Synchronous control system of gearless double-permanent-magnet air compressor motor

Abstract

The present application relates to motor drive and industrial automation control technical field, specifically relates to gearless double permanent magnet air compressor motor synchronous control system, include: data acquisition module is used for obtaining the rotor absolute space angle and phase current of first motor and second motor;Vector analysis module is used for coordinate transformation analysis to phase current, obtains current cross axis current and current direct axis current;Space domain feedforward observation module is used for respectively into corresponding first motor's first feedforward current instruction and corresponding second motor's second feedforward current instruction according to;Cross coupling control module is used for combining the adaptive weight coefficient based on motor operation state real-time calculation and converting synchronous error into first compensation current and second compensation current;Synchronous execution module is used for generating voltage instruction, generates drive signal through space vector pulse width modulation;The present application avoids unilateral torque sharp mutation, realizes high synchronous stiffness control under lower current fluctuation.

Description

Technical Field

[0001] This invention relates to the field of motor drive and industrial automation control technology, specifically to a synchronous control system for a gearless dual permanent magnet air compressor motor. Background Technology

[0002] Synchronous control of gearless dual permanent magnet motors is mainly used to control multiple motors that drive the same load. Its technical essence is to adjust the torque and speed of different motors so that the driven components maintain a stable spatial phase relationship in a continuous working cycle. This is a high-precision industrial control method.

[0003] Existing dual-motor synchronous control methods include data exchange control based on an external communication bus and master-slave follower control. These conventional control strategies typically rely on real-time error feedback during operation to infer the load and make adjustments. Both address the issue of matching the basic speed and phase of the two motors under normal, stable loads. However, for applications such as twin-screw air compressors facing deterministic, pulsating loads highly dependent on spatial position—for example, when the compression chamber rapidly enters the high-pressure zone—conventional error feedback mechanisms tend to first cause physical synchronization loss before compensation, resulting in inherent control lag. Simultaneously, external bus data exchange can lead to the continuous accumulation of position errors. Therefore, how to avoid error accumulation and achieve high-rigidity synchronous control with low current fluctuations without mechanical gears has become a problem to be solved. Summary of the Invention

[0004] The purpose of this invention is to provide a synchronous control system for a gearless dual permanent magnet air compressor motor, solving the following technical problems:

[0005] It avoids the accumulation of position errors caused by external bus switching and the phase deviation caused by feedback lag. It can also pre-map and compensate deterministic pulsating loads and dynamically allocate compensation tasks according to real-time energy status, thereby achieving high synchronous stiffness control with lower current fluctuations and equivalent electronic meshing synchronization that is more in line with physical inertia.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A synchronous control system for a gearless dual permanent magnet motor, used to control a first motor and a second motor that jointly drive the same load, includes a main control chip, which is communicatively connected to a data acquisition module, a vector analysis module, a spatial domain feedforward observation module, a cross-coupling control module, and a synchronous execution module.

[0008] The data acquisition module is used to acquire the absolute spatial angles and phase currents of the rotors of the first motor and the second motor; the vector analysis module is used to perform coordinate transformation analysis on the phase currents to acquire the current quadrature-axis current and the current direct-axis current.

[0009] The spatial domain feedforward observation module is used to obtain the expected load torque according to the rotor absolute spatial angle and the preset spatial mapping table, and convert it into a first feedforward current command corresponding to the first motor and a second feedforward current command corresponding to the second motor, respectively.

[0010] The cross-coupling control module is used to calculate the synchronization error based on the absolute spatial angle of the rotor, and combine the adaptive weighting coefficient calculated in real time based on the motor operating status to convert the synchronization error into a first compensation current and a second compensation current.

[0011] The synchronous execution module is used to fuse the first feedforward current command with the first compensation current, and the second feedforward current command with the second compensation current to generate a first current command and a second current command, respectively, and generate a voltage command with the feedback phase current closed-loop adjustment, generate a drive signal through space vector pulse width modulation, and control the inverter that is electrically connected to both the first motor and the second motor to perform synchronous drive operation on the two motors.

[0012] Optionally, the process by which the data acquisition module obtains the absolute spatial angle of the rotors and the phase current of the first motor and the second motor includes:

[0013] The absolute spatial angles of the rotors of the first motor and the second motor are obtained by absolute encoders installed on the first motor and the second motor; the phase currents of the first motor and the second motor are obtained by current sensors installed on the inverter bridge arm.

[0014] Optionally, the preset process for generating the spatial mapping table includes:

[0015] Control the first motor and the second motor to rotate at a preset calibrated speed; collect the quadrature-axis current of the first motor and the second motor; perform integral calculation on the quadrature-axis current to calibrate the load torque curve;

[0016] The load torque curve is bound to the rotor absolute spatial angle to generate the spatial mapping table and write it into non-volatile memory.

[0017] Optionally, the process by which the spatial domain feedforward observation module acquires the expected load torque and converts it into a first feedforward current command and a second feedforward current command, respectively, includes:

[0018] Using the absolute spatial angle of the rotor as the index address, address matching is performed in the spatial mapping table;

[0019] Extract the load torque value corresponding to the address matching as the expected load torque; calculate the corresponding first feedforward current command and second feedforward current command according to the expected load torque and the electromagnetic torque conversion relationship of the corresponding motor obtained in advance.

[0020] Optionally, the process by which the cross-coupling control module calculates the synchronization error includes:

[0021] Obtain the number of first pole pairs of the first motor and the number of second pole pairs of the second motor; divide the number of first pole pairs by the number of second pole pairs to obtain the virtual electronic gear ratio;

[0022] Multiply the absolute spatial angle of the rotor of the second motor by the virtual electronic gear ratio to obtain the converted spatial angle;

[0023] The synchronization error is obtained by subtracting the calculated spatial angle from the absolute spatial angle of the rotor of the first motor.

[0024] Optionally, the process of the cross-coupling control module obtaining the adaptive weight coefficient includes: calculating the rotor angular velocity based on the rotor absolute spatial angle, and combining it with the preset first moment of inertia of the first motor and the second moment of inertia of the second motor; calculating the first rotor kinetic energy based on the rotor angular velocity and the first moment of inertia, and calculating the second rotor kinetic energy based on the rotor angular velocity and the second moment of inertia;

[0025] Calculate the ratio of the first rotor kinetic energy to the second rotor kinetic energy to obtain the kinetic energy ratio; perform weight allocation based on the kinetic energy ratio to generate a first adaptive weight coefficient and a second adaptive weight coefficient.

[0026] Optionally, the process by which the cross-coupling control module converts the synchronization error into a first compensation current and a second compensation current includes:

[0027] The synchronization error is input into a preset error-torque compensation function to generate a basic compensation torque; the basic compensation torque is multiplied by the first adaptive weighting coefficient to obtain the first compensation torque;

[0028] The basic compensation torque is multiplied by the second adaptive weighting coefficient to obtain the second compensation torque;

[0029] The first compensation torque is converted into the first compensation current; the second compensation torque is converted into the second compensation current.

