Intelligent control method and system for precise positioning of motor stator

Abstract

The invention discloses an intelligent control method and system for precise positioning of a motor stator, relates to the field of intelligent control for precise positioning of motor stators, and is used for solving the problem that the positioning precision is reduced due to misjudgment of a visual system caused by factors such as an oil film on the surface of the stator, and the method comprises the following steps: determining an actual position parameter of the stator according to a stator image; comparing the actual position parameter with a preset target position parameter to obtain positioning deviation information; obtaining production process information related to the positioning deviation information; analyzing the positioning deviation information and the production process information to identify an evolution rule of the positioning deviation; according to an evolution rule of the positioning deviation, generating a compensation amount for adjusting a subsequent stator target position, and updating the subsequent stator target position based on the compensation amount; and recording the manual adjustment behavior of the operator on the target position, determining the adjustment range of the target position according to the evolution rule of the positioning deviation, and performing intervention when the manual adjustment behavior of the operator exceeds the adjustment range.

Description

Technical Field

[0001] This invention relates to the field of intelligent control for precise positioning of motor stators, and more particularly to an intelligent control method and system for precise positioning of motor stators. Background Technology

[0002] In modern industrial manufacturing, the positioning accuracy of precision components such as motor stators is crucial to product quality and production efficiency. Automated production lines rely on advanced intelligent control methods and systems, using high-resolution cameras to capture the geometric details of the stator core and guide robotic arms for accurate gripping and placement. However, before entering the precision positioning stage, the stator core undergoes pre-treatment processes such as stamping and cleaning. These preliminary steps, especially the cleaning process, are often difficult to maintain in an ideal state, resulting in uneven oil films on the stator surface that are difficult to discern with the naked eye.

[0003] This residual, uneven oil film has a significant and deceptive impact on the visual system. The edge detection method built into the computational module used in image recognition programs to find object edges is designed based on the ideal condition of a clean stator surface with uniform optical reflectivity. When a microscopic oil film exists on the stator core surface, it alters the local optical reflectivity and light scattering characteristics of the stator surface, causing originally sharp edges in the image to become blurred, artifacts to appear, or unnatural grayscale gradations to occur near the edges. When processing these abnormal optical signals, the edge detection method may misjudge them as actual offsets of the stator edges, causing slight "expansion" or "contraction" in the calculated stator profile locally, thus affecting the accuracy of its center point and rotation angle. This misjudgment is not random noise, but a systematic deviation related to the oil film distribution, directly leading to inaccurate stator center or angle information calculated by the system.

[0004] Positioning errors introduced by vision system misjudgments may be very small when positioning a single stator, but on high-speed, high-volume production lines, stators with residual oil film appear randomly in batches. The accumulated positioning deviation gradually exceeds the tolerance range that subsequent assembly processes can tolerate, severely affecting motor performance, reliability, and even leading to product scrap. To cope with this accumulated positioning deviation and the ever-increasing scrap rate, field operators often manually adjust the target position coordinates set in the program based on experience. However, this experience-based, non-data-driven manual adjustment makes the correspondence between the correction data set generated by the camera calibration program within the intelligent control system and the actual physical space more complex and disordered, weakening the system's original adaptive and self-correcting capabilities.

[0005] Ultimately, this intelligent control method and system, originally designed for precise positioning of motor stators, showed a significant decrease in the stability and reliability of its precision positioning when faced with stators of different batches and with varying degrees of surface oil film. It could not consistently achieve the sub-millimeter positioning accuracy required by the design, resulting in a persistently high scrap rate and seriously affecting production efficiency and manufacturing costs. Summary of the Invention

[0006] This application discloses an intelligent control method and system for precision positioning of motor stators, aiming to solve the problem that existing intelligent control methods and systems for precision positioning of motor stators suffer from a significant decrease in stability and reliability when faced with stators of different batches and with varying degrees of surface oil film. This results in the inability to consistently achieve the sub-millimeter positioning accuracy required by the design, leading to a persistently high scrap rate and seriously affecting production efficiency and manufacturing costs.

[0007] In a first aspect, this application discloses an intelligent control method for precise positioning of a motor stator, comprising the following steps: Acquire a stator image and determine the actual position parameters of the stator based on the stator image; The actual position parameters are compared with the preset target position parameters to obtain positioning deviation information; Obtain production process information related to this positioning deviation; The positioning deviation information and the production process information are analyzed to identify the evolution pattern of the positioning deviation; Based on the evolution of the positioning deviation, a compensation amount is generated to adjust the target position of the subsequent stator, and the target position of the subsequent stator is updated based on the compensation amount. Record the operator's manual adjustment behavior to the target position, and determine the adjustment range of the target position based on the evolution of the positioning deviation. When the operator's manual adjustment behavior exceeds the adjustment range, intervene.

[0008] This technical solution enables intelligent identification, analysis, compensation, and intervention of motor stator positioning deviations, effectively solving the problem of decreased positioning accuracy caused by factors such as oil film, and improving the automation level of the production line and product quality.

[0009] Furthermore, in some implementations, the actual position parameters are compared with preset target position parameters to obtain positioning deviation information, including: In a controlled environment, sample images of stator samples with different oil film types and distributions are acquired, and the real geometric edge position of the stator is obtained using physical measurement equipment. Specific optical features generated by the oil film in the sample images are extracted, and the pixel offset caused by the oil film to visual recognition is quantified in order to establish the mapping relationship between the optical features of the oil film and the real geometric edge position. On the production line, the stator images captured by industrial cameras are analyzed in real time to identify specific texture patterns, edge features or spot distributions related to the oil film microstructure in the stator images, so as to determine the optical characteristics of the oil film currently present on the stator surface. Based on the identified optical features of the oil film and the mapping relationship, the stator image or the initially detected edge information is corrected to infer the true geometric edge position of the stator. Based on the corrected edge information, the center coordinates and rotation angle of the stator are calculated and compared with the preset target position to obtain the positioning deviation information.

[0010] This technical solution establishes a mapping relationship between the optical features of the oil film and the actual geometric edge position, and corrects the influence of the oil film on visual recognition in real time on the production line. This allows for a more accurate inference of the actual geometric edge position of the stator, effectively overcoming the interference of the oil film on positioning accuracy and improving positioning accuracy.

[0011] In another implementation, the actual position parameters are compared with preset target position parameters to obtain positioning deviation information, including: Acquire the light intensity characteristics of ambient light intensity; these characteristics include the frequency, amplitude, and instantaneous phase of light intensity fluctuations. Synchronize the industrial camera's trigger signal with the phase of light intensity fluctuations, so that the camera can expose at a specific phase point of the light intensity fluctuations; Based on this light intensity characteristic, the driving of the LED lighting module is adjusted to generate a reverse compensation signal to stabilize the total incident light intensity on the stator surface. Based on images acquired under stable light intensity, edge recognition and positioning parameter calculation are performed to obtain positioning deviation information.

[0012] This technical solution can synchronize camera exposure with light intensity fluctuations and stabilize the total incident light intensity on the stator surface by using a reverse compensation signal, effectively eliminating the impact of ambient light intensity fluctuations on image quality and edge recognition, and further improving the stability and accuracy of positioning.

[0013] Preferably, based on the evolution law of the positioning deviation, a compensation amount is generated to adjust the target position of subsequent stators, and the target position of subsequent stators is updated based on this compensation amount, including: The positioning deviation is decomposed into the contributions of stator characteristics, equipment operating status, and environmental parameters. For the deviations contributed by the stator characteristics, the equipment operating status, and the environmental parameters, compensation amounts are generated and superimposed on the target position to update the target position.

[0014] This technical solution can decompose positioning deviations into contributions from different sources and generate targeted compensation amounts, thereby achieving more refined and accurate positioning adjustments and improving the effectiveness of compensation.

[0015] Based on this, the operator's manual adjustments to the target position are recorded, and the adjustment range of the target position is determined according to the evolution pattern of the positioning deviation. When the operator's manual adjustments exceed the adjustment range, intervention is implemented, including: Based on the deviation contributed by the stator characteristics, equipment operating status, and environmental parameters, and their evolution patterns, the adjustment range boundaries for each deviation source are set. Based on the operator's manual adjustment behavior, determine the corresponding source of deviation for adjustment; Compare the operator's adjustment amount with the adjustment range boundary of the corresponding deviation source; Intervention is initiated when the operator's adjustment exceeds the adjustment range boundary of the corresponding deviation source.

[0016] This technical solution can effectively prevent operators' experience-based adjustments from introducing new deviations by setting adjustment range boundaries for each deviation source and monitoring and intervening in the operator's manual adjustment behavior, thus maintaining the stability of the system's adaptive and self-correcting capabilities.

