first effective vector action time and the second effective vector action time reaches the preset switching carrier period and the zero vector action time is zero, the pulse width modulation signal generation unit smoothly switches to a zero-vector pulse modulation mode in which each phase bridge arm operates only once per single switching cycle.

[0017] Preferably, the phase current detection unit in the state sensing sampling module includes a Hall current sensor disposed on two phase output lines on the AC output side of the three-phase inverter bridge unit; the adaptive modulation control module also includes a coordinate transformation unit, which is used to receive the measured two-phase instantaneous phase current, calculate and output the stator current vector amplitude and spatial position angle of the controlled motor load through static two-phase coordinate transformation, and input the stator current vector amplitude and spatial position angle to the closed-loop regulation circuit to generate the target reference voltage vector.

[0018] Preferably, in the main power conversion module, the rectifier and filter unit includes a three-phase electromagnetic interference filter circuit and a three-phase uncontrolled rectifier bridge; the DC bus energy storage unit includes a high-frequency thin-film absorption capacitor and an aluminum electrolytic energy storage capacitor group connected in parallel; the three-phase inverter bridge unit includes six symmetrically arranged insulated gate bipolar transistors and fast recovery freewheeling diodes connected in reverse parallel to each transistor.

[0019] Preferably, when the target reference voltage vector is in the first sector, the vector timing dynamic reconstruction unit uses half of the updated zero vector action time as the reference time offset, and accumulates the reference time offset, the first effective vector action time and the second effective vector action time to generate the conduction comparison count values ​​of the upper bridge arm switches of phase a, phase b and phase c of the three-phase inverter bridge unit, respectively, and outputs them to the comparison register inside the pulse width modulation signal generation unit.

[0020] Preferably, the adaptive modulation control module further includes: a limit undervoltage start safety protection unit, with its signal input terminals connected to the bus fluctuation transient feedforward compensation unit and the phase current detection unit, respectively; the limit undervoltage start safety protection unit is used to apply a limiting constraint on the amplitude ramp-up slope of the target reference voltage vector when the instantaneous DC bus voltage value is lower than the preset limit undervoltage threshold and the instantaneous phase current value exceeds the preset overcurrent threshold, and to lower the inverter output fundamental frequency setpoint by a preset frequency reduction step size, so as to maintain the single-cycle phase current peak value of the three-phase inverter bridge unit in the safe operating range under power supply drop conditions.

[0021] A wide-voltage adaptive variable frequency speed control method, implemented by a wide-voltage adaptive variable frequency speed control system, includes the following steps:

[0022] Step S1: Obtain the instantaneous DC bus voltage value collected by the DC bus voltage transient sampling unit at the start of the current switching carrier cycle. When the preset bus reference voltage value remains unchanged, calculate the dimensionless transient voltage compensation gain coefficient according to the inverse relationship with the instantaneous DC bus voltage value.

[0023] Step S2: Obtain the target reference voltage vector in the stationary two-phase coordinate system. Based on the target reference voltage vector in the sector of the current sampling period, generate the basic effective vector action time. Multiply the transient voltage compensation gain coefficient as a multiplication factor into the basic effective vector action time, and calculate the output compensated first effective vector action time. Second effective vector action time and zero vector action time ;

[0024] Step S3: When the sum of the first effective vector action time and the second effective vector action time exceeds the preset switching carrier period, based on the preset switching carrier period and the first effective vector action time... and the second effective vector action time The sum is The ratio of the sums generates an overmodulation scaling factor. This overmodulation scaling factor is used to proportionally shrink and clamp the first and second effective vector action times, keeping the ratio of the first to the second effective vector action times constant, and setting the final zero vector action time to zero. When the overmodulation proportional shrinking clamp is triggered, the effective vector volt-second product lost in a single cycle is calculated and accumulated in the time domain based on the effective vector time difference before and after clamping and the instantaneous DC bus voltage value. When the sum of the effective vector action times in subsequent switching carrier cycles is less than the preset switching carrier cycle, the accumulated volt-second product loss is converted into a time compensation amount and added equally to the first and second effective vector action times in the unsaturated cycle. The total effective vector action time after the addition is capped at the preset switching carrier cycle.

[0025] Step S4: Determine the direction of the dead zone error voltage based on the polarity of the instantaneous current of each phase measured by the phase current detection unit. Based on the preset hardware dead time of the switching transistor, the difference between the turn-on and turn-off delay of the switching transistor, the difference between the saturation conduction voltage drop of the switching device and the forward voltage drop of the anti-parallel freewheeling diode, and combined with the obtained instantaneous DC bus voltage value, calculate the time-domain dead zone compensation amount of the conduction time of each phase bridge arm, and add the time-domain dead zone compensation amount to the reference modulation conduction time of each phase bridge arm.

[0026] Step S5: Based on the corrected vector action time and the reference modulation conduction time after superimposing the time-domain dead zone compensation, a space vector pulse width modulation drive pulse sequence is generated and output to the three-phase inverter bridge unit to drive the controlled switch tube to operate.

[0027] Compared with existing technologies, the wide voltage adaptive variable frequency speed control system and method of the present invention have the following advantages:

[0028] 1. In wide-voltage adaptive variable frequency speed control, by acquiring the instantaneous DC bus voltage within a single switching carrier cycle and adjusting the first effective vector action time, the second effective vector action time, and the zero vector action time accordingly, the single-cycle integral volt-second product of the inverter bridge output voltage vector can be kept constant during input voltage fluctuations. This reduces the transmission of low-frequency pulsations from the DC bus to the AC output side and maintains the symmetry of the output fundamental voltage. Consequently, it can suppress harmonic distortion of the phase current of the controlled motor, eliminate electromagnetic torque pulsations caused by power supply voltage fluctuations, and enable the motor to maintain operational stability when the input voltage changes.

[0029] 2. When the DC bus voltage drops sharply and enters an overmodulation state, the action time of the first effective vector and the action time of the second effective vector are shortened proportionally. This allows the time ratio of the two vectors to remain constant while utilizing the switching carrier cycle, thus stabilizing the spatial phase angle of the synthesized voltage vector. This eliminates the vector phase shift and phase current harmonic distortion caused by direct saturation truncation, enabling the controlled motor to maintain its maximum output torque capability even when the input power supply voltage drops sharply, and suppressing the stall phenomenon during heavy-load startup.

[0030] 3. Real-time correction of the inverter bridge dead zone compensation based on the instantaneous DC bus voltage and phase current status allows the compensation amount to be adjusted promptly with changes in bus voltage, avoiding undercompensation or overcompensation when the bus voltage fluctuates significantly due to fixed dead zone compensation. This reduces the nonlinear amplification of the inverter bridge dead zone volt-second loss caused by bus voltage fluctuations, reduces waveform distortion near the zero-crossing point of the phase current during light load and low frequency operation, and improves the quality of the inverter output waveform. Attached Figure Description

[0031] Figure 1 This is a flowchart of the adaptive space vector pulse width modulation control of the present invention;

[0032] Figure 2 This is a diagram showing the composition of the main power conversion circuit and control unit of the present invention. Detailed Implementation

[0033] 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 embodiments of this application, not all embodiments. 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.

[0034] A wide voltage adaptive variable frequency speed control system includes:

[0035] The main power conversion module includes a rectifier and filter unit, a DC bus energy storage unit, and a three-phase inverter bridge unit connected in sequence to the controlled motor load. The rectifier and filter unit is connected to the AC power supply.

[0036] The state sensing sampling module includes a DC bus voltage transient sampling unit located at both ends of the DC bus energy storage unit, and a phase current detection unit located at the output side of the three-phase inverter bridge unit.

[0037] The adaptive modulation control module has signal input terminals connected to the DC bus voltage transient sampling unit and the phase current detection unit, respectively, and its output terminal connected to the three-phase inverter bridge unit, including:

[0038] The bus fluctuation transient feedforward compensation unit is used to obtain the instantaneous voltage value of the DC bus. When the preset bus reference voltage value remains unchanged, it calculates the output transient voltage compensation gain coefficient according to the inverse relationship with the instantaneous voltage value of the DC bus.

