A method and system for detecting defects in a motor stator based on image analysis

Abstract

The application discloses a kind of motor stator defect detection method and system based on image analysis, it is related to image analysis technical field, including the surface image of motor stator is handled with characteristic analysis, determine the surface defect data of motor stator to be detected and the shape feature of motor stator to be detected;Determine the detection base point, according to the internal detection processing of motor stator to be detected to detection base point, determine the internal parameter of motor stator to be detected.The application first is by the analysis to the outside of motor stator, determine whether the outside of motor stator exists defect, if there is defect, then without needing to the inside of motor stator is analyzed, shorten the detection time of motor stator, second, again, the internal copper wire winding trend of motor stator is analyzed, determine whether the winding trend of copper wire conforms to standard, guarantee that motor stator for use is all not defective, guarantee the normal use of motor, reduce the probability that motor appears fault.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image analysis, in particular to a motor stator defect detection method and system based on image analysis. BACKGROUND

[0002] Motor stator: also a rotating part in the motor. The motor is composed of stator and rotor, which is used to realize the conversion of electric energy and mechanical energy and mechanical energy and electric energy. The motor stator is divided into motor stator and generator stator. The motor stator is divided into inner stator rotating mode and outer stator rotating mode. The core body in the middle of the motor is the rotating body in the inner stator rotating mode, which outputs torque (referring to motor) or receives energy (referring to generator). The outer stator rotating mode is that the outer body of the motor is the rotating body, which facilitates the application of various occasions.

[0003] Whether the motor stator can be put into use not only needs to consider whether there is defect outside the motor stator, but also needs to consider whether there is defect inside the motor stator. If the internal detection of the motor stator is not carried out, the motor stator will be put into use, which may cause the motor to malfunction in the use process. SUMMARY

[0004] To solve the above technical problems, a motor stator defect detection method and system based on image analysis are provided. The technical scheme solves the problem that whether the motor stator can be put into use not only needs to consider whether there is defect outside the motor stator, but also needs to consider whether there is defect inside the motor stator. If the internal detection of the motor stator is not carried out, the motor stator will be put into use, which may cause the motor to malfunction in the use process.

[0005] To achieve the above purpose, the technical scheme adopted by the present application is:

[0006] A motor stator defect detection method based on image analysis, comprising:

[0007] Obtaining the surface image of the motor stator to be detected, performing feature analysis and processing on the surface image of the motor stator, determining the surface defect data of the motor stator to be detected and the shape feature of the motor stator to be detected;

[0008] Determining the detection base point, performing internal detection processing on the motor stator to be detected according to the detection base point, and determining the internal parameters of the motor stator to be detected; wherein the internal parameters of the motor stator to be detected include the coordinates of each position of the copper wire of the motor stator to be detected and the winding direction of the copper wire of the motor stator to be detected;

[0009] According to the shape feature of the motor stator to be detected and the internal parameters of the motor stator to be detected, the model construction processing is carried out to determine the three-dimensional model of the motor stator to be detected;

[0010] respectively, the surface defect data of the motor stator to be detected and the three-dimensional model of the motor stator to be detected are compared and analyzed to determine the surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected;

[0011] The surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected are comprehensively evaluated and processed to determine the quality of the motor stator to be detected.

[0012] Preferably, the surface image of the motor stator to be detected is obtained, and the surface image of the motor stator is subjected to feature analysis processing to determine the surface defect data of the motor stator to be detected and the shape feature of the motor stator to be detected, which specifically includes the following steps:

[0013] The image of the motor stator to be detected is captured by an image capturing device to obtain the surface image of the motor stator to be detected;

[0014] The surface image of the motor stator to be detected is subjected to region analysis processing to determine the surface defect data of the motor stator to be detected;

[0015] The surface image of the motor stator to be detected is subjected to data extraction processing to obtain the shape feature of the motor stator to be detected.

