Conveyance agv drive control method for transformer calibration pipeline

Abstract

This invention relates to a method for driving and controlling AGVs used in automated instrument transformer calibration production lines, belonging to the field of drive and control technology for automated instrument transformer calibration production lines. The invention addresses the problems of severe speed jitter in existing AGV drive systems during high-speed operation and the difficulty of traditional sliding mode observers in simultaneously managing convergence speed and steady-state jitter. This invention obtains a speed switching threshold through simulation of AGV drive system parameters. When the motor speed is less than the threshold, a photoelectric encoder and dual closed-loop vector control are used. When the motor speed is greater than or equal to the threshold, voltage and current are collected, and an improved sliding mode observer is used to establish a back electromotive force to extract rotor speed and position, replacing the signal collected by the photoelectric encoder for dual closed-loop vector control. The improved sliding mode observer uses the sigmoid function as the switching function and employs an adaptive variable speed exponential approach law. This invention is applicable to the drive and control of AGVs used in instrument transformer calibration production lines.

Description

Technical Field

[0001] This invention belongs to the field of automated calibration production line drive control technology for instrument transformers. Background Technology

[0002] Low-voltage current transformers (CTs) are core supporting measuring components in various low-voltage power supply systems, enabling measurement, control, and energy metering functions. Their performance directly determines the accuracy of measurement and the reliability of protection device operation. Given their critical role, current transformers are included in the national mandatory verification measuring instruments, requiring 100% full inspection throughout the entire process, including manufacturing, delivery, and acceptance.

[0003] Currently, with the deepening of the centralized management model of power grid enterprises, instrument transformer calibration has formed an integrated model of centralized bidding, centralized procurement, centralized inspection, and unified distribution by provincial grid enterprises, resulting in a significant increase in the number and scale of instrument transformer testing. Simultaneously, against the backdrop of rapid iteration in digital information technology, enterprises have an increasingly urgent need for product quality control and improved production efficiency. Against this backdrop, all relevant stakeholders, including power grid enterprises, electricity users, and instrument transformer manufacturers, are actively promoting the automation upgrade of instrument transformer testing, and low-voltage CT batch automated calibration production lines have been gradually put into practical operation.

[0004] exist Figure 12 In the automated batch verification production line of medium and low pressure CT scanners, the design of the Automated Guided Vehicle (AGV) for transporting turnover boxes is a key link to ensure the efficient and stable operation of the production line. Its operating accuracy and reliability directly affect the efficiency and quality of the entire verification process, and the core of the AGV's operating performance depends on its motor control scheme.

[0005] Currently, photoelectric encoders or sliding diaphragm observers based on constant velocity approaching rate are commonly used in the field for AGV motor drive control. However, these methods have the following shortcomings when adapted to the working conditions of AGVs in low-pressure CT calibration production lines:

[0006] (1) Severe speed jitter in the high-speed stage: The photoelectric encoder relies on discrete triggering, when Figure 2 When the rotational speed exceeds 1000 r / min, the signal update lags and the accuracy decreases, resulting in severe jitter of the speed waveform, which affects the safety of AGV high-speed operation and positioning accuracy.

[0007] (2) Inherent defects of the sensor: mechanical installation error, vibration, and electromagnetic interference can easily cause signal noise and jumps; the hardware sampling frequency and resolution have physical limits, which cannot match the real-time requirements of high-speed FOC.

[0008] (3) Shortcomings of traditional sliding mode observers: Figure 3 In the working principle of the Sliding Mode Observer (SMO), the convergence to the sliding surface adopts a constant-rate approach law and a sign function, which presents a contradiction between the convergence speed and steady-state chattering. Under high-speed conditions, the estimation error is large and the dynamic response is slow. Summary of the Invention

[0009] This invention aims to address the problems of severe speed jitter in existing AGV drive systems during high-speed operation and the difficulty of traditional sliding mode observers in simultaneously considering convergence speed and steady-state jitter, resulting in poor control accuracy and stability. A method for controlling the drive of a conveyor AGV in a current transformer calibration production line is provided.

[0010] The AGV drive control method for a current transformer calibration production line according to the present invention includes:

[0011] Collect parameters of the AGV drive system and motor body; based on the parameters, perform dual closed-loop vector control simulation on the AGV drive system under the condition of motor speed acquisition by photoelectric encoder, and obtain the critical chattering speed corresponding to the control mode as the speed switching threshold.

[0012] The current operating speed of the motor is detected in real time, and it is determined whether the current speed of the motor is less than the speed switching threshold.

[0013] If so, a photoelectric encoder is used to collect the rotor speed and position of the motor, and the AGV drive system is controlled using a dual closed-loop vector control mode;

[0014] Otherwise, the voltage and current of the motor are collected in real time, and the back electromotive force of the motor is established using the voltage and current with an improved sliding mode observer, and the current motor rotor speed and position are extracted.