[0039] The vector timing dynamic reconstruction unit is used to generate the basic effective vector duration based on the target reference voltage vector, multiply the transient voltage compensation gain coefficient by the basic effective vector duration, and output the first effective vector duration. Second effective vector action time and zero vector action time ;

[0040] The overmodulation dynamic boundary clamping unit is used to clamp the first effective vector according to the preset switching carrier period and the first effective vector's action time when the sum of the two effective vector action times exceeds the preset switching carrier period. and the second effective vector action time The sum is The ratio of the sums generates an overmodulation scaling factor to implement proportional contraction clamping, keeping the ratio of the first effective vector action time to the second effective vector action time constant and the final zero vector action time zero;

[0041] The pulse width modulation signal generation unit is used to generate a space vector pulse width modulation drive pulse based on the corrected vector action time and output it to the three-phase inverter bridge unit.

[0042] The overmodulation volt-second loss cross-cycle accumulation compensation unit has its signal input connected to the overmodulation dynamic boundary clamping unit. When triggering the overmodulation proportional contraction clamping, the overmodulation volt-second loss cross-cycle accumulation compensation unit calculates the effective vector volt-second product lost within a single cycle based on the effective vector time difference before and after clamping and the instantaneous DC bus voltage value, and stores it in the time domain. When the sum of the effective vector action times in subsequent switching carrier cycles is less than the preset switching carrier cycle, the accumulated volt-second product loss is converted into a time compensation amount and added equally to the first and second effective vector action times of the unsaturated cycle. The total effective vector action time after the addition is capped at the preset switching carrier cycle.

[0043] Preferably, the DC bus voltage transient sampling unit includes a high-frequency voltage divider sampling circuit and a linear isolation amplifier circuit, with a signal conditioning delay of less than 2μs. The DC bus voltage transient sampling unit is synchronously connected to the symmetrical triangular carrier generation unit inside the adaptive modulation control module. The analog-to-digital conversion sampling is synchronously triggered at the apex and trough of the symmetrical triangular carrier to obtain the instantaneous DC bus voltage value, and the instantaneous DC bus voltage value is input to the bus fluctuation transient feedforward compensation unit in real time.

[0044] Preferably, the adaptive modulation control module further includes: a magnetic saturation and resonance suppression unit, with the signal input terminal connected to the bus fluctuation transient feedforward compensation unit; the magnetic saturation and resonance suppression unit is used to inject an anti-phase damping voltage component into the target reference voltage vector in the stationary two-phase coordinate system when the instantaneous voltage value of the DC bus exceeds the preset overvoltage safety threshold, and to lower the direct-axis demagnetizing current setpoint in the synchronous rotating coordinate system, thereby suppressing the high-frequency switching resonance generated by the DC bus energy storage unit and the parasitic inductance of the main circuit, and reducing the transient magnetic flux density of the stator core of the controlled motor load.

[0045] Preferably, the pulse width modulation signal generation unit outputs a seven-segment symmetrical space vector pulse sequence; when the sum of the first effective vector action time and the second effective vector action time reaches the preset switching carrier period and the zero vector action time is zero, the pulse width modulation signal generation unit smoothly switches to a zero-vector pulse modulation mode in which each phase bridge arm operates only once per single switching cycle.

[0046] Preferably, the phase current detection unit in the state sensing sampling module includes a Hall current sensor disposed on two phase output lines on the AC output side of the three-phase inverter bridge unit; the adaptive modulation control module also includes a coordinate transformation unit, which is used to receive the measured two-phase instantaneous phase current, calculate and output the stator current vector amplitude and spatial position angle of the controlled motor load through static two-phase coordinate transformation, and input the stator current vector amplitude and spatial position angle to the closed-loop regulation circuit to generate the target reference voltage vector.

[0047] Preferably, in the main power conversion module, the rectifier and filter unit includes a three-phase electromagnetic interference filter circuit and a three-phase uncontrolled rectifier bridge; the DC bus energy storage unit includes a high-frequency thin-film absorption capacitor and an aluminum electrolytic energy storage capacitor group connected in parallel; the three-phase inverter bridge unit includes six symmetrically arranged insulated gate bipolar transistors and fast recovery freewheeling diodes connected in reverse parallel to each transistor.

[0048] Preferably, when the target reference voltage vector is in the first sector, the vector timing dynamic reconstruction unit uses half of the updated zero vector action time as the reference time offset, and accumulates the reference time offset, the first effective vector action time and the second effective vector action time to generate the conduction comparison count values ​​of the upper bridge arm switches of phase a, phase b and phase c of the three-phase inverter bridge unit, respectively, and outputs them to the comparison register inside the pulse width modulation signal generation unit.

[0049] Preferably, the adaptive modulation control module further includes: a limit undervoltage start safety protection unit, with its signal input terminals connected to the bus fluctuation transient feedforward compensation unit and the phase current detection unit, respectively; the limit undervoltage start safety protection unit is used to apply a limiting constraint on the amplitude ramp-up slope of the target reference voltage vector when the instantaneous DC bus voltage value is lower than the preset limit undervoltage threshold and the instantaneous phase current value exceeds the preset overcurrent threshold, and to lower the inverter output fundamental frequency setpoint by a preset frequency reduction step size, so as to maintain the single-cycle phase current peak value of the three-phase inverter bridge unit in the safe operating range under power supply drop conditions.

[0050] A wide-voltage adaptive variable frequency speed control method, implemented by a wide-voltage adaptive variable frequency speed control system, includes the following steps:

[0051] Step S1: Obtain the instantaneous DC bus voltage value collected by the DC bus voltage transient sampling unit at the start of the current switching carrier cycle. When the preset bus reference voltage value remains unchanged, calculate the dimensionless transient voltage compensation gain coefficient according to the inverse relationship with the instantaneous DC bus voltage value.

[0052] Step S2: Obtain the target reference voltage vector in the stationary two-phase coordinate system. Based on the target reference voltage vector in the sector of the current sampling period, generate the basic effective vector action time. Multiply the transient voltage compensation gain coefficient as a multiplication factor into the basic effective vector action time, and calculate the output compensated first effective vector action time. Second effective vector action time and zero vector action time ;

[0053] Step S3: When the sum of the first effective vector action time and the second effective vector action time exceeds the preset switching carrier period, based on the preset switching carrier period and the first effective vector action time... and the second effective vector action time The sum is The ratio of the sums generates an overmodulation scaling factor. This overmodulation scaling factor is used to proportionally shrink and clamp the first and second effective vector action times, keeping the ratio of the first to the second effective vector action times constant, and setting the final zero vector action time to zero. When the overmodulation proportional shrinking clamp is triggered, the effective vector volt-second product lost in a single cycle is calculated and accumulated in the time domain based on the effective vector time difference before and after clamping and the instantaneous DC bus voltage value. When the sum of the effective vector action times in subsequent switching carrier cycles is less than the preset switching carrier cycle, the accumulated volt-second product loss is converted into a time compensation amount and added equally to the first and second effective vector action times in the unsaturated cycle. The total effective vector action time after the addition is capped at the preset switching carrier cycle.

[0054] Step S4: Determine the direction of the dead zone error voltage based on the polarity of the instantaneous current of each phase measured by the phase current detection unit. Based on the preset hardware dead time of the switching transistor, the difference between the turn-on and turn-off delay of the switching transistor, the difference between the saturation conduction voltage drop of the switching device and the forward voltage drop of the anti-parallel freewheeling diode, and combined with the obtained instantaneous DC bus voltage value, calculate the time-domain dead zone compensation amount of the conduction time of each phase bridge arm, and add the time-domain dead zone compensation amount to the reference modulation conduction time of each phase bridge arm.

[0055] Step S5: Based on the corrected vector action time and the reference modulation conduction time after superimposing the time-domain dead zone compensation, a space vector pulse width modulation drive pulse sequence is generated and output to the three-phase inverter bridge unit to drive the controlled switch tube to operate.