[0030] Optionally, the process of the synchronous execution module generating the first current command and the second current command includes:

[0031] Subtract the first compensation current from the first feedforward current command to generate the first current command;

[0032] The second current command is generated by adding the second feedforward current command to the second compensation current, which has the opposite polarity to the first compensation current.

[0033] Optionally, before converting the first compensation current and the second compensation current, the cross-coupling control module further includes a condition determination process for the synchronization error: setting a synchronization error threshold; determining whether the absolute value of the synchronization error is greater than or equal to the synchronization error threshold;

[0034] If the absolute value of the synchronization error is greater than or equal to the synchronization error threshold, then the calculation process of the adaptive weight coefficient is initiated and the first compensation current and the second compensation current are generated.

[0035] If the absolute value of the synchronization error is less than the synchronization error threshold, then the first compensation current and the second compensation current are set to zero.

[0036] The beneficial effects of this invention are:

[0037] 1. The control system of the present invention uses a common main control chip to communicate with core units such as the data acquisition module and the synchronous execution module, and directly performs synchronous drive operation on two motors. By allowing two motors to share the same main control chip to complete synchronous control, the position error caused by external bus data exchange is avoided from the physical architecture, ensuring the real-time performance of data detection and command issuance to meet the preset control cycle response requirements.

[0038] 2. This invention obtains the expected load torque based on the rotor's absolute spatial angle and a preset spatial mapping table through a spatial domain feedforward observation module, and converts it into a feedforward current command in advance. This mechanism maps the deterministic pulsating load unique to the air compressor to the current given side in advance, realizing advance compensation for periodic loads and effectively avoiding the physical step loss phenomenon that is easy to occur when relying solely on error feedback.

[0039] 3. This invention utilizes a cross-coupling control module to calculate the synchronization error and combines it with an adaptive weighting coefficient based on real-time calculation of the rotor kinetic energy of dual motors to convert the error into a compensation current for differential correction. This strategy not only achieves equivalent electronic meshing synchronization under conditions without mechanical gears, but also dynamically allocates compensation tasks according to the real-time energy state of the motor, avoiding abrupt changes in torque on one side and achieving high synchronization stiffness control under low current fluctuations. Attached Figure Description

[0040] The invention will now be further described with reference to the accompanying drawings.

[0041] Figure 1 This is a schematic diagram of the synchronous control system of the gearless dual permanent magnet motor provided in the embodiments of this application. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Please see Figure 1 A synchronous control system for gearless dual permanent magnet motors, used to control a first motor and a second motor that jointly drive the same load, includes a main control chip, which is communicatively connected to a data acquisition module, a vector analysis module, a spatial domain feedforward observation module, a cross-coupling control module, and a synchronous execution module;

[0044] The data acquisition module is used to acquire the absolute spatial angles and phase currents of the rotors of the first motor and the second motor; the vector analysis module is used to perform coordinate transformation analysis on the phase currents to acquire the current quadrature-axis current and the current direct-axis current.

[0045] The spatial domain feedforward observation module is used to obtain the expected load torque according to the rotor absolute spatial angle and the preset spatial mapping table, and convert it into a first feedforward current command corresponding to the first motor and a second feedforward current command corresponding to the second motor, respectively.

[0046] The cross-coupling control module is used to calculate the synchronization error based on the absolute spatial angle of the rotor, and combine the adaptive weighting coefficient calculated in real time based on the motor operating status to convert the synchronization error into a first compensation current and a second compensation current.

[0047] The synchronous execution module is used to fuse the first feedforward current command with the first compensation current, and the second feedforward current command with the second compensation current to generate a first current command and a second current command, respectively. It also generates a voltage command by combining the feedback phase current closed-loop adjustment with the feedback phase current closed-loop adjustment. The module generates a drive signal through space vector pulse width modulation to control the inverter, which is electrically connected to both the first motor and the second motor, to perform synchronous drive operation on the two motors.

[0048] This embodiment provides a synchronous control mechanism for a gearless dual permanent magnet motor. Specifically, the application is a twin-screw air compressor, in which the male rotor is driven by a first permanent magnet synchronous motor and the female rotor is driven by a second permanent magnet synchronous motor. The two motors share the same main control chip to complete synchronous control, avoiding the accumulation of position errors caused by external bus exchanges.

[0049] The entire control process includes: acquisition, analysis, feedforward, coupling compensation and execution steps, so that the two compression rotors maintain a stable spatial phase relationship during the continuous working cycle of intake, compression and exhaust.

[0050] Specifically, within each control cycle, the data acquisition module reads the absolute spatial angles of the two rotors and the three-phase currents of the two motors.

[0051] The vector analysis module performs coordinate transformation on the three-phase currents to obtain their respective current quadrature-axis current and current direct-axis current. The quadrature-axis current corresponds to the electromagnetic torque component, and the direct-axis current corresponds to the flux linkage adjustment component.

[0052] The main control chip sends the absolute spatial angle of the rotor to the spatial domain feedforward observation module, accesses the pre-calibrated spatial mapping table with the angle as the index, and obtains the expected load torque that should be overcome at this spatial position; then, based on the motor's own torque constant, the expected load torque is converted into the first feedforward current command and the second feedforward current command respectively.

[0053] Based on this, the cross-coupling control module does not directly adopt the master-slave follow method, but constructs the synchronization error based on the absolute spatial angles of the two motors; if the spatial angle of the first motor at the current moment is recorded as 30°, and the corresponding spatial angle of the second motor after proportional conversion is 29.7°, then the synchronization error is 0.3°.

[0054] After the synchronization error is processed by the compensation function, which is an adjustment function with a built-in proportional coefficient or a nonlinear limiting rule, a set of compensation amounts for restoring the phase relationship is obtained. Considering that the energy states of the two motors are different under different speeds and inertia conditions, the main control chip also calculates an adaptive weighting coefficient based on the current operating state to determine which motor should undertake more compensation tasks. For example, if the current kinetic energy of the first motor is greater than that of the second motor, the compensation proportional coefficient allocated to the second motor is increased to reduce the impact of torque mutation on the motor with higher kinetic energy.

[0055] Subsequently, the synchronous execution module merges the first feedforward current command with the first compensation current, and merges the second feedforward current command with the second compensation current to form two final current commands. These commands are then adjusted in a closed loop with the feedback current. Specifically, a PI control algorithm can be used. The final current command is compared with the feedback current by subtraction, and the deviation value is input to the PI controller for proportional-integral calculation. Two sets of voltage commands are output and generated as inverter drive signals through space vector pulse width modulation.