[0017] To improve the solution, the method also includes: Capture operator hand movements and user interface images; Based on the operator's hand movements and the image of the operating interface, the contact point, movement trajectory, and dwell time of the operator's hand on the operating interface are identified, and the adjustment amount of the operator is determined by combining the changes in the target position parameters displayed on the operating interface.

[0018] This technical solution can capture operator hand movements and interface images, and combine them with changes in interface parameters to more accurately identify and quantify the operator's adjustments, providing reliable data support for subsequent adjustment range judgment and intervention.

[0019] After the operator adjusts the behavior, the method also includes: Assess the effectiveness of the operator's adjustment behavior and adjust the adjustment range boundaries of each deviation source based on the assessment results.

[0020] This technical solution enables the effectiveness evaluation of operator adjustment behaviors and dynamically adjusts the adjustment range boundaries based on the evaluation results. This allows the system to learn from the operator's experience, further optimize adjustment strategies, and improve the system's adaptability.

[0021] Furthermore, assess the effectiveness of the operator's adjustment behavior and adjust the adjustment range boundaries of each deviation source based on the assessment results, including: After the operator makes a change, the actual positioning accuracy data of the stator is continuously collected in the subsequent production process; Real-time monitoring of occasional disturbances in the production process; Within a specific time window after the operator's adjustment action, the actual positioning accuracy data is subjected to time-domain filtering to filter out the instantaneous impact of this occasional interference on the positioning accuracy. By comparing the changes in positioning accuracy before and after operator adjustments, after time-domain filtering, the true impact of operator adjustment behavior on positioning accuracy is quantified. Based on this actual impact, the adjustment range boundaries for each deviation source are adjusted.

[0022] This technical solution enables the continuous collection of positioning accuracy data, monitoring of occasional interference, and time-domain filtering, thereby more accurately quantifying the real impact of operator adjustments on positioning accuracy and providing a more reliable basis for optimizing the adjustment range boundaries.

[0023] As a technological improvement, by comparing the changes in positioning accuracy before and after operator adjustments, after time-domain filtering, the true impact of operator adjustments on positioning accuracy is quantified, including: Obtain the reference positioning accuracy data of the current batch of stators; Based on the benchmark positioning accuracy data, batch normalization processing is performed on the positioning accuracy data before and after adjustment; By comparing the changes in positioning accuracy before and after operator adjustments following batch normalization, the true impact of operator adjustment behavior on positioning accuracy is quantified.

[0024] This technical solution can eliminate the impact of inherent differences between stators in different batches on positioning accuracy assessment through batch normalization, making the quantification of the true impact of operator adjustment behavior more accurate and further improving the reliability of the assessment.

[0025] Secondly, this application also discloses an intelligent control system for precise positioning of a motor stator, the system comprising: The actual position determination module is used to acquire a stator image and determine the actual position parameters of the stator based on the stator image. The deviation comparison module is used to compare the actual position parameters with the preset target position parameters to obtain positioning deviation information. The information acquisition module is used to acquire production process information related to the positioning deviation information; The analysis module is used to analyze the positioning deviation information and the production process information to identify the evolution pattern of the positioning deviation. The processing module is used to generate a compensation amount for adjusting the target position of the subsequent stator according to the evolution law of the positioning deviation, and update the target position of the subsequent stator based on the compensation amount. The intervention module records the operator's manual adjustment behavior to the target position and determines the adjustment range of the target position based on the evolution of the positioning deviation. When the operator's manual adjustment behavior exceeds the adjustment range, intervention is performed.

[0026] This technical solution provides a system for implementing the aforementioned intelligent control method for precise positioning of motor stators. Through modular design, it enables intelligent control of stator positioning, thereby improving production efficiency and product quality.

[0027] Beneficial Effects: The intelligent control method for precise positioning of motor stators disclosed in this application obtains positioning deviation information by acquiring stator images and determining actual position parameters, comparing them with preset target position parameters. Based on this, production process information related to the positioning deviation information is acquired and analyzed to identify the evolution pattern of the positioning deviation. According to this evolution pattern, a compensation amount is generated and the target position of the subsequent stator is updated. Simultaneously, the operator's manual adjustment behavior is recorded, and the adjustment range is determined according to the evolution pattern of the positioning deviation. Intervention is initiated when the operator's adjustment exceeds the range. This method effectively solves the problem in the prior art where visual system misjudgments caused by factors such as oil film on the stator surface lead to decreased positioning accuracy and high product scrap rates. Through intelligent analysis and compensation of positioning deviations, and intelligent intervention in operator manual adjustments, this application can significantly improve the stability and reliability of precise positioning of motor stators, ensuring sub-millimeter positioning accuracy, thereby reducing product scrap rates and improving production efficiency and manufacturing cost-effectiveness. Attached Figure Description

[0028] Figure 1 This is a schematic flowchart of an intelligent control method for precise positioning of a motor stator provided in an embodiment of the present invention; Figure 2 This is a schematic flowchart of another intelligent control method for precise positioning of motor stator provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a motor stator precision positioning intelligent control system provided in an embodiment of the present invention. Detailed Implementation

[0029] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0030] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0031] First, let's introduce the terminology used in this application.

[0032] "Stator image" refers to two-dimensional or three-dimensional image data of a motor stator obtained through vision sensors such as industrial cameras. These images contain information such as the stator's geometry and surface features.

[0033] "Actual position parameters" refer to the precise position and orientation information of the stator in the current coordinate system extracted from the stator image through image processing and analysis, such as the center coordinates (X, Y) and rotation angle (θ).

[0034] "Target position parameters" refer to the ideal position and attitude that the stator is expected to achieve, either pre-set or dynamically adjusted by the system.

[0035] "Positioning deviation information" is the difference between the actual position parameters and the target position parameters, which quantifies the degree to which the stator deviates from the ideal position.

[0036] "Production process information" encompasses various data related to stator production, such as stator batch information, parameters of upstream processes (e.g., stamping, cleaning), equipment operating status (e.g., temperature, pressure, vibration), and environmental parameters (e.g., light intensity, humidity). This information is crucial for understanding the causes and evolution of positioning deviations.

[0037] The following specific embodiments will provide a detailed introduction and explanation of the intelligent control method for precise positioning of motor stator provided in this application.

[0038] Reference Figure 1This invention provides an intelligent control method for precise positioning of a motor stator, comprising the following steps: S1, acquire the stator image, and determine the actual position parameters of the stator based on the stator image.

[0039] High-resolution industrial cameras can be used to photograph the stator on the production line, acquiring its two-dimensional image. Then, traditional image processing algorithms, such as Canny edge detection, Sobel operator, or Prewitt operator, are used to extract the stator's edge contours from the image. Based on these edge contours, the stator's center coordinates and rotation angle can be further calculated, thus obtaining the stator's actual position parameters. Alternatively, a 3D vision system can be used to acquire the stator's 3D point cloud data through structured light projection or binocular stereo vision technology. Then, a point cloud registration algorithm (such as ICP algorithm) is used to match the acquired point cloud data with a pre-set stator CAD model, thereby accurately determining the stator's actual position and orientation parameters in 3D space.

[0040] S2. Compare the actual position parameters with the preset target position parameters to obtain positioning deviation information.

[0041] For example, if the actual position parameters are represented as (X_actual, Y_actual, θ_actual) and the target position parameters are represented as (X_target, Y_target, θ_target), then the positioning deviation information can be represented as (ΔX, ΔY, Δθ), where ΔX = X_actual - X_target, ΔY = Y_actual - Y_target, and Δθ = θ_actual - θ_target. These deviation values ​​quantify the translational and rotational deviations of the stator in the X and Y directions.

[0042] S3. Obtain production process information related to positioning deviation information.

[0043] Specifically, this can be achieved by integrating multiple sensors and data acquisition systems. For example, pressure and speed parameters of upstream stamping equipment, and cleaning fluid concentration, temperature, and spray pressure of cleaning equipment can be obtained from the PLC (Programmable Logic Controller) or SCADA (Supervisory Control and Data Acquisition) system on the production line. Simultaneously, environmental sensors can be deployed to monitor ambient light intensity, temperature, and humidity in the production workshop in real time. Furthermore, batch information and production history of the current stator can be obtained through RFID tags or QR code scanning. This data will be aggregated and correlated with corresponding positioning deviation information.

[0044] S4. Analyze the positioning deviation information and production process information to identify the evolution pattern of the positioning deviation.

[0045] For example, historical positioning deviation data and corresponding production process information can be input into time series analysis models, such as ARIMA models or LSTM networks, to predict future trends in positioning deviation. Clustering algorithms (such as K-means) can also be used to classify different batches of stators and analyze the common characteristics of positioning deviation across batches. Furthermore, regression analysis can be employed to establish mathematical models between positioning deviation and various production process parameters, thereby identifying key factors affecting positioning deviation and their degree of influence. For instance, analysis may reveal that when the cleaning fluid temperature falls below a certain threshold, positioning deviation increases significantly, potentially indicating an exacerbation of oil film residue issues.