[0056] Example 1: This example's wide-voltage adaptive variable frequency speed control system is used to drive a high-reliability precision temperature-controlled compressor. The main power conversion module includes a rectifier and filter unit, a DC bus energy storage unit, and a three-phase inverter bridge unit connected in sequence to the controlled motor load. The rated operating line voltage of the AC power supply is 380V, which can adapt to voltage fluctuations from 304V to 456V. The rectifier and filter unit includes a three-phase electromagnetic interference filter circuit and a three-phase uncontrolled rectifier bridge. The DC bus energy storage unit includes a high-frequency thin-film absorption capacitor and an aluminum electrolytic energy storage capacitor group connected in parallel. The nominal reference voltage of the DC bus is 540V, and an overvoltage safety threshold is preset. The three-phase inverter bridge unit includes six symmetrically arranged insulated-gate bipolar transistors and fast recovery freewheeling diodes connected in reverse parallel to each transistor. (Switching hardware...) Dead time, turn-on delay, turn-off delay, transistor saturation conduction voltage drop, and fast recovery freewheeling diode forward conduction voltage drop are all used as hardware parameters for dead time compensation calculation. The state sensing sampling module includes a DC bus voltage transient sampling unit and a phase current detection unit. The DC bus voltage transient sampling unit is located at both ends of the DC bus energy storage unit and includes a high-frequency voltage divider sampling circuit and a linear isolation amplifier circuit, with a signal conditioning delay of less than 2μs. The phase current detection unit is located on two phase output lines on the AC output side of the three-phase inverter bridge unit and uses a Hall current sensor to collect the phase current. The DC bus voltage transient sampling unit is synchronously connected to the symmetrical triangular carrier generation unit inside the adaptive modulation control module. Analog-to-digital conversion sampling is triggered at the apex and trough of the symmetrical triangular carrier to obtain the instantaneous DC bus voltage value. The adaptive modulation control module is configured with the corresponding switching carrier period. In terms of hardware connections, the input of the three-phase electromagnetic interference filter circuit is connected to the AC power supply, and its output is connected to the AC input of the three-phase uncontrolled rectifier bridge. The DC positive and negative outputs of the three-phase uncontrolled rectifier bridge are connected to the DC bus energy storage unit. The high-frequency voltage divider sampling circuit is connected in parallel across the two ends of the DC bus. The output of the linear isolation amplifier circuit is connected to the analog-to-digital conversion pin of the microprocessor inside the adaptive modulation control module through the analog signal sampling channel. The AC output of the three-phase inverter bridge unit is connected to the stator winding of the controlled motor load through the three-phase power supply bus. The output of the phase current detection unit is connected to the current sampling input pin of the microprocessor. Inside the microprocessor, the bus fluctuation transient feedforward compensation unit, the vector timing dynamic reconstruction unit, the overmodulation dynamic boundary clamping unit, and the pulse width modulation signal generation unit transmit control data in sequence. The six drive signals output by the pulse width modulation signal generation unit are connected to the gate control terminals of the six insulated gate bipolar transistors in the three-phase inverter bridge unit after passing through the drive isolation amplifier circuit.

[0057] When transient fluctuations occur in the AC power grid, the bus fluctuation transient feedforward compensation unit receives the instantaneous DC bus voltage value collected during the current switch carrier cycle. At the preset bus reference voltage value When it remains unchanged, it is based on the instantaneous voltage value of the DC bus. The dimensionless transient voltage compensation gain coefficient is calculated in an inverse relationship. : ,in, To preset the bus reference voltage value, This is the instantaneous voltage value of the DC bus. This is the transient voltage compensation gain coefficient, which is based on the preset bus reference voltage value. When kept constant, the transient voltage compensation gain coefficient DC bus instantaneous voltage value The instantaneous voltage value of the DC bus is inversely proportional to the DC bus voltage value. Reduce transient voltage compensation gain coefficient The corresponding increase is used to achieve volt-second integral balance of the output voltage in a single switching cycle. The vector timing dynamic reconstruction unit receives the target reference voltage vector in the stationary two-phase coordinate system, generates two sets of basic effective vector action times based on the target reference voltage vector in the sector where the current sampling period is located, and then... Multiplying the time by the multiplication factor to the corresponding fundamental effective vector action time yields the first effective vector action time. Second effective vector action time and zero vector action time : ,in, and These represent the effective time of action of the fundamental vector obtained from the solution in the stationary two-phase coordinate system. To preset the switching carrier period, and These are the first and second effective vector action times after compensation, respectively. To compensate for the zero vector action time, the bus coupling dead zone effect dynamic compensation unit simultaneously receives the instantaneous current of each phase measured by the phase current detection unit. Based on the phase current polarity, it determines the direction of the dead zone error voltage. It then uses the hardware dead time of the switching transistor, the difference between turn-on and turn-off delays, the difference between the transistor's saturation conduction voltage drop and the fast recovery freewheeling diode's forward conduction voltage drop, and the DC bus instantaneous voltage value of the current switching cycle to calculate the time-domain dead zone compensation. In a single-phase arm of the three-phase inverter bridge, when the phase current flows out of the arm in the forward direction, the corresponding switching transistor turns on; when the phase current flows into the arm in the reverse direction, the corresponding freewheeling diode forms a freewheeling path. The difference between the transistor's saturation conduction voltage drop and the freewheeling diode's forward conduction voltage drop will form a corresponding volt-second deviation within the switching carrier cycle. Based on the current DC bus instantaneous voltage value, this volt-second deviation is converted into a conduction time correction amount, and the resulting time-domain dead zone compensation amount is directly superimposed on the reference modulation conduction time of each phase arm.

[0058] When a deep undervoltage disturbance occurs at the power supply input, the instantaneous DC bus voltage can drop to around 360V. If the sum of the first effective vector action time and the second effective vector action time exceeds the preset switching carrier period, the overmodulation dynamic boundary clamping unit will adjust the timing according to the preset switching carrier period. The ratio of the sums generates the modulation scaling factor. : Subsequently, proportional contraction clamping was applied to the first and second effective vector action times respectively to obtain the final effective vector action time: ,in, and These are the final first effective vector action time and the final second effective vector action time after clamping, respectively. The final zero vector action time, since the first effective vector action time and the second effective vector action time use the same overmodulation scaling factor, their time ratio remains unchanged; in the coordinated control mechanism of single-cycle transient compensation and dynamic contraction clamping, the instantaneous DC bus voltage value When a voltage drop occurs, the bus fluctuation transient feedforward compensation unit calculates the transient voltage compensation gain coefficient based on an inverse proportional relationship. The effective vector action time is amplified by a single-cycle feedforward amplification to keep the single-cycle volt-second integral of the output phase voltage constant, blocking the transmission of low-frequency pulsations from the DC bus side to the AC side and preventing the weakening of the stator composite magnetic flux; when deep undervoltage causes the sum of the effective vector action times to exceed the preset switching carrier period... When overmodulation occurs, the overmodulation dynamic boundary clamping unit utilizes the overmodulation scaling factor. Time of action of the first effective vector Second effective vector action time Implement proportional contraction and maintain and The ratio equals and The ratio is constant. While compressing the zero vector action time to zero to fully release the output voltage amplitude capability, the spatial phase angle of the target reference voltage vector remains unchanged, eliminating the vector phase shift caused by saturation truncation. The two work together to maintain the electromagnetic torque capability of the motor output in terms of amplitude and avoid distortion of the stator current rotation trajectory in terms of spatial phase. This achieves synchronous suppression of electromagnetic torque collapse and phase current distortion under input undervoltage and overmodulation conditions. The overmodulation volt-second loss cross-cycle accumulation compensation unit calculates the effective vector volt-second product of the current switching carrier cycle loss based on the effective vector time difference before and after clamping and the instantaneous DC bus voltage value. It accumulates and stores this loss in the time domain. In the subsequent unsaturated period when the power supply voltage recovers and the sum of the first and second effective vector action times is less than the preset switching carrier cycle, the accumulated volt-second product loss is converted into a time compensation amount and added equally to the first and second effective vector action times of that cycle. The total effective vector action time after the addition does not exceed the preset switching carrier cycle.