[0056] To facilitate understanding, a simplified operating condition simulation can be performed; assume that within a certain control cycle, the expected load torque given by the space mapping table at the current angular position is 8 N·m;

[0057] If the torque-current conversion factor of the first motor is 1 N·m / A, then the first feedforward current command can be 4A; the second motor can also be converted to 4A to share the total load; at this time, the synchronization error generates a basic compensation current of 1A through the compensation function, and after combining the adaptive weight, the compensation value allocated to the first motor is 0.4A, and the compensation value allocated to the second motor is 0.6A; the synchronization execution module corrects the two feedforward commands in opposite directions accordingly, thereby reducing the target speed value of one motor and increasing the target speed value of the other motor, and finally restoring the angle difference to the preset proportional relationship;

[0058] Under abnormal operating conditions, if the encoder data of a certain channel is lost within the sampling period, the main control chip can temporarily maintain the effective angle of the previous cycle and apply a limit to the current compensation value to prevent the current surge caused by a sudden increase in error; if the sampling of a certain phase current is abnormal, the corresponding current loop will switch to protection mode and only retain zero torque or limited torque output; if the current position is not matched in the spatial mapping table, the interpolation result of the adjacent angle will be used as the replacement value to ensure the continuity of the feedforward link.

[0059] For example, during the process of a twin-screw air compressor going from no load to full load, when the compression chamber enters the exhaust zone, the gas reaction force on the rotor will suddenly increase at a specific spatial position.

[0060] At this time, the spatial domain feedforward observation module will provide a larger feedforward current in advance at this angle to avoid the two motors having to wait until they actually lose synchronism before they start to compensate; if the machining deviation or lubrication fluctuation causes the instantaneous load of one of the rotors to exceed the preset reference value, the cross-coupling control module will then perform differential compensation on the two current commands to restore the phase relationship within the preset control cycle.

[0061] The purpose of this step is to pre-map the spatially related deterministic pulsating load in the air compressor to the current-given side and leave the random error to small-amplitude cross-compensation processing, thereby achieving high synchronous stiffness control with low current fluctuations.

[0062] In a preferred embodiment of the present invention, the process by which the data acquisition module acquires the absolute spatial angle of the rotors of the first motor and the second motor and the phase current includes: acquiring the absolute spatial angle of the rotors of the first motor and the second motor by means of an absolute encoder installed on the first motor and the second motor; and acquiring the phase current of the first motor and the second motor by means of a current sensor installed on the inverter bridge arm.

[0063] This embodiment provides a data acquisition mechanism for dual-motor synchronous control. Specifically, in order to ensure that the twin-screw air compressor has stable phase detection accuracy throughout the entire speed range, this embodiment uses two types of raw quantities as closed-loop basic inputs: one type is the rotor absolute spatial angle output by the absolute encoder installed on the shaft ends of the two motors, and the other type is the three-phase current sampling value output by the current sensor arranged on the dual inverter bridge arm.

[0064] Specifically, the shaft ends of the first motor and the second motor are respectively equipped with single-turn or multi-turn absolute encoders; the encoder outputs a frame of angle data after each control cycle. After reading the data, the main control chip can directly obtain the mechanical spatial position of the corresponding rotor. The mechanical spatial position is calibrated and defined with the preset physical assembly meshing point of the rotors of the first motor and the second motor as the unified zero point coordinates, without relying on incremental counting to backtrack. Therefore, even if the equipment loses power and is powered on again, the current rotor position can be restored immediately.

[0065] Correspondingly, Hall current sensors or shunt resistor sampling units installed on the inverter bridge arms are used to collect the two-phase or three-phase current of their respective motors; the main control chip can reconstruct the complete three-phase current information based on the sampling topology; for example, if the first motor measures 3A for phase A and -1A for phase B at a certain moment, it can be deduced that phase C is -2A; the second motor completes the collection in the same way.

[0066] For example, the current output of the first motor encoder can be recorded as 120.2°, and the output of the second motor encoder can be recorded as 78.6°; at the same time, the three-phase current of the first motor can be sampled as {2.1A, -0.9A, -1.2A}, and the three-phase current of the second motor can be sampled as {1.8A, -0.4A, -1.4A}; these quantities are latched in the same control cycle and then input into the subsequent analysis stage to avoid the distortion of spatial error calculation caused by mixing data from different timestamps;

[0067] In an abnormal protection mechanism, if the encoder signal does not change for several consecutive cycles and the current command is not zero, it can be determined that there is a code jam or communication abnormality. At this time, the main control chip will prohibit further increase of torque command and trigger current limiting protection. If the bridge arm current sampling exceeds the range or the bias drift occurs, zero-point calibration will be performed first or the sampling channel will be switched to redundant sampling channel.

[0068] If individual sampling points are contaminated by switching noise, median filtering or holding the previous effective value is used for that period to prevent noise from being mistaken for real load fluctuations.

[0069] For example, at the instant the air compressor switches from standby to loading state, the absolute encoder can directly tell the meshing angle of the two compression rotors, and the current sensor synchronously feeds back the actual torque output. The main control chip obtains the actual spatial position of the rotor and the actual electromagnetic torque output of the motor based on this, so that the subsequent feedforward and coupling compensation have a real input basis.

[0070] The purpose of this step is to provide spatial position and electromagnetic current quantities under a unified time reference for the synchronous control of dual motors, thereby enabling reliable input for subsequent angle matching, load inference, and current closed-loop regulation.

[0071] In a preferred embodiment of the present invention, the process of generating the preset space mapping table includes: controlling the first motor and the second motor to rotate at a preset calibrated speed; collecting the cross-axis current of the first motor and the second motor; performing an integral operation on the cross-axis current to calibrate the load torque curve; binding the load torque curve with the absolute spatial angle of the rotor to generate the space mapping table and writing it into a non-volatile memory.

[0072] This embodiment provides a calibration generation mechanism for a spatial mapping table; specifically, during the initial configuration or parameter reset phase of the twin-screw air compressor, a correspondence between the rotor spatial position and the load torque is established through a special calibration condition, so that the expected load information at the current position can be obtained without an additional torque sensor during operation;

[0073] Specifically, if the load is inferred solely from error feedback during operation, it is easy to lose synchronization before compensation occurs when the compression chamber rapidly enters the high-pressure zone, resulting in inherent control lag.

[0074] Therefore, in this embodiment, two motors are driven to complete at least one full rotation at a preset calibrated speed, and the quadrature-axis current data at each spatial angular position is collected. Since the quadrature-axis current is approximately proportional to the electromagnetic torque, the main control chip can integrate or accumulate the quadrature-axis current sequence in angular order, thereby recovering the load torque curve of the air compressor as the spatial position changes within one revolution.

[0075] To further explain, the above integration operation is preferably piecewise integration or bin-by-bin accumulation within the angle domain, rather than performing a single accumulation without boundaries on the entire current sequence. Specifically, the spatial angle of one cycle can be divided into multiple angle intervals, and the corresponding sampled cross-axis current can be accumulated, averaged, or equivalently converted within each angle interval. Then, the representative load torque of that angle interval can be obtained by combining the torque-current conversion relationship.

[0076] The technical implications of this processing are: on the one hand, it retains the processing method of integral operation to suppress instantaneous sampling noise and switching ripple; on the other hand, it ensures that the output results still correspond to the local load level at each spatial location, without losing angular resolution due to continuous accumulation of integer cycles.