[0015] The calculated and extracted current motor rotor speed and position are used as feedback quantities to replace the motor rotor speed and position collected by the photoelectric encoder. This feedback is then connected to a dual closed-loop vector control loop to control the AGV drive system until the current motor speed is detected to drop below the speed switching threshold. In this case, the system switches to a control mode that uses the photoelectric encoder to collect motor speed data.

[0016] This invention achieves hybrid precision control of the BLDC motor across its entire speed range. For the actual operating conditions of the AGV in the low-pressure CT calibration production line, a segmented control strategy is adopted: when the motor rotor speed is below 890 r / min (corresponding to the AGV moving at low speed and stopping precisely near the calibration station, coordinating with the loading, unloading, and calibration of the low-pressure CT turnover box), an FOC + photoelectric encoder solution is used to ensure reliable connection between the low-pressure CT and the calibration equipment; when the speed exceeds 890 r / min (corresponding to the AGV's long-distance rapid transfer between different calibration areas on the production line), it automatically switches to the SMO algorithm, effectively solving problems such as speed jitter and accuracy reduction caused by the sampling characteristics of the photoelectric encoder at high speeds, ensuring the smoothness of the AGV's high-speed transport.

[0017] This invention makes dual improvements to the SMO (Sliding Mode Approach) mechanism. It employs adaptive sliding mode gain to balance convergence speed and chatter suppression, replacing the discontinuous sign function with a continuous sigmoid function. This effectively solves the inherent contradiction in traditional sliding mode approach laws between fast convergence and steady-state chattering: when the system state is far from the sliding surface, the adaptive gain increases accordingly, ensuring the system has a fast approach speed and dynamic response capability, quickly adapting to changes in operating conditions, and avoiding the impact of AGV start-up delay and speed switching lag on production line efficiency; when the system state approaches the sliding surface, the gain decreases adaptively, significantly reducing high-frequency chattering near the sliding surface while ensuring stable system convergence, ensuring smooth AGV operation and providing a stable working environment for low-pressure CT calibration. The comparison shows that the improved SMO scheme effectively suppresses speed chattering in the high-speed range, reducing chattering to 1.3 r / min, a 65% reduction compared to the original. At the same time, it reduces the rotor position estimation error, with the maximum error reduced by 0.31 rad and the steady-state error reduced from 0.13 to 0.09, achieving smooth and reliable control of BLDC across the entire speed range. Attached Figure Description

[0018] Figure 1 This is a flowchart of the method described in this invention;

[0019] Figure 2 This is a waveform diagram of the speed feedback for FOC control.

[0020] Figure 3 Here is a block diagram of the SMO structure;

[0021] Figure 4 The equivalent circuit model diagram of BLDC is shown below;

[0022] Figure 5 This is a graph showing the relationship between the rotor position and time in a traditional SMO.

[0023] Figure 6 To improve the SMO rotor position versus time curve;

[0024] Figure 7 A comparison chart of rotor position errors;

[0025] Figure 8 Comparison of feedback speed waveforms at 1000 r / min;

[0026] Figure 9 Comparison of velocity waveforms with increasing speed feedback;

[0027] Figure 10 A comparison chart of rotor speed errors;

[0028] Figure 11 A comparison chart of speed feedback waveforms across the entire speed range;

[0029] Figure 12 This is a schematic diagram of the calibration process for a low-pressure CT automated calibration system. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0031] Specific Implementation Method 1: Combination Figure 1 This embodiment describes an AGV drive control method for a current transformer calibration production line, comprising:

[0032] Collect parameters of the AGV drive system and motor body; based on the parameters, perform dual closed-loop vector control simulation on the AGV drive system under the condition of motor speed acquisition by photoelectric encoder, and obtain the critical chattering speed corresponding to the control mode as the speed switching threshold.

[0033] The current operating speed of the motor is detected in real time, and it is determined whether the current speed of the motor is less than the speed switching threshold.

[0034] If so, a photoelectric encoder is used to collect the rotor speed and position of the motor, and the AGV drive system is controlled using a dual closed-loop vector control mode;

[0035] Otherwise, the voltage and current of the motor are collected in real time, and the back electromotive force of the motor is established using the voltage and current with an improved sliding mode observer, and the current motor rotor speed and position are extracted.

[0036] The calculated and extracted current motor rotor speed and position are used as feedback quantities to replace the motor rotor speed and position collected by the photoelectric encoder. This feedback is then connected to a dual closed-loop vector control loop to control the AGV drive system until the current motor speed is detected to drop below the speed switching threshold. In this case, the system switches to a control mode that uses the photoelectric encoder to collect motor speed data.

[0037] Furthermore, in this invention, the speed switching threshold is 890 r / min.

[0038] Furthermore, this invention also includes converting the real-time acquired motor voltage and current to... The process of establishing a coordinate system, and the results obtained through this process The current and voltage in the coordinate system are the inputs of the improved sliding mode observer.

[0039] Furthermore, in this invention, the back electromotive force of the motor is:

[0040] ;

[0041] In the formula, It is the electric angular velocity of the motor. , For DC inductors and AC inductors, and They represent Current in coordinate system and This indicates the establishment of observations by the sliding mode observer. Back electromotive force in coordinate system; The position angle of the rotor, It is a magnetic flux.