[0059] When an overvoltage surge occurs at the power input terminal and the instantaneous DC bus voltage exceeds the preset overvoltage safety threshold, the magnetic saturation and resonance suppression unit injects an anti-phase damping voltage component into the target reference voltage vector in the stationary two-phase coordinate system. Simultaneously, it lowers the direct-axis demagnetizing current setpoint in the synchronous rotating coordinate system to suppress high-frequency switching resonance caused by the parasitic inductance of the DC bus energy storage unit and the main circuit, and to reduce the transient magnetic flux density of the stator core of the controlled motor load. Specifically, the microprocessor calculates the difference between the instantaneous DC bus voltage and the preset overvoltage safety threshold. When this difference is greater than 0V... The difference is multiplied by a preset voltage damping coefficient to obtain a damping adjustment scalar. Based on the collected stator phase current, the two orthogonal components of the stator current vector in the stationary two-phase coordinate system are calculated to form a unit damping vector opposite to the direction of the stator current vector. The damping adjustment scalar is then multiplied by the two orthogonal components of the unit damping vector to obtain the anti-phase damping voltage component. Subsequently, the two orthogonal components of the anti-phase damping voltage component are algebraically added to the components on the corresponding axes of the target reference voltage vector. The reconstructed target reference voltage vector is then sent to the space vector pulse width modulation timing solution.

[0060] When the target reference voltage vector is in the first sector, the vector timing dynamic reconstruction unit uses half of the updated zero vector action time as the reference time offset. It then accumulates the reference time offset, the first effective vector action time, and the second effective vector action time according to the switching sequence of the current sector, forming the conduction comparison count values ​​of the upper bridge arm switches of phases a, b, and c of the three-phase inverter bridge unit. These counts are then written into the comparison register inside the pulse width modulation signal generation unit. The pulse width modulation signal generation unit generates a seven-segment symmetrical spatial vector pulse sequence based on this. When the sum of the first and second effective vector action times reaches the preset switching carrier period and the zero vector action time is zero, the loading timing of the comparison register is updated at the end of the current carrier period and when the symmetrical triangular carrier reaches its trough. The zero vector action time in the original seven-segment pulse sequence is set to zero, and the zero-vector pulse modulation timing sequence is reconstructed according to the first and second effective vector action times. This ensures that each bridge arm of the three phases only performs one switching action within a single switching carrier period, smoothly switching from the seven-segment symmetrical spatial vector pulse sequence to the zero-vector pulse modulation mode.

[0061] When the controlled compressor is in a -40℃ low temperature environment and withstands a starting pressure difference of 15kg to 20kg, the three-phase inverter bridge unit is driven according to the above control method. In the verification test, the controlled compressor is placed in a -40℃ high and low temperature test chamber and kept at a constant temperature for 4 hours. An initial starting pressure difference of 18kg is established at both ends of the compressor's suction and discharge valves. At the moment of starting trigger, the AC grid simulated power supply reduces the power supply line voltage from 380V to 304V in a step. The transient power analyzer records the voltage, current and torque data of the entire starting process. During the dynamic transition period when the motor accelerates from standstill to 1800r / min, the peak value of the stator phase current remains within 1.42 times the rated phase current. The measured value of the single-cycle phase voltage volt-second product is 12685V·μs, and the fluctuation deviation relative to the nominal reference operating condition is 0.12%. The output electromagnetic torque of the motor is maintained above 16.8N·m throughout the entire process, and the measured torque ripple rate is 5.2%. No overcurrent tripping or stall shutdown occurred during the test.

[0062] Example 2: This example uses a power conversion test platform to test a wide-voltage adaptive variable frequency speed control system. The input of the test platform is connected to an AC grid simulation power supply with harmonic generation and transient sag programming functions. The output shaft of the controlled motor is rigidly connected to an AC power dynamometer on the same axis. The phase current, phase voltage, and DC bus instantaneous voltage signals are synchronously acquired by a transient power analyzer. The analyzer has a simulation bandwidth of 10MHz and a synchronous sampling rate of 50MS / s. When the voltage at both ends of the DC bus energy storage unit continuously pulsates due to the operation of the inverter switch within a single switching carrier cycle, the DC bus voltage transient sampling unit uses the hardware timer inside the adaptive modulation control module to align the sampling trigger time to the apex and basal point of the symmetrical triangular carrier wave to avoid the period of strong common-mode interference of the switch and to obtain the DC bus instantaneous voltage value. During the test, the rated line voltage of the AC grid simulation power supply is set to 380V. At the same time, Gaussian white noise with a signal-to-noise ratio of 20dB is superimposed on the power supply circuit, and a 50Hz power frequency harmonic component with an amplitude ratio of 5% is continuously injected.

[0063] The test setup includes two independent branches: a control group and a sample group for this embodiment. The control group uses the average bus voltage obtained through low-pass filtering for space vector calculation, while the sample group uses the instantaneous DC bus voltage value for transient feedforward compensation of bus fluctuations. The test conditions sequentially include nominal steady state, mild undervoltage drop, and deep extreme drop across the saturation boundary. The target reference voltage vector amplitude of the controlled motor is set to 220V, the spatial phase angle is set to 30°, and the preset switching carrier period is used. The value is 100 μs.

[0064] Under nominal steady-state conditions, the DC bus voltage is around 540V. The actual measured instantaneous DC bus voltage value is... The voltage is 539.8V. When performing timing calculations based on the sector where the target reference voltage vector is located, the effective duration of both sets of basic vectors is 35.28μs. The bus fluctuation transient feedforward compensation unit is based on... The transient voltage compensation gain coefficient was calculated. The effective time of the first unclamped vector after compensation is 1.0004. The value is 35.29 μs. At this point, the sum of the action times of the two effective vectors has not reached the switching carrier period. The overmodulation dynamic boundary clamping unit does not perform shrink clamping, and the overmodulation scaling factor... Keep it at 1.0000, the final effective vector action time The value is 35.29 μs.

[0065] After the AC power grid simulates a power supply switching to a mild undervoltage drop condition... The voltage dropped to 431.5V, accompanied by test noise fluctuations. The bus fluctuation transient feedforward compensation unit calculated the voltage based on this instantaneous voltage value. The value is 1.2514. Based on this, the vector timing dynamic reconstruction unit corrects the effective vector application time to obtain... The effective vector duration is 44.15 μs. At this point, the total effective vector duration is 88.30 μs, which is still less than the preset switching carrier period of 100 μs. Therefore... Keep it at 1.0000. The value remains at 44.15 μs, and the overmodulated dynamic boundary clamping unit does not operate.

[0066] The output voltage of the AC grid analog power supply was reduced and measured under deep extreme drop conditions. The voltage is 358.4V. After calculation using the same transient feedforward compensation relationship, Increased to 1.5067, resulting in The effective vector interaction time is 53.16 μs, and the theoretical sum of the interaction times of the two effective vectors reaches 106.32 μs, which exceeds the preset switching carrier period of 100 μs. The overmodulation dynamic boundary clamping unit calculates this based on the above. The value is 0.9406, and the action time of the first effective vector and the action time of the second effective vector are subjected to proportional contraction clamping to maintain their time ratio unchanged; after clamping, The zero vector action time is 50.00 μs. During this period, the overmodulation volt-second loss cross-cycle accumulation compensation unit records the volt-second accumulation loss formed by the change in effective vector action time before and after clamping, and performs compensation addition when the bus voltage rises and the sum of effective vector action time is less than the preset switching carrier period again.

