Full-speed-domain dual-mode redundancy fault-tolerant control method for permanent magnet synchronous motor of electric vehicle

Abstract

A full-speed-domain dual-mode redundancy fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle is characterized by comprising the steps that S1, a full-speed-domain dual-mode redundancy fault-tolerant control system of the permanent magnet synchronous motor is constructed; s2, a weighted switching module obtains the motor speed omega; s3a: when omega < lt >; when [omega] 1, a high-frequency injection module injects a dynamic amplitude high-frequency square wave signal to a d axis according to a load working condition, and extracts a rotor position in a zero low-speed domain; s3b: when [omega] gt; when omega is 2, the sliding mode observer module estimates the position of the rotor in a medium-high speed domain based on an improved super-spiral sliding mode algorithm of an adaptive inductance parameter; s3c, when omega belongs to [omega 1, omega 2], a weighted switching module performs dynamic weight distribution on the rotor positions from the high-frequency injection module and the sliding mode observer module according to the motor speed omega, and comprehensively extracts the motor rotor position; and S4, the motor driving module realizes full-speed domain control of the permanent magnet synchronous motor according to the position of the motor rotor. The method has the advantages that full-speed-domain seamless switching control can be achieved when the encoder breaks down.

Description

Technical Field

[0001] The present invention relates to the technical field of motor control, and in particular to a full-speed domain dual-mode redundant fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle. Background Art

[0002] In the current electric vehicle market, high-efficiency, small size, light weight, and low-cost motor systems are core elements for achieving product competitiveness. With the acceleration of global urbanization and stricter environmental protection policies, the market demand for electric vehicles as zero-emission, highly flexible short-distance travel tools continues to surge. However, traditional electric vehicle drive systems rely on position sensors such as photoelectric encoders or resolvers to obtain rotor information. These sensors are prone to failure under complex operating conditions (such as high temperature, humidity, and vibration), resulting in reduced system reliability. For example, during the rainy season or on bumpy roads, encoder disconnection or signal distortion problems are frequent, which not only increases maintenance costs but also may pose safety hazards. In addition, the introduction of high-precision and high-reliability sensors has significantly increased system costs, limiting the popularity of electric vehicles in price-sensitive markets. To ensure user safety, a backup motor control solution that does not require physical sensors and can operate stably across the full speed range is needed to meet users' urgent needs for highly reliable and low-cost electric vehicles.

[0003] In the field of sensorless control technology, high-frequency signal injection and fundamental back-EMF observation are currently the mainstream solutions. The high-frequency injection method injects a high-frequency voltage signal into the motor, leveraging the saliency effect to achieve accurate position estimation in the low-speed range. However, it cannot cover high-speed operation scenarios and suffers from significant additional losses and noise. While the fundamental back-EMF observation method is suitable for medium and high-speed ranges, its accuracy drops sharply at low speeds due to the small back-EMF amplitude. More seriously, existing technologies lack robustness to sampling errors (such as DC bias) and parameter variations (such as stator resistance temperature drift), resulting in poor low-speed control stability and large torque fluctuations. For example, in electric vehicles with frequent starting and stopping or climbing conditions, traditional sensorless control systems are prone to angle estimation errors, leading to acceleration lag or power interruption, seriously affecting the driving experience and safety. Therefore, developing a sensorless control technology that seamlessly switches across all speed ranges and has strong anti-interference capabilities has become a key breakthrough in improving electric vehicle performance. Summary of the Invention

[0004] The present invention provides a full-speed dual-mode redundant fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle, which can achieve seamless switching control in the full-speed domain when an encoder fails, effectively improving the robustness and safety of the motor system.

[0005] To achieve the above-mentioned object, the present invention provides a full-speed domain dual-mode redundant fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle, the key of which is to include the following steps:

[0006] S1: Constructing a full-speed domain dual-mode redundant fault-tolerant control system for a permanent magnet synchronous motor, wherein the full-speed domain dual-mode redundant fault-tolerant control system is provided with a motor drive module, the motor drive module is connected to a high-frequency injection module, a synovial observer module and a weighted switching module, and the high-frequency injection module and the synovial observer module are respectively connected to the weighted switching module;

[0007] S2: The weighted switching module obtains speed information ω of the permanent magnet synchronous motor in real time, and determines the operating speed range of the permanent magnet synchronous motor according to the speed information ω;

[0008] S3a: When the motor speed ω<the speed lower limit threshold ω1, the permanent magnet synchronous motor operates in the zero low speed domain. The high frequency injection module injects a dynamic amplitude high frequency square wave signal into the d axis according to the load condition, extracts the rotor position information in the zero low speed domain, and transmits it to the weighted switching module; the weighted switching module converts the high frequency injection method weight w h Set to 1 to output the rotor position information to the motor drive module;

[0009] S3b: When the motor speed ω> the speed upper limit threshold ω2, the permanent magnet synchronous motor operates in the medium and high speed domain, and the synovial observer module extracts the back electromotive force based on the improved super-helical synovial algorithm with adaptive inductance parameters, estimates the rotor position information in the medium and high speed domain, and transmits it to the weighted switching module; the weighted switching module converts the synovial observer weight w s Set to 1 to output the rotor position information to the motor drive module;

[0010] S3c: When the motor speed ω∈[ω1,ω2], the permanent magnet synchronous motor operates in the switching speed domain, and the weighted switching module dynamically distributes the rotor position information from the high-frequency injection module and the sliding film observer module according to the motor speed ω. The weighted switching module allocates the high-frequency injection weight w h Set to Synovial observer weight w s Set to n is the speed, and the comprehensive output rotor position information is given to the motor drive module;

[0011] S4: The motor drive module implements full-speed domain dual-mode redundant fault-tolerant control of the permanent magnet synchronous motor according to the motor rotor position information.