[0077] To avoid the misinterpretation that two independent and non-synthetic torque curves must be obtained separately, it should be noted that the calibration results in this embodiment can be represented as the joint load curve of the cross-axis currents of the two motors under the same spatial reference, or as the equivalent total load curve after conversion according to the load sharing relationship; in other words, the original values ​​of the two cross-axis currents are collected during the calibration stage, while the target load torque value after synthesis, smoothing and conversion can be written into the spatial mapping table for direct table lookup during the operation stage.

[0078] Furthermore, to ensure that the spatial mapping table generated during calibration is consistent with the angle definition used in the table lookup process and the synchronization error calculation process during operation, the rotor absolute spatial angle used to bind the load torque curve in this embodiment adopts a unified spatial reference domain; preferably, the original mechanical absolute angle output by the encoder is first converted within the same domain in the main control chip, and then used as the angle reference for calibration binding.

[0079] Alternatively, during parameter configuration, the angle quantities involved in calibration and operation can be uniformly defined as the same spatial reference quantity. In this way, the index angle, addressing angle, and synchronization error comparison angle in the spatial mapping table will remain consistent in meaning, avoiding the deviation caused by defining the table with one angle during the calibration stage and looking up the table with another angle during the operation stage.

[0080] For example, suppose the mechanical angle of one revolution is roughly divided into four calibration intervals: 0°, 90°, 180°, and 270°; at the calibration speed, the average cross-axis current of the first motor near these four positions is 2A, 5A, 3A, and 6A, respectively, and that of the second motor is 2A, 4A, 3A, and 5A, respectively.

[0081] After segmenting and accumulating the two cross-axis currents, the load peak areas around 90° and 270° can be identified. Combined with the torque conversion factor, the four angle points can be mapped to load curves of, for example, 4 N·m, 9 N·m, 6 N·m, and 11 N·m. Then, these torque values ​​are bound to the corresponding absolute spatial angles to form an indexable spatial mapping table, which is written into non-volatile memory for direct table lookup during subsequent operation.

[0082] Furthermore, the spatial mapping table can be stored using a fixed angle step size; for example, storing one load value every 1°, then there are 360 ​​data points in one circle; if higher precision is considered, a 0.5° step size can also be used in conjunction with linear interpolation; for the same model of equipment, a basic template can be formed after multiple prototype calibrations, and then small-scale corrections can be made during the trial operation of the whole machine.

[0083] In an anomaly protection mechanism, if the speed fluctuation exceeds the preset range during the calibration period, for example, the target calibration speed is 300 r / min but the actual fluctuation exceeds ±5%, then the data of the current cycle will not be written into the mapping table and recalibration is required.

[0084] If the collected values ​​at certain angle points deviate from the surrounding trend due to instantaneous electromagnetic disturbances, a moving average can be taken for multiple neighboring points; if obvious angle breakpoints appear after integer integration, the main control chip performs smoothing according to the principle of beginning and end closure to prevent sudden changes at the 0° / 360° boundary when running the table lookup.

[0085] For example, in the factory test of a twin-screw air compressor, the equipment rotates at a low speed and uniform speed for one revolution, and the main control chip records the cross-axis current changes of the male rotor and female rotor in the intake zone, closed compression zone and exhaust zone.

[0086] Calibration revealed that the load peak near the exhaust zone was significantly higher than in other areas, so this characteristic was incorporated into the spatial mapping table; when the equipment is subsequently put into operation on site, the main control chip can increase the feedforward current in advance before approaching this location.

[0087] The purpose of this step is to pre-encode the inherent periodic spatial load characteristics of the air compressor into the memory, thereby achieving advance compensation for deterministic pulsating loads and reducing the dependence on high-gain feedback during operation.

[0088] In a preferred embodiment of the present invention, the process by which the spatial domain feedforward observation module acquires the expected load torque and converts the expected load torque into a first feedforward current command and a second feedforward current command respectively includes: using the rotor absolute spatial angle as the index address, performing address matching in the spatial mapping table; extracting the load torque value corresponding to the address matching as the expected load torque; and calculating the corresponding first feedforward current command and second feedforward current command respectively based on the expected load torque and the pre-acquired electromagnetic torque conversion relationship of the corresponding motor.

[0089] This embodiment provides a spatial domain feedforward observation mechanism; specifically, after completing the mapping table calibration, the main control chip looks up the table according to the current rotor absolute spatial angle in each control cycle, and directly converts the expected load torque at that angle into the feedforward current command of the two motors.

[0090] In summary, the existing spatial mapping table scheme lacks an execution mechanism for converting current commands into operational phases, resulting in a missing closed-loop execution link in the actual control chain; therefore, this embodiment further uses the absolute spatial angle as the index address.

[0091] For example, if the current angle of the first motor is 135.2°, and the system uses a 1° step size for storage, then the two adjacent entries at 135° and 136° can be located first. If the corresponding load torques in the table are 8.0 N·m and 8.4 N·m respectively, then the expected load torque at 135.2° can be obtained by linear interpolation as approximately 8.08 N·m. After that, the main control chip allocates and converts the expected load torque into a feedforward current command based on the electromagnetic torque conversion relationship between the two motors.

[0092] For ease of understanding, we can assume that the total expected load torque is 8 N·m, and the system is set to share the load with two motors in a 1:1 ratio under this condition, so each motor will bear 4 N·m; if the torque-current conversion relationship of the first motor is 1 N·m to 1 A, then the first feedforward current command is 4 A.

[0093] If the conversion relationship of the second motor is 1 N·m corresponds to 0.8 A, then the second feedforward current command is 5 A; in this way, even if the parameters of the two motors are not exactly the same, they can obtain their own adapted feedforward command, instead of simply copying the same current value.

[0094] Under certain operating conditions, asymmetrical load sharing can also be carried out according to the preset load distribution coefficient; for example, considering thermal balance or rated current difference, the first motor is allowed to bear 40% and the second motor is allowed to bear 60%; if the expected load torque is still 8 N·m, the torque command of the two paths is 3.2 N·m and 4.8 N·m respectively, and then the corresponding feedforward current command is calculated respectively.

[0095] In an anomaly protection mechanism, if the current angle index exceeds the end of the mapping table, the addressing continues from near 0° in an angle rollback manner; if some addresses in the mapping table have not been written with valid values, the average of the valid values ​​before and after is taken or the nearest neighbor value is used as a substitute; if the converted feedforward current exceeds the allowable upper limit of the motor or inverter, the limiting is performed and the saturation flag is recorded, so that the cross-coupling compensation module can reduce the compensation slope in the future and avoid loss of control after the instructions are superimposed.