[0067] After completing the control parameter tests for the three input disturbance levels, the transient power analyzer continued to extract the single-cycle phase voltage characteristic volt-second product and phase current total harmonic distortion rate during the mild undervoltage drop and deep extreme drop. The AC power dynamometer simultaneously recorded the output shaft torque ripple rate. Under the mild undervoltage drop condition, the control group used the average bus voltage to calculate the modulation timing, and the single-cycle phase voltage characteristic volt-second product was 10156.4 V·μs, the phase current total harmonic distortion rate was 8.35%, and the output shaft torque ripple rate was 12.4%. In this embodiment, the sample group used the DC bus instantaneous voltage value to correct the effective vector action time for a single cycle, and the single-cycle phase voltage characteristic volt-second product was 12698.5 V·μs, the phase current total harmonic distortion rate was 3.12%, and the output shaft torque ripple rate was 4.6%. The single-cycle phase voltage characteristic volt-second product under this condition is close to the theoretical value of 12700.8 V·μs under the nominal steady state.

[0068] Under extreme drop conditions, the control group experienced pulse width truncation after the effective vector modulation time crossed the carrier period boundary. The measured single-cycle phase voltage characteristic volt-second product was 8431.2 V·μs, the total harmonic distortion rate of the phase current was 17.82%, and the output shaft torque ripple rate was 29.8%. In this embodiment, after the overmodulation dynamic boundary clamping unit proportionally reduced the action time of the two effective vectors, the single-cycle phase voltage characteristic volt-second product was 11985.3 V·μs, the total harmonic distortion rate of the phase current was 4.95%, and the output shaft torque ripple rate was 6.8%. During this process, the ratio of the action time of the two effective vectors remained unchanged, and the phase relationship of the spatially synthesized voltage vector remained stable.

[0069] Example 3: The main power conversion module in this example includes a rectifier and filter unit, a DC bus energy storage unit, and a three-phase inverter bridge unit. The three-phase inverter bridge unit is connected to the controlled motor load. The rectifier and filter unit uses a three-phase uncontrolled rectifier bridge. The DC bus energy storage unit is used to store and smooth the rectified DC voltage. The three-phase inverter bridge unit uses an insulated-gate bipolar transistor as the controlled switch. The state sensing and sampling module includes a DC bus voltage transient sampling unit and a phase current detection unit. The DC bus voltage transient sampling unit obtains the voltage signal at both ends of the DC bus through isolated sampling. The phase current detection unit uses Hall current sensors set on the two-phase output lines of the AC output side of the three-phase inverter bridge unit to collect the instantaneous phase current of the two phases. The digital signal microprocessor carries the adaptive modulation control module and is connected to the DC bus voltage transient sampling unit, the phase current detection unit, and the three-phase inverter bridge unit.

[0070] To balance closed-loop regulation and transient feedforward control within a single-cycle switching carrier, the digital signal microprocessor employs a two-channel processing approach for the DC bus voltage signal. One channel performs low-pass filtering on the acquired DC bus voltage signal to obtain the effective bus voltage, which is then used for setting the adjustment parameters of the speed loop and current loop, as well as calculating the field weakening voltage limit. The other channel directly calls upon the instantaneous DC bus voltage values ​​obtained at the apex and trough of the symmetrical triangular carrier wave. The transient voltage compensation gain coefficient is calculated by the bus fluctuation transient feedforward compensation unit, and then the vector timing dynamic reconstruction unit corrects the first effective vector action time, the second effective vector action time, and the zero vector action time within the current switching carrier cycle accordingly. Thus, the bus voltage signal used for closed-loop regulation and the instantaneous DC bus voltage value used for single-cycle vector timing reconstruction enter their respective control channels.

[0071] During closed-loop regulation, the photoelectric encoder acquires the rotor mechanical angular position of the controlled motor load and obtains the actual speed accordingly. The speed regulation unit compares the target speed setpoint with the actual speed, and generates the quadrature-axis current reference through proportional-integral regulation. After the phase current detection unit obtains the instantaneous phase current of the two phases, the coordinate transformation unit performs a stationary two-phase coordinate transformation, calculates the stator current vector amplitude and spatial position angle, and inputs the stator current vector amplitude and spatial position angle into the closed-loop regulation loop. The closed-loop regulation loop obtains the current feedback quantity based on this, and generates the direct-axis voltage reference and quadrature-axis voltage reference by combining the speed regulation result. These are then synthesized into a target reference voltage vector in the stationary two-phase coordinate system. This target reference voltage vector also serves as the input for the vector timing dynamic reconstruction unit to perform spatial vector action time calculation.

[0072] When the effective bus voltage changes, the current regulation unit adjusts the proportional-integral regulation parameters of the direct-axis current loop and quadrature-axis current loop accordingly to adapt the closed-loop current regulation to the current bus power supply status. The stator voltage vector limiting and field weakening control unit simultaneously determine the current allowable output voltage amplitude based on the effective bus voltage. When the speed increases or the bus voltage decreases, causing the target voltage vector to reach the corresponding output voltage limit, the direct-axis current setpoint is lowered, and the current loop continues to generate the direct-axis voltage setpoint and quadrature-axis voltage setpoint. This slow regulation process uses the filtered effective bus voltage, while the single-cycle correction of the effective vector action time still directly uses the instantaneous DC bus voltage value. The two are respectively applied to the closed-loop regulation and space vector pulse width modulation timing calculation.

[0073] During the dead-time compensation process of the three-phase inverter bridge unit, after the phase current detection unit measures the instantaneous current of each phase, it determines the dead-time error voltage direction of the corresponding bridge arm according to the phase current polarity. The digital signal microprocessor simultaneously reads the preset hardware dead-time of the switching transistor, the difference between the turn-on and turn-off delay of the switching transistor, the saturation conduction voltage drop of the switching device, and the difference in forward voltage drop of the anti-parallel freewheeling diode. Combined with the instantaneous DC bus voltage value obtained in the current switching carrier cycle, the volt-second deviation formed by the dead-time, switching delay, and device conduction voltage drop is converted into the time-domain dead-time compensation amount of the conduction time of each phase bridge arm. The obtained time-domain dead-time compensation amount is superimposed on the reference modulation conduction time of the corresponding bridge arm according to the compensation direction determined by the phase current polarity, and directly participates in the dead-time compensation calculation with the instantaneous DC bus voltage value.

[0074] After dead-zone compensation is completed, the adaptive modulation control module generates a space vector pulse width modulation drive pulse sequence using the corrected vector action time and the reference modulation conduction time of each phase bridge arm after superimposing the time-domain dead-zone compensation. The drive pulse is then output to the three-phase inverter bridge unit to control the operation of each controlled switch. When a transient voltage drop occurs in the power supply voltage, the bus fluctuation transient feedforward compensation unit directly updates the transient voltage compensation gain coefficient using the DC bus instantaneous voltage value obtained from the current switching carrier cycle. The vector timing dynamic reconstruction unit then updates the first effective vector action time, the second effective vector action time, and the zero vector action time. At the same time, the closed-loop regulation loop continues to use the filtered effective bus voltage to complete the speed, current, and field weakening regulation, realizing independent decoupling control of the single-cycle modulation timing correction and the closed-loop operation state regulation channel.

