Rotary transformer observation mechanism for pure electric commercial vehicle

Abstract

The invention relates to the technical field of vehicle control, and discloses a pure electric commercial vehicle resolver observation mechanism which comprises the following steps: in a whole vehicle driving state, judging whether to enter an observation mode or not, if the rotating speed of a motor is greater than an observation threshold value, entering the observation mode; if the rotating speed of the motor is smaller than the observation threshold, the observation mode is not entered or quitted; after entering the observation mode, calculating according to an LO observation model to obtain a back electromotive force observed quantity; inputting the back electromotive force observed quantity into the phase-locked loop model, and calculating to obtain a target observed quantity; meanwhile, a real signal is acquired and decoded by the rotary transformer; the target observed quantity and the real signal both comprise an angle and an angular velocity; and comparing the real signal with the target observed quantity to judge whether fault early warning is needed or not and judge whether the real signal meets the observation expectation or not. The real state of the driving system can be accurately observed, the robustness and reliability are high, and the operation efficiency is high.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle control, and particularly relates to a resolver observation mechanism for pure electric commercial vehicles. Background Art

[0002] As one of the core sensors of the drive system of pure electric vehicles, the resolver's function is to obtain the motor rotor angle and speed information in real time by analyzing the excitation signal and the feedback sine / cosine signals, providing key inputs for the vector control algorithm. The controller of the drive system decodes by collecting the resolver electrical signals to obtain control information such as the angular velocity and angle of the drive motor, and then adjusts the motor torque output according to the vehicle's intention to meet the normal driving requirements of the vehicle.

[0003] At present, the resolver observation technology for pure electric commercial vehicles has developed to a certain extent. Some advanced technologies can collect resolver electrical signals more accurately and perform preliminary analysis and processing. There are also various types of resolver observation systems on the market, which meet the needs of different vehicle models and application scenarios to a certain extent. However, the existing technologies still have many deficiencies. For example, the existing resolver observation systems have poor adaptability under complex working conditions. In the field of commercial vehicles, vehicles often face complex working conditions (such as heavy load climbing, frequent start-stop, long-distance high-speed driving, etc.), and the requirements for the dynamic response accuracy of motor torque, system stability, and energy efficiency are much higher than those of passenger vehicles. However, resolver signals are easily affected by factors such as electromagnetic interference, mechanical vibration, and sensor aging, resulting in signal errors or fluctuations, which cause the cumulative error of the decoded angle, and then lead to motor torque fluctuations, reduce the response speed and torque tracking accuracy of the system, and further affect the vehicle's overall control effect. For example, when the angle deviation exceeds ±2°, the motor output torque error can reach more than 5%, directly affecting the vehicle's power performance and driving range; while signal jumps or noise interference may trigger the protection mechanism of the control unit, causing safety hazards such as power interruption.

[0004] Therefore, designing a highly reliable and robust resolver observation mechanism plays an irreplaceable role in improving the control accuracy of the drive system of commercial vehicles and ensuring functional safety under complex working conditions. Summary of the Invention

[0005] The present invention aims to provide a resolver observation mechanism for pure electric commercial vehicles, which can accurately observe the true state of the drive system, has strong robustness and reliability, and high operating efficiency.

[0006] The basic solution provided by the present invention is as follows: A resolver observation mechanism for pure electric commercial vehicles, including the following steps:

[0007] When the vehicle is in a driving state, it is judged whether to enter the observation mode. If the motor speed is greater than the observation threshold, the observation mode is entered. If the motor speed is less than the observation threshold, the observation mode is not entered or exited.

[0008] After entering the observation mode, the back electromotive force observation value is calculated according to the LO observation model. The back electromotive force observation value is input into the phase-locked loop model, and the target observation value is calculated. At the same time, the real signal is collected and decoded by the resolver. Both the target observation value and the real signal include the angle and the angular velocity.

