A control method for an induction motor drive system based on a current source inverter

Abstract

The present invention discloses a control method for an induction motor drive system based on a current source inverter, and belongs to the technical field of motor control. The drive system is composed of a two-stage current source inverter and a voltage source inverter, and is provided with a position observer module. The two-stage current source inverter includes two thyristor-based current source inverters, which are 30° out of phase with each other and are used to provide active power to the motor; the voltage source inverter is a three-phase bridge inverter based on an insulated gate bipolar transistor, which is powered by a pre-charged capacitor for reactive power compensation; the position observer module obtains the motor speed based on a positionless sensor. The control method includes a control method for a two-stage current source inverter and a control method for a voltage source inverter, and the two are performed synchronously. While realizing the driving of an induction motor by a current source inverter, the present invention effectively reduces the size, weight and cost of the system at the topological level, and effectively reduces the output current harmonics of the current source inverter.

Description

Technical Field

[0001] The present invention relates to the technical field of motor control, and in particular to a control method for an induction motor drive system based on a current source inverter. Background Art

[0002] In the field of low-altitude economy, the development of new aircraft such as drones and electric vertical take-off and landing aircraft has attracted increasing attention. For these new aircraft, the performance of the electric drive system is one of the key factors that determine their performance. New aircraft have extremely high performance requirements for the electric drive system, especially in terms of efficiency, dynamic response, safety and reliability.

[0003] In the early days, due to the low efficiency of inductor energy storage and the low frequency of switching tube devices, the inductance value of the current source inverter was relatively large and the efficiency of the entire system was low, and the electric drive system of the motor mainly used a voltage source inverter. However, with the development of high-temperature superconducting materials, the efficiency of inductor energy storage has increased. In particular, with the rapid development of wide-bandgap semiconductor devices SiC and GaN in recent years, the selected value of inductance has decreased, and the motor drive system based on the current source inverter has become a new research hotspot. Compared with the use of voltage source inverters to drive motors, the use of current source inverters has the advantages of faster dynamic response, stronger current limiting capability, higher short-circuit protection reliability, and less electromagnetic interference. In addition, it has regenerative braking and four-quadrant operation capabilities, and is more suitable for wide-range torque and speed control, which can better meet the complex and changeable power requirements of new aircraft. Due to its simple structure, strong reliability, large starting torque, and good speed regulation performance, the induction motor has a huge application space in the electric drive of new aircraft.

[0004] However, as far as the prior art is concerned, there are still certain defects in driving induction motors with current source inverters. First, due to factors such as the stacking effect, the current source inverter will generate large harmonics when outputting current, which may affect the operating stability of the motor, resulting in increased motor heating, decreased efficiency, and even affecting the service life of the motor. Secondly, when the current source inverter drives the induction motor, its power factor may be affected to a certain extent. Thyristor (SCR) is currently the most preferred choice for current source inverters due to its ruggedness and availability under high voltage and high current ratings. However, since SCR is a semi-controlled device, in most applications, its shutdown requires a forced commutation circuit. When operating at an advanced power factor, the SCR in the current source inverter can be turned off by load commutation. For synchronous motors, load switching can be achieved by operating the motor at an advanced power factor through overexcitation. However, for induction motors, due to the limitations of their working principles and operating characteristics, they usually do not have the ability to operate at an advanced power factor. Therefore, the current source inverter based on SCR load rectification cannot be directly used to drive induction motors. In this regard, current scholars mostly use hybrid technology to realize current source inverter driving induction motor, but there are still certain problems in terms of system size, weight and cost, which is not suitable for application scenarios of new aircraft. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes a control method for an induction motor drive system based on a current source inverter, which can effectively reduce the size, weight and cost of the system at the topological level while realizing the driving of the induction motor by the current source inverter, and effectively reduce the output current harmonics of the current source inverter.

[0006] The technical solution to achieve the purpose of the present invention is:

[0007] A control method for an induction motor drive system based on a current source inverter, wherein the drive system is composed of a two-stage current source inverter at one end of the stator winding of the induction motor and a voltage source inverter at the other end of the stator winding, and is provided with a position observer module;

[0008] The control method includes a control method of the two-stage current source inverter and a control method of the voltage source inverter;

[0009] The control method of the two-stage current source inverter comprises the following steps:

[0010] Step S1-1, obtain the motor speed observation value from the position observer module Input speed controller, and motor reference speed ω ref By comparison, we can get the speed difference ω slip ;

[0011] Step S1-2: according to the speed difference ω slip , obtain the DC reference current i from the current reference module dref ;

[0012] Step S1-3: Obtain the DC current i input to CSI-A and CSI-B d1 、i d2 , input current controller, the current controller adjusts the trigger angle of the two front controlled rectifiers to convert the DC current i d1 、i d2 Maintain at reference value

[0013] Step S1-4: According to the motor speed observation value and the speed difference ω obtained in step S1-1 slip , obtain the frequency information of the two-stage current source inverter, input it into the frequency control module, and obtain the gating pulse signals of CSI-A and CSI-B;

[0014] The control of the voltage source inverter is performed synchronously with the control of the two-stage current source inverter, and includes the following steps:

[0015] Step S2-1: Based on the strobe pulse signal of CSI-A and the DC current i d1 , determine the fundamental component i of the three-phase current of the induction motor maf 、i mbf 、i mcf ;

[0016] Step S2-2: The fundamental wave component i of the three-phase current of the induction motor maf 、i mbf 、i mcf Input coordinate transformation module, perform Clark transformation, convert to αβ coordinate system, and output i mfα 、i mfβ , used for unit vector generation;