[0075] Example 4: This example combines Figures 1 to 2 This paper describes the wide-voltage adaptive variable frequency speed control system and method, such as... Figure 1 As shown, in step S1, the instantaneous DC bus voltage value collected by the DC bus voltage transient sampling unit at the start of the current switching carrier cycle is obtained. With the preset bus reference voltage value remaining unchanged, the dimensionless transient voltage compensation gain coefficient is calculated based on its inverse relationship with the instantaneous DC bus voltage value. In step S2, the target reference voltage vector in the stationary two-phase coordinate system is obtained. Based on the effective vector action time generated in the sector where the target reference voltage vector is located in the current sampling cycle, the transient voltage compensation gain coefficient is multiplied by the effective vector action time to calculate and output the compensated first effective vector action time. Second effective vector action time and zero vector action time Step S3: When the sum of the first effective vector action time and the second effective vector action time exceeds the preset switching carrier period, based on the preset switching carrier period and the first effective vector action time... and the second effective vector action time The sum is The ratio of the sums generates an overmodulation scaling factor. This factor is used to proportionally shrink and clamp the first and second effective vector action times, maintaining a constant ratio between them, and setting the final zero vector action time to zero. When the proportional overmodulation shrinking clamp is triggered, the effective vector volt-second product lost within a single cycle is calculated and accumulated in the time domain based on the effective vector time difference before and after clamping and the instantaneous DC bus voltage. If the sum of the effective vector action times in subsequent switching carrier cycles is less than a preset switching carrier cycle, the accumulated volt-second product loss is converted into a time compensation amount and added equally to the first and second effective vector action times in the unsaturated cycle. The total effective vector action time after the addition is capped at the preset switching carrier period. In step S4, the direction of the dead zone error voltage is determined based on the polarity of the instantaneous current of each phase measured by the phase current detection unit. Based on the preset hardware dead time of the switching transistor, the difference between the turn-on and turn-off delay of the switching transistor, the difference between the saturation conduction voltage drop of the switching device and the forward voltage drop of the anti-parallel freewheeling diode, and the obtained instantaneous DC bus voltage value, the time-domain dead zone compensation amount of the conduction time of each phase bridge arm is calculated, and the time-domain dead zone compensation amount is superimposed on the reference modulation conduction time of each phase bridge arm. In step S5, based on the corrected vector action time and the reference modulation conduction time after superimposing the time-domain dead zone compensation amount, a space vector pulse width modulation drive pulse sequence is generated and output to the three-phase inverter bridge unit to drive the controlled switching transistor to operate.

[0076] like Figure 2As shown, the wide-voltage adaptive variable frequency speed control system includes a main power conversion module, a state sensing sampling module, a magnetic saturation and resonance suppression unit, a bus fluctuation transient feedforward compensation unit, a vector timing dynamic reconstruction unit, an overmodulation dynamic boundary clamping unit, and an overmodulation volt-second loss cross-cycle accumulation compensation unit. The main power conversion module includes a three-phase electromagnetic interference filter circuit, a three-phase uncontrolled rectifier bridge, a DC bus energy storage unit, and a three-phase inverter bridge unit connected in sequence. The state sensing sampling module includes a DC bus voltage transient sampling unit and a phase current detection unit, wherein the DC bus voltage transient sampling unit includes a high-frequency voltage divider sampling circuit and a linear isolation amplifier circuit. The bus fluctuation transient feedforward compensation unit acquires the instantaneous DC bus voltage value and calculates the output transient voltage compensation gain coefficient by combining it with a preset bus reference voltage value. The sequence dynamic reconstruction unit outputs the first effective vector action time, the second effective vector action time, and the zero vector action time based on the basic effective vector action time and the transient voltage compensation gain coefficient; the magnetic saturation and resonance suppression unit outputs the anti-phase damping voltage component and adjusts the direct-axis demagnetizing current setpoint; the overmodulation dynamic boundary clamping unit generates an overmodulation scaling factor when the sum of the first effective vector action time and the second effective vector action time exceeds the preset switching carrier period, and performs proportional shrinking clamping on the first effective vector action time and the second effective vector action time; the overmodulation volt-second loss cross-cycle accumulation compensation unit calculates the effective vector volt-second product of storage loss when triggering the proportional shrinking clamping of overmodulation, and converts it into time compensation amount in subsequent unsaturated periods and adds it to the first effective vector action time and the second effective vector action time in equal amounts.

[0077] Example 5: When the system is powered on for the first time or a new controlled motor load is connected, the adaptive modulation control module enters the modulation pre-modulation offline parameter self-test calibration process. The state sensing sampling module performs static zero-point and hardware gain calibration on the DC bus voltage transient sampling unit and the phase current detection unit on the AC output side, respectively. When the adaptive modulation control module is in the pulse blockade state, it continuously collects the DC bus voltage divider signal and the phase current sensor bias level for 1024 switching carrier cycles, calculates the arithmetic average value as the static zero drift bias compensation amount of the corresponding channel and stores it in the internal memory. When the analog-to-digital conversion sampling is triggered synchronously at the apex and trough of the symmetrical triangular carrier, the corresponding static zero drift is deducted.

[0078] To address the device parameter drift caused by changes in junction temperature of the controlled switching transistors in the three-phase inverter bridge unit, the system establishes a temperature compensation gradient lookup table for the saturation on-state voltage drop of the switching devices and the forward on-state voltage drop of the fast recovery freewheeling diode. A temperature sensor mounted on the heat dissipation substrate of the three-phase inverter bridge unit collects the current substrate temperature. The microprocessor estimates the real-time junction temperature by combining the thermal resistance parameters of the device junction to the substrate. Linear interpolation is then used between adjacent calibration nodes to calculate the saturation on-state voltage drop of the switching devices and the forward on-state voltage drop of the fast recovery freewheeling diode corresponding to the current junction temperature. The calculation results are input during operation. The calculation process for the dead zone compensation involves setting five reference temperature calibration nodes at 25℃, 50℃, 75℃, 100℃, and 125℃ using a temperature compensation gradient lookup table. At the 25℃ calibration node, the reference value for the saturation on-state voltage drop of the switching device is 1.80V, and the reference value for the forward on-state voltage drop of the fast recovery freewheeling diode is 1.20V. As the temperature increases, the saturation on-state voltage drop of the switching device increases by a temperature coefficient of 0.002V per degree Celsius, while the forward on-state voltage drop of the fast recovery freewheeling diode decreases by a temperature coefficient of 0.0015V per degree Celsius.

[0079] When the controlled motor is under light load or in the phase current zero-crossing region, the bus coupling dead-zone effect dynamic compensation unit processes the polarity misjudgment caused by high-frequency ripple through the current polarity hysteresis discrimination window. The microprocessor compares the absolute value of the instantaneous phase current with the preset zero-crossing threshold current. When the absolute value of the instantaneous phase current is lower than the preset zero-crossing threshold current, the phase current polarity sign of the previous switching carrier cycle is maintained, and the dead-zone error voltage direction is determined by this polarity sign. When the absolute value of the instantaneous phase current exceeds the preset zero-crossing threshold current, the current phase current polarity is then determined according to the current phase current polarity. The polarity flag is updated, and the preset zero-crossing threshold current is 0.05 times the effective value of the rated phase current of the motor. The inverter high-frequency switching ripple superimposed on the phase current fundamental wave and the peak value of the simulated sampling thermal noise are in the range of 0.02 to 0.03 times the rated phase current of the motor. When the zero-crossing threshold current coefficient is less than 0.03 times, the high-frequency ripple will cause the polarity flag to flip frequently near the zero point. When it exceeds 0.08 times, it will prolong the polarity holding time in the current zero-crossing region. Therefore, this embodiment uses 0.05 times as the zero-crossing threshold current coefficient.

[0080] During the dead-zone compensation process, the microprocessor determines the direction of the dead-zone error voltage based on the polarity of the instantaneous current in each phase measured by the phase current detection unit. It then uses the real-time updated saturation conduction voltage drop of the switching device, the forward conduction voltage drop of the fast recovery freewheeling diode, the preset hardware dead-zone time of the switching transistor, the difference between the turn-on and turn-off delay of the switching transistor, and the instantaneous DC bus voltage value of the current switching carrier cycle to calculate the time-domain dead-zone compensation amount. The resulting time-domain dead-zone compensation amount is superimposed on the reference modulation conduction time of each phase bridge arm to compensate for the deviation caused by device temperature changes and current zero-crossing polarity changes in the actual conduction time of the bridge arm.

[0081] The magnetic saturation and resonance suppression unit adjusts the direct-axis demagnetizing current setting using a voltage response coefficient. This voltage response coefficient is determined based on the nominal capacitance of the DC bus energy storage unit and the direct-axis inductance of the motor stator. The calibration relationship is the square root of the ratio of the nominal capacitance of the DC bus energy storage unit to the direct-axis inductance of the motor stator. When a transient overvoltage pulse occurs at the DC bus terminal, the magnetic saturation and resonance suppression unit determines the downward adjustment of the direct-axis demagnetizing current setting based on the deviation between the instantaneous DC bus voltage value and the preset overvoltage safety threshold. Simultaneously, an anti-phase damping voltage component is injected into the target reference voltage vector in a stationary two-phase coordinate system, so that the two adjustments act on the high-frequency oscillation of the DC bus and the change in the magnetic flux of the controlled motor stator core, respectively.