[0096] For example, when the twin-screw air compressor is running to the position where the exhaust window is about to open, the mapping table entry corresponding to the absolute spatial angle will give a higher expected load torque; the main control chip can directly convert the expected load into two feedforward current commands and output them to the corresponding current loop closed-loop control module before the speed drops or the angle deviation increases, so that the male and female rotors maintain a preset speed fluctuation rate in the load peak area;

[0097] The purpose of this step is to directly map the spatial position to the current input, thereby achieving advance compensation for known periodic loads and reducing phase deviation caused by feedback lag.

[0098] In a preferred embodiment of the present invention, the process of calculating the synchronization error by the cross-coupling control module includes: obtaining the first number of pole pairs of the first motor and the second number of pole pairs of the second motor; dividing the first number of pole pairs by the second number of pole pairs to obtain the virtual electronic gear ratio; and multiplying the absolute spatial angle of the rotor of the second motor by the virtual electronic gear ratio to obtain the converted spatial angle.

[0099] The synchronization error is obtained by subtracting the calculated spatial angle from the absolute spatial angle of the rotor of the first motor.

[0100] This embodiment provides a synchronization error calculation mechanism; specifically, in the gearless dual-motor direct drive structure, the spatial matching relationship of the two compression rotors does not depend on physical gears, but rather establishes a calculable angle conversion relationship through a virtual electronic gear ratio, and then the main control chip calculates the synchronization error in real time;

[0101] Specifically, if we only directly compare the original mechanical angles of the two motors without considering the number of pole pairs and the corresponding spatial mapping relationship, the normal proportional relationship will be misjudged as a loss of step under different motor parameters.

[0102] Therefore, in this embodiment, the number of the first pole pairs of the first motor and the number of the second pole pairs of the second motor are read first to obtain the virtual electronic gear ratio; this ratio is used to convert the absolute spatial angle of the rotor of the second motor to the same reference domain as the first motor, and then subtract it from the current angle of the first motor to obtain the synchronization error;

[0103] To further explain, in order to ensure that the above conversion relationship holds in a physical sense, the absolute spatial angle of the rotor in this embodiment is defined by a unified reference domain before entering the virtual electronic gear ratio calculation; preferably, the encoder provides the mechanical absolute angle of its own rotor, and the main control chip then converts it into a unified spatial angle representation value for control comparison according to its own pole pair number.

[0104] Alternatively, the acquired quantity can be directly defined as a spatial angle quantity that has undergone same-domain conversion in the system parameter configuration; the rotor absolute spatial angle in the calculation context of this embodiment is not an original angle value with arbitrary meaning, but a same spatial reference quantity that has met the comparability requirements;

[0105] In this way, when the first pole pair and the second pole pair are subsequently introduced to form a virtual electronic gear ratio, what is actually done is the proportional conversion of the two motors under a unified control reference domain, rather than directly multiplying and comparing the original mechanical angles that are not comparable to each other.

[0106] In the simplified operating condition simulation, assuming the first motor has 4 pole pairs and the second motor has 2 pole pairs, the virtual electronic gear ratio is 2. If the current absolute spatial angle of the first motor is 150° and the absolute spatial angle of the second motor is 74.5°, then the converted spatial angle of the second motor is 149°, and the synchronization error is 1°. If the error is positive, it means that the phase of the first side is slightly ahead; if it is negative, it means that the phase of the second side is more ahead after conversion. The main control chip sends this error to the compensation stage, instead of simply using one as the absolute spindle and the other as a passive follower.

[0107] Furthermore, the calculation of synchronization error can be performed using angle modulo operation to ensure that the result always falls within a preset range, such as -180° to 180°. In this way, when the angle crosses the 0° or 360° boundary, there will be no false large error due to the apparent numerical jump. For example, when the angle of the first motor is 1° and the equivalent angle of the second motor is 359°, the actual error should be processed as 2° instead of -358°.

[0108] Furthermore, in the application of twin-screw air compressors, if the geometric meshing relationship of the two compression rotors also includes the initial phase offset during assembly, a pre-calibrated phase reference offset value can be superimposed after the spatial angle is calculated, and then compared with the angle of the first motor.

[0109] This phase reference offset value belongs to the initial phase parameter of the system and is used to characterize the target phase difference that should be maintained during normal synchronous operation after the equipment is assembled. With this method, the main control chip compares the deviation between the current relative phase and the target relative phase, thereby avoiding misjudging the fixed assembly offset as a continuous synchronization error.

[0110] Under abnormal operating conditions, if the pole-log parameter is misconfigured after maintenance, the main control chip can detect the long-term mismatch between the mapping table and the synchronization error through parameter self-check, and then enter the prohibited loading state; if the angle jump variable is greater than the set allowable range of the mechanical system within a certain period, it is considered a sampling anomaly and will not be used immediately for compensation calculation, but will be replaced by the predicted value of the adjacent period; if the conversion error is close to the boundary value, it will first be folded back and normalized, and then sent to the subsequent compensation function to avoid direction reversal at the boundary point;

[0111] For example, when the twin-screw air compressor is running at high speed, even if the male and female rotors have different electrical angle change rates due to differences in motor structure, the main control chip can still use a virtual electronic gear ratio to unify the two to the same comparison basis; in this way, when a phase lag occurs on one side due to an increase in local compression resistance, the system can identify the real phase deviation and make corrections.

[0112] The purpose of this step is to establish a quantifiable spatial ratio between the two motors, thereby achieving equivalent electronic meshing synchronization under conditions without mechanical gears.

[0113] In a preferred embodiment of the present invention, the process of the cross-coupling control module obtaining the adaptive weight coefficient includes: calculating the rotor angular velocity based on the rotor absolute spatial angle, and combining it with the preset first moment of inertia of the first motor and the second moment of inertia of the second motor;

[0114] Calculate the ratio of the first rotor kinetic energy to the second rotor kinetic energy to obtain the kinetic energy ratio; perform weight allocation based on the kinetic energy ratio to generate a first adaptive weight coefficient and a second adaptive weight coefficient.

[0115] This embodiment provides an adaptive weighting coefficient generation mechanism. Specifically, under actual air compressor operating conditions, although the two motors drive the same load together, their instantaneous speed, inertia, and energy state are not always symmetrical. If the compensation is always allocated according to a fixed ratio, it is easy to generate unnecessary current surges on one side. Therefore, this embodiment dynamically calculates the weighting coefficients based on the rotor kinetic energy state of the two motors.

[0116] Specifically, the main control chip calculates the angular velocity based on the change in the absolute spatial angle of the rotor over multiple consecutive control cycles; for example, if the angle of the first motor increases from 100.0° to 101.8° in two adjacent cycles, and the sampling period is 1ms, the corresponding angular velocity can be calculated.

[0117] The second motor obtains its angular velocity in the same way; combined with the preset motor rotational inertia, the rotor kinetic energy of the two motors is calculated respectively; the side with larger kinetic energy should not be subjected to too drastic correction to avoid causing additional oscillations; the side with smaller kinetic energy can undertake more fine-tuning tasks; the main control chip obtains the corresponding first adaptive weighting coefficient and second adaptive weighting coefficient according to the ratio of the kinetic energy of the two sides.