[0016] Preferably, the surface image of the motor stator to be detected is subjected to region analysis processing to determine the surface defect data of the motor stator to be detected, which specifically includes the following steps:

[0017] The surface image of the motor stator to be detected is subjected to grid division processing to obtain a grid of an analysis unit;

[0018] The grid of the analysis unit is subjected to grid content analysis processing;

[0019] If there are redundant lines in the grid, and the lines intersect at a point and extend outward, output the grid protrusion information, and record the protrusion position and the number of protrusions;

[0020] If there are redundant lines in the grid, and the lines intersect at a point and extend inward, output the grid area depression information, and record the depression position and the number of depressions;

[0021] If there are no redundant lines in the grid, output the grid normal information.

[0022] Preferably, the detection base point is determined, and the internal parameters of the motor stator to be detected are determined by internal detection processing of the motor stator to be detected based on the detection base point, which specifically includes the following steps:

[0023] The surface center point of the motor stator to be detected is set as the detection base point, and a three-dimensional rectangular coordinate system is constructed with the detection base point as the coordinate origin.

[0024] placing the motor stator to be detected in a three-dimensional rectangular coordinate system, and recording coordinate data of the shape features of the motor stator to be detected;

[0025] based on the acoustic wave technology, sequentially performing internal detection processing on the motor stator to be detected from the detection base point, and acquiring the position coordinates of the copper wires of the motor stator to be detected;

[0026] performing data extraction processing on the database system to acquire the production process of the motor stator;

[0027] performing data extraction processing on the production process of the motor stator to determine the winding direction of the internal copper wires of the motor stator;

[0028] based on the winding direction of the internal copper wires of the motor stator, performing series processing on the position coordinates of the copper wires of the motor stator to be detected to determine the winding direction of the copper wires of the motor stator to be detected.

[0029] Preferably, the model construction processing according to the shape features of the motor stator to be detected and the internal parameters of the motor stator to be detected to determine the three-dimensional model of the motor stator to be detected specifically comprises the following steps:

[0030] selecting a model construction starting point in the modeling software;

[0031] placing the detection base point at the model construction starting point, and constructing the internal structure model of the motor stator to be detected in the modeling software according to the position coordinates of the copper wires of the motor stator to be detected and the winding direction of the copper wires of the motor stator to be detected;

[0032] based on the coordinate data of the shape features of the motor stator to be detected, performing secondary construction processing on the internal structure model of the motor stator to be detected to determine the three-dimensional model of the motor stator to be detected.

[0033] Preferably, the comparative analysis of the surface defect data of the motor stator to be detected and the three-dimensional model of the motor stator to be detected to determine the surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected specifically comprises the following steps:

[0034] judging the surface defect data of the motor stator to be detected and the set surface defect threshold value;

[0035] if the total number of depressions and protrusions on the surface of the motor stator to be detected is greater than or equal to the set surface defect threshold value, outputting that the surface quality of the motor stator to be detected is poor, etc.;

[0036] if the total number of depressions and protrusions on the surface of the motor stator to be detected is less than the set surface defect threshold value, outputting that the surface quality of the motor stator to be detected is excellent;

[0037] The model construction processing is performed on the standard motor stator to obtain a three-dimensional model of the standard motor stator.

[0038] Based on the three-dimensional model of the standard motor stator, the three-dimensional model of the motor stator to be detected is subjected to comparative analysis processing to determine the internal quality score of the motor stator to be detected.

[0039] Preferably, the model construction processing on the standard motor stator to obtain a three-dimensional model of the standard motor stator specifically includes the following steps:

[0040] The image shooting device is used to shoot images of the standard motor stator to obtain surface images of the standard motor stator;

[0041] The data extraction processing is performed on the surface images of the standard motor stator to obtain the shape features of the standard motor stator;

[0042] The center of the surface of the standard motor stator is placed in a three-dimensional rectangular coordinate system, and the placement direction of the standard motor stator is consistent with the placement direction of the motor stator to be detected, and the coordinate data of the shape features of the standard motor stator are recorded;

[0043] Based on the acoustic wave technology, the internal detection processing is sequentially performed on the standard motor stator from the detection base point to obtain the position coordinates of the copper wire of the standard motor stator;

[0044] Based on the winding direction of the internal copper wire of the motor stator, the position coordinates of the copper wire of the standard motor stator are subjected to series processing to determine the winding direction of the copper wire of the standard motor stator;

[0045] In the modeling software, the model is constructed according to the winding direction of the copper wire of the standard motor stator, the position coordinates of the copper wire of the standard motor stator, and the coordinate data of the shape features of the standard motor stator recorded to obtain the three-dimensional model of the standard motor stator.