[0046] S5. Based on the evolution law of positioning deviation, generate compensation amount for adjusting the target position of subsequent stators, and update the target position of subsequent stators based on the compensation amount.

[0047] For example, if analysis reveals that the positioning deviation exhibits periodic fluctuations and is related to the wear cycle of a certain equipment component, a periodic compensation amount can be generated based on this periodicity and superimposed on the target position. If the positioning deviation is found to be linearly related to ambient temperature, a linear compensation amount can be generated based on the real-time ambient temperature. These compensation amounts can be fixed values ​​or dynamically adjusted functions. For instance, when the positioning deviation is detected to be continuously drifting in a certain direction, the system can calculate a reverse compensation amount and add it to the current target position parameters, thereby fine-tuning the target position of the subsequent stator to counteract this drift.

[0048] S6. Record the operator's manual adjustment behavior to the target position, and determine the adjustment range of the target position according to the evolution law of the positioning deviation. When the operator's manual adjustment behavior exceeds the adjustment range, intervene.

[0049] For example, the system can record the specific values ​​and times when the operator modifies the target position parameters on the HMI (Human-Machine Interface). Simultaneously, based on the evolution of historical positioning deviations, the system can dynamically calculate a reasonable adjustment range. For instance, if historical data shows that the positioning deviation typically fluctuates within ±0.1mm, the system can set the operator's adjustment range to ±0.15mm. When the operator's adjustment exceeds this range, the system can issue a warning, prompting the operator to confirm, or automatically roll back to the recommended adjustment value to prevent empirical adjustments from introducing greater system instability.

[0050] The overall working principle of this application lies in achieving adaptive and optimized precise positioning of the motor stator by constructing a closed-loop intelligent control system. First, the actual position of the stator is acquired through a vision system and compared with the target position to obtain the positioning deviation. This deviation is not merely a simple numerical difference, but the basis for the system's subsequent decisions. Next, the system not only focuses on the current deviation but also delves deeper into production process information related to the deviation, such as upstream cleaning parameters and environmental conditions. This information is crucial for understanding "why the deviation occurs" and "how it evolves." Through comprehensive analysis of the positioning deviation and production process information, the system can identify the inherent evolutionary pattern of the positioning deviation, such as whether it is periodic, trend-based, or related to specific production conditions. This step is one of the core innovations of this application, elevating positioning control from simple feedback adjustment to intelligent control based on prediction and causal analysis. Once the evolutionary pattern is identified, the system can proactively generate compensation quantities to adjust the subsequent target position of the stator. This means that the system no longer passively waits for the deviation to occur before correcting it, but can predict and adjust in advance, thereby maintaining the stator positioning accuracy at a higher level. Furthermore, this application introduces intelligent management of operator manual adjustment behavior. Traditionally, operator experience-based adjustments can lead to system instability. This application records operator adjustment behavior and, combined with the evolution of positioning deviation, dynamically sets a reasonable adjustment range. When the operator's adjustment exceeds this range, the system intervenes, thus preserving operator experience while avoiding the negative impact of improper adjustments on system stability. The entire process forms an intelligent closed loop, enabling the system to continuously learn, adapt, and optimize the stator's positioning accuracy.

[0051] Compared with existing technologies, the core innovation of this application lies in its in-depth exploration of the evolution law of positioning deviation and the intelligent compensation mechanism based on this law, as well as the intelligent intervention in the operator's manual adjustment behavior. Traditional methods often only stop at simple feedback adjustment of the current positioning deviation, failing to fully consider the causes and evolution trends of the deviation, making it difficult for the system to maintain high accuracy in the face of complex and ever-changing production environments. For example, when there is an oil film on the stator surface, existing vision systems may misjudge the edges, leading to systematic deviations. Although operators may make manual adjustments, such empirical adjustments lack data support and may complicate the relationship between system calibration data and actual physical space, weakening the system's adaptive capabilities.

[0052] This application, by introducing the acquisition and analysis of production process information, can identify the intrinsic relationship between positioning deviation and stator characteristics, equipment operating status, environmental parameters, etc., thereby revealing the evolution law of positioning deviation. For example, it can discover the correlation between oil film residue and parameters such as cleaning fluid temperature and spray pressure, thus understanding the influence mechanism of oil film on visual positioning. Based on these evolution laws, the system can generate more accurate and forward-looking compensation amounts, proactively adjusting the target position of subsequent stators, rather than merely passively correcting deviations that have already occurred. Furthermore, the intelligent intervention of the operator's manual adjustment behavior effectively solves the system disorder problem that may be caused by experience-based adjustments. By setting a dynamic adjustment range and intervening when the range is exceeded, this application ensures the overall stability and reliability of the system while retaining the value of operator experience. Therefore, this application can significantly improve the accuracy, stability, and intelligence level of motor stator positioning, reduce scrap rate, and increase production efficiency, representing a significant advancement compared to existing technologies.

[0053] Traditional intelligent control methods for precise positioning of motor stators typically rely on visual recognition using stator images captured by industrial cameras when comparing the actual stator position parameters with preset target position parameters. However, in actual production environments, oil films often remain on the stator surface. These oil films alter the optical properties of the stator surface, causing blurring, edge distortion, or light spots in the images captured by the industrial camera, thus affecting the accuracy of visual recognition and leading to deviations in the determined actual stator position parameters. If this problem is not addressed, comparisons based on inaccurate actual position parameters will result in inaccurate positioning deviation information, affecting the effectiveness of subsequent compensation generation and target position updates, ultimately reducing the overall accuracy of motor stator precision positioning. To address this, this application proposes a more accurate method for acquiring positioning deviation information. By establishing a mapping relationship between the optical characteristics of the oil film and the true geometric edge position, and by identifying and correcting the influence of the oil film on visual recognition in real time on the production line, a more accurate true geometric edge position of the stator can be obtained, thereby improving the accuracy of the positioning deviation information.

[0054] In one possible design, such as Figure 2 As shown, in order to obtain positioning deviation information, this application may further include the following steps: S101. Under controlled conditions, sample images are acquired for stator samples with different oil film types and distributions. The actual geometric edge position of the stator is obtained using physical measurement equipment. Specific optical features generated by the oil film in the sample image are extracted, and the pixel offset caused by the oil film to visual recognition is quantified to establish a mapping relationship between the optical features of the oil film and the actual geometric edge position.

[0055] Specifically, under controlled conditions, a large amount of image data containing oil film information can be obtained by acquiring sample images of stator samples with different oil film types and distributions. Simultaneously, using high-precision physical measurement equipment, such as a coordinate measuring machine or laser scanner, the true geometric edge positions of these stator samples can be accurately obtained. By comparing the sample images with the true geometric edge positions, specific optical features produced by the oil film in the sample images can be extracted, such as the impact of oil film thickness, distribution, and refractive index on image brightness, contrast, texture, and edge sharpness. Furthermore, the pixel offset caused by these oil film optical features on visual recognition can be quantified; that is, the pixel distance between the edge identified in the image and the true geometric edge due to the presence of the oil film. Therefore, a mapping relationship between the oil film optical features and the true geometric edge positions can be established. This mapping relationship can be a lookup table, a mathematical model, or a machine learning-based prediction model, with the aim of predicting or correcting the true geometric edge positions of the stator based on the oil film features identified in the images.

[0056] S102. On the production line, analyze the stator images captured by the industrial camera in real time, identify specific texture patterns, edge features or spot distributions related to the oil film microstructure in the stator images, so as to determine the optical characteristics of the oil film currently present on the stator surface.

[0057] Specifically, image processing algorithms can be used to analyze these stator images in real time, identifying specific texture patterns, edge features, or spot distributions related to the oil film microstructure. These features are the visual manifestations of the oil film's presence; for example, a thin oil film may appear as a faint spot or blurred edges, while a thick oil film may form obvious textures or reflective areas. By identifying these features, the optical characteristics of the oil film currently present on the stator surface can be determined.

[0058] S103. Based on the identified optical features and mapping relationship of the oil film, the stator image or the edge information initially detected is corrected to infer the true geometric edge position of the stator.

[0059] For example, if the mapping relationship indicates that a certain oil film feature will cause the edge to shift inward by X pixels, then the detected edge will be corrected outward by X pixels during image processing. The purpose of this correction is to eliminate or reduce the interference of the oil film on visual recognition, thereby more accurately inferring the true geometric edge position of the stator.