[0009] The real signal is compared with the target observation value to judge whether the angular acceleration difference between the two is greater than the first fault threshold. If so, a fault warning is output. If not, it is judged whether the angular velocity difference between the two is greater than the second fault threshold. If so, a fault warning is output. If not, it is judged whether the position increment difference between the two is greater than the third fault threshold. If so, a fault warning is output. If not, it is determined that the real signal meets the observation expectation, and the real signal is output to the vehicle control loop.

[0010] Further, when the vehicle meets the power-on condition and the high-voltage system is in the ready state, it is determined that the vehicle can enter the driving state.

[0011] Further, in the LO observation model, the voltage of the vehicle motor in the stationary coordinate system is set as the real input, and the current of the vehicle motor in the stationary coordinate system is set as the real output.

[0012] Further, after the back electromotive force observation value is calculated, it is also subjected to low-pass filtering processing.

[0013] Further, during the operation of the vehicle, with T as the judgment period, the comparison between the real signal and the target observation value is cyclically triggered.

[0014] Further, the value of T is calibrated to be 1 to 5 sampling periods. The value of the sampling period is calibrated and selected according to the characteristics of the vehicle system, according to the DTC fault definition level, or according to experimental data.

[0015] Further, when a fault warning is output, the corresponding fault warning flag bit is set to 1, and the fault warning information is reported to the vehicle control system.

[0016] Further, when setting the first fault threshold, the second fault threshold, and the third fault threshold, based on comprehensive factors, the error between the observation system and the actual system is calibrated for fault segmentation, so as to select the first fault threshold, the second fault threshold, and the third fault threshold in segments. The comprehensive factors include system response time, system reception time, signal resolution, and noise.

[0017] The working principle and advantages of the present invention are as follows:

[0018] A resolver observation mechanism for a pure electric commercial vehicle according to the present invention can accurately observe the true state of the drive system, has strong robustness and reliability, and has a high operating efficiency. The key points are as follows:

[0019] First, this solution comprehensively applies the LO observer and the phase-locked loop algorithm. Without adding new sensors or other hardware, it can use the output of the real system to perform error feedback correction to construct a closed-loop observer, so that it can finally accurately observe the true state of the system. Specifically, during the process of the dynamic driving speed of the whole vehicle reaching the speed threshold (i.e., the observation threshold), based on the LO observer, using the input and output of the voltage and current signals of the motor system stator in the α-β stationary coordinate system and the dynamic model, the state of the resolver can be accurately estimated, and it can be judged whether there are errors or fluctuations in the true output (such as signal distortion, drift, etc.). Furthermore, it can assist the vehicle control system to handle the fluctuation situation and correct the error in time, and can effectively avoid the further expansion of the fault or causing the system to shut down.

[0020] Moreover, compared with the prior art solutions that use the LO observer or the phase-locked loop algorithm alone, this solution has higher stability. Since the LO observer is prone to steady-state errors when calculating the back electromotive force observation value, and the single phase-locked loop algorithm is prone to lock loss under low-speed conditions, in this solution, the back electromotive force observation value output by the LO observer is selected as the input of the phase-locked loop algorithm. Through the coupling of the algorithm model, the limitations of a single algorithm can be effectively broken through, the observation error can be reduced, and the observation efficiency can be improved.

[0021] Second, this solution is provided with a multi-layer progressive fault diagnosis architecture, adopting a three-level judgment mechanism of angular acceleration difference, angular velocity difference, and position increment difference. Compared with single-threshold detection (such as only the angle difference > 0.2 rad), the fault recognition accuracy can be greatly improved. And this solution filters transient interference through the angular acceleration difference, which can reduce the false alarm rate of faults. At the same time, the hierarchical threshold setting can assist in distinguishing different fault types such as mechanical jamming (sudden change in angular acceleration) and signal decoding error (angular velocity drift), and has strong functionality. In addition, this solution also specially judges the position increment difference. By comparing the cumulative error between the observed value and the true value, the cumulative error at low speed can be effectively identified, which helps to improve the fault detection rate.