[0017] Step S2-3, obtaining the voltage v of each phase terminal of the two-stage current source inverter sa 、v sb 、v sc , input the coordinate transformation module, and input the unit vector generated in step S2-2, and convert v sa 、v sb 、v sc Perform Clark transform and Park transform, convert to dq coordinate system, and output V sd 、V sq ;

[0018] Step S2-4: convert the d-axis component of the voltage V sdInput reference module, output double-stage current source inverter voltage q-axis component reference value V sqref ;

[0019] Step S2-5: V obtained in step S2-3 sq and V obtained in step S2-4 sqref Input Q-axis controller, output voltage source inverter voltage q-axis component V vq ;

[0020] Step S2-6, obtaining the voltage source inverter capacitor voltage V c , input to the D-axis controller, and the reference capacitor voltage V cref In contrast, the output voltage source inverter voltage d-axis component V vd ;

[0021] Step S2-7: Substitute the unit vector obtained in step S2-2 and the V obtained in step S2-5 vq and V obtained in step S2-6 vd Input the coordinate inverse transformation module, perform Park inverse transformation and Clark inverse transformation, and output the voltage source inverter three-phase modulation signal V vma 、V vmb 、V vmc ;

[0022] Step S2-8: V vma 、V vmb 、V vmc Input pulse module, output gate pulse signal of voltage source inverter IGBT.

[0023] The two-stage current source inverter includes two SCR-based current source inverters: CSI-A and CSI-B, wherein the SCR is a thyristor; the CSI-A and CSI-B are connected in parallel, both in series with a SCR-based controllable rectifier, and in series with a DC link inductor, and are powered by two 50Hz AC power supplies respectively; the CSI-A and CSI-B both operate in a 120° conduction mode, with a 30° phase difference between the two, and a quasi-square wave output current is generated in each inverter, and the output current has a multi-level distribution, realizing multi-level control, and is used to provide active power to the motor. The voltage source inverter is a three-phase bridge inverter based on IGBT, powered by a pre-charged capacitor, and is only used to provide reactive power; the IGBT is an insulated gate bipolar transistor. The position observer module is based on a positionless sensor, and uses a combination of a multilayer perceptron neural network and a sliding mode observation to obtain the motor speed for closed-loop control.

[0024] Furthermore, during the normal operation of the motor, the power factor of the two-stage current source inverter is positive, and the fundamental component i of the induction motor current is mfLeading two-stage current source inverter terminal voltage v s , the leading phase angle is β; the voltage source inverter is used for reactive power compensation to ensure that the leading phase angle β of the two-stage current source inverter remains unchanged; the phasor of the motor current fundamental component I mf Lagging behind the motor voltage phasor V m , the power factor angle is The phasor I of the fundamental component of the CSI-B current s2f The phase I that lags behind the fundamental component of the CSI-A current s1f , the lagging phase angle is 30°; the phasor V of the terminal voltage of the two-stage current source inverter s The motor voltage phasor V m 、Voltage phasor of voltage source inverter V v The following relationship exists:

[0025]

[0026] The leading phase angle β is determined according to system requirements, and the following relationship exists:

[0027] β=(15+γ)°

[0028] The selection of the commutation angle γ takes into account the SCR turn-off time, commutation inductance and DC current.

[0029] Furthermore, the induction motor driven by the drive system has regeneration capability; when the motor speed drops to a lower value or during speed reversal, the regeneration control is activated, and the induction motor acts as a generator to perform external work; during the regeneration operation, the motor voltage phasor V m and the phase of the motor current fundamental component I mf The phase angle between is the power factor angle of the induction motor under normal operating conditions, I mf The phase quantity V that leads the voltage at the two-stage current source inverter s , the leading phase angle is (π-β), β is V under normal operating conditions of the induction motor m with I mf The phase angle between .

[0030] Further, in the control method of the two-stage current source inverter, the position observer module in step S1-1 is composed of an observation module and a neural network speed estimation module;

[0031] The observation module adopts a discrete second-order sliding mode observer, which is composed of a Clark transformation module, a current observer, a sliding mode surface, a sliding mode observation module, and a rotor flux observation module. The construction steps are as follows:

[0032] Step S1-1-A1, establish a Clark transformation module to transform the stator current and voltage parameters of the induction motor from the natural coordinate system to the αβ coordinate system. The mathematical model of the Clark transformation module is as follows:

[0033] [i mα i mβ i0] T =T 3s / 2s [i ma i mb i mc ] T

[0034] [V mα V mβ V0] T =T 3s / 2s [V ma V mb V mc ] T

[0035] Among them, T 3s / 2s is the Clark transformation matrix, i ma 、i mb 、i mc are the axis components of the induction motor stator current in the natural coordinate system, i mα 、i mβ are the axis components of the induction motor stator current in the αβ coordinate system, i0 is the zero-sequence current; V ma 、V mb 、V mc are the axis components of the induction motor stator voltage in the natural coordinate system, V mα 、V mβ They are the axis components of the induction motor stator voltage in the αβ coordinate system, V0 is the zero-sequence voltage;

[0036] Step S1-1-A2: Build the dynamic model of the induction motor as follows:

[0037]

[0038] Among them, ω r is the rotor speed of the induction motor, i mα 、i mβ are the axis components of the induction motor stator current in the αβ coordinate system, V mα 、V mβ are the axis components of the induction motor stator voltage in the αβ coordinate system, ψ a , β The components of the rotor flux of the induction motor in the αβ coordinate system are induced respectively; η is the rotor time constant, k1=k2R s , R r is the rotor resistance, L r is the rotor inductance, R s is the stator resistance, L s is the stator inductance, L m For mutual induction;