[0012] Through the above design, the present invention proposes a dual-mode redundant control scheme that integrates dynamic amplitude high-frequency square wave injection technology and an improved super-spiral synonautic observer with adaptive inductance parameters that introduces the Tanh function. The system seamlessly switches to sensorless mode when the encoder fails through a dynamic weight switching strategy: the low-speed domain adopts a dynamic amplitude high-frequency injection method to better extract the rotor position by dynamically adjusting the square wave amplitude; the medium and high-speed domains adopt a super-spiral synonautic observer back electromotive force model based on adaptive inductance parameters to estimate the position, and the error caused by the magnetic saturation effect is reduced through adaptive inductance parameters; the motor efficiency and speed regulation range are optimized through a full-domain MAP current distribution strategy. The present invention realizes seamless fault-tolerant control in the full-speed domain, solves the problems of poor low-speed stability, insufficient high-speed coverage and sensor dependence of traditional solutions, significantly improves the robustness and safety of the system, and is suitable for fields that are sensitive to reliability and cost, such as electric vehicles.

[0013] Preferably, the high-frequency injection module is provided with a high-frequency signal generating unit, a high-frequency signal processing unit, a PARK conversion unit, a first phase-locked loop unit and a polarity identification unit;

[0014] The high-frequency signal processing unit is provided with a differential operator and an envelope detector, the input end of the differential operator is connected to the motor drive module, the output end of the differential operator is connected to the first phase-locked loop unit via the envelope detector, and the output end of the differential operator is also connected to the polarity identification unit via the PARK conversion unit; the high-frequency signal generation unit is connected to the motor drive module.

[0015] The high-frequency signal generating unit is used to generate the high-frequency signal required to meet the motor rotor position prediction, determine the frequency and amplitude of the high-frequency signal, and adjust the high-frequency square wave amplitude according to the actual motion situation. Perform dynamic adjustments, such as increasing the high-frequency square wave amplitude simultaneously when facing high load conditions and high operating current, to achieve better response signal effects.

[0016] The high frequency signal processing unit is used to separate the high frequency component i by differential operation. αhf 、i βhf , and calculate the envelope signal I αh , I βh ;

[0017] The first phase-locked loop unit is used to utilize the error signal Output rotor position through PI regulation

[0018] The polarity identification unit is used to determine the initial polarity of the rotor by comparing the d-axis current amplitude through positive and negative bias voltage injection.

[0019] Preferably, the high-frequency injection module extracts rotor position information in the zero-speed domain, and the specific steps are as follows:

[0020] A1: The high-frequency signal generation unit dynamically adjusts the square wave amplitude according to the load conditions Inject a dynamic amplitude high-frequency square wave signal into the d-axis;

[0021] A2: The differential operator obtains the response current I from the motor drive module in the αβ two-phase stationary coordinate system α , I β , and calculate and separate the high-frequency response current component I in the response current αhf , I βhf , the calculation expression is as follows:

[0022] I αhf =(I α -I α z -1 ) / 2

[0023] I βhf =(I β -I β z -1 ) / 2

[0024] Among them, z -1 Indicates a delay of one cycle;

[0025] The differential operator converts the high frequency response current component I αhf , I βhf Passed to the envelope detector and PARK transformation unit;

[0026] A3: The envelope detector responds to the high frequency current component I αhf , I βhf Perform envelope detection to obtain the high-frequency response component envelope I αh , I βh , the expression is as follows:

[0027]

[0028] in, Indicates the amplitude of high-frequency square wave;

[0029] The envelope detector envelops the high frequency response component I αh , I βh passing it to the first phase-locked loop unit;

[0030] The first phase-locked loop unit is configured to generate a high-frequency response component envelope I αh , I βh , the rotor position information is obtained through PI regulation calculation The calculation expression is as follows:

[0031]

[0032] After simplification and per-unit processing, it is expressed as:

[0033]

[0034] Among them, L q Indicates the d-axis inductance, L d represents the q-axis inductance, ε represents the rotor position error, and the rotor position information when ε approaches 0 is finally output after PI adjustment.

[0035] A4: The PARK conversion unit converts the high frequency response current component I αhf Converted into d-axis high-frequency response current I dh , and passed to the polarity identification unit;

[0036] The polarity identification unit uses the principle of magnetic saturation effect to inject a high-frequency voltage U into the rotor during the initial position check. dh , change the d-axis voltage bias direction and compare the d-axis high-frequency response current amplitudes of the two cases at the same time to determine the rotor polarity;

[0037] When the forward bias amplitude is greater than the reverse bias amplitude, the obtained position is the rotor N pole, and the high-frequency injection module directly outputs the rotor position information For the weighted switching module, the expression is:

[0038] |I dh (U dh )|>|I dh (-U dh )|, then

[0039] When the forward bias amplitude is less than the reverse bias amplitude, the obtained position is the rotor S pole, and the high frequency injection module is used to calculate the rotor position information. After position compensation, it is passed to the weight switching module. The expression is as follows:

[0040] |I dh (U dh )|<|I dh (-U dh )|, then

[0041] As a preference: the high frequency injection module is based on the q-axis current I q The load condition is divided into three intervals, namely low load area, medium load area and high load area;

[0042] When the q-axis current I q<20A is the low load area, and the high frequency square wave amplitude corresponding to the low load area

[0043] When the q-axis current 20A≤I q ≤40A is the medium load area, and the high-frequency square wave amplitude corresponding to the medium load area

[0044] When the q-axis current I q >40A is the high load area, and the high-frequency square wave amplitude corresponding to the high load area

[0045] Preferably, the synovial film observer module is provided with an improved super-spiral synovial film observer with adaptive inductance parameters and a second phase-locked loop unit, the input end of the improved super-spiral synovial film observer is connected to the motor drive module, and the output end of the improved super-spiral synovial film observer is connected to the weighted switching module via the second phase-locked loop unit.