[0060] S104. Based on the corrected edge information, calculate the center coordinates and rotation angle of the stator, and compare them with the preset target position to obtain the positioning deviation information.

[0061] Based on the corrected edge information, the center coordinates and rotation angle of the stator can be accurately calculated; these parameters are the actual position parameters of the stator. By comparing these actual position parameters with the preset target position parameters, more accurate positioning deviation information can be obtained.

[0062] As a specific implementation, suppose that on a motor stator production line, after the stator has undergone cleaning or oiling processes, a thin film of lubricating oil remains on its surface. When an industrial camera captures an image of the stator, this oil film causes the stator edges to appear blurred in the image and produces faint reflective spots at certain angles.

[0063] First, a series of stator samples with oil films of varying thicknesses and distributions were prepared under a controlled laboratory environment. The true geometric edge positions of each sample were acquired using a high-precision laser rangefinder. Simultaneously, images of these samples were captured using an industrial camera identical to those used on the production line. These images were analyzed using image processing software to extract optical features such as the degree of edge blurring caused by the oil film, the intensity and position of the light spot, and these features were compared with the true edges obtained by the laser rangefinder. The pixel offset caused by the oil film was quantified; for example, it was found that an oil film of a certain thickness would cause the edge to shift inward by 2 pixels. Based on this, a lookup table was established to record the mapping relationship between different oil film optical features and their corresponding pixel offsets.

[0064] On an actual production line, when a stator enters the positioning station, an industrial camera captures its image. The image analysis system processes this image in real time, identifying oil film features on the stator surface, such as detecting an edge blur of intensity X and a light spot with intensity Y. Based on a pre-established mapping relationship, the system determines that the current oil film features will cause the edge to shift inward by 1.8 pixels. Subsequently, the system corrects the initially detected stator edge information, extending the edge outward by 1.8 pixels. Based on this corrected edge information, the center coordinates and rotation angle of the stator are accurately calculated. Finally, these corrected actual position parameters are compared with preset target position parameters to obtain more accurate positioning deviation information. For example, if the uncorrected image might show a stator offset of 0.5mm, after oil film compensation correction, the actual offset might be accurately determined to be 0.3mm, thus avoiding a 0.2mm measurement error caused by oil film interference.

[0065] This application's solution effectively addresses the issue of decreased positioning accuracy in traditional visual positioning methods when oil film interference is present by introducing an oil film compensation mechanism. Specifically, firstly, a mapping relationship between the optical features of the oil film and the actual geometric edge position is established under controlled conditions. This step is fundamental to understanding how the oil film affects visual recognition. Precisely quantifying the pixel offset caused by the oil film provides data support for subsequent real-time correction. Secondly, the optical features of the oil film in the stator image are identified in real-time on the production line, ensuring timely perception of the oil film's impact under the current production conditions. Finally, based on the identified oil film features and the preset mapping relationship, the stator image or initially detected edge information is corrected. This correction process directly counteracts the interference of the oil film on visual recognition, allowing the subsequently calculated stator center coordinates and rotation angle to more accurately reflect the stator's true geometric position. It is precisely this precise identification and compensation of oil film interference that makes the obtained positioning deviation information more reliable, thus providing a solid foundation for subsequent intelligent control and compensation quantity generation.

[0066] Through the above technical solution, this application can significantly improve the accuracy and robustness of motor stator precision positioning. Traditional positioning methods often suffer from large positioning errors due to image distortion when an oil film is present on the stator surface, leading to unstable positioning accuracy. This application, however, effectively eliminates the interference of the oil film on visual recognition by establishing a quantitative model of the oil film's influence and performing real-time correction, making the acquired actual stator position parameters closer to the true values. As a result, the accuracy of positioning deviation information is greatly improved, avoiding misjudgments and over-adjustments caused by oil film interference. This ensures the accuracy of subsequent compensation quantity generation and the effectiveness of target position updates, ultimately achieving high-precision and stable positioning of the motor stator in complex production environments.

[0067] Traditional intelligent control methods for precise positioning of motor stators often rely on stator images acquired by industrial cameras when comparing actual position parameters with preset target position parameters. However, fluctuations in ambient light intensity at the production site, such as uneven lighting caused by changes in external light sources, power frequency fluctuations, or equipment heating, can degrade stator image quality, affecting the accuracy of edge recognition and leading to errors in the calculated actual position parameters. This reduces the reliability of positioning deviation information. If these problems are not addressed, the accuracy of positioning deviation information will be severely affected, potentially leading to inaccurate subsequent compensation and adjustments, ultimately impacting the final positioning accuracy of the motor stator and product quality. Therefore, this application proposes a method for optimizing the acquisition of positioning deviation information through intelligent control of ambient light intensity to improve positioning accuracy and stability.

[0068] In this regard, this application further proposes the following steps for comparing the actual position parameters with the preset target position parameters to obtain positioning deviation information: S201. Obtain the light intensity characteristics of ambient light intensity.

[0069] Among them, light intensity characteristics include the frequency, amplitude, and instantaneous phase of light intensity fluctuations.

[0070] Specifically, the light intensity characteristics of ambient light can be obtained through light sensors or specialized light intensity monitoring equipment.

[0071] S202. Synchronize the industrial camera trigger signal with the phase of light intensity fluctuation, so that the camera can expose at a specific phase point of the light intensity fluctuation.

[0072] The purpose is to ensure that image acquisition is performed at the moment when the light intensity is most stable or predictable. For example, when the ambient light intensity fluctuates periodically, the camera exposure can be triggered when the light intensity reaches its peak, trough, or a specific stable phase, thereby avoiding image acquisition when the light intensity changes rapidly and reducing image blurring or uneven brightness caused by changes in light intensity.

[0073] S203. Based on the light intensity characteristics, adjust the driving of the LED lighting module to generate a reverse compensation signal to stabilize the total incident light intensity on the stator surface.

[0074] Specifically, based on the detected characteristics of ambient light intensity fluctuations, for example, when the ambient light intensity increases, the driving current of the LED lighting module is reduced to decrease its luminous intensity; conversely, when the ambient light intensity decreases, the driving current of the LED lighting module is increased to increase its luminous intensity. This reverse compensation mechanism aims to counteract the impact of ambient light intensity fluctuations on the stator surface illumination, ensuring that the stator surface is always under relatively stable and uniform illumination conditions, with the purpose of providing high-quality input for subsequent image processing.

[0075] S204. Based on the image acquired under stable light intensity, perform edge recognition and location parameter calculation to obtain location deviation information.

[0076] Under stable light intensity conditions, the contrast and brightness uniformity of the image are significantly improved. This greatly facilitates the accurate identification of the stator's geometric edges and the accurate calculation of positioning parameters such as the stator's center coordinates and rotation angle, thereby obtaining more reliable and accurate positioning deviation information.

[0077] This application's solution effectively solves the problems of image quality degradation and positioning accuracy loss caused by ambient light intensity fluctuations in traditional positioning methods by intelligently sensing, synchronizing, and actively compensating for ambient light intensity. Specifically, firstly, by acquiring light intensity characteristics such as frequency, amplitude, and instantaneous phase, the system can comprehensively understand the dynamic variation law of light intensity. Secondly, by synchronizing the industrial camera trigger signal with the phase of light intensity fluctuations, image acquisition can be performed at specific phase points where the light intensity is relatively stable or predictable, thus avoiding exposure during drastic light intensity changes and effectively reducing image blurring or uneven brightness caused by light intensity variations. Furthermore, based on the real-time acquired light intensity characteristics, the system can intelligently adjust the drive of the LED lighting module to generate a compensation signal opposite to the direction of ambient light intensity fluctuations, thereby actively counteracting the influence of ambient light intensity on the stator surface illumination and ensuring that the stator surface is always in a constant and uniform lighting environment. It is precisely because of the above synergistic effect that the subsequently acquired stator images have higher clarity and stability, providing a solid foundation for accurate edge recognition and positioning parameter calculation, and ultimately significantly improving the accuracy and reliability of positioning deviation information.

[0078] Through the above technical solution, this application effectively overcomes the negative impact of ambient light intensity fluctuations on image acquisition and positioning accuracy in the intelligent control method for precise positioning of motor stators. Compared to solutions that do not consider ambient light intensity fluctuations, this application significantly improves the quality and stability of stator images by intelligently synchronizing camera exposure with light intensity phase and actively compensating for ambient light intensity. Therefore, when performing edge recognition and positioning parameter calculation, more accurate actual stator position parameters can be obtained, thereby greatly improving the accuracy of positioning deviation information. This improvement not only enhances the robustness of the entire positioning control system, enabling it to maintain high-precision operation under changing environmental conditions, but also provides a more reliable data foundation for subsequent positioning deviation analysis, compensation amount generation, and target position updates, ultimately ensuring high-precision positioning of the motor stator and the stability of product quality.