[0039] Step S1-1-A3: Extract the sliding mode function from the induction motor dynamic model as follows:

[0040]

[0041] Based on the above sliding mode function, the current observer model is designed as follows:

[0042]

[0043] in, and Observe the current for the stator;

[0044] Step S1-1-A4, define the sliding surface as follows:

[0045]

[0046] s α 、s β is the sliding surface function;

[0047] Design the Lyapunov candidate function as:

[0048]

[0049] According to Lyapunov stability theory, the candidate function satisfies the following condition a:

[0050]

[0051] Among them, Γ is a positive definite matrix, Γ1 and Γ2 are constants;

[0052] According to the above condition a, the candidate function satisfies the following condition b:

[0053]

[0054] According to condition b, the Lyapunov condition is defined as follows:

[0055]

[0056] Step S1-1-A5: Establish a sliding mode function T according to the current observer model described in step S1-1-A2 and the Lyapunov condition described in step S1-1-A3. α 、T β The observation model is as follows:

[0057]

[0058] in, T α 、T β Observed value of

[0059] According to the above observation model, the reverse Euler method is used to calculate and obtain its discrete form, and the sliding mode observation module is established as follows:

[0060]

[0061] Among them, T s is the system sampling time;

[0062] Step S1-1-A6, establish the rotor flux observation module as follows:

[0063]

[0064] in, is the rotor flux ψ a , β Observed value of

[0065] The neural network speed estimation module is constructed using a multilayer perceptron neural network, which includes an input layer, a hidden layer, and an output layer, and is composed of 4 input units, 20 neurons, 100 weighting factors, 21 biases, and an output unit; the input layer is used to inject feedback signals; the hidden layer processes the data and sends the results to the output layer; the four input features are the motor reference speed ω ref , Observed value of rotor flux And the stator current i of the α axis in the αβ coordinate system mα , the output feature is the motor speed observation value The neurons in the hidden layer are designed based on the tangent-sigmoid activation function;

[0066] The mathematical model of the neural network speed estimation module is as follows:

[0067]

[0068] Among them, μ i is the input feature vector, f j (μ i ,w i,j) is the activation function of the jth neuron in the hidden layer, w i,j is the jth weighting factor of the ith input, w k is the kth weighting factor of the output layer, δ j is the bias of the jth neuron in the hidden layer, is the bias for the neurons in the output layer;

[0069] The steps of estimating the speed of the induction motor using the position observer module include:

[0070] Step S1-1-B1, obtain the three-phase current i of the induction motor ma 、i mb 、i mc And the three-phase voltage V ma 、V mb 、V mc , input Clark transformation module, output stator current i in αβ coordinate system mα 、i mβ and stator voltage V mα 、V mβ ;

[0071] Step S1-1-B2: The stator current i in the αβ coordinate system obtained in step S1-1-B1 is mα 、i mβ and stator voltage V mα 、V mβ Input current observer, output stator observation current and

[0072] Step S1-1-B3: The stator observation current obtained in step S1-1-B2 is and Input sliding surface, output sliding surface function s α 、s β Enter the sliding mode observation module and establish T α 、T β Observations of

[0073] Step S1-1-B4: Input the current observer and repeat steps S1-1-B3. When the trajectory of the system follows the sliding surface, the stator current observer and The value of will converge to the actual stator current i mα 、i mβ , at this time T α 、T β Observed value Also with T α 、T β same;

[0074] Step S1-1-B5: The converged Input rotor flux observation module, output rotor flux observation value

[0075] Step S1-1-B6: The stator current i obtained in step S1-1-B1 is mα , rotor flux observation value obtained in step S1-1-B5 And the motor reference speed ω ref Input neural network speed estimation module, output motor speed observation value

[0076] Furthermore, in the control method of the two-stage current source inverter, in the frequency control module described in step S1-4, the gating pulse signals of CSI-A and CSI-B are given a phase difference of 30°.

[0077] Furthermore, in the control method of the voltage source inverter, when the induction motor is in normal operation, the reference module in step S2-4 is given by the following formula:

[0078] V sqref = -tan(β)×V sd

[0079] Where β is the fundamental component of the induction motor current i mf Relative to the voltage v of the two-stage current source inverter s The leading phase angle of

[0080] In the induction motor regenerative operation state, the reference module described in step S2-4 is given by the following formula:

[0081] V sqref =tan(β)×V sd

[0082] The Q-axis controller in step S2-5 and the D-axis controller in step S2-6 both use PI controllers.

[0083] Furthermore, the control method includes a pre-charging method for the voltage source inverter capacitor, wherein the pre-charging of the capacitor is achieved by triggering the SCR of the upper bridge arm and the lower bridge arm of any two phases of CSI-A and CSI-B, and there is no need to provide a trigger signal to the IGBT; the pre-charging method is: controlling the DC current to pass through the SCR of the upper bridge arm and the lower bridge arm of any two phases of CSI-A and CSI-B, the motor winding, and the anti-parallel diode of the IGBT to form a closed loop circuit to pre-charge the capacitor, and when the capacitor voltage V c Reach its reference value V crefThe drive system starts to work normally.