[0042] Furthermore, in this invention, the reaching law of the improved sliding mode observer is:

[0043] ;

[0044] In the formula, ; ; >1, >1, , Represents the sliding surface matrix function;

[0045] ;

[0046] ;

[0047] In the formula, and These are the approach rates of the isotropic term and the exponential term, respectively; It is the rate of convergence. It is the sign function, and 'a' is the coefficient of the switching function;

[0048] ;

[0049] in, and These represent the current values ​​estimated by the observer. and These represent the α-axis stator current observation error and the β-axis stator current observation error, respectively.

[0050] Furthermore, in this invention, a phase-locked loop is used to obtain the back electromotive force of the motor. and Extract the rotor speed and position of the motor in the current AGV drive system.

[0051] In recent years, with the steady growth of electricity load demand, the market demand and production volume of low-voltage CT have increased significantly year by year. To meet the expanding market demand and address the problems of low efficiency, poor accuracy, and high labor intensity associated with traditional manual and semi-automatic verification, automated batch verification lines for low-voltage CT have been successively put into operation and continuously iterated and optimized. The efficient operation of these automated batch verification lines heavily relies on the stable and precise drive of AGVs (Automated Guided Vehicles) for transporting turnover boxes. As the core execution unit for transporting low-voltage CT turnover boxes within the production line, the AGV needs to complete operations such as low-speed stable docking and high-speed rapid transfer between different verification stations. The performance of its drive system directly determines the verification efficiency, verification accuracy, and operational reliability of the production line. By adding the AGV drive system of this invention to the existing automated low-voltage CT verification production line system, without large-scale modifications to the original production line, the operating efficiency of the transformer verification production line can be effectively improved, AGV downtime due to malfunctions can be reduced, and the system can be adapted to the large-scale and efficient requirements of automated batch verification of low-voltage CT.

[0052] The core advantage of this invention lies in achieving precise hybrid control of the BLDC motor across its entire speed range. For the actual operating conditions of the AGV in the low-pressure CT calibration production line, a segmented control strategy is adopted: when the motor rotor speed is below 890 r / min (corresponding to the AGV moving at low speed and precisely stopping near the calibration station, coordinating with the loading, unloading, and calibration of the low-pressure CT turnover box), an FOC + photoelectric encoder solution is used to ensure reliable connection between the low-pressure CT and the calibration equipment; when the speed exceeds 890 r / min (corresponding to the AGV's long-distance rapid transfer between different calibration areas on the production line), it automatically switches to the SMO algorithm, effectively solving problems such as speed jitter and accuracy reduction caused by the sampling characteristics of the photoelectric encoder at high speeds, ensuring the smoothness of the AGV's high-speed transport.

[0053] For the specific application scenarios and operating conditions of low-pressure CT batch automated verification production lines, the AGV (Automated Guided Vehicle) for transporting turnover boxes, as the core equipment for material transfer in the production line, must accurately stop or dock at designated verification stations and loading / unloading nodes to complete the automated loading and unloading tasks of the turnover boxes. Since even slight positional deviations in the low-pressure CT verification production line can lead to turnover box loading / unloading failures, the AGV's motor drive control system must possess sufficient control precision to ensure reliable speed adjustment and positioning control in every operation. Simultaneously, the performance of the motor drive control system directly determines the AGV's operational smoothness and speed stability. This is not only crucial for ensuring smooth transmission and precise docking of CT turnover boxes but also a core prerequisite for improving the overall operational efficiency of the low-pressure CT batch verification production line.

[0054] Considering the operational characteristics and cost control requirements of low-pressure CT batch automated verification production lines, the core drive motor of the AGV drive system is selected as a BLDC motor. The BLDC motor, in dual-closed-loop control mode, can obtain excellent speed and torque response curves, eliminating the need for a complex three-closed-loop control structure. This simplifies system design and reduces costs while ensuring control accuracy.

[0055] Based on the actual working conditions of the low-voltage CT automated inspection production line, the maximum weight of goods transported by the turnover box conveyor AGV is 200kg, while the AGV itself weighs 150kg (including all accessories and the battery). Therefore, the selection analysis and calculation of the drive motor are as follows:

[0056] (1) The diameter of the drive wheel is Let the maximum travel speed of the AGV be... =0.75m / s, therefore, the rotational speed of the drive wheel can be calculated as:

[0057] (1)

[0058] The reduction ratio of the motor's gearbox is: (2)

[0059] Therefore, the motor drive speed can be obtained as: (3)

[0060] When an AGV is traveling on a road, its drive system needs to provide sufficient driving force. Used to overcome frictional resistance Acceleration resistance Equal resistance.

[0061] ① Frictional resistance:

[0062] The AGV's wheel system is made of polyurethane (PU); the coefficient of dynamic friction between the PU drive wheel and the ground can be set to [value missing]. Based on this, its rolling friction force can be calculated:

[0063] (4)

[0064] In the formula, For the quality of the AGV itself, Maximum load capacity, Let be the acceleration due to gravity, and take . .