[0079] In some preferred embodiments, this application is implemented as follows: Suppose that on the motor stator production line, due to aging of workshop lighting fixtures or slight fluctuations in grid voltage, the ambient light intensity in the stator inspection area exhibits periodic fluctuations, with a frequency of approximately 50Hz and an amplitude variation within ±10%. To address this fluctuation, a high-sensitivity light sensor is first installed at the stator inspection station. This sensor collects ambient light intensity data in real time, and the signal processing unit analyzes the data to obtain the frequency, amplitude, and instantaneous phase of the light intensity fluctuations. For example, the system identifies a relatively stable low-intensity phase within each cycle of the light intensity.

[0080] Secondly, the trigger signal of the industrial camera used to capture stator images is synchronized with the instantaneous phase of light intensity fluctuations. Specifically, based on the phase information fed back from the light sensor, the control system precisely triggers the industrial camera to perform exposure when the light intensity fluctuation reaches its stable trough phase. This ensures that each image acquisition is completed at a moment when the light intensity is relatively stable, avoiding image blurring or uneven brightness caused by rapid changes in light intensity.

[0081] Simultaneously, an auxiliary light source consisting of LED lighting modules is configured in the stator detection area. The control system dynamically adjusts the drive current of the LED lighting modules based on the real-time detected ambient light intensity characteristics. For example, when the ambient light intensity begins to increase, the control system correspondingly reduces the brightness of the LED lighting modules, generating a reverse compensation signal; when the ambient light intensity decreases, it increases the brightness of the LED lighting modules. In this way, the LED auxiliary light source works in synergy with the ambient light intensity to keep the total incident light intensity on the stator surface at a preset constant level, for example, within the range of 5000 lux ± 1%.

[0082] Ultimately, under such stable and uniform lighting conditions, the stator images acquired by the industrial camera exhibit extremely high clarity and contrast. Image processing algorithms can more accurately identify the stator's geometric edges and precisely calculate actual position parameters such as the stator's center coordinates and rotation angle. By comparing these actual position parameters with preset target position parameters, highly accurate and reliable positioning deviation information can be obtained, providing a solid data foundation for subsequent intelligent control and adjustments.

[0083] In some embodiments described above, an intelligent control method for precise positioning of a motor stator is proposed. This method identifies the evolution pattern of positioning deviation, generates a compensation amount for adjusting the target position of subsequent stators, and updates the target position of subsequent stators based on the compensation amount. However, in actual production processes, positioning deviations are often the result of multiple factors, such as manufacturing tolerances of the stator itself, wear or drift of the equipment, and changes in ambient temperature or humidity. If only a single, undifferentiated compensation amount is generated, it may not be able to accurately and effectively correct deviations from different sources, resulting in poor compensation effects. Furthermore, it may negatively impact the correction of one type of deviation while correcting another, thus limiting further improvements in positioning accuracy and system robustness.

[0084] In response, this application further proposes a more refined compensation amount generation and update mechanism. Based on the evolution of the positioning deviation, a compensation amount is generated to adjust the target position of subsequent stators, and the target position of subsequent stators is updated based on the compensation amount, including: S301. Decompose the positioning deviation into the contributions of stator characteristics, equipment operating status and environmental parameters.

[0085] Specifically, decomposing the positioning deviation into contributions from stator characteristics, equipment operating status, and environmental parameters means identifying the specific contributions of different factors to the total positioning deviation through data analysis and modeling.

[0086] Stator characteristics can be understood as the inherent properties of the stator formed during the manufacturing process, such as minor deviations in its geometric dimensions, material uniformity, and surface roughness. These characteristics may cause the stator to exhibit specific and predictable deviation patterns during positioning. Equipment operating status refers to the real-time working conditions of the positioning equipment during operation, such as the wear and tear of mechanical parts, the temperature of the drive motor, sensor drift, and the tightness of the clamps. Changes in these states may cause positioning accuracy to change slowly or periodically over time or with usage frequency. Environmental parameters refer to external conditions at the production site, such as ambient temperature, humidity, air pressure, and vibration. Changes in these parameters may cause positioning deviations by affecting the thermal expansion and contraction of stator materials, deformation of equipment components, or sensor performance. In practical applications, machine learning algorithms, such as multivariate regression models, neural networks, or support vector machines, can be used to train historical positioning deviation data, stator batch information, equipment sensor data (such as temperature sensors, vibration sensors, and current sensors), and environmental sensor data (such as thermometers, hygrometers, and barometers). This establishes a mapping relationship between positioning deviations and these factors, enabling accurate decomposition of the total positioning deviation.

[0087] S302. For the deviations contributed by stator characteristics, equipment operating status and environmental parameters, generate compensation amounts respectively, add them to the target position, and update the target position.

[0088] This means that after decomposing the contribution of each deviation source, the system will independently calculate a compensation value for each deviation source. For example, for deviations caused by stator characteristics, a static or quasi-static compensation amount can be generated based on batch or specific measurement data of the stator; for deviations caused by equipment operating status, a dynamically adjusted compensation amount can be generated based on real-time monitoring data of the equipment to offset the effects of wear or drift; for deviations caused by environmental parameters, a real-time response compensation amount can be generated based on real-time feedback from environmental sensors. After these independent compensation amounts are generated, they will be precisely superimposed onto the preset target position. The superposition method can be vector superposition, that is, the compensation amounts are accumulated on the X, Y, Z axes and rotation angles respectively, thereby forming a new subsequent stator target position that has been corrected by multiple sources.

[0089] This application's solution decomposes complex positioning deviations into traceable components contributed by different factors and generates targeted compensation amounts, enabling more accurate identification and resolution of the root causes of positioning inaccuracies. This meticulous differentiation of deviation sources allows the system to avoid over-compensation or under-compensation problems that can result from a "one-size-fits-all" compensation strategy. For example, when positioning deviations are primarily caused by equipment wear, the system can focus on adjusting compensation amounts related to equipment operating conditions, rather than incorrectly adjusting compensation amounts related to stator characteristics or the environment, thus ensuring the effectiveness and relevance of the compensation. This refined compensation mechanism allows the system to more effectively suppress various interference factors, significantly improving the overall accuracy and stability of motor stator positioning.

[0090] Through the above technical solution, this application enables a deeper understanding and more precise control of motor stator positioning deviation. Since the compensation amount is generated specifically for the source of deviation, it effectively eliminates or reduces the impact of multi-source interference on positioning accuracy, significantly improving positioning accuracy and repeatability. Furthermore, this decomposition and independent compensation strategy also makes the system more adaptable and robust in the face of complex and changing production environments, thereby reducing scrap rates and improving production efficiency and product quality.

[0091] In some preferred embodiments, a specific example is given below. Assume that on a motor stator production line, the evolution of positioning deviation shows that the deviation is not only related to the batch of the stator itself, but also to the usage time of the positioning fixture and fluctuations in the workshop ambient temperature. Specifically, different batches of stators may have fixed, slight positioning offsets due to differences in manufacturing tolerances; after prolonged use, the clamping force or positioning surface of the positioning fixture may experience slight wear, causing the positioning point to gradually drift; and changes in the workshop ambient temperature may cause thermal expansion and contraction of the stator or positioning mechanism, resulting in instantaneous positioning deviations.

[0092] In this context, the method of this application first decomposes the total positioning deviation into three parts: the deviation contributed by the stator batch, the deviation contributed by fixture wear, and the deviation contributed by ambient temperature. For example, by analyzing historical data, a model can be established that, when the total positioning deviation is detected as ΔX, can calculate that ΔX_stator is caused by the stator batch, ΔX_fixture is caused by fixture wear, and ΔX_ambienterprise is caused by ambient temperature.

[0093] Subsequently, the system generates compensation amounts for each of the three decomposed deviation components. For example, for the deviation contributed by the stator batch, a static compensation amount can be found in a preset compensation table or calculated using a simple linear model based on the current stator batch information; for the deviation contributed by fixture wear, a dynamic compensation amount that changes slowly over time can be generated based on the cumulative usage time of the fixture or by monitoring the wear sensor data of the fixture in real time; for the deviation contributed by ambient temperature, a compensation amount that is adjusted in real time can be calculated using a thermal expansion model based on the real-time collected workshop temperature data.

[0094] Finally, these independent compensation amounts (e.g., C_stator, C_fixture, and C_environment in the X direction) are superimposed onto the target X position of the next stator. If the original target X position is X_target, then the updated target X position will become X_target + C_stator + C_fixture + C_environment. In this way, the system can accurately and independently correct positioning deviations from different sources, thereby ensuring that subsequent stators can be positioned more precisely.