[0084] Compared with the prior art, the control method of the induction motor drive system based on the current source inverter described in the present invention has the following beneficial effects:

[0085] 1. The induction motor drive system proposed by the present invention adopts a voltage source inverter to compensate reactive power, thereby solving the problem that the current source inverter based on SCR load rectification cannot be directly used to drive the induction motor;

[0086] 2. The induction motor drive system proposed in the present invention can provide rated torque in the full speed range, and will not cause problems such as insufficient back electromotive force at low speeds leading to inverter commutation failures, and does not require a separate control strategy to start and run the motor at low speeds;

[0087] 3. Compared with the conventional induction motor drive topology structure that uses a mixture of current source inverter and voltage source inverter, in the induction motor drive system topology structure proposed by the present invention, the voltage source inverter does not require a separate interface inductor and DC power supply, which can effectively reduce the size, weight and cost of the system;

[0088] 4. The double-stage current source inverter proposed in the present invention is composed of two current source inverters with a phase difference of 30°, and can significantly reduce the fifth and seventh harmonics by outputting a multi-level current waveform;

[0089] 5. The control method of the induction motor drive system proposed in the present invention is a novel closed-loop control scheme, which includes the coordinated control of a two-stage current source inverter and a voltage source inverter, and can realize a smooth transition of the induction motor from a normal working mode to a regeneration mode and from a regeneration mode to a normal working mode;

[0090] 6. The present invention proposes a position observer module for induction motors, which obtains the motor speed for closed-loop control based on the position sensorless technology, thereby improving the robustness of the system and effectively reducing the system cost. The position observer module adopts a combination of machine learning and sliding mode observation, and introduces a neural network to solve the jitter problem of the traditional sliding mode observer, thereby avoiding the problems of limited observer bandwidth and weakened dynamic performance of the drive system caused by the traditional sliding mode observer introducing filtering to solve the jitter problem. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] Figure 1 A topological diagram of an induction motor drive system based on a current source inverter proposed by the present invention;

[0092] Figure 2 A schematic diagram of a multi-level current output by a double-stage current source inverter proposed in the present invention;

[0093] Figure 3 A control strategy diagram of a two-stage current source inverter proposed by the present invention;

[0094] Figure 4 A control strategy diagram of a voltage source inverter proposed by the present invention;

[0095] Figure 5 This is a structural block diagram of the position observer module proposed by the present invention;

[0096] Figure 6 A neural network configuration diagram of a neural network speed estimation module in a position observer module proposed by the present invention;

[0097] Figure 7 This is a schematic diagram of the capacitor pre-charging solution proposed by the present invention. DETAILED DESCRIPTION

[0098] The present invention is described below based on the embodiments, but the present invention is not limited to the specific embodiments. In the detailed description of the present invention below, some specific details are described in detail. For those skilled in the art, the present invention can be fully understood without the description of these details.

[0099] A control method for an induction motor drive system based on a current source inverter, wherein the drive system topology is as follows: Figure 1 As shown, it consists of a two-stage current source inverter at one end of the induction motor stator winding and a voltage source inverter at the other end of the stator winding, and is accompanied by a position observer module.

[0100] The control method includes a control method of the two-stage current source inverter and a control method of the voltage source inverter;

[0101] The control method of the two-stage current source inverter is as follows: Figure 3 As shown, it includes the following steps:

[0102] Step S1-1, obtain the motor speed observation value from the position observer module Input speed controller, and motor reference speed ω ref By comparison, we can get the speed difference ω slip ;

[0103] Step S1-2: according to the speed difference ω slip , obtain the DC reference current i from the current reference module dref ;

[0104] Step S1-3: Obtain the DC current i input to CSI-A and CSI-B d1 、i d2, input current controller, the current controller adjusts the trigger angle of the two front controlled rectifiers to convert the DC current i d1 、i d2 Maintain at reference value

[0105] Step S1-4: According to the motor speed observation value and the speed difference ω obtained in step S1-1 slip , obtain the frequency information of the two-stage current source inverter, input it into the frequency control module, and obtain the gating pulse signals of CSI-A and CSI-B;

[0106] The control of the voltage source inverter is performed synchronously with the control of the two-stage current source inverter, such as Figure 4 As shown, the following steps are included:

[0107] Step S2-1: Based on the strobe pulse signal of CSI-A and the DC current i d1 , determine the fundamental component i of the three-phase current of the induction motor maf 、i mbf 、i mcf ;

[0108] Step S2-2: The fundamental wave component i of the three-phase current of the induction motor maf 、i mbf 、i mcf Input coordinate transformation module, perform Clark transformation, convert to αβ coordinate system, and output i mfα 、i mfβ , used for unit vector generation;

[0109] Step S2-3, obtaining the voltage v of each phase terminal of the two-stage current source inverter sa 、v sb 、v sc , input the coordinate transformation module, and input the unit vector generated in step S2-2, and convert v sa 、v sb 、v sc Perform Clark transform and Park transform, convert to dq coordinate system, and output V sd 、V sq ;

[0110] Step S2-4: convert the d-axis component of the voltage V sd Input reference module, output double-stage current source inverter voltage q-axis component reference value V sqref ;

[0111] Step S2-5: V obtained in step S2-3 sq and V obtained in step S2-4 sqrefInput Q-axis controller, output voltage source inverter voltage q-axis component V vq ;

[0112] Step S2-6, obtaining the voltage source inverter capacitor voltage V c , input to the D-axis controller, and the reference capacitor voltage V cref In contrast, the output voltage source inverter voltage d-axis component V vd ;

[0113] Step S2-7: Substitute the unit vector obtained in step S2-2 and the V obtained in step S2-5 vq and V obtained in step S2-6 vd Input the coordinate inverse transformation module, perform Park inverse transformation and Clark inverse transformation, and output the voltage source inverter three-phase modulation signal V vma 、V vmb 、V vmc ;

[0114] Step S2-8: V vma 、V vmb 、V vmc Input pulse module, output gate pulse signal of voltage source inverter IGBT.