[0046] The improved super spiral synovial observer adopts hyperbolic tangent function instead of sign function to design equivalent feedback signal To eliminate the vibration phenomenon of the synovial observer.

[0047] The improved super spiral sliding film observer is based on the motor inductance parameter L s With q-axis current I q The change curve of the gain K under different working conditions is designed, and the real-time acquisition I q The current determines the different working conditions of the motor, thereby dynamically adjusting the inductance parameter L s The value of φ can be selected to reduce the error caused by magnetic saturation effect.

[0048] The second phase-locked loop unit passes the error signal Adjust the output rotor position

[0049] Preferably, the synovial observer module estimates the rotor position information in the medium and high speed domain, comprising the following steps:

[0050] B1: The improved super-helical synovial observer obtains the response current I from the motor drive module in the αβ two-phase stationary coordinate system α , I β and the response voltage U α 、U β , and then the synovial equivalent feedback signal is obtained by integrating the hyperbolic tangent function And the equivalent feedback signal passing it to the second phase-locked loop unit;

[0051] The mathematical model expression of the improved super spiral synovial observer is:

[0052]

[0053] in, represents the α-axis observation current; represents the β-axis observation current; R s Indicates stator resistance; U α Indicates the α-axis voltage; U β represents the β-axis voltage; It represents the synovial equivalent feedback of the α-axis; Indicates the equivalent feedback of the synovial membrane on the β axis; N indicates the feedback gain; S α represents the sliding mode term of the α axis; S β represents the sliding mode term of the β axis; I α Indicates the actual current of α axis; I β Indicates the actual current of β axis; Represents the adaptive inductance parameter;

[0054]

[0055] Among them, L s is the inductance parameter; K is the sliding film gain; K1 is the average change gain of medium load; K2 is the average change gain of high load, Tanh() is the hyperbolic tangent function,

[0056] Equivalent feedback signal It is generated by the synovial term through integration operation and is expressed as follows:

[0057]

[0058] Where M is the feedback gain;

[0059] B2: The second phase-locked loop unit generates an equivalent feedback signal according to the Adjust error signal S err , and output error signal S err Rotor position information when it approaches 0 For the weighted switching module, the expression is:

[0060] As an example, the weighted dynamic weight allocation strategy of the weighted switching module in each operating speed domain is as follows:

[0061] Zero low-speed domain: high-frequency injection method weight w h =1, synovial observer weight w s =0; weighted switching module outputs rotor position information

[0062] Switching speed domain: High frequency injection method weight Synovial observer weight Where n is the speed, ω = 2πn; the weighted switching module outputs the rotor position information

[0063] Medium and high speed domain: high frequency injection method weight w h =0, synovial observer weight w s =1; weighted switching module outputs rotor position information

[0064] The weighted switching module distributes the weights of the high-frequency injection method and the synovial observer in sections according to the motor speed, thereby achieving a smooth transition of the full-speed range sensorless control.

[0065] Preferably, the weighted switching module is also connected to the encoder of the permanent magnet synchronous motor, and the weighted switching module detects the working status of the encoder in real time. When the encoder is detected to be disconnected or faulty, the weighted switching module seamlessly enables the high-frequency injection module and the synovial observer module to detect the rotor position of the permanent magnet synchronous motor.

[0066] The weighted switching module monitors the encoder's operating status in real time and seamlessly switches to dual-mode redundant control mode if the encoder fails. In dual-mode redundant control mode, a weighted switching approach is used to transition between the high-frequency injection method and the sliding film observer method, smoothly combining the two motor sensorless control algorithms.

[0067] The beneficial effects of the present invention are as follows: the present invention adopts a motor FOC control system with a global MAP current distribution strategy and has a dual-mode switching function for position detection, which can prevent the dangers caused by encoder failure while meeting the efficient operation of electric vehicles in multiple working conditions; the dual-mode redundant control combining a dynamic high-frequency injection method and an improved super-spiral sliding film observer can expand the operating speed range of the backup system to the full speed range, thereby improving the safety and robustness of the electric vehicle motor system. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 This is the principle block diagram of the full-speed domain dual-mode redundant fault-tolerant control system for the permanent magnet synchronous motor of an electric vehicle;

[0069] Figure 2 This is a logic block diagram of the high-frequency injection module in the embodiment;

[0070] Figure 3 1. It is a logic block diagram of a phase-locked loop in a high-frequency injection module in an embodiment;

[0071] Figure 4 This is a flow chart for identifying the initial position polarity of an electric vehicle motor in an embodiment;

[0072] Figure 5 This is a logic block diagram of a synovial membrane observer module in an embodiment;

[0073] Figure 6 is the inductance L of the electric vehicle motor in the embodiment s Change curve chart;

[0074] Figure 7 is a logic block diagram of a phase-locked loop in a synovial membrane observer module in an embodiment;

[0075] Figure 8 This is a weight distribution block diagram of the weighted switching module in the embodiment;

[0076] Figure 9 This is a block diagram of the dynamic switching of the electric vehicle position detection module in the embodiment;

[0077] Figure 10 This is a diagram showing the effect of dynamic amplitude high frequency injection in the embodiment;