[0065] ② Acceleration resistance:

[0066] The calculation formula for the AGV's acceleration from a standstill to a preset speed during operation is as follows:

[0067] (5)

[0068] In the formula, acceleration =0.25m / .

[0069] The maximum resistance encountered by the car during driving is:

[0070] (6)

[0071] Therefore, the motor's output torque can be calculated:

[0072] (7)

[0073] In the formula, r is the radius of the drive wheel, and n is the number of drive motors. , Let the total transmission efficiency from the motor output shaft through the reducer to the drive wheel be taken as... Then the output torque .

[0074] According to the rated power of the motor With total driving force The relationship between them:

[0075] (8)

[0076] Then the output power .

[0077] Based on the above analysis and calculation results, and to meet the operational requirements of the AGV in the low-pressure CT batch automated verification production line, the selected motor in this study is a BLDC motor manufactured by Ningbo Zhongda Lide Transmission Equipment Co., Ltd., model Z4BLD200-24GU-20S, with a power of 200W, a rated voltage of 24V, a rated speed of 2000r / min, and a rated output torque of 0.77. .

[0078] To facilitate research on BLDC control methods and adapt to the control requirements of AGVs in low-voltage CT calibration production lines, this invention adjusts the mathematical model of the BLDC motor to obtain... Figure 4 Equivalent circuit. The equivalent circuit is adapted to the analysis scenario of AGV drive motor in low-voltage CT verification production line, which can simplify subsequent circuit characteristic analysis and motor performance calculation, and lay the foundation for the design and optimization of control algorithm.

[0079] To meet the control requirements of AGVs in low-pressure CT verification production lines and ensure their positioning accuracy and transport efficiency, this invention's BLDC motor drive scheme first employs a field-oriented control (FOC) algorithm to control the BLDC motor, and uses a photoelectric encoder to provide feedback on rotor position and speed. However, in the actual operation of low-pressure CT verification production lines, the AGV needs to flexibly switch speeds according to the verification rhythm. When the motor operates at high speed, the rotor position changes rapidly, causing the update frequency and accuracy of the photoelectric encoder output to be difficult to match the dynamic response requirements of the motor, ultimately leading to AGV jitter. Furthermore, low-pressure CT verification production lines contain numerous electrical devices with strong electromagnetic interference, making the photoelectric encoder susceptible to interference and signal loss. To address these issues with photoelectric encoders during high-speed motor operation, this invention considers the advantages of a sensorless motor motor (SMO) such as high-speed response, strong real-time performance, excellent anti-interference capabilities, and low cost and maintenance. Therefore, it selects an SMO to replace the photoelectric encoder at high speeds to perform sensorless FOC control of the BLDC motor.

[0080] The sliding mode observer is designed based on a mathematical model of the motor, combined with the operating parameters of the AGV drive motor in the low-pressure CT calibration production line, and then transformed through the Clarke transformation of the motor stator. The voltage and current in the coordinate system establish the back electromotive force of the motor, ensuring that the observer can operate based on real and accurate motor operating data. The motor voltage equation in the coordinate system is:

[0081] (9)

[0082] In the formula, It is the electric angular velocity of the motor. For stator winding resistance, , For direct-axis inductors and quadrature-axis inductors, and express System voltage in the coordinate system.

[0083] The BLDC used in this invention includes .in , Sliding mode observations are as follows:

[0084] (10)

[0085] The back electromotive force contains information about the electric angle and electric angular velocity, and the voltage equation is transformed into... The equation for the motor current in the coordinate system is:

[0086] (11)

[0087] In order to obtain , The sliding mode observer can be designed as follows:

[0088] (12)

[0089] In the formula, , The observer estimates the current value. , The control inputs to the observer are respectively, and can be represented as:

[0090] (13)

[0091] In the formula, It is a positive value for sliding mode gain. , It is an observation error, and the magnitude of the error directly affects the stability of AGV operation, which needs to be strictly controlled.

[0092] Therefore, the state equation for the stator current estimation error is:

[0093] (14)

[0094] Sliding mode gain of equation (13) The observed in determinant (12) , Will it converge to the actual value? , And how quickly it converges to the actual value. , In other words, whether the constructed observer system can be stabilized on the sliding surface. This, in turn, affects the operational stability of the AGV within the low-pressure CT verification production line. To determine the stability of the observer system, a sliding mode surface matrix function is first defined. :

[0095] (15)

[0096] According to Lyapunov's theorem and These represent the α-axis stator current observation error and the β-axis stator current observation error, respectively.

[0097] Based on the control stability requirements of the AGV drive motor in the low-pressure CT verification production line, the stability condition of the SMO can be derived as follows:

[0098] (16)

[0099] in, and These are the positive values ​​of the sliding mode gain h, respectively. Components and Quantity, For stator winding resistance, It is a symbolic function. , yes Observation error in the coordinate system.