[0115] The two-stage current source inverter comprises two SCR-based current source inverters: CSI-A and CSI-B, wherein the SCR is a thyristor; the CSI-A and CSI-B are connected in parallel, both are connected in series with a SCR-based controllable rectifier, and are connected in series with a DC link inductor, and are powered by two 50Hz AC power supplies respectively; the CSI-A and CSI-B both operate in a 120° conduction mode, and the phase difference between the two is 30°, so that a quasi-square wave output current is generated in each inverter; Figure 2 As shown, the output current has a multi-level distribution, realizing multi-level control, and is used to provide active power to the motor. The voltage source inverter is a three-phase bridge inverter based on IGBT, powered by a pre-charged capacitor, and is only used to provide reactive power; the IGBT is an insulated gate bipolar transistor. The position observer module is based on a position sensorless, and uses a combination of a multilayer perceptron neural network and a sliding mode observation to obtain the motor speed for closed-loop control.

[0116] Furthermore, during the normal operation of the motor, the power factor of the two-stage current source inverter is positive, and the fundamental component i of the induction motor current is mf Leading two-stage current source inverter terminal voltage v s , the leading phase angle is β; the voltage source inverter is used for reactive power compensation to ensure that the leading phase angle β of the two-stage current source inverter remains unchanged; the phasor of the motor current fundamental component I mf Lagging behind the motor voltage phasor Vm , the power factor angle is The phasor I of the fundamental component of the CSI-B current s2f The phase I that lags behind the fundamental component of the CSI-A current s1f , the lagging phase angle is 30°; the phasor V of the terminal voltage of the two-stage current source inverter s The motor voltage phasor V m 、Voltage phasor of voltage source inverter V v The following relationship exists:

[0117]

[0118] The leading phase angle β is determined according to system requirements, and the following relationship exists:

[0119] β=(15+γ)°

[0120] The selection of the commutation angle γ takes into account the SCR turn-off time, commutation inductance and DC current.

[0121] Furthermore, the induction motor driven by the drive system has regeneration capability; when the motor speed drops to a lower value or during speed reversal, the regeneration control is activated, and the induction motor acts as a generator to perform external work; during the regeneration operation, the motor voltage phasor V m and the phase of the motor current fundamental component I mf The phase angle between is the power factor angle of the induction motor under normal operating conditions, I mf The phase quantity V that leads the voltage at the two-stage current source inverter s , the leading phase angle is (π-β), β is V under normal operating conditions of the induction motor m with I mf The phase angle between .

[0122] Furthermore, in the control method of the two-stage current source inverter, the position observer module in step S1-1 is composed of an observation module and a neural network speed estimation module, and its structure is as follows: Figure 5 As shown;

[0123] The observation module adopts a discrete second-order sliding mode observer, which is composed of a Clark transformation module, a current observer, a sliding mode surface, a sliding mode observation module, and a rotor flux observation module. The construction steps are as follows:

[0124] Step S1-1-A1, establish a Clark transformation module to transform the stator current and voltage parameters of the induction motor from the natural coordinate system to the αβ coordinate system. The mathematical model of the Clark transformation module is as follows:

[0125] [i mα imβ i0] T =T 3s / 2s [i ma i mb i mc ] T

[0126] [V mα V mβ V0] T =T 3s / 2s [V ma V mb V mc ] T

[0127] Among them, T 3s / 2s is the Clark transformation matrix, i ma 、i mb 、i mc are the axis components of the induction motor stator current in the natural coordinate system, i mα 、i mβ are the axis components of the induction motor stator current in the αβ coordinate system, i0 is the zero-sequence current; V ma 、V mb 、V mc are the axis components of the induction motor stator voltage in the natural coordinate system, V mα 、V mβ They are the axis components of the induction motor stator voltage in the αβ coordinate system, V0 is the zero-sequence voltage;

[0128] Step S1-1-A2: Build the dynamic model of the induction motor as follows:

[0129]

[0130] Among them, ω r is the rotor speed of the induction motor, i mα 、i mβ are the axis components of the induction motor stator current in the αβ coordinate system, V mα 、V mβ are the axis components of the induction motor stator voltage in the αβ coordinate system, ψ a , β The components of the rotor flux of the induction motor in the αβ coordinate system are induced respectively; η is the rotor time constant, k1=k2R s , R r is the rotor resistance, L r is the rotor inductance, R s is the stator resistance, L sis the stator inductance, L m For mutual induction;

[0131] Step S1-1-A3: Extract the sliding mode function from the induction motor dynamic model as follows:

[0132]

[0133] Based on the above sliding mode function, the current observer model is designed as follows:

[0134]

[0135] in, and Observe the current for the stator;

[0136] Step S1-1-A4, define the sliding surface as follows:

[0137]

[0138] s α 、s β is the sliding surface function;

[0139] Design the Lyapunov candidate function as:

[0140]

[0141] According to Lyapunov stability theory, the candidate function satisfies the following condition a:

[0142]

[0143] Among them, Γ is a positive definite matrix, Γ1 and Γ2 are constants, and Γ1=3, Γ2=4 are selected;

[0144] According to the above condition a, the candidate function satisfies the following condition b:

[0145]

[0146] According to condition b, the Lyapunov condition is defined as follows:

[0147]