[0095] In some embodiments described above, while generating compensation amounts and updating the target position based on the evolution of positioning deviations, and recording operator manual adjustments for intervention, positioning deviations in actual production processes can be caused by a combination of factors, such as stator characteristics, equipment operating status, and environmental parameters. Setting only a uniform adjustment range may not effectively distinguish and manage the reasonable adjustment ranges corresponding to different deviation sources, resulting in an insufficiently refined intervention mechanism. This could even limit reasonable fine-tuning by operators at specific deviation sources or fail to promptly prevent unreasonable adjustments. Therefore, this application further proposes a more refined intervention mechanism designed to intelligently determine the reasonable range of operator manual adjustments based on the source of the positioning deviation, thereby improving the accuracy and stability of positioning control.

[0096] The above records the operator's manual adjustments to the target position and determines the adjustment range of the target position based on the evolution of the positioning deviation. When the operator's manual adjustments exceed the adjustment range, intervention is carried out, including: S401. Based on the deviation contributed by stator characteristics, equipment operating status, and environmental parameters, and their evolution patterns, set the adjustment range boundaries for each deviation source.

[0097] Specifically, the system continuously analyzes historical positioning deviation data and, in conjunction with production process information, identifies the typical fluctuation range and trends of the deviations caused by stator characteristics, equipment operating status, and environmental parameters. Based on these analysis results, the system sets upper and lower limits for each deviation source—such as stator characteristic deviation, equipment operating status deviation, and environmental parameter deviation—that can be manually adjusted by the operator, forming the adjustment range boundaries for each deviation source. These boundaries are dynamic and can be adaptively adjusted according to the evolution of the deviation.

[0098] S402. Based on the operator's manual adjustment behavior, determine the corresponding deviation source for adjustment.

[0099] Specifically, when an operator manually adjusts the target position via a human-machine interface or other means, the system intelligently analyzes and determines which type or types of deviation sources the operator's adjustment is primarily targeting, by combining current real-time production data, sensor information, and the decomposition results of positioning deviations. For example, if the system detects that the main deviation is caused by equipment wear, and the operator makes an adjustment, the system will associate this adjustment with the deviation in the equipment's operating status.

[0100] S403. Compare the operator adjustment amount with the adjustment range boundary of the corresponding deviation source.

[0101] The system will compare the manual adjustment amount actually entered by the operator with the adjustment range boundary previously set for that specific deviation source. For example, if the operator adjusts a certain value and the system determines that the adjustment is for a deviation in the equipment's operating status, then the adjustment amount will be compared with the adjustment range boundary for the equipment's operating status deviation.

[0102] It should be noted that the method for determining the operator's adjustment amount includes: capturing the operator's hand movements and the operation interface image; based on the operator's hand movements and the operation interface image, identifying the contact point, movement trajectory, and dwell time of the operator's hand on the operation interface, and combining this with the changes in the target position parameters displayed on the operation interface to determine the operator's adjustment amount.

[0103] Specifically, capturing operator hand movements and interface images refers to acquiring real-time 3D pose data of the operator's hands and screen images of the interface using visual sensors (e.g., high-speed cameras or depth sensors) located in the operating area. The interface images can include visual elements such as target position parameters, adjustment buttons, and sliders displayed on the human-computer interaction interface. Furthermore, based on the operator's hand movements and interface images, identifying the operator's hand contact points, movement trajectories, and dwell times on the interface can be achieved using image processing and machine learning algorithms. For example, analyzing hand movement images using a gesture recognition model can determine the interaction points between the operator's fingers and virtual or physical controls on the interface; tracking changes in hand position over time can construct its movement trajectory; and calculating the time the hand remains stationary in a specific area or control can determine the dwell time. This information collectively depicts the detailed process of the operator's adjustments. In addition, determining the operator's adjustment amount by combining changes in target position parameters displayed on the interface refers to correlating the identified operator interaction behavior with the real-time updated target position parameters on the interface. For example, when an operator changes the target position parameters by dragging a slider or clicking a button, the system records the starting and ending values ​​of the parameter change and, combined with the operator's hand gesture recognition results, accurately calculates the actual adjustment amount applied by the operator. This adjustment amount can be an absolute value or an increment relative to the original target position.

[0104] S404. When the operator's adjustment exceeds the adjustment range boundary of the corresponding deviation source, intervention shall be carried out.

[0105] Once the system detects that an operator's manual adjustment exceeds the reasonable range allowed for that specific deviation source, it will immediately activate the corresponding intervention mechanism. This intervention may include, but is not limited to, issuing audible and visual alarms to the operator, displaying warning messages on the operating interface, restricting further adjustment inputs by the operator, or automatically rolling back the out-of-range adjustment to within the boundary value, in order to prevent negative impacts on the production process due to excessive or inappropriate manual adjustments.

[0106] This application's solution decomposes positioning deviation into contributions from stator characteristics, equipment operating status, and environmental parameters, and sets independent adjustment range boundaries for these different deviation sources. This allows for more precise evaluation and management of operator manual adjustments. When an operator makes a manual adjustment, the system intelligently identifies the deviation source corresponding to their adjustment intention and compares the adjustment amount with the reasonable adjustment range for that specific deviation source. This effectively prevents operators from making adjustments beyond the reasonable range due to insufficient experience or misjudgment, thus preventing unnecessary interference with the production process or the introduction of new positioning errors. This refined intervention mechanism ensures the effectiveness and safety of manual adjustments, allowing the system to maintain the advantages of automated control while fully utilizing operator experience for local optimization, thus avoiding potential negative impacts.

[0107] Through the above technical solution, this application enables intelligent management and intervention of operator manual adjustment behavior. Compared to simply setting a uniform adjustment range, this solution can provide customized adjustment boundaries for different deviation sources based on the specific origin of the positioning deviation, thereby significantly improving the accuracy and effectiveness of manual adjustments. This not only effectively prevents operators from making unreasonable or excessive adjustments and reduces the risk of human error, but also helps maintain the stability and positioning accuracy of the production process, ultimately improving the overall control level and production efficiency of motor stator precision positioning.

[0108] In some preferred embodiments, a specific example is given below. Assume a production line for precision positioning of motor stators. The system continuously monitors the actual position parameters of the stator and compares them with preset target position parameters to obtain positioning deviation information. Simultaneously, the system acquires production process information related to the positioning deviation information and analyzes the positioning deviation information and production process information to identify the evolution pattern of the positioning deviation. Further, the system decomposes the positioning deviation into components contributed by stator characteristics, equipment operating status, and environmental parameters. For example, stator characteristic deviation may be related to the material differences between 100 batches of stators, equipment operating status deviation may be related to the wear of the positioning fixture, and environmental parameter deviation may be related to fluctuations in workshop temperature.

[0109] Based on historical data and evolution patterns, the system sets adjustment range boundaries for these three types of deviation sources. For example, the adjustment range for stator characteristic deviation may be ±0.05mm, the adjustment range for equipment operating status deviation may be ±0.02mm, and the adjustment range for environmental parameter deviation may be ±0.01mm.

[0110] When an operator notices a slight fluctuation in positioning accuracy and manually adjusts the target position by 0.03mm, the system, based on current production data and deviation decomposition results, determines that the adjustment is primarily to compensate for deviations caused by equipment operating conditions. At this point, the system compares the operator's 0.03mm adjustment with the ±0.02mm boundary of the equipment operating condition deviation adjustment range. Since 0.03mm exceeds the ±0.02mm range, the system will intervene immediately, for example, by displaying a pop-up message to the operator stating, "The equipment operating condition deviation adjustment exceeds the recommended range. Please check the equipment operating status or contact an engineer." The system may also restrict the effectiveness of the adjustment or suggest that the operator limit the adjustment to within 0.02mm.

[0111] Conversely, if the operator adjusts by 0.04mm, and the system determines that this adjustment is for stator characteristic deviation, then since 0.04mm is within the adjustment range of ±0.05mm for stator characteristic deviation, the system will allow the adjustment to take effect without intervention. In this way, the solution of this application can achieve intelligent judgment and precise intervention of the operator's manual adjustment behavior, thereby effectively improving the control accuracy and stability of the motor stator precision positioning.

[0112] In some of the aforementioned implementations, while mechanisms have been proposed to record operator manual adjustments to the target position and determine the adjustment range based on the evolution of positioning deviation, and to intervene when the operator's manual adjustments exceed this range, these mechanisms may have certain limitations. Specifically, preset adjustment range boundaries may not fully adapt to the complex dynamic changes during production. For example, while some operator adjustments may exceed the preset range under certain circumstances, they may actually have a positive effect on improving positioning accuracy; conversely, some adjustments within the range may be ineffective. This static adjustment range setting may limit the system's effective learning and utilization of operator experience, and may also lead to unnecessary interventions or failure to promptly correct potential optimization opportunities.