[0100] That is, suitable This value allows the observed current to converge to the actual current, reducing the amplitude of the observed current's fluctuations near the actual current to a certain extent, thereby improving the accuracy of angle extraction. Furthermore... After satisfying equation (16), in equation (15) , It will converge to 0, that is, the current observation error is 0, at which point equation (14) can be simplified to:

[0101] (17)

[0102] Furthermore, the motor rotor speed and position required for FOC closed-loop control can be obtained by using a phase-locked loop (PLL).

[0103] Analysis of formula (16) shows that the rated current and peak current of the AGV drive motor in the low-voltage CT calibration production line have been determined according to the production line's transfer requirements. Therefore, the absolute value of the difference between the observed current and the actual current is... Since it is a bounded quantity, the gain of the sliding mode observer is... A lower bound exists. Therefore, through proper design... This allows the sliding mode observer system to eventually stabilize. However, further analysis of the convergence process reveals that the current algorithm has flaws and cannot meet the high-performance control requirements of pipelines.

[0104] If sliding mode gain If the requirement of equation (16) is satisfied, then This will cause the observed current Converging to the actual current That is, to make these two current variables converge to the sliding surface. Above. But after convergence to the sliding surface, the sliding controller input in equation (13) Due to current observation error There are positive and negative changes, after After symbolic function processing, Will Switching between them caused It is no longer continuous. The specific implementation process is as follows:

[0105] (1) If at a certain moment for (and at this time) (If the value is relatively large), according to equation (14), As the observed current becomes negative, it decreases over time, and The larger the current, the faster it decreases.

[0106] (2) Subsequently Switch to ,but It becomes a positive value, and the observed current increases again over time. Switch back to .

[0107] (3) The above process is repeated continuously, and the fluctuation amplitude of the observed current gradually decreases. However, after reaching the sliding surface, since the value of h is fixed, the observed current will eventually fluctuate slightly near the actual current. This fluctuation will be transmitted to the AGV drive system, causing the AGV to shake during operation, which will affect the transfer accuracy and efficiency of the low-voltage CT verification line.

[0108] In summary, the larger the value of h, the faster the system state converges to the sliding surface, enabling a rapid response to the AGV's speed switching requirements. However, the system will experience severe chattering after reaching the sliding surface, affecting the AGV's operational stability. At the same time, the sign switching function's step change from -1 to 1 causes the sliding motion trajectory to shuttle back and forth across the sliding switching surface, resulting in high-frequency chattering in the traditional SMO algorithm, which cannot meet the core requirements of "high-speed response and stable operation" for AGVs in low-pressure CT verification production lines.

[0109] To address the problems of high-speed chattering, sensor dependence, the contradiction between sliding mode observer convergence and chattering, and unreliable sensorless switching in existing technologies, this invention provides a FOC control method for a sliding mode observer (SMO) with improved reaching law. In the sliding mode observer, the design of the reaching law dominates the entire process of system state convergence to the sliding surface, and its architecture directly affects the control effect and the overall stability of the system. Traditional SMO algorithms use a constant-rate reaching law, the specific mathematical expression of which is:

[0110] (18)

[0111] In the formula, It is the rate of convergence. It is the rate of approach. It is a symbolic function.

[0112] To address the problems of traditional SMOs, the switching function is first optimized. This invention uses the following sigmoid function as the continuous switching function:

[0113] (19)

[0114] This function is strictly monotonically increasing and continuously differentiable throughout the real number domain, ensuring the continuity of the control system's state equations on the sliding surface. This not only suppresses chattering in the control system but also significantly improves the dynamic stability and robustness of the sliding mode observer.

[0115] To address the core technical requirements of rotor position and speed control in automated calibration lines for low-voltage current transformers, such as dynamic response, steady-state accuracy, and chatter suppression, this invention proposes a novel variable speed exponential approach law to optimize SMO performance. This solves the technical problem that traditional sliding mode control struggles to balance convergence speed and steady-state chatter in automated calibration scenarios, thereby improving the stability and accuracy of the calibration process.

[0116] This invention designs an adaptive gain to replace the original coefficients, enabling the SMO to adaptively adjust its convergence speed according to changes in the system operating state of the automated calibration line for low-voltage current transformers. This adapts to dynamic rotor speed switching and load fluctuations in the production line. The novel variable speed exponential approach law is as follows:

[0117] (20)

[0118] In the formula, ; ; , >1, .

[0119] In the new approach law expression, combined with the operating characteristics of the automated calibration pipeline for low-voltage current transformers, its adaptive gain adjustment mechanism is specifically manifested as follows: when the system's operating trajectory is far from the sliding mode surface... The gain coefficient of the constant velocity term approximately approaches 0, at which point it is b / This coefficient, at larger values, significantly amplifies the effect of the constant-velocity approaching term. Combined with the exponential approaching term, it can substantially increase the system's approach speed to the sliding surface. As the system approaches the sliding surface, the gain coefficient of the exponential term gradually transitions and stabilizes. The system primarily moves towards the sliding surface with a smooth, constant approach rate. And as... As the constant-rate gain coefficient approaches 0, it decreases continuously and smoothly, eventually converging to b / This adaptive gain adjustment mechanism can effectively reduce the high-frequency switching amplitude of the control input near the sliding surface, thereby significantly suppressing the chattering phenomenon inherent in sliding mode control.