[0118] To further explain, this embodiment preferably uses the reverse sharing principle to generate weights, that is, the side with greater kinetic energy is assigned a smaller weight, and the side with less kinetic energy is assigned a larger weight;

[0119] In practice, the first adaptive weight coefficient can be determined according to the proportion of the second rotor kinetic energy to the total kinetic energy of the two rotors, and the second adaptive weight coefficient can be determined according to the proportion of the first rotor kinetic energy to the total kinetic energy of the two rotors. After processing in this way, the two weights are naturally complementary. The technical meaning is that the side with larger moment of inertia and kinetic energy is allocated a smaller compensation amount, and the side with smaller moment of inertia and kinetic energy is allocated a larger compensation amount, so that the compensation control strategy adapts to the physical inertia distribution of the system.

[0120] Let's take a simplified working condition as an example. Assume the moment of inertia of the first motor is 0.02 and the moment of inertia of the second motor is 0.01. At a certain moment, the angular velocity of the first motor is 100 after conversion, and the angular velocity of the second motor is 80. Then, the kinetic energies on both sides can be compared by multiplying the moment of inertia by half the square of the angular velocity. We get that the first side is about 100 and the second side is about 32. At this time, the kinetic energy ratio is about 100:32.

[0121] To avoid applying excessive perturbation to the high-energy side, a smaller first weight and a larger second weight can be assigned, for example, 0.24 and 0.76, so that more compensation tasks fall on the low-energy side. The above example values ​​correspond to the aforementioned reverse sharing principle, where the first side has higher kinetic energy, so the first adaptive weight coefficient is smaller; the second side has lower kinetic energy, so the second adaptive weight coefficient is larger.

[0122] Compared with the fixed 50%:50% compensation method, this method of distributing according to kinetic energy state can reduce the reverse torque impact on the high energy side, and is particularly suitable for working conditions with rapid load fluctuations and small mechanical clearance between the two rotors.

[0123] Under abnormal operating conditions, if the calculated angular velocity on one side is abnormally close to zero while the other side is still in normal operation, it indicates that there may be a stall or measurement error. In this case, the allocation will no longer be based on the current kinetic energy ratio, but will switch to conservative weighting. If the kinetic energy on both sides is lower than the preset lower threshold, for example, if the equipment is on the verge of stopping, the default neutral weighting can be used to avoid numerical amplification caused by the divisor being lower than the minimum allowable overflow threshold. If the calculated weights deviate from 1, normalization processing will be performed to ensure that the sum of the two compensation amounts is controlled.

[0124] For example, when an air compressor suddenly switches from a medium load to a high load state, the male rotor side may maintain higher kinetic energy due to its larger inertia, while the rotational speed of the female rotor side is affected by the compression pulsation and decreases at a greater rate. In this embodiment, the compensation ratio of the motor corresponding to the female rotor is automatically increased, and the reverse torque effect on the large inertia system of the male rotor is reduced, so that the synchronous recovery process meets the preset smoothness.

[0125] The purpose of this step is to dynamically allocate compensation tasks based on the real-time energy status of the two motors, thereby achieving a synchronous correction process that is more in line with physical inertia.

[0126] In a preferred embodiment of the present invention, the process by which the cross-coupling control module converts the synchronization error into a first compensation current and a second compensation current includes: inputting the synchronization error into a preset error-torque compensation function to generate a basic compensation torque; and multiplying the basic compensation torque by the first adaptive weighting coefficient to obtain the first compensation torque.

[0127] The basic compensation torque is multiplied by the second adaptive weighting coefficient to obtain the second compensation torque; the first compensation torque is converted into the first compensation current; and the second compensation torque is converted into the second compensation current.

[0128] This embodiment provides a conversion mechanism from synchronization error to compensation current. Specifically, after obtaining the synchronization error and two adaptive weights, the main control chip first uses a preset compensation function to generate a basic compensation torque, then allocates it according to the weights as a first compensation torque and a second compensation torque, and finally converts it into two compensation currents respectively.

[0129] In detail, knowing only the magnitude of the synchronization error is insufficient to directly drive the inverter, because the current loop actually receives the current command rather than the angle command. Therefore, this embodiment introduces a compensation function to achieve the mapping from the error domain to the torque domain. The compensation function can be a linear function, a piecewise linear function, or a nonlinear function with limiting. For example, when the synchronization error is small, a high-sensitivity slope is used to quickly eliminate small deviations; when the synchronization error is large, it enters the limiting region to prevent the current command from being too large instantaneously.

[0130] For example, if the synchronization error is 0.5°, and the compensation function is set to output a basic compensation torque of 2 N·m for every 1° error, then the basic compensation torque is 1 N·m; if the first adaptive weight coefficient is 0.3 and the second adaptive weight coefficient is 0.7 obtained from the previous step, then the first compensation torque is 0.3 N·m and the second compensation torque is 0.7 N·m.

[0131] Based on the torque-current conversion relationship, if 1 N·m of the first motor corresponds to 1 A, then the first compensation current is 0.3 A; if 1 N·m of the second motor corresponds to 0.8 A, then the second compensation current is 0.875 A; thus, the conversion from spatial error to two-way current compensation is completed.

[0132] Furthermore, in order to maintain consistency with the differential fusion logic in the subsequent synchronous execution module, the first compensation current and the second compensation current output by the cross-coupling control module in this embodiment are preferably expressed as compensation amplitudes corresponding to their respective compensation torques.

[0133] The final direction of application is determined by the synchronous error sign during the synchronous execution phase; that is, when the synchronous error indicates that the first motor is ahead, the first compensation current in the synchronous execution module is a subtraction term of the first feedforward current command, and the second compensation current is an addition term of the second feedforward current command; when the synchronous error direction is reversed, the action directions of the two compensations are synchronously reversed.

[0134] By determining the compensation amplitude in this step and then determining the action polarity in subsequent steps, the problem of duplicate assignment or inconsistent symbol conventions between the cross-coupling control module and the synchronous execution module can be avoided.

[0135] Furthermore, the compensation function can be configured with a dead zone, for example, the output is zero when the absolute value of the synchronization error is less than 0.05°, in order to reduce frequent current swings caused by encoder quantization noise; it can also be configured with a slope switching range, for example, a larger slope is used from 0.05° to 0.5°, and a limiting process is used after exceeding 0.5°, in order to balance sensitivity and safety.

[0136] In an abnormal protection mechanism, if any result of the weighted distribution of the basic compensation torque exceeds the upper limit of the corresponding motor's allowable torque, the upper limit is limited, and the remaining part is no longer forcibly transferred to the other side to avoid total compensation overshoot; if the torque current conversion coefficient on one side changes due to field weakening conditions, the compensation current is recalculated using the current operating parameters; if a sudden jump occurs in the compensation function input, a change rate limit is added to the output compensation torque to prevent the inverter from being subjected to an excessively steep command step.