[0046] Preferably, the comparative analysis processing on the three-dimensional model of the motor stator to be detected based on the three-dimensional model of the standard motor stator specifically includes the following steps:

[0047] The three-dimensional model of the standard motor stator and the three-dimensional model of the motor stator to be detected are subjected to coincidence degree analysis processing;

[0048] If the three-dimensional model of the motor stator to be detected and the three-dimensional model of the standard motor stator completely coincide, the internal quality of the motor stator to be detected is output as excellent;

[0049] If the three-dimensional model of the motor stator to be detected and the three-dimensional model of the standard motor stator do not completely coincide, the internal quality of the motor stator to be detected is output as poor.

[0050] Preferably, the surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected are comprehensively evaluated and processed to determine the quality of the motor stator to be detected, which specifically includes the following steps:

[0051] The surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected are comprehensively evaluated and processed;

[0052] If the surface quality of the motor stator to be detected is poor, the quality of the motor stator to be detected is output as poor;

[0053] If the internal quality of the motor stator to be detected is poor, the quality of the motor stator to be detected is output as poor;

[0054] If the surface quality of the motor stator to be detected is excellent and the internal quality of the motor stator to be detected is excellent, the quality of the motor stator to be detected is output as excellent.

[0055] Further, an image analysis-based motor stator defect detection system is proposed for implementing the image analysis-based motor stator defect detection method as described above, which includes:

[0056] An intelligent analysis terminal is configured to control each module to evaluate and process the surface quality and internal quality of the motor stator to be detected and determine the quality of the motor stator to be detected, and is configured to control data transmission and information interaction between each module;

[0057] A database system is configured to store the production process of the motor stator;

[0058] An image shooting device is configured to shoot images of the motor stator to be detected and the standard motor stator to obtain surface images of the motor stator to be detected and the standard motor stator;

[0059] An image processing module is configured to extract features from the surface images of the motor stator to be detected and the standard motor stator to obtain the shape features of the motor stator to be detected and the standard motor stator;

[0060] A grid analysis module is configured to detect defects of the grid to be analyzed to determine the surface defect data of the motor stator to be detected;

[0061] A coordinate determination module is configured to analyze coordinates of the motor stator to be detected and the standard motor stator to determine the position coordinates of the copper wire of the motor stator to be detected and the position coordinates of the copper wire of the standard motor stator;

[0062] A model construction module is configured to construct a three-dimensional model of a standard motor stator and a three-dimensional model of the motor stator to be detected.

[0063] A quality assessment module is configured to comprehensively evaluate the surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected, and determine the quality of the motor stator to be detected.

[0064] Compared with the prior art, the motor stator defect detection method and system based on image analysis have the following beneficial effects:

[0065] Firstly, the motor stator is analyzed externally to determine whether the motor stator has defects, and if there are defects, the internal defects of the motor stator do not need to be analyzed, thereby shortening the detection time of the motor stator. BRIEF DESCRIPTION OF DRAWINGS

[0066] Figure 1 A flowchart of steps S100-S500 in the motor stator defect detection method based on image analysis is provided.

[0067] Figure 2 A structure block diagram of the motor stator defect detection system based on image analysis is provided. DETAILED DESCRIPTION

[0068] The following description is provided to enable any person skilled in the art to practice the present application. The preferred embodiments in the following description are only examples of the present application, and other obvious modifications can be made by those skilled in the art.