[0022] Thirdly, this solution has a high degree of generality and adaptability. Specifically, since pure electric commercial vehicles of different brands and models have significant differences in drive system structure, electrical parameters, etc., it is required that the resolver observation mechanism has a high degree of generality and adaptability, and can be flexibly adjusted and optimized according to the characteristics of different vehicles. This is also one of the R & D difficulties of the resolver observation mechanism. In response to this, this solution can calibrate relevant thresholds, enabling the fault diagnosis architecture to match different vehicle characteristics accordingly.

[0023] Fourthly, this solution has strong dynamic real-time monitoring capabilities and high fault response efficiency. This solution establishes a dynamic mapping relationship between the speed threshold and the observation mode by dynamically judging whether the motor speed reaches the observation threshold, and triggers the observation mode only during driving, solving the problem that traditional static detection cannot capture the distortion of the resolver signal under dynamic working conditions. For example, when a commercial vehicle climbs a slope or accelerates suddenly, the sudden change in the motor load is likely to cause an increase in the noise of the resolver signal. Existing technologies mostly rely on fixed-period detection (such as once every 10 seconds), resulting in detection blind spots. However, this solution can achieve millisecond-level response through real-time speed threshold judgment, significantly improving system safety.

[0024] Moreover, there is no control interruption problem in this solution (traditional solutions need to switch to a backup sensor after detecting a fault, which is prone to cause a 10 - 20ms control interruption). This technology directly outputs the real resolver signal to the control loop in a non-fault state through "observation expectation compliance verification", with extremely small signal switching delay, effectively avoiding the driving force fluctuation caused by signal switching. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic flowchart of the first embodiment of a resolver observation mechanism for a pure electric commercial vehicle according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] The following is a more detailed description through specific embodiments:

[0027] Embodiment 1

[0028] The embodiment is basically as shown in the attached Figure 1 figures: A resolver observation mechanism for a pure electric commercial vehicle includes the following steps:

[0029] When the vehicle is in a driving state, it is judged whether to enter the observation mode - if the motor speed is greater than the observation threshold, enter the observation mode; if the motor speed is less than the observation threshold, do not enter the observation mode or exit the observation mode (that is, when the vehicle is in the observation mode and the motor speed is less than the observation threshold, it is to exit the observation mode).

[0030] Here, the observation threshold can be determined by calibration. When the whole vehicle meets the power-on condition and the high-voltage system is in the ready state, it is determined that the vehicle can enter the driving state.

[0031] After entering the observation mode, the back electromotive force observation quantity is calculated according to the LO observation model; the back electromotive force observation quantity is input into the phase-locked loop model, and the target observation quantity is calculated; at the same time, the real signal is collected and decoded by the resolver; both the target observation quantity and the real signal include the angle and the angular velocity.

[0032] In the LO observation model, it is set that the voltage of the vehicle motor in the stationary coordinate system is the real input, and the current of the vehicle motor in the stationary coordinate system is the real output.

[0033] Specifically, according to the voltage equation of the motor in the stationary coordinate system (α-β coordinate system), we can get:

[0034] u α is the α-axis voltage in the stationary coordinate system; u β is the β-axis voltage in the stationary coordinate system; R is the winding resistance; i β is the β-axis current in the stationary coordinate system; i α is the α-axis current in the stationary coordinate system; L s is the winding inductance (surface-mounted); if it is an embedded motor, the inductances in the stationary coordinate system are L β 、L α ; e β is the β-axis back electromotive force in the stationary coordinate system; e α is the α-axis back electromotive force in the stationary coordinate system.

[0035] After calculating the back electromotive force observation quantity, it is also subjected to low-pass filtering, which helps to filter out high-frequency interference in the signal data, improve the quality and reliability of the signal; it can also filter out system interference.