[0078] Figure 11 This is a comparison diagram of the optimization effect of the hyperbolic tangent function in the embodiment;

[0079] Figure 12 This is a diagram showing the rotor position identification effect in the embodiment;

[0080] Figure 13 This is a diagram showing the effect of closed-loop speed control of the system in the embodiment. DETAILED DESCRIPTION

[0081] The present invention will be further described in detail below with reference to the accompanying drawings and specific examples. The following examples or drawings are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0082] A full-speed domain dual-mode redundant fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle comprises the following steps:

[0083] S1: Construct a full-speed domain dual-mode redundant fault-tolerant control system for permanent magnet synchronous motors, such as Figure 1 As shown, the full-speed domain dual-mode redundant fault-tolerant control system is provided with a motor drive module, the motor drive module is connected to the high-frequency injection module, the synovial observer module and the weighted switching module, and the high-frequency injection module and the synovial observer module are respectively connected to the weighted switching module;

[0084] The high-frequency injection module is provided with a high-frequency signal generating unit, a high-frequency signal processing unit, a PARK conversion unit, a first phase-locked loop unit and a polarity identification unit;

[0085] The high-frequency signal processing unit is provided with a differential operator and an envelope detector, the input end of the differential operator is connected to the motor drive module, the output end of the differential operator is connected to the first phase-locked loop unit via the envelope detector, and the output end of the differential operator is also connected to the polarity identification unit via the PARK conversion unit; the high-frequency signal generation unit is connected to the motor drive module.

[0086] The synovial film observer module is provided with an improved super-spiral synovial film observer with adaptive inductance parameters and a second phase-locked loop unit. The input end of the improved super-spiral synovial film observer is connected to the motor drive module, and the output end of the improved super-spiral synovial film observer is connected to the weighted switching module via the second phase-locked loop unit. Figure 1 In the figure, SOM stands for improved super spiral synovial membrane observer.

[0087] S2: The weighted switching module obtains speed information ω of the permanent magnet synchronous motor in real time, and determines the operating speed range of the permanent magnet synchronous motor according to the speed information ω;

[0088] S3a: When the motor speed ω<the speed lower limit threshold ω1, the permanent magnet synchronous motor operates in the zero low speed domain. The high frequency injection module injects a dynamic amplitude high frequency square wave signal into the d axis according to the load condition, extracts the rotor position information in the zero low speed domain, and transmits it to the weighted switching module; the weighted switching module converts the high frequency injection method weight w h Set to 1 to output the rotor position information to the motor drive module;

[0089] S3b: When the motor speed ω> the speed upper limit threshold ω2, the permanent magnet synchronous motor operates in the medium and high speed domain, and the synovial observer module extracts the back electromotive force based on the improved super-helical synovial algorithm with adaptive inductance parameters, estimates the rotor position information in the medium and high speed domain, and transmits it to the weighted switching module; the weighted switching module converts the synovial observer weight w s Set to 1 to output the rotor position information to the motor drive module;

[0090] S3c: When the motor speed ω∈[ω1,ω2], the permanent magnet synchronous motor operates in the switching speed domain, and the weighted switching module dynamically distributes the rotor position information from the high-frequency injection module and the sliding film observer module according to the motor speed ω. The weighted switching module allocates the high-frequency injection weight w h Set to Synovial observer weight w s Set to n is the speed, and the comprehensive output rotor position information is given to the motor drive module;

[0091] S4: The motor drive module implements full-speed domain dual-mode redundant fault-tolerant control of the permanent magnet synchronous motor according to the motor rotor position information.

[0092] like Figure 1As shown, the motor drive module is provided with a first multiplier, the first input end of the first multiplier obtains the actual speed information N from the weighted switching module or the motor encoder, the second input end of the first multiplier obtains the reference speed Nref from the host computer, the output end of the first multiplier is connected to the global MAP allocation table through the first PI regulator and the power limit module, and the global MAP allocation table allocates the d-axis reference current The first input terminal of the second multiplier is supplied with the actual current I of the d-axis of the permanent magnet synchronous motor. d ; The global MAP allocation table allocates q-axis reference current The first input terminal of the third multiplier is supplied with the actual current I of the q-axis of the permanent magnet synchronous motor. q The output of the second multiplier outputs the d-axis voltage U through the second PI regulator d The first input terminal of the fourth multiplier is connected to the output terminal of the high-frequency signal generating unit, and the output terminal of the fourth multiplier is connected to the PARK inverse transformation unit; the output terminal of the third multiplier outputs the q-axis voltage U through the third PI regulator. q To the PARK inverse conversion unit, the PARK inverse conversion unit converts the dq axis voltage U d 、U q Perform PARK inverse transformation to obtain the αβ axis voltage U α 、U β And transmit it to the space vector pulse width modulation unit SVPWM, the space vector pulse width modulation unit SVPWM generates a control signal and transmits it to the three-phase inverter circuit; the three-phase inverter circuit generates a three-phase stator current (i a ,i b ,i c ) and transmitted to the permanent magnet synchronous motor.

[0093] The CLARKE inverse transformation unit extracts the three-phase stator current (i a ,i b ,i c ), and convert it into αβ axis current I α , I β , and then transmit it to the PARK conversion unit, high frequency injection module and synovial observer module; the PARK conversion unit converts the αβ axis current I α , I β Converted into dq axis actual current I d , I q .

[0094] The motor drive module adopts the speed and current double closed loop vector control mode. d 、iq The current distribution stage uses a global MAP allocation strategy based on the operating conditions of the electric vehicle to ensure optimal motor performance. Normally, the rotor position information in vector control is provided by an encoder. However, if an encoder fails, the system dynamically switches to a dual-mode redundant fault-tolerant control system to provide position information.