[0113] In response, this application further proposes that, after the operator's adjustment behavior occurs, the above method also includes: assessing the effectiveness of the operator's adjustment behavior and adjusting the adjustment range boundaries of each deviation source based on the assessment results. Specifically, "assessing the effectiveness of the operator's adjustment behavior" refers to monitoring and analyzing the actual changes in stator positioning accuracy during subsequent production after the operator's manual adjustment, in order to determine whether the adjustment behavior has achieved the expected optimization effect. For example, the effectiveness of the adjustment can be quantified by comparing indicators such as the average positioning deviation, pass rate, or stability of the stator over a period of time before and after the adjustment. Here, effectiveness can be understood as the degree of positive impact of the operator's adjustment behavior on improving or maintaining stator positioning accuracy.

[0114] Assess the effectiveness of operator adjustments and adjust the adjustment range boundaries for each deviation source based on the assessment results, specifically including: S501. After the operator makes an adjustment, continuously collect the actual positioning accuracy data of the stator in the subsequent production process.

[0115] The actual positioning accuracy data reflects the stator's actual positioning performance after operator adjustments.

[0116] S502. Real-time monitoring of occasional disturbances in the production process.

[0117] For example, sensors can be used to monitor instantaneous abnormal fluctuations in parameters such as power supply voltage, ambient temperature, humidity, and vibration. Among these, sporadic interference can be understood as external or internal factors that occur randomly during the production process, have a short duration, but may have an instantaneous impact on positioning accuracy.

[0118] S503. Within a specific time window after the operator's adjustment behavior occurs, perform time-domain filtering on the actual positioning accuracy data to filter out the instantaneous impact of occasional interference on positioning accuracy.

[0119] Specifically, time-domain filtering can be performed on the collected actual positioning accuracy data. The purpose of time-domain filtering is to filter out the instantaneous impact of occasional interference on positioning accuracy, thereby obtaining a more stable positioning accuracy trend that better reflects the true situation. Various algorithms can be used for time-domain filtering, such as moving average filtering, Kalman filtering, and exponential smoothing; the choice depends on the specific interference characteristics and the system's response speed requirements. Filtering effectively removes spikes and noise from the data, making subsequent evaluations more accurate.

[0120] S504. By comparing the changes in positioning accuracy before and after operator adjustment, after time-domain filtering, the true impact of operator adjustment behavior on positioning accuracy is quantified.

[0121] This comparison can eliminate the illusions caused by occasional interference, thus providing a more objective assessment of the effectiveness of operator adjustments. For example, the impact can be quantified by calculating indicators such as the difference in the mean positioning accuracy before and after the adjustment, changes in the standard deviation, or changes in the slope of the trend line.

[0122] S505. Adjust the adjustment range boundaries of each deviation source based on the actual impact.

[0123] Specifically, if the operator's adjustments prove effective and have a sustained positive impact, the adjustment range boundaries for the corresponding deviation source can be appropriately relaxed or optimized; conversely, if the adjustments are ineffective or have a negative impact, it may be necessary to tighten or reassess the adjustment range boundaries.

[0124] This application's solution, through continuous data collection and real-time monitoring of occasional interference, provides a comprehensive information foundation for evaluating operator adjustment behavior. Furthermore, by performing time-domain filtering on the actual positioning accuracy data within a specific time window, the instantaneous impact of occasional interference on positioning accuracy is effectively filtered out, thus avoiding misjudgments caused by noise or short-term fluctuations. It is precisely because the true impact of operator adjustment behavior on positioning accuracy can be accurately quantified that adjustments to the boundaries of each deviation source's adjustment range can be based on more reliable and accurate criteria, thereby ensuring the rationality and effectiveness of the adjustments.

[0125] The above technical solutions significantly improve the accuracy and reliability of evaluating the effectiveness of operator adjustments, avoiding the misleading influence of occasional interference on the evaluation results. This makes the updating of the adjustment range boundaries of each deviation source more precise and stable, thereby enhancing the robustness and adaptability of the entire intelligent control method for precise positioning of the motor stator, ensuring the long-term stability and high precision of the production process.

[0126] In some preferred embodiments, it is assumed that on a motor stator production line, an operator manually adjusts a parameter related to stator characteristics, such as stator pressing force, to improve positioning accuracy. After the adjustment, the system continuously collects actual stator positioning accuracy data for subsequent production, such as stator center coordinate deviation obtained through a vision inspection system. Simultaneously, the system monitors occasional disturbances in the production environment in real time, for example, monitoring equipment vibration through vibration sensors and ambient light intensity fluctuations through photoelectric sensors. Within one hour of the operator's adjustment (a specific time window), the collected positioning accuracy data is processed using a moving average filter to smooth out positioning data glitches caused by instantaneous vibrations or light intensity fluctuations. Subsequently, the adjusted positioning accuracy data after filtering is compared with historical filtered data before adjustment. If the root mean square error (RMSE) of the stator positioning accuracy after adjustment is found to continuously decrease by 15%, it is quantified as a significant positive impact of the operator's adjustment behavior on positioning accuracy. Based on this real impact, the system will adjust the adjustment range boundaries related to the source of stator characteristic deviation accordingly. For example, the upper limit of the allowable adjustment for the source of deviation will be appropriately widened to allow operators more room for maneuver when making similar effective adjustments in the future, while ensuring that it is within a reasonable range.

[0127] In some embodiments described above in this application, the true impact of operator adjustment behavior on positioning accuracy is quantified by comparing the changes in positioning accuracy before and after operator adjustment, after time-domain filtering. Specifically, to more accurately quantify the true impact of operator adjustment behavior on positioning accuracy, this application further proposes the following scheme.

[0128] The above compares the changes in positioning accuracy before and after operator adjustments, after time-domain filtering, to quantify the true impact of operator adjustments on positioning accuracy, including: S601. Obtain the reference positioning accuracy data of the stator in the current batch.

[0129] Specifically, reference positioning accuracy data refers to the positioning accuracy data measured for the current batch of stators before operator adjustments or under a certain standard production condition. This data reflects the inherent positioning characteristics or average positioning level of the current batch of stators without operator adjustments. Reference positioning accuracy data can be obtained by measuring the positioning accuracy of a certain number of stators before operator adjustments and performing statistical analysis, such as calculating the average deviation and standard deviation.

[0130] S602. Perform batch normalization processing on the positioning accuracy data before and after adjustment based on the reference positioning accuracy data.

[0131] Batch normalization can be understood as a data preprocessing method aimed at eliminating or reducing inherent positioning accuracy differences between different batches of stators caused by factors such as materials, processes, and equipment wear. Specifically, batch normalization can be implemented in various ways. For example, the positioning accuracy data before and after adjustment can be subtracted from or divided by the reference positioning accuracy data of the current batch of stators to obtain the relative change or the normalized accuracy value. Alternatively, the difference or ratio between the adjusted positioning accuracy data and the reference positioning accuracy data can be calculated to eliminate systematic deviations between batches.

[0132] S603. Compare the changes in positioning accuracy before and after operator adjustment after batch normalization to quantify the real impact of operator adjustment behavior on positioning accuracy.

[0133] In practical applications, the change in positioning accuracy before and after operator adjustments, after batch normalization, refers to the actual improvement or deterioration in positioning accuracy caused by operator adjustments, after eliminating inherent batch differences. By comparing these normalized data, the effectiveness of operator adjustments can be assessed more accurately, avoiding misjudgments caused by inherent batch differences.

[0134] This application's solution addresses the interference of inherent positioning accuracy differences between different batches of stators when evaluating the effectiveness of operator adjustments. By introducing benchmark positioning accuracy data and performing batch normalization, it solves the problem of interference in evaluation results caused by inherent differences in stator positioning accuracy between different batches. In actual production, even under the same operator adjustments, the initial positioning accuracy of stators from different batches may fluctuate due to subtle differences in raw materials, production environment, and equipment operating status. Without batch normalization, directly comparing the positioning accuracy before and after adjustment might misjudge inherent batch differences as the effect of operator adjustments, leading to inaccurate evaluations of the effectiveness of operator adjustments. By obtaining benchmark positioning accuracy data for the current batch of stators and using this as a reference to perform batch normalization on the positioning accuracy data before and after adjustment, the positioning accuracy data between different batches becomes comparable. Specifically, the benchmark positioning accuracy data provides a "zero point" or "standard" for the current batch of stators, and all subsequent positioning accuracy data are measured relative to this "zero point." Therefore, the true impact of operator adjustments on positioning accuracy—that is, the accuracy changes caused by the operator adjustments themselves—can be more clearly separated, avoiding the confusion caused by inherent batch differences. This approach ensures that the assessment of the effectiveness of operator adjustments is based on their impact on the actual performance of the current batch of stators, rather than random fluctuations between batches.