[0082] The switching action of the three-phase inverter bridge unit will excite a high-frequency series resonance between the DC bus energy storage unit and the parasitic inductance of the main circuit. The oscillation frequency is in the range of tens of kilohertz to megahertz. The high-frequency thin-film absorption capacitor connected in parallel in the DC bus energy storage unit provides a low-impedance path for the high-frequency oscillation current. The adaptive modulation control module injects an anti-phase damping voltage component into the target reference voltage vector within a single switching carrier cycle to adjust the corresponding switching timing. The direct-axis demagnetizing current in the synchronous rotating coordinate system is adjusted according to the fundamental control bandwidth. The transient magnetic flux density of the stator core of the controlled motor is reduced by the direct-axis demagnetizing magnetomotive force on the stator side.

[0083] Example 6: When a precision temperature-controlled compressor is starting under heavy load with a large differential pressure, and the input AC power grid experiences a limit drop, the adaptive modulation control module activates the limit undervoltage start safety protection unit. The DC bus voltage transient sampling unit and the phase current detection unit on the AC output side of the three-phase inverter bridge unit respectively acquire the instantaneous DC bus voltage value and the instantaneous phase current value of the current switching carrier cycle. When the instantaneous DC bus voltage value is lower than the preset low-voltage threshold and the instantaneous phase current value exceeds the preset overcurrent threshold, the limit undervoltage start safety protection unit limits the amplitude rise slope of the target reference voltage vector in the stationary two-phase coordinate system. At the same time, it lowers the inverter output fundamental frequency setpoint periodically according to the preset frequency reduction step size, so that the peak value of the single-cycle phase current remains within the safe operating range of the controlled motor load.

[0084] In this embodiment, the preset frequency reduction step size is calibrated according to the ratio of the rated fundamental frequency of the controlled motor load to the number of frequency reduction attenuation filter cycles, where the rated fundamental frequency is 50Hz, the number of frequency reduction attenuation filter cycles is 500, and the preset ramp-up step size of the target reference voltage vector is set to 0.5V. This value is determined based on the stator phase inductance of the controlled motor load and the maximum allowable phase current change rate of the three-phase inverter bridge unit. Within each switching carrier cycle, the limiting controller calculates the amplitude of the current target reference voltage vector and the actual output reference voltage of the previous switching carrier cycle. The absolute difference between the voltage vector amplitudes is used. When the absolute difference is greater than 0.5V, the target reference voltage vector amplitude of the current cycle is limited to the actual output amplitude of the previous cycle plus 0.5V, and the voltage component in the stationary two-phase coordinate system is recalculated based on the limited amplitude and the current spatial position angle. When the absolute difference is less than or equal to 0.5V, the target reference voltage vector amplitude calculated at the moment is used. At the same time, the inverter output fundamental frequency setpoint is updated according to the preset frequency reduction step size in each switching carrier cycle, so that the output fundamental frequency decreases cycle by cycle.

[0085] When the inverter modulation enters the deep overmodulation region and triggers proportional shrinkage clamping, the overmodulation volt-second loss cross-cycle accumulation compensation unit maintains a sliding time-domain accumulation queue through an internal register. Based on the first effective vector action time before and after clamping, the second effective vector action time, and the current DC bus instantaneous voltage value, it calculates the effective vector volt-second accumulation loss generated in that switching carrier cycle, and continuously accumulates the volt-second accumulation loss for each cycle; the effective vector volt-second accumulation loss for the k-th switching carrier cycle is denoted as... The calculation formula is: ;in, The timing sequence number of the switching carrier period. For the first The effective vector volt-second product of the loss for each switching carrier cycle; the accumulated volt-second product loss currently stored in the cross-cycle volt-second loss accumulator register is denoted as... The accumulator is set to a saturation upper limit threshold. ,in, The saturation limit threshold for the accumulator is set beforehand. Under prolonged undervoltage conditions, the accumulation result is limited to this threshold range. The anti-saturation boundary formula is as follows: ;in, This register stores the accumulated volt-second deficit from the previous switching carrier cycle; when the system detects a shutdown control command or protection shutdown signal, the cross-cycle volt-second deficit accumulation register is forcibly cleared. During normal recovery operation, when the accumulated volt-second loss is completely recovered by subsequent unsaturated cycles, the cross-cycle volt-second loss accumulation register is also cleared. Reset to 0.

[0086] When the input voltage recovers, the instantaneous DC bus voltage rises, and the system exits the overmodulation region, the overmodulation volt-second deficit cross-cycle accumulation compensation unit performs compensation within the unsaturated switching carrier cycle when the sum of the first effective vector action time and the second effective vector action time is less than the preset switching carrier cycle. The criterion for the microprocessor to perform compensation timing in the current switching carrier cycle is: and When the current period is determined to meet the compensation condition, the remaining modulation time of the current switching carrier period is used as the compensation time boundary to calculate the maximum time margin allowed for compensation in the current switching carrier period. Then calculate the maximum permissible volt-second product for a single cycle in the current period. Compare the accumulated volt-second deficit stored in the accumulator register. With the maximum permissible volt-second product in a single cycle Determine the actual release of the replenishment volt-second product in the current cycle. The actual released volt-second product is then converted into a single-vector time compensation amount that is equally distributed to the two effective vectors. The single-vector time compensation amount is then added equally to the first effective vector action time and the second effective vector action time to obtain the final effective vector action time after compensation. ; And the total effective vector action time after addition Strictly based on the preset switching carrier period This is the upper limit threshold; after this addition is completed, update the accumulated volt-second loss amount stored in the current cycle of the cross-cycle volt-second loss accumulation register: .

[0087] When the volt-second product deficit temporarily stored in the accumulator register is greater than the volt-second product allowed to be replenished in the current switching carrier cycle, only the portion not exceeding the maximum allowed volt-second product replenishment in a single cycle is released in this cycle, and the corresponding time compensation amount is added equally to the first effective vector action time and the second effective vector action time; the unreleased volt-second product deficit continues to be stored in the sliding time-domain accumulator queue and deferred to subsequent switching carrier cycles that meet the replenishment conditions. As each unsaturated switching carrier cycle continuously performs the above replenishment, the accumulated volt-second product deficit decreases cycle by cycle until the corresponding cross-cycle compensation is completed.

[0088] The embodiments of this application have been described above with reference to the accompanying drawings. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. This application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make various equivalent modifications under the guidance of this application without departing from the spirit of this application and the scope of protection of this invention, all of which fall within the scope of protection of this application.

Claims

1. A wide-voltage adaptive variable frequency speed control system, characterized in that, include: The main power conversion module includes a rectifier and filter unit, a DC bus energy storage unit, and a three-phase inverter bridge unit connected in sequence to the controlled motor load. The rectifier and filter unit is connected to the AC power supply. The state sensing sampling module includes a DC bus voltage transient sampling unit located at both ends of the DC bus energy storage unit, and a phase current detection unit located at the output side of the three-phase inverter bridge unit. The adaptive modulation control module has signal input terminals connected to the DC bus voltage transient sampling unit and the phase current detection unit, respectively, and its output terminal connected to the three-phase inverter bridge unit, including: The bus fluctuation transient feedforward compensation unit is used to obtain the instantaneous voltage value of the DC bus. When the preset bus reference voltage value remains unchanged, it calculates the output transient voltage compensation gain coefficient according to the inverse relationship with the instantaneous voltage value of the DC bus. The vector timing dynamic reconstruction unit is used to generate the basic effective vector duration based on the target reference voltage vector, multiply the transient voltage compensation gain coefficient by the basic effective vector duration, and output the first effective vector duration. Second effective vector action time and zero vector action time ; The overmodulation dynamic boundary clamping unit is used to clamp the first effective vector according to the preset switching carrier period and the first effective vector's action time when the sum of the two effective vector action times exceeds the preset switching carrier period. and the second effective vector action time The sum is The ratio of the sums generates an overmodulation scaling factor to implement proportional contraction clamping, keeping the ratio of the first effective vector action time to the second effective vector action time constant and the final zero vector action time zero; The pulse width modulation signal generation unit is used to generate a space vector pulse width modulation drive pulse based on the corrected vector action time and output it to the three-phase inverter bridge unit. The overmodulation volt-second loss cross-cycle accumulation compensation unit has its signal input connected to the overmodulation dynamic boundary clamping unit. When triggering the overmodulation proportional contraction clamping, the overmodulation volt-second loss cross-cycle accumulation compensation unit calculates the effective vector volt-second product lost within a single cycle based on the effective vector time difference before and after clamping and the instantaneous DC bus voltage value, and stores it in the time domain. When the sum of the effective vector action times in subsequent switching carrier cycles is less than the preset switching carrier cycle, the accumulated volt-second product loss is converted into a time compensation amount and added equally to the first and second effective vector action times of the unsaturated cycle. The total effective vector action time after the addition is capped at the preset switching carrier cycle.