[0095] Global MAP allocation strategy: In the low speed area of ​​the permanent magnet synchronous motor, in order to ensure the optimization of the motor operation efficiency, the maximum torque current ratio (MTPA) control strategy is used to find the optimal current distribution (i d 、i q ) to maximize the torque generated per unit current. As the motor speed increases, measures must be taken to reduce the motor's magnetic flux to overcome the increase in back EMF, a strategy known as flux weakening control. This allows the motor to operate at higher speeds while sacrificing torque performance without exceeding the inverter voltage limit. Global MAP control encompasses MTPA control and deep flux weakening control. Based on the MTPA curve and multiple sets of optimal flux weakening operating points calibrated under actual motor operation, an algorithm generates a three-dimensional global MAP control table for torque, speed, and current, covering the dq-axis current distribution strategy at all torque and speed levels.

[0096] like Figure 8 As shown, the weighted dynamic weight allocation strategy of the weighted switching module in each operating speed domain is as follows:

[0097] Zero low-speed domain: high-frequency injection method weight w h =1, synovial observer weight w s =0; weighted switching module outputs rotor position information

[0098] Switching speed domain: High frequency injection method weight Synovial observer weight Where n is the speed, ω = 2πn; the weighted switching module outputs the rotor position information

[0099] Medium and high speed domain: high frequency injection method weight w h =0, synovial observer weight w s =1; weighted switching module outputs rotor position information

[0100] The high frequency injection module extracts the rotor position information in the zero low speed domain, combined with Figure 2 The high-frequency square wave signal injection method shown in the figure injects a high-frequency square wave signal with a certain frequency into the d-axis when realizing the position prediction function, and decouples the position signal based on the extracted high-frequency response current. In addition, when predicting the initial position, polarity identification is required after the position prediction to compensate and correct the predicted angle. The specific steps are as follows:

[0101] A1: The high-frequency signal generation unit dynamically adjusts the square wave amplitude according to the load conditions Inject a dynamic amplitude high-frequency square wave signal into the d-axis;

[0102] A2: The differential operator obtains the response current I from the motor drive module in the αβ two-phase stationary coordinate system α , I β , and calculate and separate the high-frequency response current component I in the response current αhf , I βhf , the calculation expression is as follows:

[0103] I αhf =(I α -I α z -1 ) / 2

[0104] I βhf =(I β -I β z -1 ) / 2

[0105] Among them, z -1 Indicates a delay of one cycle;

[0106] The differential operator converts the high frequency response current component I αhf , I βhf Passed to the envelope detector and PARK transformation unit;

[0107] A3: The envelope detector responds to the high frequency current component I αhf , I βhf Perform envelope detection to obtain the high-frequency response component envelope I αh , I βh , the expression is as follows:

[0108]

[0109] in, Indicates the amplitude of high-frequency square wave;

[0110] The envelope detector envelops the high frequency response component I αh , I βh passing it to the first phase-locked loop unit;

[0111] The logic block diagram of the first phase-locked loop unit is as follows: Figure 3 As shown, the first phase-locked loop unit is based on the high frequency response component envelope I αh , I βh, the rotor position information is obtained through PI regulation calculation The calculation expression is as follows:

[0112]

[0113] After simplification and per-unit processing, it is expressed as:

[0114]

[0115] Among them, L q Indicates the d-axis inductance, L d represents the q-axis inductance, ε represents the rotor position error, and the rotor position information when ε approaches 0 is finally output after PI adjustment

[0116] A4: The PARK conversion unit converts the high frequency response current component I αhf Converted into d-axis high-frequency response current I dh , and passed to the polarity identification unit;

[0117] like Figure 4 As shown, the polarity identification unit uses the principle of magnetic saturation effect to inject a high-frequency voltage U when checking the initial position of the rotor. dh , change the d-axis voltage bias direction and compare the d-axis high-frequency response current amplitudes of the two cases at the same time to determine the rotor polarity;

[0118] When the forward bias amplitude is greater than the reverse bias amplitude, the obtained position is the rotor N pole, and the high-frequency injection module directly outputs the rotor position information For the weighted switching module, the expression is:

[0119] |I dh (U dh )|>|I dh (-U dh )|, then

[0120] When the forward bias amplitude is less than the reverse bias amplitude, the obtained position is the rotor S pole, and the high frequency injection module is used to calculate the rotor position information. After position compensation, it is passed to the weight switching module. The expression is as follows:

[0121] |I dh (U dh )|<|I dh (-U dh )|, then

[0122] After the high-frequency injection module injects a high-frequency square wave into the d-axis, the rotor position information needs to be extracted through the high-frequency response current. Under high-frequency injection, the voltage equation of the permanent magnet synchronous motor can be expressed as follows (the subscript h represents the high-frequency component):

[0123]

[0124] The relationship between the rotor rotating shaft system and the actual shaft system voltage can be expressed as;

[0125]

[0126] Among them, U d Indicates the d-axis voltage; U q represents the q-axis voltage; and Indicates the estimated high-frequency square wave variable under the dq axis; Indicates the actual angle θ and the estimated angle Similarly, the dq axis response current is converted to the αβ two-phase stationary coordinate system:

[0127]

[0128] Taking the derivatives of both sides at the same time, combined with the conclusion of the above formula, we can deduce:

[0129]

[0130] After simplification, the high-frequency response current is expressed, where L0 = (L d +L q ) / 2, L1=(L q -L d ) / 2:

[0131]

[0132] Among them, ω c represents the frequency of the high-frequency injection signal;

[0133] when When , that is, when the position prediction has no error, we can get:

[0134]

[0135] It can be deduced that the envelope of the high-frequency response current in the two-phase stationary coordinate system contains the rotor position information.