[0135] Through the above technical solution, this application can more accurately and objectively quantify the true impact of operator adjustment behavior on the positioning accuracy of the motor stator. By eliminating the interference of inherent positioning accuracy differences between different batches of stators, the reliability and accuracy of the evaluation results are significantly improved. This enables the system to more accurately determine the effectiveness of operator adjustments, thus providing a more reliable basis for subsequent adjustments to the adjustment range boundaries, avoiding misjudgments caused by batch differences, and further optimizing the adaptability and robustness of the intelligent control method.

[0136] like Figure 3 As shown in the figure, this embodiment of the invention also provides an intelligent control system for precise positioning of a motor stator. The system includes: The actual position determination module is used to acquire stator images and determine the actual position parameters of the stator based on the stator images; The deviation comparison module is used to compare the actual position parameters with the preset target position parameters to obtain positioning deviation information; The information acquisition module is used to acquire production process information related to positioning deviation information; The analysis module is used to analyze positioning deviation information and production process information to identify the evolution pattern of positioning deviation; The processing module is used to generate a compensation amount for adjusting the target position of the subsequent stator according to the evolution law of the positioning deviation, and update the target position of the subsequent stator based on the compensation amount. The intervention module records the operator's manual adjustment behavior to the target position and determines the adjustment range of the target position based on the evolution of the positioning deviation. When the operator's manual adjustment behavior exceeds the adjustment range, intervention is performed.

[0137] This application also provides a computer-readable storage medium. All or part of the processes in the above method embodiments can be executed by a computer program instructing related hardware. This program can be stored in the computer-readable storage medium, and when executed, it can include the processes of the above method embodiments. The computer-readable storage medium can be an internal storage unit of the task execution device (including a data sending end and / or a data receiving end) of any of the foregoing embodiments, such as the hard disk or memory of the task execution device. The computer-readable storage medium can also be an external storage device of the terminal device, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the terminal device. Further, the computer-readable storage medium can include both the internal storage unit of the task execution device and an external storage device. The computer-readable storage medium is used to store the computer program and other programs and data required by the task execution device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0138] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0139] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, essentially, or the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0140] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be covered within the scope of protection of this application.

Claims

1. A method for intelligent control of precise positioning of a motor stator, characterized in that, include: Acquire a stator image and determine the actual position parameters of the stator based on the stator image; The actual position parameters are compared with the preset target position parameters to obtain positioning deviation information; Obtain production process information related to the positioning deviation information; The positioning deviation information and the production process information are analyzed to identify the evolution pattern of the positioning deviation; Based on the evolution law of the positioning deviation, a compensation amount is generated to adjust the target position of the subsequent stator, and the target position of the subsequent stator is updated based on the compensation amount. Record the operator's manual adjustment behavior to the target position, and determine the adjustment range of the target position according to the evolution law of the positioning deviation. When the operator's manual adjustment behavior exceeds the adjustment range, intervene.

2. The intelligent control method for precise positioning of a motor stator according to claim 1, characterized in that, The step of comparing the actual position parameters with the preset target position parameters to obtain positioning deviation information includes: In a controlled environment, sample images of stator samples with different oil film types and distributions are acquired, and the real geometric edge position of the stator is obtained using physical measurement equipment. Specific optical features generated by the oil film in the sample images are extracted, and the pixel offset caused by the oil film to visual recognition is quantified in order to establish the mapping relationship between the optical features of the oil film and the real geometric edge position. On the production line, the stator images captured by industrial cameras are analyzed in real time to identify specific texture patterns, edge features or spot distributions related to the oil film microstructure in the stator images, so as to determine the optical characteristics of the oil film currently present on the stator surface. Based on the identified optical features of the oil film and the mapping relationship, the stator image or the initially detected edge information is corrected to infer the true geometric edge position of the stator; Based on the corrected edge information, the center coordinates and rotation angle of the stator are calculated and compared with the preset target position to obtain the positioning deviation information.

3. The intelligent control method for precise positioning of a motor stator according to claim 1, characterized in that, The step of comparing the actual position parameters with the preset target position parameters to obtain positioning deviation information includes: Acquire the light intensity characteristics of ambient light intensity; the light intensity characteristics include the frequency, amplitude, and instantaneous phase of light intensity fluctuations; Synchronize the industrial camera's trigger signal with the phase of light intensity fluctuations, so that the camera can expose at a specific phase point of the light intensity fluctuations; Based on the light intensity characteristics, the driving of the LED lighting module is adjusted to generate a reverse compensation signal to stabilize the total incident light intensity on the stator surface. Based on images acquired under stable light intensity, edge recognition and positioning parameter calculation are performed to obtain positioning deviation information.

4. The intelligent control method for precise positioning of a motor stator according to claim 1, characterized in that, The step of generating a compensation amount for adjusting the target position of subsequent stators based on the evolution law of the positioning deviation, and updating the target position of subsequent stators based on the compensation amount, includes: The positioning deviation is decomposed into the contributions of stator characteristics, equipment operating status, and environmental parameters. For the deviations contributed by the stator characteristics, the equipment operating status, and the environmental parameters, compensation amounts are generated respectively and superimposed on the target position to update the target position.

5. The intelligent control method for precise positioning of a motor stator according to claim 4, characterized in that, The system records the operator's manual adjustments to the target position and determines the adjustment range of the target position based on the evolution of the positioning deviation. When the operator's manual adjustments exceed the adjustment range, intervention is initiated, including: Based on the deviations contributed by the stator characteristics, equipment operating status, and environmental parameters, and their evolution patterns, the adjustment range boundaries for each deviation source are set. Based on the operator's manual adjustment behavior, determine the corresponding source of deviation for adjustment; Compare the operator's adjustment amount with the adjustment range boundary of the corresponding deviation source; Intervention is performed when the operator's adjustment exceeds the adjustment range boundary of the corresponding deviation source.

6. The intelligent control method for precise positioning of a motor stator according to claim 5, characterized in that, The method further includes: Capture operator hand movements and user interface images; Based on the operator's hand movements and the operation interface image, the contact point, movement trajectory, and dwell time of the operator's hand on the operation interface are identified, and the adjustment amount of the operator is determined in combination with the changes in the target position parameters displayed on the operation interface.

7. The intelligent control method for precise positioning of a motor stator according to claim 5, characterized in that, After the operator adjusts the behavior, the method further includes: The effectiveness of the operator's adjustment behavior is evaluated, and the adjustment range boundaries of each deviation source are adjusted based on the evaluation results.

8. The intelligent control method for precise positioning of a motor stator according to claim 7, characterized in that, The evaluation of the effectiveness of the operator's adjustment behavior, and the adjustment range boundaries of each deviation source based on the evaluation results, include: After the operator makes a change, the actual positioning accuracy data of the stator is continuously collected in the subsequent production process; Real-time monitoring of occasional disturbances in the production process; Within a specific time window after the operator's adjustment behavior occurs, the actual positioning accuracy data is subjected to time-domain filtering to filter out the instantaneous impact of the occasional interference on the positioning accuracy. By comparing the changes in positioning accuracy before and after operator adjustments, after time-domain filtering, the true impact of operator adjustment behavior on positioning accuracy is quantified. Based on the actual impact, the adjustment range boundaries of each deviation source are adjusted.

9. The intelligent control method for precise positioning of a motor stator according to claim 8, characterized in that, The method of quantifying the true impact of operator adjustment behavior on positioning accuracy by comparing the changes in positioning accuracy before and after operator adjustment, after time-domain filtering, includes: Obtain the reference positioning accuracy data of the current batch of stators; The positioning accuracy data before and after adjustment are batch normalized based on the reference positioning accuracy data. By comparing the changes in positioning accuracy before and after operator adjustments following batch normalization, the true impact of operator adjustment behavior on positioning accuracy is quantified.

10. A precise positioning intelligent control system for a motor stator, characterized in that, The system includes: The actual position determination module is used to acquire a stator image and determine the actual position parameters of the stator based on the stator image. The deviation comparison module is used to compare the actual position parameters with the preset target position parameters to obtain positioning deviation information; The information acquisition module is used to acquire production process information related to the positioning deviation information; The analysis module is used to analyze the positioning deviation information and the production process information to identify the evolution pattern of the positioning deviation. The processing module is used to generate a compensation amount for adjusting the target position of the subsequent stator according to the evolution law of the positioning deviation, and update the target position of the subsequent stator based on the compensation amount. The intervention module is used to record the operator's manual adjustment behavior to the target position, and determine the adjustment range of the target position according to the evolution law of the positioning deviation. When the operator's manual adjustment behavior exceeds the adjustment range, intervention is performed.

Images