[0036] Compare the real signal with the target observation quantity. According to the dynamic characteristics of the system, judge whether the angular acceleration difference between the two is greater than the first fault threshold. If so, output a fault warning; if not, judge whether the angular velocity difference between the two is greater than the second fault threshold. If so, output a fault warning; if not, judge whether the position increment difference between the two is greater than the third fault threshold. If so, output a fault warning; if not, it is determined that the real signal meets the observation expectation, and the real signal is output to the vehicle control loop.

[0037] Here, this solution adopts a three - level progressive fault judgment logic. Among them, the judgment based on the angular velocity difference can capture sudden - change faults (such as instantaneous interference of the resolver decoding chip, signal jumps caused by magnetic circuit saturation, etc.); the judgment based on the angular velocity difference can identify continuous deviations (such as periodic errors caused by mechanical structures, parameter drift errors, etc.); the judgment based on the position increment difference can detect cumulative errors (such as angular recurrence deviations caused by mechanical wear, cumulative integral errors of the decoding algorithm). This hierarchical diagnosis associates fault types with physical roots, can effectively reduce the false alarm rate (for example, instantaneous interference only triggers a warning instead of directly reporting an error), and at the same time can improve the missed - detection tolerance ability through multi - dimensional verification. The diagnostic accuracy is significantly improved compared with the single - threshold method.

[0038] When outputting a fault warning, mark the corresponding fault warning position as 1 and report the fault warning information to the vehicle control system.

[0039] During the operation of the vehicle, with T as the judgment period, the comparison between the real signal and the target observable is cyclically triggered. The value of T is calibrated to be 1 - 5 sampling periods; the value of the sampling period is calibrated according to the characteristics of the vehicle system, calibrated according to the DTC fault definition level, or calibrated according to experimental data. Specifically, in practical applications, for serious faults or faults that may affect system safety, a shorter judgment period (1 - 2 sampling periods) should be selected; for minor faults or faults that do not affect system safety, a slightly longer judgment period (3 - 5 sampling periods) can be selected, while considering system resource limitations and avoiding too many and too short judgment periods.

[0040] When setting the first fault threshold, the second fault threshold, and the third fault threshold, based on comprehensive factors, the error between the observation system and the actual system is calibrated for fault segmentation to select the first fault threshold, the second fault threshold, and the third fault threshold by segments; the comprehensive factors include system response time, system reception time, signal resolution, and noise. Such a setting can fully meet the fault judgment requirements and enhance the robustness and reliability of the system.

[0041] A resolver observation mechanism for a pure - electric commercial vehicle provided in this embodiment can accurately observe the real state of the drive system, has strong robustness and reliability, and high operating efficiency.

[0042] Embodiment Two

[0043] A resolver observation mechanism for a pure - electric commercial vehicle adjusts the phase - locked loop model on the basis of Embodiment One.

[0044] After entering the observation mode, calculate the back - electromotive - force observable according to the LO observation model; input the back - electromotive - force observable into the phase - locked loop model and calculate the target observable; specifically, it includes the following sub - steps:

[0045] The back electromotive force observation value e is calculated based on the LO observation model α 、e β , and through Clarke transformation, the e d 、e q components in the rotating coordinate system are obtained.

[0046] Calculate the deviation between e q and the reference value (such as 0) to obtain the phase error Δθ; Δθ = arctan(e q / e d ).

[0047] Set the gain coefficient Kp of the phase-locked loop model, and establish the dynamic relationship between the gain coefficient Kp and the angle observation error Δθ;

[0048]

[0049] In this embodiment, the base value K p_base = 0.5, the error weight coefficient α = 0.8, and the error change rate weight β = 0.3. When the angle error Δθ > 0.2°, Kp is automatically increased to more than 1.2; when Δθ > 0.5°, the upper limit of Kp is locked at 2.0.

[0050] Based on the gain coefficient, perform phase-locked loop operation according to the following formula, and output the target observation value.