[0136] In electric vehicle control systems, considering chip performance limitations, an orthogonal phase-locked loop is designed to determine the rotor position. Error control through phase-locked loop Set to 0 to achieve rotor position prediction.

[0137] Contains position error information The extracted information is ε, and a phase-locked loop can be designed to extract the information, which can be expressed as:

[0138]

[0139] After simplification and per-unit processing, it is expressed as

[0140]

[0141] Design the phase-locked loop so that ε=0 through PI, which means The accurate rotor position can be obtained

[0142] The polarity identification unit is used to determine the rotor polarity in the motor initial position detection function. Using the principle of magnetic saturation effect, a high-frequency voltage U is injected during the rotor initial position detection. dh By changing the d-axis voltage bias direction and comparing the d-axis high-frequency response current amplitudes in the two cases at the same time, the rotor polarity can be determined. When the forward bias amplitude is larger, the obtained position is the rotor N pole; otherwise, it is the S pole, and a compensation value π needs to be added.

[0143] The high frequency injection module is based on the q-axis current I q The load condition is divided into three intervals, namely low load area, medium load area and high load area;

[0144] When the q-axis current I q <20A is the low load area, and the high frequency square wave amplitude corresponding to the low load area

[0145] When the q-axis current 20A≤I q ≤40A is the medium load area, and the high-frequency square wave amplitude corresponding to the medium load area

[0146] When the q-axis current I q >40A is the high load area, and the high-frequency square wave amplitude corresponding to the high load area

[0147] Design a high-frequency square wave signal with a certain frequency and dynamic amplitude, according to I q The current size divides the load condition into three intervals, namely low, medium and high load areas. q The current value determines the working condition and obtains different high-frequency square wave amplitudes In this way, a better high-frequency response current effect can be obtained.

[0148] Specific working conditions are divided as follows Figure 6 As shown, the working condition is divided into three intervals: low, medium and high load areas, with 20A and 40A as the boundaries; the high-frequency square wave amplitude Select the optimal voltage amplitude according to different working conditions. like Figure 10 shown.

[0149] The synovial observer module estimates the rotor position information in the medium and high speed domain, including the following steps:

[0150] B1: If Figure 5 As shown, the improved super-helical synovial observer obtains the response current I from the motor drive module in the αβ two-phase stationary coordinate system. α , I β and the response voltage U α 、U β , and then the synovial equivalent feedback signal is obtained by integrating the hyperbolic tangent function And the equivalent feedback signal passing it to the second phase-locked loop unit;

[0151] In the αβ two-phase stationary coordinate system, the mathematical model of the permanent magnet synchronous motor is usually expressed as:

[0152]

[0153] where e α and e β It represents the back electromotive force component of the permanent magnet synchronous motor in the αβ two-phase stationary coordinate system.

[0154] According to the above mathematical model, an improved super-helical sliding film observer with adaptive inductance parameters based on equivalent feedback is designed to further eliminate the chattering phenomenon of the sliding film observer method. Figure 11 As shown in the figure, after the hyperbolic tangent function (Tanh) is introduced to replace the sign function, the chattering phenomenon of the equivalent back electromotive force is significantly improved.

[0155] Introducing the hyperbolic tangent function (Tanh) instead of the sign function, we have:

[0156]

[0157] in

[0158] At this time, the equivalent feedback idea is introduced, assuming is the equivalent feedback signal. By eliminating the noise in the above formula through integration operation, we have:

[0159]

[0160] According to the above formula, the improved super-spiral sliding film observer with adaptive inductance parameters based on equivalent feedback is designed as follows (N is the feedback gain):

[0161]

[0162] According to the improved supercoil SMO constructed above, the parameter L s However, since electric vehicles often require higher load capacity, they will need a larger I q Current, the magnetic saturation effect generated by the large current will make the inductor L s Rapidly decreases, affecting the accuracy of the results, so according to L s The change curve of L s According to the three load conditions of electric vehicles: low, medium and high, the adaptive inductor parameters are designed To correct L s The impact of errors:

[0163] Adaptive inductance parameters according to motor Change curve, such as Figure 6 As shown, the electric vehicle operating conditions are divided into low load area, medium load area and high load area, among which the low load area Approximately the original L s , medium and high load area The gains K1 and K2 are taken according to the average slope after the divided areas, and the following formula is obtained:

[0164]

[0165] Among them, K1 and K2 are the average change gains of medium load and high load, according to L s Get the change curve diagram.

[0166] Set the adaptive inductor parameters The improved super spiral synovial membrane observer is introduced. The structure of the entire synovial membrane observer is as follows Figure 5 As shown, the feedback gain M, N and the synovial gain K are continuously adjusted according to the back electromotive force effect. The model expression is as follows:

[0167]

[0168] in, represents the α-axis observation current; represents the β-axis observation current; R s Indicates stator resistance; U α Indicates the α-axis voltage; U β represents the β-axis voltage; It represents the synovial equivalent feedback of the α-axis; Indicates the equivalent feedback of the synovial membrane on the β axis; N indicates the feedback gain; S α represents the sliding mode term of the α axis; S β represents the sliding mode term of the β axis; I αIndicates the actual current of α axis; I β Indicates the actual current of β axis; Represents the adaptive inductor parameter.