2. The wide voltage adaptive variable frequency speed control system according to claim 1, characterized in that, The DC bus voltage transient sampling unit includes a high-frequency voltage divider sampling circuit and a linear isolation amplifier circuit, with a signal conditioning delay of less than 2μs. The DC bus voltage transient sampling unit is synchronously connected to the symmetrical triangular carrier generation unit inside the adaptive modulation control module. The analog-to-digital conversion sampling is triggered synchronously at the apex and trough of the symmetrical triangular carrier to obtain the instantaneous DC bus voltage value, and the instantaneous DC bus voltage value is input to the bus fluctuation transient feedforward compensation unit in real time.

3. The wide voltage adaptive variable frequency speed control system according to claim 1, characterized in that, The adaptive modulation control module also includes: a magnetic saturation and resonance suppression unit, with the signal input terminal connected to the bus fluctuation transient feedforward compensation unit; the magnetic saturation and resonance suppression unit is used to inject an anti-phase damping voltage component into the target reference voltage vector in the stationary two-phase coordinate system when the instantaneous voltage value of the DC bus exceeds the preset overvoltage safety threshold, and to lower the direct-axis demagnetizing current setpoint in the synchronous rotating coordinate system, thereby suppressing the high-frequency switching resonance generated by the DC bus energy storage unit and the parasitic inductance of the main circuit, and reducing the transient magnetic flux density of the stator core of the controlled motor load.

4. The wide voltage adaptive variable frequency speed control system according to claim 1, characterized in that, The pulse width modulation signal generation unit outputs a seven-segment symmetrical space vector pulse sequence; when the sum of the first effective vector action time and the second effective vector action time reaches the preset switching carrier period and the zero vector action time is zero, the pulse width modulation signal generation unit smoothly switches to a zero-vector pulse modulation mode in which each phase bridge arm operates only once per single switching cycle.

5. The wide voltage adaptive variable frequency speed control system according to claim 1, characterized in that, The phase current detection unit in the state sensing sampling module includes a Hall current sensor installed on two phase output lines on the AC output side of the three-phase inverter bridge unit; the adaptive modulation control module also includes a coordinate transformation unit, which receives the measured two-phase instantaneous phase current, calculates the stator current vector amplitude and spatial position angle of the controlled motor load through static two-phase coordinate transformation, and inputs the stator current vector amplitude and spatial position angle to the closed-loop regulation circuit to generate the target reference voltage vector.

6. The wide voltage adaptive variable frequency speed control system according to claim 1, characterized in that, In the main power conversion module, the rectifier and filter unit includes a three-phase electromagnetic interference filter circuit and a three-phase uncontrolled rectifier bridge; the DC bus energy storage unit includes a high-frequency thin-film absorption capacitor and an aluminum electrolytic energy storage capacitor group connected in parallel; the three-phase inverter bridge unit includes six symmetrically arranged insulated gate bipolar transistors and fast recovery freewheeling diodes connected in reverse parallel to each transistor.

7. The wide voltage adaptive variable frequency speed control system according to claim 1, characterized in that, When the target reference voltage vector is in the first sector, the vector timing dynamic reconstruction unit uses half of the updated zero vector action time as the reference time offset, and accumulates the reference time offset, the first effective vector action time and the second effective vector action time to generate the conduction comparison count values ​​of the upper bridge arm switches of phase a, phase b and phase c of the three-phase inverter bridge unit, respectively, and outputs them to the comparison register inside the pulse width modulation signal generation unit.

8. A wide voltage adaptive variable frequency speed control system according to claim 1, characterized in that, The adaptive modulation control module also includes: a limit undervoltage start safety protection unit, whose signal input terminals are respectively connected to the bus fluctuation transient feedforward compensation unit and the phase current detection unit; the limit undervoltage start safety protection unit is used to apply amplitude limiting constraint to the amplitude ramping slope of the target reference voltage vector when the instantaneous DC bus voltage value is lower than the preset limit low voltage threshold and the instantaneous phase current value exceeds the preset overcurrent threshold, and to reduce the inverter output fundamental frequency setpoint by a preset frequency reduction step size, so as to maintain the single-cycle phase current peak value of the three-phase inverter bridge unit in the safe operating range under power supply drop conditions.

9. A wide-voltage adaptive variable frequency speed control method, implemented by the wide-voltage adaptive variable frequency speed control system according to claim 1, characterized in that, Includes the following steps: Step S1: Obtain the instantaneous DC bus voltage value collected by the DC bus voltage transient sampling unit at the start of the current switching carrier cycle. When the preset bus reference voltage value remains unchanged, calculate the dimensionless transient voltage compensation gain coefficient according to the inverse relationship with the instantaneous DC bus voltage value. Step S2: Obtain the target reference voltage vector in the stationary two-phase coordinate system. Based on the target reference voltage vector in the sector of the current sampling period, generate the basic effective vector action time. Multiply the transient voltage compensation gain coefficient as a multiplication factor into the basic effective vector action time, and calculate the output compensated first effective vector action time. Second effective vector action time and zero vector action time ; Step S3: When the sum of the first effective vector action time and the second effective vector action time exceeds the preset switching carrier period, based on the preset switching carrier period and the first effective vector action time... and the second effective vector action time The sum is The ratio of the sums generates an overmodulation scaling factor. The overmodulation scaling factor is used to apply proportional contraction clamping to the first effective vector action time and the second effective vector action time, keeping the ratio of the first effective vector action time to the second effective vector action time constant, and setting the final zero vector action time to zero. When the overmodulation proportional contraction clamp is triggered, the effective vector volt-second product lost in a single cycle is calculated and accumulated in the time domain based on the effective vector time difference before and after clamping and the instantaneous DC bus voltage value. When the sum of the effective vector action time in subsequent switching carrier cycles is less than the preset switching carrier cycle, the accumulated volt-second product loss is converted into a time compensation amount and added equally to the first and second effective vector action times of the unsaturated cycle. The total effective vector action time after the addition is capped at the preset switching carrier cycle. Step S4: Determine the direction of the dead zone error voltage based on the polarity of the instantaneous current of each phase measured by the phase current detection unit. Based on the preset hardware dead time of the switching transistor, the difference between the turn-on and turn-off delay of the switching transistor, the difference between the saturation conduction voltage drop of the switching device and the forward voltage drop of the anti-parallel freewheeling diode, and combined with the obtained instantaneous DC bus voltage value, calculate the time-domain dead zone compensation amount of the conduction time of each phase bridge arm, and add the time-domain dead zone compensation amount to the reference modulation conduction time of each phase bridge arm. Step S5: Based on the corrected vector action time and the reference modulation conduction time after superimposing the time-domain dead zone compensation, a space vector pulse width modulation drive pulse sequence is generated and output to the three-phase inverter bridge unit to drive the controlled switch tube to operate.

Images