[0051]

[0052] Among them, is the estimated angular velocity of the motor, and θ test is the estimated angle of the motor; K i is the integral coefficient, which is fixed at 0.1 in this embodiment to avoid integral saturation.

[0053] The resolver observation mechanism of a pure electric commercial vehicle provided in this embodiment, compared with Embodiment 1, the LO observation model and the phase-locked loop model can be better coupled, and can better meet the adaptability requirements of dynamic working conditions, and has higher operating efficiency (such as fast response to sudden load changes).

[0054] The above are only embodiments of the present invention. Common knowledge such as specific structures and characteristics known in the art has not been described in detail herein. Those of ordinary skill in the art know all the common general technical knowledge in the technical field to which the invention pertains before the filing date or the priority date, are able to obtain all the prior art in this field, and have the ability to apply conventional experimental means before this date. Those of ordinary skill in the art can, under the inspiration given in this application, combine their own abilities to complete and implement this solution. Some typical well-known structures or well-known methods should not become an obstacle for those of ordinary skill in the art to implement this application. It should be noted that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can still be made, and these should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicality of the patent.

Claims

1. A resolver observation mechanism for a pure electric commercial vehicle, characterized in that: The following steps are involved: When the vehicle is in driving state, determine whether to enter the observation mode - if the motor speed is greater than the observation threshold, enter the observation mode; if the motor speed is less than the observation threshold, do not enter or exit the observation mode; After entering the observation mode, the back electromotive force observation quantity is calculated according to the LO observation model; The back electromotive force observation is input into the phase-locked loop model, and the target observation is calculated; at the same time, the real signal is collected and decoded by the rotary transformer; the target observation and the real signal both include angle and angular velocity; Compare the real signal with the target observation, and determine whether the angular acceleration difference between the two is greater than the first fault threshold. If so, output a fault warning; if not, determine whether the angular velocity difference between the two is greater than the second fault threshold. If so, output a fault warning; if not, determine whether the position increment difference between the two is greater than the third fault threshold. If so, output a fault warning; if not, determine that the real signal meets the observation expectations, and output the real signal to the vehicle control loop.

2. A resolver observation mechanism for a pure electric commercial vehicle according to claim 1, characterized in that: When the vehicle meets the power-on conditions and the high-voltage system is in the ready state, it is determined that the vehicle can enter the driving state.

3. A resolver observation mechanism for a pure electric commercial vehicle according to claim 1, characterized in that: In the LO observation model, the voltage of the vehicle motor in the stationary coordinate system is set as the real input, and the current of the vehicle motor in the stationary coordinate system is set as the real output.

4. A resolver observation mechanism for a pure electric commercial vehicle according to claim 1, characterized in that: After the back electromotive force observable is calculated, it is also subjected to low-pass filtering.

5. A resolver observation mechanism for a pure electric commercial vehicle according to claim 1, characterized in that: During the operation of the vehicle, T is used as the judgment period to cyclically trigger the comparison between the real signal and the target observation.

6. A resolver observation mechanism for a pure electric commercial vehicle according to claim 5, characterized in that: The value of T is calibrated to 1 to 5 sampling cycles; the value of the sampling cycle is selected based on the calibration of the vehicle system characteristics, based on the calibration of the DTC fault definition level, or based on the calibration of experimental data.

7. A resolver observation mechanism for a pure electric commercial vehicle according to claim 1, characterized in that: When a fault warning is output, the corresponding fault warning mark position is set to 1, and the fault warning information is reported to the vehicle control system.

8. A resolver observation mechanism for a pure electric commercial vehicle according to claim 1, characterized in that: When setting the first fault threshold, the second fault threshold and the third fault threshold, based on comprehensive factors, the error between the observation system and the actual system is calibrated in sections to select the first fault threshold, the second fault threshold and the third fault threshold in sections; The comprehensive factors include system response time, system receiving time, signal resolution, and noise.

Images