[0148] Step S1-1-A5: Establish a sliding mode function T according to the current observer model described in step S1-1-A2 and the Lyapunov condition described in step S1-1-A3. α 、T β The observation model is as follows:

[0149]

[0150] in, T α 、T β Observed value of

[0151] According to the above observation model, the reverse Euler method is used to calculate and obtain its discrete form, and the sliding mode observation module is established as follows:

[0152]

[0153] Among them, T s is the system sampling time, T s =1μs;

[0154] Step S1-1-A6, establish the rotor flux observation module as follows:

[0155]

[0156] in, is the rotor flux ψ a , β Observed value of

[0157] The neural network speed estimation module is constructed using a multi-layer perceptron neural network, and the configuration is as follows Figure 6 The multilayer perceptron neural network includes an input layer, a hidden layer, and an output layer, which is composed of 4 input units, 20 neurons, 100 weighting factors, 21 biases, and an output unit; the input layer is used to inject feedback signals; the hidden layer processes the data and sends the results to the output layer; the four input features are the motor reference speed ω ref , Observed value of rotor flux And the stator current i of the α axis in the αβ coordinate system mα , the output feature is the motor speed observation value The neurons in the hidden layer are designed based on the tangent-sigmoid activation function;

[0158] The mathematical model of the neural network speed estimation module is as follows:

[0159]

[0160] Among them, μ i is the input feature vector, f j (μ i ,w i,j ) is the activation function of the jth neuron in the hidden layer, w i,j is the jth weighting factor of the ith input, w k is the kth weighting factor of the output layer, δ j is the bias of the jth neuron in the hidden layer, is the bias for the neurons in the output layer;

[0161] The steps of estimating the speed of the induction motor using the position observer module include:

[0162] Step S1-1-B1, obtain the three-phase current i of the induction motor ma 、i mb 、i mc And the three-phase voltage V ma 、V mb 、V mc , input Clark transformation module, output stator current i in αβ coordinate system mα 、i mβ and stator voltage V mα 、V mβ ;

[0163] Step S1-1-B2: The stator current i in the αβ coordinate system obtained in step 5-2-1 is mα 、i mβ and stator voltage V mα 、V mβ Input current observer, output stator observation current and

[0164] Step S1-1-B3: The stator observation current obtained in step S1-1-B2 is and Input sliding surface, output sliding surface function s α 、s β Enter the sliding mode observation module and establish T α 、T β Observations of

[0165] Step S1-1-B4: Input the current observer and repeat steps S1-1-B3. When the trajectory of the system follows the sliding surface, the stator current observer and The value of will converge to the actual stator current i mα 、i mβ , at this time T α 、T β Observed value Also with T α 、T β same;

[0166] Step S1-1-B5: The converged Input rotor flux observation module, output rotor flux observation value

[0167] Step S1-1-B6: The stator current i obtained in step S1-1-B1 is mα , rotor flux observation value obtained in step S1-1-B5 And the motor reference speed ω ref Input neural network speed estimation module, output motor speed observation value

[0168] Furthermore, in the control method of the two-stage current source inverter, in the frequency control module described in step S1-4, the gating pulse signals of CSI-A and CSI-B are given a phase difference of 30°.

[0169] Furthermore, in the control method of the voltage source inverter, when the induction motor is in normal operation, the reference module in step S2-4 is given by the following formula:

[0170] V sqref = -tan(β)×V sd

[0171] Where β is the fundamental component of the induction motor current i mf Relative to the voltage v of the two-stage current source inverter s The leading phase angle of

[0172] In the induction motor regenerative operation state, the reference module described in step S2-4 is given by the following formula:

[0173] V sqref =tan(β)×V sd

[0174] The Q-axis controller in step S2-5 and the D-axis controller in step S2-6 both use PI controllers.

[0175] Furthermore, the control method includes a pre-charging method for the capacitor of the voltage source inverter, wherein the pre-charging of the capacitor is achieved by triggering the SCR of the upper bridge arm and the lower bridge arm of any two phases of CSI-A and CSI-B, without providing a trigger signal to the IGBT. Figure 7 As shown, the pre-charging method is: control the DC current to pass through the SCR, motor winding, and anti-parallel diode of the upper bridge arm and the lower bridge arm of any two phases of CSI-A and CSI-B to form a closed loop to pre-charge the capacitor. c Reach its reference value V cref The drive system starts to work normally.