[0137] For example, if the female rotor is detected to lag behind the male rotor by 0.5° after conversion near the angle zone where the exhaust pulsation of the twin-screw air compressor is strongest, the main control chip first converts the deviation into the basic compensation torque, and then allocates different sizes of compensation current according to the current kinetic energy of both sides, so that the side that is more suitable to undertake the adjustment task can undertake more correction.

[0138] The purpose of this step is to convert the synchronization error into a compensation command that can be directly executed by the current loop, thereby achieving a closed-loop connection from spatial deviation to electromagnetic correction.

[0139] In a preferred embodiment of the present invention, the process of the synchronous execution module generating the first current command and the second current command includes: subtracting the first compensation current from the first feedforward current command to generate the first current command; and adding the second compensation current, which has the opposite polarity to the first compensation current, to the second feedforward current command to generate the second current command.

[0140] This embodiment provides a current command fusion execution mechanism; specifically, after both the feedforward current and the compensation current have been obtained, the main control chip does not simply superimpose the two compensations, but generates the final first current command and the second current command according to the coupling relationship of opposite polarities, so as to ensure that the phase error is differentially corrected.

[0141] Specifically, if the synchronization error indicates that the first motor is slightly ahead of the second motor after conversion, the correction strategy should be to moderately reduce the torque output of the first motor and moderately increase the torque output of the second motor. Based on this, this embodiment subtracts the first compensation current from the first feedforward current command to generate the first current command; at the same time, the second compensation current with the opposite polarity to the first compensation current is added to the second feedforward current command to generate the second current command. The opposite polarity represents the differential mode adjustment characteristic in the differential correction logic, that is, the output directions on both sides are different.

[0142] To further explain, in order to avoid the second compensation current being misunderstood as an independent addend that is always positive, in this embodiment, both the first compensation current and the second compensation current are preferably calculated according to the compensation amplitude, and their final direction of action is uniformly determined by the sign of the synchronization error; when it is determined that the first motor is ahead, the first compensation current corresponds to the direction of reducing the current command on the first side, and the second compensation current corresponds to the direction of increasing the current command on the second side.

[0143] When it is determined that the second motor is ahead, the above directions are reversed as a whole; in other words, the second compensation current with the opposite polarity of the first compensation current emphasizes that the second compensation is opposite to the first compensation in the direction of action, rather than requiring the first compensation current and the second compensation current to use the same symbol convention in their numerical expressions.

[0144] For example, suppose that in a certain cycle, the first feedforward current command is 4A and the second feedforward current command is 4A; after the synchronization error is calculated, the first compensation current is 0.3A and the second compensation current is 0.7A, and it is currently determined that the first motor is ahead;

[0145] The first current command is 4A-0.3A=3.7A, and the second current command is 4A+0.7A=4.7A. If the direction of the synchronization error reverses in the next cycle, it means that the second motor is relatively ahead. Then the polarity of the two compensation paths will also reverse synchronously. In the end, the first current command may be higher than the feedforward value and the second current command may be lower than the feedforward value.

[0146] The core of this fusion method is that the feedforward part is used to bear the common load, and the compensation part is used to redistribute the instantaneous torque share between the two motors; in this way, the basic current level required by the total load will not be destroyed, and the synchronization relationship can be quickly restored through differential correction.

[0147] Furthermore, the main control chip can first limit the amplitude and constrain the rate of change of the feedforward component and the compensation component respectively before the final synthesis is performed. The technical advantages of this processing method are: the feedforward link prioritizes ensuring the continuity of load drive, and the compensation link prioritizes ensuring the smoothness of synchronous recovery. Even when there is a large feedforward demand in the high-pressure exhaust area, the current loop will not be saturated due to the instantaneous surge of the compensation component.

[0148] In an abnormal protection mechanism, if the final current command of any of the merged paths exceeds the rated upper limit of the inverter, it will be limited to the maximum allowable value, and the other side will be synchronously limited to prevent excessive imbalance in the total torque distribution.

[0149] If the final current command of a certain path is reduced to a negative value due to compensation, and the current operating condition does not allow reverse electromagnetic braking, then it is clamped to the minimum allowable value; if the feedforward currents on both sides are already close to the limit, then the compensation component is appropriately compressed to ensure the continuity of the basic load drive.

[0150] For example, when the twin-screw air compressor approaches the high-pressure exhaust zone, both motors require a large feedforward current to maintain normal compression. If the male rotor is detected to be slightly ahead at this time, this embodiment will not simply increase the total current on both sides, but will slightly reduce the current of the motor corresponding to the male rotor and increase the current setpoint of the motor corresponding to the female rotor, thereby restoring the spatial phase between the rotors while basically satisfying the total load.

[0151] The purpose of this step is to decouple and integrate the feedforward load bearing and the synchronization error correction, so as to achieve the synchronous completion of total torque supply and phase correction.

[0152] In a preferred embodiment of the present invention, before the cross-coupling control module converts the first compensation current and the second compensation current, it further includes a condition determination process for synchronization error: setting a synchronization error threshold; determining whether the absolute value of the synchronization error is greater than or equal to the synchronization error threshold;

[0153] If the absolute value of the synchronization error is greater than or equal to the synchronization error threshold, the calculation process of the adaptive weight coefficient is initiated and the first compensation current and the second compensation current are generated; if the absolute value of the synchronization error is less than the synchronization error threshold, the first compensation current and the second compensation current are assigned to zero.

[0154] This embodiment provides a compensation enable mechanism with threshold determination; specifically, in dual-motor synchronous control, if cross-coupling compensation is triggered immediately for all minor errors, encoder quantization noise, current sampling noise and mechanical micro-vibration will be amplified into frequent compensation current jitter; therefore, this embodiment first performs condition determination on the synchronization error before generating the compensation current.

[0155] Specifically, the main control chip presets a synchronization error threshold, for example, 0.1°. In each control cycle, it first calculates the absolute value of the synchronization error and then compares it with the threshold. If the absolute value of the synchronization error is greater than or equal to the threshold, it is considered that the current deviation has exceeded the natural fluctuation range. The main control chip starts the adaptive weight calculation process and further generates the first compensation current and the second compensation current.

[0156] If the absolute value of the synchronization error is less than the threshold, the two compensation currents are directly set to zero, and only the spatial domain feedforward and current closed-loop operation are retained.

[0157] For example, if the current synchronization error is 0.04° and the threshold is 0.1°, the system determines that the deviation does not need to enter cross compensation, and the first compensation current and the second compensation current are both 0A.

[0158] If the synchronization error rises to 0.18° in the next cycle, the system will start calculating the kinetic energy ratio and weight, and output two compensation currents, such as 0.2A and 0.5A. This can prevent the compensation module from frequently operating within the error range below the preset threshold.

[0159] Furthermore, a dual-threshold strategy with hysteresis can be set; for example, the compensation start threshold is 0.1° and the exit threshold is 0.06°; in this way, when the error fluctuates slightly near the critical point, the phenomenon of repeated start and stop compensation will not occur.