[0120] Finally, to verify the stability and feasibility of the novel reaching rate described in this invention, the Lyapunov function is designed as follows:

[0121] (twenty one)

[0122] Differentiation yields:

[0123] (twenty two)

[0124] Using formula (18) Replacing it with a new type of exponential reaching law, we get:

[0125] (twenty three)

[0126] Therefore, there is According to the Lyapunov stability criterion, the novel variable speed exponential approach law proposed in this invention satisfies the sliding mode arrival condition. The SMO controller designed based on this approach law has good stability. Combined with the full speed range control requirements of the AGV drive motor in the low-pressure CT calibration production line, the following detailed analysis of the rotor position and speed control performance further verifies that the improved SMO can adapt to the long-term and stable operation requirements of the low-pressure CT calibration production line.

[0127] (1) Regarding rotor position control: The AGV needs to precisely stop at the calibration station to avoid AGV deviation due to rotor position estimation errors, which would affect the docking accuracy between the low-pressure CT and the calibration equipment. In the rotor position control stage of the low-pressure CT calibration production line, through comparison... Figure 5 , Figure 6 As can be seen, the SMO based on the novel approach law improvement of this invention significantly shortens the time for the estimated rotor position to converge to the actual position, exhibiting a faster dynamic response speed and the ability to quickly track the dynamic changes in the rotor speed of the motor in the production line; at the same time, the estimation curve of the improved SMO has a higher degree of overlap with the actual position curve, indicating that it has a stronger ability to track the rotor position in real time, and the obtained rotor position accuracy is higher, which can effectively improve the metrological accuracy of low-pressure CT automated verification and meet the requirements of the production line for the accuracy of verification data.

[0128] To further illustrate the superiority of the improved SMO proposed in this invention in the automated verification scenario of low-pressure CT, through... Figure 7A detailed comparative analysis of rotor position estimation errors reveals that traditional SMOs have a high initial position error peak of up to 0.65 rad during the startup phase, and the error convergence process is relatively slow. This can easily lead to AGV position deviation in the early stages of calibration, affecting the docking accuracy between the low-pressure CT and the calibration equipment, and consequently causing calibration data deviation. After entering the steady-state calibration condition, the position error continues to fluctuate around 0.13 rad, with a significant jitter phenomenon. This can cause slight AGV shaking, affecting the stability of calibration signal acquisition and reducing the reliability of calibration data.

[0129] In comparison, the improved SMO proposed in this invention exhibits superior estimation performance, accurately meeting the core requirements of accuracy and stability in low-pressure CT automated verification. It not only effectively suppresses the initial error peak to 0.34 rad, significantly reducing position deviation during startup and accelerating error convergence, but also enables the AGV to stabilize at the workstation more quickly, allowing the system to rapidly enter a stable verification state and effectively improving the verification efficiency of the production line. Under steady-state verification conditions, the position estimation error of the improved SMO can be stably controlled within 0.09 rad, effectively suppressing waveform jitter and ensuring smooth AGV docking. This provides a stable operating environment for low-pressure CT verification, guaranteeing the reliability and consistency of verification data.

[0130] (2) Regarding rotor speed control: The stability of the AGV drive motor's rotor speed directly affects the smoothness of low-pressure CT transport and the calibration accuracy. Speed ​​fluctuations can cause AGV vibration, potentially leading to collisions and damage to the low-pressure CT, and interfering with signal acquisition by the calibration equipment, thus reducing calibration accuracy. Simultaneously, the production line requires the AGV to rapidly switch speeds between different workstations, placing high demands on the dynamic response speed of the motor's rotor speed. Figure 8 It can be seen that when the rotor speed of the BLDC motor switches to 1000 r / min, the speed jitter amplitude of the SMO before and after the improvement is 3.7 r / min and 1.3 r / min, respectively, with a jitter amplitude reduction of 65%, and the jitter phenomenon is significantly suppressed. The system's stable convergence time is shortened from 0.053 s of the traditional SMO to 0.034 s, and the system's convergence speed and dynamic response performance are significantly improved. Under this condition, the speed overshoot corresponding to the photoelectric encoder is 56 r / min. Compared with the traditional SMO, the overshoot of the improved SMO is further reduced from 43 r / min to 31 r / min, a reduction of 45%. Excessive speed overshoot can cause sudden changes in AGV speed, which can easily cause low-pressure CT shaking and deviation, affecting the safety of transportation and the accuracy of subsequent verification. The improved SMO effectively suppresses the overshoot problem caused by sudden changes in speed, avoids the impact of speed fluctuations on the safety of low-pressure CT transportation and the accuracy of verification, and adapts to the needs of high-speed transportation conditions on the assembly line.