[0176] Finally, it should be noted that the specific embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to this embodiment, a person skilled in the art should understand that the technical solutions described in the foregoing specific embodiments may still be modified, or some or all of the technical features may be replaced by equivalents, but these modifications and replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A control method for an induction motor drive system based on a current source inverter, characterized in that: The drive system consists of a two-stage current source inverter at one end of the induction motor stator winding and a voltage source inverter at the other end of the stator winding, and is accompanied by a position observer module; The control method includes a control method of the two-stage current source inverter and a control method of the voltage source inverter; The control method of the two-stage current source inverter comprises the following steps: Step S1-1, obtain the motor speed observation value from the position observer module Input speed controller, and motor reference speed ω ref By comparison, we can get the speed difference ω slip ; Step S1-2: according to the speed difference ω slip , obtain the DC reference current i from the current reference module dref ; Step S1-3: Obtain the DC current i input to CSI-A and CSI-B d1 、i d2 , input current controller, the current controller adjusts the trigger angle of the two front controlled rectifiers to convert the DC current i d1 、i d2 Maintain at reference value Step S1-4: According to the motor speed observation value and the speed difference ω obtained in step S1-1 slip , obtain the frequency information of the two-stage current source inverter, input it into the frequency control module, and obtain the gating pulse signals of CSI-A and CSI-B; The control of the voltage source inverter is performed synchronously with the control of the two-stage current source inverter, and includes the following steps: Step S2-1: Based on the strobe pulse signal of CSI-A and the DC current i d1 , determine the fundamental component i of the three-phase current of the induction motor maf 、i mbf 、i mcf ; Step S2-2: The fundamental wave component i of the three-phase current of the induction motor maf 、i mbf 、i mcf Input coordinate transformation module, perform Clark transformation, convert to αβ coordinate system, and output i mfα 、i mfβ , used for unit vector generation; Step S2-3, obtaining the voltage v of each phase terminal of the two-stage current source inverter sa 、v sb 、v sc , input the coordinate transformation module, and input the unit vector generated in step S2-2, and convert v sa 、v sb 、v sc Perform Clark transform and Park transform, convert to dq coordinate system, and output V sd 、V sq ; Step S2-4: convert the d-axis component of the voltage V sd Input reference module, output double-stage current source inverter voltage q-axis component reference value V sqref ; Step S2-5: V obtained in step S2-3 sq and V obtained in step S2-4 sqref Input Q-axis controller, output voltage source inverter voltage q-axis component V vq ; Step S2-6, obtaining the voltage source inverter capacitor voltage V c , input to the D-axis controller, and the reference capacitor voltage V cref In contrast, the output voltage source inverter voltage d-axis component V vd ; Step S2-7: Substitute the unit vector obtained in step S2-2 and the V obtained in step S2-5 vq and V obtained in step S2-6 vd Input the coordinate inverse transformation module, perform Park inverse transformation and Clark inverse transformation, and output the voltage source inverter three-phase modulation signal V vma 、V vmb 、V vmc ; Step S2-8: V vma 、V vmb 、V vmc Input pulse module, output gate pulse signal of voltage source inverter IGBT; The two-stage current source inverter comprises two SCR-based current source inverters: CSI-A and CSI-B, wherein the SCR is a thyristor; the CSI-A and CSI-B are connected in parallel, both are connected in series with a SCR-based controllable rectifier, and are connected in series with a DC link inductor, and are powered by two 50Hz AC power supplies respectively; the CSI-A and CSI-B both operate in a 120° conduction mode, with a phase difference of 30°, and a quasi-square wave output current is generated in each inverter, and the output current has a multi-level distribution, realizing multi-level control, and is used to provide active power to the motor; the voltage source inverter is an IGBT-based three-phase bridge inverter, powered by a pre-charged capacitor, and is only used to provide reactive power, and the IGBT is an insulated gate bipolar transistor; the position observer module is based on a position sensorless, and adopts a combination of a multilayer perceptron neural network and a sliding mode observation to obtain the motor speed for closed-loop control.

2. The control method of the induction motor drive system based on the current source inverter according to claim 1, characterized in that: During the normal operation of the motor, the power factor of the two-stage current source inverter is positive, and the fundamental component i of the induction motor current is mf Leading two-stage current source inverter terminal voltage v s , the leading phase angle is β; the voltage source inverter is used for reactive power compensation to ensure that the leading phase angle β of the two-stage current source inverter remains unchanged; the phasor of the motor current fundamental component I mf Lagging behind the motor voltage phasor V m , the power factor angle is The phasor I of the fundamental component of the CSI-B current s2f The phase I that lags behind the fundamental component of the CSI-A current s1f , the lagging phase angle is 30°; the phasor V of the terminal voltage of the two-stage current source inverter s The motor voltage phasor V m 、Voltage phasor of voltage source inverter V v The following relationship exists: The leading phase angle β is determined according to system requirements, and the following relationship exists: β=(15+γ)° The selection of the commutation angle γ takes into account the SCR turn-off time, commutation inductance and DC current.

3. The control method of an induction motor drive system based on a current source inverter according to claim 1, characterized in that: The induction motor driven by the drive system has regeneration capability; when the motor speed drops to a lower value or during speed reversal, the regeneration control is activated, and the induction motor acts as a generator to perform external work; during the regeneration operation, the motor voltage phasor V m and the phase of the motor current fundamental component I mf The phase angle between is the power factor angle of the induction motor under normal operating conditions, I mf The phase quantity V that leads the voltage at the two-stage current source inverter s , the leading phase angle is (π-β), β is V under normal operating conditions of the induction motor m with I mf The phase angle between .