[0160] Although this embodiment can complete the basic function using a single threshold, adding hysteresis in high-speed scenarios helps to further smooth the system response;

[0161] In an anomaly protection mechanism, if the synchronization error remains within the lower limit tolerance range of the preset threshold for a long period of time but shows a continuous increasing trend, the main control chip can statistically analyze the error change rate over multiple consecutive cycles and lower the threshold or issue an early warning when necessary. If the threshold setting is too small, causing compensation to be triggered frequently, it can be recalibrated after being identified through the operation log.

[0162] If the threshold is set too high and the actual deviation is not processed in time, it can be readjusted during the equipment commissioning stage based on the rotor safety clearance and allowable electrical angle deviation. In the event of a sudden large error, such as an absolute value greater than or equal to the set safety limit, ordinary compensation will no longer be performed, but the torque limiting or shutdown protection process will be directly initiated.

[0163] For example, during stable loading operation of a twin-screw air compressor, there may always be angular jitter between the male and female rotors with an amplitude less than the set allowable range. By introducing an error threshold, the system can regard these minor fluctuations that do not affect the safety of mechanical fit as normal disturbances and not generate additional compensation current. The cross-coupling compensation link is only activated when compression pulsation, lubrication changes, or processing errors cause the phase deviation to actually increase.

[0164] The purpose of this step is to filter out small errors in the noise and quantization levels, thereby enabling on-demand triggering of compensation actions, reducing high-frequency current jitter, and improving system stability.

[0165] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A synchronous control system for a gearless dual permanent magnet motor, used to control a first motor and a second motor that jointly drive the same load, characterized in that, It includes a main control chip, which is communicatively connected to a data acquisition module, a vector analysis module, a spatial domain feedforward observation module, a cross-coupling control module, and a synchronous execution module; The data acquisition module is used to acquire the absolute spatial angles and phase currents of the rotors of the first motor and the second motor; the vector analysis module is used to perform coordinate transformation analysis on the phase currents to acquire the current quadrature-axis current and the current direct-axis current. The spatial domain feedforward observation module is used to obtain the expected load torque according to the rotor absolute spatial angle and the preset spatial mapping table, and convert it into a first feedforward current command corresponding to the first motor and a second feedforward current command corresponding to the second motor, respectively. The cross-coupling control module is used to calculate the synchronization error based on the absolute spatial angle of the rotor, and combine the adaptive weighting coefficient calculated in real time based on the motor operating status to convert the synchronization error into a first compensation current and a second compensation current. The synchronous execution module is used to fuse the first feedforward current command with the first compensation current, and the second feedforward current command with the second compensation current to generate a first current command and a second current command, respectively, and generate a voltage command with the feedback phase current closed-loop adjustment, generate a drive signal through space vector pulse width modulation, and control the inverter that is electrically connected to both the first motor and the second motor to perform synchronous drive operation on the two motors. The preset process for generating the spatial mapping table includes: The first motor and the second motor are controlled to rotate at a preset calibrated speed; the cross-axis current of the first motor and the second motor is collected; the cross-axis current is integrated to calibrate the load torque curve; The load torque curve is bound to the rotor absolute spatial angle to generate the spatial mapping table and write it into non-volatile memory; The process by which the cross-coupling control module obtains the adaptive weighting coefficients includes: calculating the rotor angular velocity based on the rotor's absolute spatial angle, and combining this with a preset first moment of inertia of the first motor and a second moment of inertia of the second motor; calculating the first rotor kinetic energy based on the rotor angular velocity and the first moment of inertia, and calculating the second rotor kinetic energy based on the rotor angular velocity and the second moment of inertia; Calculate the ratio of the first rotor kinetic energy to the second rotor kinetic energy to obtain the kinetic energy ratio; perform weight allocation based on the kinetic energy ratio to generate a first adaptive weight coefficient and a second adaptive weight coefficient.

2. The synchronous control system for the gearless dual permanent magnet motor according to claim 1, characterized in that, The process by which the data acquisition module acquires the absolute spatial angles and phase currents of the rotors of the first motor and the second motor includes: The absolute spatial angles of the rotors of the first motor and the second motor are obtained by absolute encoders installed on the first motor and the second motor; the phase currents of the first motor and the second motor are obtained by current sensors installed on the inverter bridge arm.

3. The synchronous control system for the gearless dual permanent magnet motor according to claim 1, characterized in that, The process by which the spatial domain feedforward observation module acquires the expected load torque and converts the expected load torque into the first feedforward current command and the second feedforward current command respectively includes: Using the absolute spatial angle of the rotor as the index address, address matching is performed in the spatial mapping table; Extract the load torque value corresponding to the address matching as the expected load torque; calculate the corresponding first feedforward current command and second feedforward current command according to the expected load torque and the electromagnetic torque conversion relationship of the corresponding motor obtained in advance.

4. The synchronous control system for the gearless dual permanent magnet motor according to claim 1, characterized in that, The process by which the cross-coupling control module calculates the synchronization error includes: Obtain the number of first pole pairs of the first motor and the number of second pole pairs of the second motor; divide the number of first pole pairs by the number of second pole pairs to obtain the virtual electronic gear ratio; Multiply the absolute spatial angle of the rotor of the second motor by the virtual electronic gear ratio to obtain the converted spatial angle; The synchronization error is obtained by subtracting the calculated spatial angle from the absolute spatial angle of the rotor of the first motor.

5. The synchronous control system for the gearless dual permanent magnet motor according to claim 1, characterized in that, The process by which the cross-coupling control module converts the synchronization error into the first compensation current and the second compensation current includes: inputting the synchronization error into a preset error-torque compensation function to generate a basic compensation torque; and multiplying the basic compensation torque by the first adaptive weighting coefficient to obtain the first compensation torque. The basic compensation torque is multiplied by the second adaptive weighting coefficient to obtain the second compensation torque; The first compensation torque is converted into the first compensation current; the second compensation torque is converted into the second compensation current.

6. The synchronous control system for the gearless dual permanent magnet motor according to claim 5, characterized in that, The process by which the synchronous execution module generates the first current command and the second current command includes: Subtract the first compensation current from the first feedforward current command to generate the first current command; The second current command is generated by adding the second feedforward current command to the second compensation current, which has the opposite polarity to the first compensation current.

7. The synchronous control system for the gearless dual permanent magnet motor according to claim 1, characterized in that, Before converting the first compensation current and the second compensation current, the cross-coupling control module further includes a condition determination process for the synchronization error: setting a synchronization error threshold; determining whether the absolute value of the synchronization error is greater than or equal to the synchronization error threshold; If the absolute value of the synchronization error is greater than or equal to the synchronization error threshold, then the calculation process of the adaptive weight coefficient is initiated and the first compensation current and the second compensation current are generated. If the absolute value of the synchronization error is less than the synchronization error threshold, then the first compensation current and the second compensation current are set to zero.

Images