[0131] exist Figure 9Under other speed switching conditions, the SMO algorithm significantly reduces overshoot compared to the photoelectric encoder (overshoot 23 r / min), preventing speed fluctuations from affecting the AGV's positioning accuracy. Specifically, the improved SMO reduces the overshoot from 7.5 r / min in the traditional SMO to 4.2 r / min, a reduction of 82%, effectively suppressing dynamic overshoot and ensuring smooth and accurate AGV transport between different workstations. During the steady-state operation of the motor, the speed jitter amplitude under the photoelectric encoder is 8.5 r / min, while the speed jitter amplitude of the improved SMO is only 1.3 r / min, a reduction of 85%. Furthermore, combined with… Figure 10 This further demonstrates that the improved SMO effectively mitigates the chattering problem inherent in traditional sliding mode observers, providing stable rotational speed support for automated calibration of low-pressure CT and ensuring the accuracy of calibration data.

[0132] Based on the performance verification results of the rotor position and speed control described above, analysis from the two core dimensions of overshoot and chatter suppression shows that the improved SMO described in this invention can effectively solve the problems of large overshoot, significant chatter, and insufficient accuracy in the high-speed operation phase of the FOC+ position sensor solution. Furthermore, compared with the traditional SMO, the improved SMO has superior dynamic response and steady-state operation performance, and can adapt to various operating conditions in the production line, such as start-up, steady-state operation, and speed switching, thereby improving verification efficiency and accuracy. In addition, during the continuous acceleration of the BLDC motor, the maximum response time of speed control in the high-speed section is only 0.036 s, indicating that the improved SMO has good speed regulation robustness.

[0133] After obtaining a reliable and improved sliding mode observer, in order to further adapt to the full speed range control requirements of the low-pressure CT calibration production line and ensure that the AGV can obtain stable and accurate speed and position feedback under high-speed, medium- and low-speed conditions, thus ensuring the smooth progress of the calibration work, it is also necessary to determine the switching threshold between sensorless and traditional sensor-based (photoelectric encoder) systems. Figure 11 Quantitative analysis revealed that when the motor rotor speed exceeded 890 r / min, the standard deviation of the photoelectric encoder speed feedback was significantly greater than that of the improved SMO algorithm. This indicates that above this speed, the encoder could no longer meet the high-precision verification requirements of low-pressure CT, causing AGV operation vibration. Based on the combined waveform feature analysis and standard deviation quantitative calculation results, the final speed switching threshold for the SMO algorithm was determined to be 890 r / min. When the motor rotor speed exceeds 890 r / min, the system automatically switches to the improved SMO algorithm for speed and position feedback, ensuring stable and accurate speed and position feedback for the production line under different speed conditions.

[0134] To verify the actual performance and algorithmic defects of the FOC control algorithm used in this invention, while reducing development risks, improving R&D efficiency, and verifying the core control logic, this invention builds an FOC control system simulation model in the Matlab / Simulink simulation environment based on the mathematical model of the BLDC motor and the FOC control principle. By analyzing the SVPWM module sector judgment value N and saddle wave Tcmx generated by the simulation, the correctness of the FOC control algorithm is verified, laying a theoretical and simulation foundation for the subsequent physical implementation of the drive system.

[0135] In low-pressure CT calibration production lines, the position control of AGVs directly determines the accuracy and safety of transporting turnover boxes. The AGV position can be accurately calculated by integrating the motor speed in the time domain. Therefore, by verifying the speed stability under the FOC control algorithm through simulation, the AGV position can be reliably determined, ensuring safe operation. Based on this requirement, this invention first verifies the adaptability of the FOC + photoelectric encoder solution to the AGV driving scenario in a production line, focusing on analyzing the speed waveform fed back by the photoelectric encoder. Simulation results show that the AGV exhibits good speed stability at low and medium speeds, but significant jitter occurs at high speeds, directly demonstrating the control limitations of the FOC + photoelectric encoder solution in high-speed scenarios.

[0136] To address the aforementioned high-speed chattering issue, this invention optimizes the original simulation model by removing the photoelectric encoder speed measurement module from the original FOC control model and replacing it with a speed feedback module of SMO+PLL. A sensorless FOC control simulation model is then built to verify the control effect of the traditional SMO solution. High-speed range speed switching tests and analysis of rotor speed and position waveforms show that the traditional SMO algorithm still cannot meet the high-speed control requirements of AGVs, necessitating algorithm improvement.

[0137] To verify the effectiveness of the improved SMO algorithm, this invention reconstructs a sensorless control simulation model of FOC equipped with the improved SMO, and compares and analyzes the rotor position and speed output waveforms before and after the improvement. This invention sets the rotor speed of the BLDC motor to the maximum required speed of the production line, 1300 r / min, and verifies the position observation performance of the improved SMO algorithm through rotor position waveform simulation: comparing the estimated position and actual rotor position deviations of the traditional SMO and the improved SMO, the results show that the estimated curve of the improved SMO has a higher degree of overlap with the actual position curve, stronger real-time tracking capability, and better rotor position acquisition accuracy.