4. The control method of the induction motor drive system based on the current source inverter according to claim 1, characterized in that: In the control method of the two-stage current source inverter, the position observer module in step S1-1 is composed of an observation module and a neural network speed estimation module; The observation module adopts a discrete second-order sliding mode observer, which is composed of a Clark transformation module, a current observer, a sliding mode surface, a sliding mode observation module, and a rotor flux observation module. The construction steps are as follows: Step S1-1-A1, establish a Clark transformation module to transform the stator current and voltage parameters of the induction motor from the natural coordinate system to the αβ coordinate system. The mathematical model of the Clark transformation module is as follows: [i mα I mβ i0] T =T 3s / 2s [i ma I mb I mc ] T [V mα V mβ V0] T =T 3s / 2s [V ma V mb V mc ] T Among them, T 3s / 2s is the Clark transformation matrix, i ma 、i mb 、i mc are the axis components of the induction motor stator current in the natural coordinate system, i mα 、i mβ are the axis components of the induction motor stator current in the αβ coordinate system, i0 is the zero-sequence current; V ma 、V mb 、V mc are the axis components of the induction motor stator voltage in the natural coordinate system, V mα 、V mβ They are the axis components of the induction motor stator voltage in the αβ coordinate system, V0 is the zero-sequence voltage; Step S1-1-A2: Build the dynamic model of the induction motor as follows: Among them, ω r is the rotor speed of the induction motor, i mα 、i mβ are the axis components of the induction motor stator current in the αβ coordinate system, V mα 、V mβ are the axis components of the induction motor stator voltage in the αβ coordinate system, ψ a , β The components of the rotor flux of the induction motor in the αβ coordinate system are induced respectively; η is the rotor time constant, k1=k2R s , R r is the rotor resistance, L r is the rotor inductance, R s is the stator resistance, L s is the stator inductance, L m For mutual induction; Step S1-1-A3: Extract the sliding mode function from the induction motor dynamic model as follows: Based on the above sliding mode function, the current observer model is designed as follows: in, and Observe the current for the stator; Step S1-1-A4, define the sliding surface as follows: s α 、s β is the sliding surface function; Design the Lyapunov candidate function as: According to Lyapunov stability theory, the candidate function satisfies the following condition a: Among them, Γ is a positive definite matrix, Γ1 and Γ2 are constants; According to the above condition a, the candidate function satisfies the following condition b: According to condition b, the Lyapunov condition is defined as follows: Step S1-1-A5: Establish a sliding mode function T according to the current observer model described in step S1-1-A2 and the Lyapunov condition described in step S1-1-A3. α , T β The observation model is as follows: in, T α , T β Observed value of According to the above observation model, the reverse Euler method is used to calculate and obtain its discrete form, and the sliding mode observation module is established as follows: Among them, T s is the system sampling time; Step S1-1-A6, establish the rotor flux observation module as follows: in, is the rotor flux ψ a , β Observed value of The neural network speed estimation module is constructed using a multilayer perceptron neural network, which includes an input layer, a hidden layer, and an output layer, and is composed of 4 input units, 20 neurons, 100 weighting factors, 21 biases, and an output unit; the input layer is used to inject feedback signals; the hidden layer processes the data and sends the results to the output layer; the four input features are the motor reference speed ω ref , Observed value of rotor flux And the stator current i of the α axis in the αβ coordinate system mα , the output feature is the motor speed observation value The neurons in the hidden layer are designed based on the tangent-sigmoid activation function; The mathematical model of the neural network speed estimation module is as follows: Among them, μ i is the input feature vector, f j (μ i ,w i,j ) is the activation function of the jth neuron in the hidden layer, w i,j is the jth weighting factor of the ith input, w k is the kth weighting factor of the output layer, δ j is the bias of the jth neuron in the hidden layer, is the bias for the neurons in the output layer; The steps of estimating the speed of the induction motor using the position observer module include: Step S1-1-B1, obtain the three-phase current i of the induction motor ma 、i mb 、i mc And the three-phase voltage V ma 、V mb 、V mc , input Clark transformation module, output stator current i in αβ coordinate system mα 、i mβ and stator voltage V mα 、V mβ ; Step S1-1-B2: The stator current i in the αβ coordinate system obtained in step S1-1-B1 is mα 、i mβ and stator voltage V mα 、V mβ Input current observer, output stator observation current and Step S1-1-B3: The stator observation current obtained in step S1-1-B2 is and Input sliding surface, output sliding surface function s α 、s β Enter the sliding mode observation module and establish T α 、T β Observations of Step S1-1-B4: Input the current observer and repeat steps S1-1-B3. When the trajectory of the system follows the sliding surface, the stator current observer and The value of will converge to the actual stator current i mα 、i mβ , at this time T α , T β Observed value Also with T α , T β same; Step S1-1-B5: The converged Input rotor flux observation module, output rotor flux observation value Step S1-1-B6: The stator current i obtained in step S1-1-B1 is mα , rotor flux observation value obtained in step S1-1-B5 And the motor reference speed ω ref Input neural network speed estimation module, output motor speed observation value 5. The control method of an induction motor drive system based on a current source inverter according to claim 1, characterized in that: In the control method of the two-stage current source inverter, in the frequency control module described in step S1-4, the gating pulse signals of CSI-A and CSI-B are given a phase difference of 30°.

6. The control method of an induction motor drive system based on a current source inverter according to claim 1, characterized in that: In the control method of the voltage source inverter, when the induction motor is in normal operation, the reference module in step S2-4 is given by the following formula: V sqref =-tan(β)×V sd Where β is the fundamental component of the induction motor current i mf Relative to the voltage v of the two-stage current source inverter s The leading phase angle of In the induction motor regenerative operation state, the reference module described in step S2-4 is given by the following formula: V sqref =tan(β)×V sd The Q-axis controller in step S2-5 and the D-axis controller in step S2-6 both use PI controllers.

7. The control method of an induction motor drive system based on a current source inverter according to claim 1, characterized in that: The control method includes a pre-charging method for a voltage source inverter capacitor, wherein the pre-charging of the capacitor is achieved by triggering the SCR of the upper bridge arm and the lower bridge arm of any two phases of CSI-A and CSI-B, and there is no need to provide a trigger signal to the IGBT; the pre-charging method is: controlling a direct current to pass through the SCR of the upper bridge arm and the lower bridge arm of any two phases of CSI-A and CSI-B, the motor winding, and the anti-parallel diode of the IGBT to form a closed loop, and pre-charging the capacitor; when the capacitor voltage V c Reach its reference value V cref The drive system starts to work normally.

Images