[0169] B2: The logic block diagram of the second phase-locked loop unit is as follows Figure 7 As shown, the second phase-locked loop unit is based on the equivalent feedback signal Adjust error signal S err , and output error signal S err Rotor position information when it approaches 0 For the weighted switching module, the expression is:

[0170] The second phase-locked loop unit is used to converge and determine the rotor position. According to the mathematical model of the permanent magnet synchronous motor and the improved super-helical sliding film observer with adaptive inductance parameters based on equivalent feedback, the equivalent back electromotive force is the same as the back electromotive force, and there is also a relationship:

[0171]

[0172] Among them, ψ f represents magnetic linkage;

[0173] Designing a Phase-Locked Loop like Figure 6 As shown, when S err When the phase-locked loop PI adjusts to 0, there is Accurate prediction of the rotor position is achieved.

[0174] The weighted switching module is also connected to the encoder of the permanent magnet synchronous motor. The weighted switching module detects the working status of the encoder in real time. When the encoder is detected to be disconnected or faulty, the weighted switching module seamlessly enables the high-frequency injection module and the synovial observer module to detect the rotor position of the permanent magnet synchronous motor.

[0175] like Figure 9 As shown in the figure, during normal operation of the electric vehicle motor, an encoder is used to provide rotor position information for vector control, while maintaining the computational operation of the sensorless algorithm. The system monitors the encoder's operating status in real time. If a fault such as a wire break occurs, an encoder fault interrupt is triggered, the current motor status (speed, angle) is saved, and the saved status is compared with the current status in sensorless control mode to further determine the encoder's condition. If an encoder wire break or fault is determined, encoder signal acquisition is stopped, seamlessly switching to sensorless mode, and then entering the corresponding sensorless control mode based on the current speed.

[0176] The present invention provides a full-speed dual-mode redundant fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle. The method adopts a motor FOC control system with a full-domain MAP current distribution strategy and has a dual-mode switching function for position detection. It can prevent the dangers caused by encoder failure while meeting the efficient operation of electric vehicles in multiple working conditions.

[0177] The present invention adopts dual-mode redundant control combining dynamic amplitude high-frequency injection method and improved super-spiral sliding film observer with adaptive inductance parameters to expand the operating speed range of the backup system to the full speed domain. The final rotor position prediction effect is as follows: Figure 12 As shown in the figure, the predicted rotor position and the actual rotor position are basically consistent, with the maximum error below 0.8°, which occurs in the switching speed domain of weighted switching. The control effect after the system speed is closed is as follows: Figure 13 As shown, the maximum error from the target speed is within 50RPM, which meets the requirements for normal operation of the electric vehicle and improves the safety and robustness of the electric vehicle motor system.

[0178] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A full-speed domain dual-mode redundant fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle, characterized in that: The following steps are involved: S1: Constructing a full-speed domain dual-mode redundant fault-tolerant control system for a permanent magnet synchronous motor, wherein the full-speed domain dual-mode redundant fault-tolerant control system is provided with a motor drive module, the motor drive module is connected to a high-frequency injection module, a synovial observer module and a weighted switching module, and the high-frequency injection module and the synovial observer module are respectively connected to the weighted switching module; S2: The weighted switching module obtains speed information ω of the permanent magnet synchronous motor in real time, and determines the operating speed range of the permanent magnet synchronous motor according to the speed information ω; S3a: When the motor speed ω<the speed lower limit threshold ω1, the permanent magnet synchronous motor operates in the zero low speed domain. The high frequency injection module injects a dynamic amplitude high frequency square wave signal into the d axis according to the load condition, extracts the rotor position information in the zero low speed domain, and transmits it to the weighted switching module; the weighted switching module converts the high frequency injection method weight w h Set to 1 to output the rotor position information to the motor drive module; S3b: When the motor speed ω> the speed upper limit threshold ω2, the permanent magnet synchronous motor operates in the medium and high speed domain, and the synovial observer module extracts the back electromotive force based on the improved super-helical synovial algorithm with adaptive inductance parameters, estimates the rotor position information in the medium and high speed domain, and transmits it to the weighted switching module; the weighted switching module converts the synovial observer weight w s Set to 1 to output the rotor position information to the motor drive module; S3c: When the motor speed ω∈[ω1,ω2], the permanent magnet synchronous motor operates in the switching speed domain, and the weighted switching module dynamically distributes the rotor position information from the high-frequency injection module and the sliding film observer module according to the motor speed ω. The weighted switching module allocates the high-frequency injection weight w h Set to Synovial observer weight w s Set to n is the speed, and the comprehensive output rotor position information is given to the motor drive module; S4: The motor drive module implements full-speed domain dual-mode redundant fault-tolerant control of the permanent magnet synchronous motor according to the motor rotor position information.

2. The full-speed domain dual-mode redundant fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle according to claim 1, characterized in that: The high-frequency injection module is provided with a high-frequency signal generating unit, a high-frequency signal processing unit, a PARK conversion unit, a first phase-locked loop unit and a polarity identification unit; The high-frequency signal processing unit is provided with a differential operator and an envelope detector, the input end of the differential operator is connected to the motor drive module, the output end of the differential operator is connected to the first phase-locked loop unit via the envelope detector, and the output end of the differential operator is also connected to the polarity identification unit via the PARK conversion unit; the high-frequency signal generation unit is connected to the motor drive module.