[0138] After verifying the effectiveness of the improved SMO algorithm in suppressing chatter and improving convergence speed at the rotor position dimension, this invention further analyzes the algorithm optimization effect from the rotor speed dimension, intuitively demonstrating its improvement in chatter suppression and dynamic response performance. To adapt to the full-speed range operation requirements of the AGV in the low-pressure CT calibration production line, the initial speed of the BLDC motor was set to 1000 r / min, and the speed in the high-speed section was increased in a stepwise manner to simulate the AGV gradually increasing from a high-speed transport state to the maximum required speed of the production line, covering complex speed switching scenarios. By comparing the rotor speed error, overshoot, and chatter amplitude before and after steady state at each speed, the control superiority of the improved SMO algorithm is clearly demonstrated.

[0139] This invention combines an improved SMO algorithm with a sensor-based FOC control scheme to construct a hybrid control mode: the motor uses a photoelectric encoder for closed-loop starting at low speeds, and automatically switches to sensorless control via the improved SMO at medium and high speeds, balancing the accuracy of low- and medium-speed control with the stability of high-speed operation. Simultaneously, to ensure a smooth and shock-free AGV switching process, the speed switching threshold of the improved SMO needs to be determined.

[0140] To determine the optimal speed switching threshold, this invention conducted a BLDC full-speed-range uniform acceleration simulation experiment (the improved SMO uses an IF open-loop start-up method to solve the problem of insufficient back EMF and inability to obtain accurate rotor information in the stationary and low-speed ranges), and analyzed the rotor speed feedback waveform. When the motor's rotor speed exceeds 900 r / min, the speed feedback jitter of the photoelectric encoder is significantly better than that of the improved SMO. Based on this, this invention uses 860 r / min as the base speed and increases the speed sequentially to 940 r / min with a fixed step size, covering the key high-speed operation range. Speed ​​feedback data from the photoelectric encoder and the improved SMO are collected at each speed node, and the standard deviation of the two sets of data is calculated using the Bessel formula to quantitatively evaluate the speed fluctuation degree and steady-state accuracy of the two schemes. Finally, the optimal speed switching threshold is determined to be 890 r / min, ensuring that the AGV can obtain stable and accurate speed and position feedback under full-speed conditions, adapting to the operational requirements of the low-pressure CT calibration production line.

[0141] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A method for driving and controlling a conveyor AGV in a current transformer calibration production line, characterized in that, include: Collect parameters of the AGV drive system and motor body; Based on the parameters, a dual-closed-loop vector control simulation was performed on the AGV drive system under the condition of motor speed acquisition by photoelectric encoder to obtain the critical chattering speed corresponding to the control mode, which was used as the speed switching threshold. The current operating speed of the motor is detected in real time, and it is determined whether the current speed of the motor is less than the speed switching threshold. If so, a photoelectric encoder is used to collect the rotor speed and position of the motor, and the AGV drive system is controlled using a dual closed-loop vector control mode; Otherwise, the voltage and current of the motor are collected in real time, and the back electromotive force of the motor is established using the voltage and current with an improved sliding mode observer, and the current motor rotor speed and position are extracted. The calculated and extracted current motor rotor speed and position information is used as feedback to replace the motor rotor speed and position information collected by the photoelectric encoder. This feedback is then connected to a dual closed-loop vector control loop to control the AGV drive system until the current motor speed is detected to have fallen below the speed switching threshold.

2. The method for driving and controlling a conveyor AGV in a transformer calibration production line according to claim 1, characterized in that, The speed switching threshold is 890 r / min.

3. The method for driving and controlling a conveyor AGV in a transformer calibration production line according to claim 1, characterized in that, This also includes converting the real-time acquired motor voltage and current to... Coordinate system acquisition The process of current and voltage in a coordinate system, the process obtained Current and voltage in the coordinate system are used as inputs to the improved sliding mode observer.

4. The method for driving and controlling a conveyor AGV in a transformer calibration production line according to claim 1, characterized in that, The back electromotive force of an electric motor: ； In the formula, It is the electric angular velocity of the motor. , For DC inductors and AC inductors, and They represent and This indicates the establishment of observations by the sliding mode observer. Back electromotive force in coordinate system; The position angle of the rotor, It is a magnetic flux.

5. The method for driving and controlling a conveyor AGV in a transformer calibration production line according to claim 1, characterized in that, The approach law of the improved sliding mode observer is: ； In the formula, ; ; >1, >1, , Represents the sliding surface matrix function; ； ； In the formula, , These are the approach rates of the isotropic term and the exponential term, respectively; It is the rate of convergence. It is the sign function, and 'a' is the coefficient of the switching function; ； in, and These represent the current values ​​estimated by the observer. and These represent the α-axis stator current observation error and the β-axis stator current observation error, respectively.

6. The method for driving and controlling a conveyor AGV in a transformer calibration production line according to claim 1, characterized in that, Using a phase-locked loop to extract the back electromotive force of the motor and Extract the rotor speed of the motor in the current AGV drive system.

Images