3. The full-speed domain dual-mode redundant fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle according to claim 2, characterized in that: The high-frequency injection module extracts rotor position information in the zero-speed domain. The specific steps are as follows: A1: The high-frequency signal generation unit dynamically adjusts the square wave amplitude according to the load conditions Inject a dynamic amplitude high frequency square wave signal into the d-axis; A2: The differential operator obtains the response current I from the motor drive module in the αβ two-phase stationary coordinate system α , I β , and calculate and separate the high-frequency response current component I in the response current αhf , I βhf , the calculation expression is as follows: I αhf =(I α -I α z -1 ) / 2 I βhf =(I β -I β z -1 ) / 2 Among them, z -1 Indicates a delay of one cycle; The differential operator converts the high frequency response current component I αhf , I βhf Passed to the envelope detector and PARK transformation unit; A3: The envelope detector responds to the high frequency current component I αhf , I βhf Perform envelope detection to obtain the high-frequency response component envelope I αh , I βh , the expression is as follows: in, Indicates the amplitude of high-frequency square wave; The envelope detector envelops the high frequency response component I αh , I βh passing it to the first phase-locked loop unit; The first phase-locked loop unit is configured to generate a high-frequency response component envelope I αh , I βh , the rotor position information is obtained through PI regulation calculation The calculation expression is as follows: After simplification and per-unit processing, it is expressed as: Among them, L q Indicates the d-axis inductance, L d represents the q-axis inductance, ε represents the rotor position error, and the rotor position information when ε approaches 0 is finally output after PI adjustment A4: The PARK conversion unit converts the high frequency response current component I αhf Converted into d-axis high-frequency response current I dh , and passed to the polarity identification unit; The polarity identification unit uses the principle of magnetic saturation effect to inject a high-frequency voltage U into the rotor during the initial position check. dh , change the d-axis voltage bias direction and compare the d-axis high-frequency response current amplitudes of the two cases at the same time to determine the rotor polarity; When the forward bias amplitude is greater than the reverse bias amplitude, the obtained position is the rotor N pole, and the high-frequency injection module directly outputs the rotor position information For the weighted switching module, the expression is: |I dh (U dh )|>|I dh (-U dh )|, then When the forward bias amplitude is less than the reverse bias amplitude, the obtained position is the rotor S pole, and the high frequency injection module is used to calculate the rotor position information. After position compensation, it is passed to the weight switching module. The expression is as follows: |I dh (U dh )|<|I dh (-U dh )|, then 4. The full-speed domain dual-mode redundant fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle according to claim 3, characterized in that: The high frequency injection module is based on the q-axis current I q The load condition is divided into three intervals, namely low load area, medium load area and high load area; When the q-axis current I q <20A is the low load area, and the high frequency square wave amplitude corresponding to the low load area When the q-axis current 20A≤I q ≤40A is the medium load area, and the high-frequency square wave amplitude corresponding to the medium load area When the q-axis current I q >40A is the high load area, and the high-frequency square wave amplitude corresponding to the high load area 5. The full-speed domain dual-mode redundant fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle according to claim 1, characterized in that: The synovial film observer module is provided with an improved super-spiral synovial film observer with adaptive inductance parameters and a second phase-locked loop unit. The input end of the improved super-spiral synovial film observer is connected to the motor drive module, and the output end of the improved super-spiral synovial film observer is connected to the weighted switching module via the second phase-locked loop unit.

6. The full-speed domain dual-mode redundant fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle according to claim 5, characterized in that: The synovial observer module estimates the rotor position information in the medium and high speed domain, including the following steps: B1: The improved super-spiral synovial observer obtains the response current I from the motor drive module in the αβ two-phase stationary coordinate system α , I β and the response voltage U α 、U β , and then the synovial equivalent feedback signal is obtained by integrating the hyperbolic tangent function And the equivalent feedback signal passing it to the second phase-locked loop unit; The mathematical model expression of the improved super spiral synovial observer is: in, represents the α-axis observation current; represents the β-axis observation current; R s Indicates stator resistance; U α Indicates the α-axis voltage; U β represents the β-axis voltage; It represents the synovial equivalent feedback of the α-axis; Indicates the equivalent feedback of the synovial membrane on the β axis; N indicates the feedback gain; S α represents the sliding mode term of the α axis; S β represents the sliding mode term of the β axis; I α Indicates the actual current of α axis; I β Indicates the actual current of β axis; Represents the adaptive inductance parameter; Among them, L s is the inductance parameter; K is the sliding film gain; K1 is the average change gain of medium load; K2 is the average change gain of high load, Tanh() is the hyperbolic tangent function, Equivalent feedback signal It is generated by the synovial term through integration operation and is expressed as follows: Where, M is the feedback gain; B2: The second phase-locked loop unit generates an equivalent feedback signal according to the Adjust error signal S err , and output error signal S err Rotor position information when it approaches 0 For the weighted switching module, the expression is:

7. The full-speed domain dual-mode redundant fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle according to claim 1, characterized in that: The weighted dynamic weight allocation strategy of the weighted switching module in each operating speed domain is as follows: Zero low-speed domain: high-frequency injection method weight w h =1, synovial observer weight w s =0; weighted switching module outputs rotor position information Switching speed domain: High frequency injection method weight Synovial observer weight Where n is the speed, ω = 2πn; the weighted switching module outputs the rotor position information Medium and high speed domain: high frequency injection method weight w h =0, synovial observer weight w s =1; weighted switching module outputs rotor position information 8. The full-speed domain dual-mode redundant fault-tolerant control method for a permanent magnet synchronous motor of an electric vehicle according to claim 1, characterized in that: The weighted switching module is also connected to the encoder of the permanent magnet synchronous motor. The weighted switching module detects the working status of the encoder in real time. When the encoder is detected to be disconnected or faulty, the weighted switching module seamlessly enables the high-frequency injection module and the synovial observer module to detect the rotor position of the permanent magnet synchronous motor.

Images