[0069] Referring to Figure 1 The motor stator defect detection method based on image analysis includes:

[0070] S100, acquiring a surface image of a motor stator to be detected, performing feature analysis on the surface image of the motor stator, and determining surface defect data of the motor stator to be detected and the shape features of the motor stator to be detected;

[0071] S200, determining a detection base point, performing internal detection processing on the motor stator to be detected according to the detection base point, and determining internal parameters of the motor stator to be detected; wherein the internal parameters of the motor stator to be detected include coordinates of each position of the copper wire of the motor stator to be detected and the winding direction of the copper wire of the motor stator to be detected;

[0072] S300, model construction processing is performed according to the external shape feature of the motor stator to be detected and the internal parameter of the motor stator to be detected, and a three-dimensional model of the motor stator to be detected is determined;

[0073] S400, the surface defect data of the motor stator to be detected and the three-dimensional model of the motor stator to be detected are compared and analyzed respectively, and the surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected are determined;

[0074] S500, the surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected are comprehensively evaluated, and the quality of the motor stator to be detected is determined;

[0075] The skilled in the art can understand that if the surface of the motor stator has defects, since the motor stator is installed in the motor, and the motor will vibrate during use, which may cause the defects on the outside of the motor stator to expand, and even cause the motor stator to be damaged, thereby causing the motor to malfunction, in addition, if the surface of the motor stator is concave, it means that the oxide layer on the surface of the motor stator is damaged, and with the passage of time, rust will appear on the surface of the motor stator, thereby shortening the service life of the motor stator, and the problem of internal copper wire winding of the motor stator will reduce the efficiency of the motor, thereby increasing the production cost, therefore, when analyzing the defects of the motor stator, not only the surface of the motor stator needs to be analyzed, but also the internal defects of the motor stator need to be analyzed, so that the probability of failure of the motor during use can be reduced.

[0076] Embodiment 1

[0077] Step S100, acquiring the surface image of the motor stator to be detected, performing feature analysis processing on the surface image of the motor stator, and determining the surface defect data of the motor stator to be detected and the external shape feature of the motor stator to be detected, which specifically includes the following steps:

[0078] S101, acquiring the surface image of the motor stator to be detected by image shooting device;

[0079] S102, performing region analysis processing on the surface image of the motor stator to be detected, and determining the surface defect data of the motor stator to be detected;

[0080] S103, performing data extraction processing on the surface image of the motor stator to be detected, and acquiring the external shape feature of the motor stator to be detected;

[0081] During the production of the motor stator, its external shape may be affected by other factors, resulting in deformation of the external shape of part of the motor stators, and these motor stators with external deformation cannot be put into use.

[0082] The step S102 of performing region analysis processing on the surface image of the motor stator to be detected to determine the surface defect data of the motor stator to be detected specifically includes the following steps.

[0083] S1021, performing grid division processing on the surface image of the motor stator to be detected to obtain a grid of a unit to be analyzed;

[0084] S1022, performing grid content analysis processing on the grid of the unit to be analyzed;

[0085] S1023, if there are redundant lines in the grid and the lines intersect at a point and extend outward, outputting grid protrusion information and recording the protrusion position and the number of protrusions;

[0086] S1024, if there are redundant lines in the grid and the lines intersect at a point and extend inward, outputting grid area depression information and recording the depression position and the number of depressions;

[0087] S1025, if there are no redundant lines in the grid, outputting grid normal information;

[0088] It can be understood that whether it is a depression or a protrusion is an irregular image, but all the lines converge at a point, so the surface image of the motor stator to be detected is divided into a unit grid, and each grid is analyzed to determine whether there is a protrusion or a depression in the grid. Therefore, the lines in the image of the protrusion extend outward and converge at a point, and the lines in the image of the depression extend inward and converge at a point, so according to the different convergence directions of the lines, it is judged whether it is a depression or a protrusion.

[0089] Embodiment 2

[0090] The step S200 of determining a detection base point and performing internal detection processing on the motor stator to be detected based on the detection base point to determine the internal parameters of the motor stator to be detected specifically includes the following steps.

[0091] S201, setting the surface center point of the motor stator to be detected as the detection base point, and constructing a three-dimensional rectangular coordinate system with the detection base point as the coordinate origin;

[0092] S202, placing the motor stator to be detected in the three-dimensional rectangular coordinate system and recording the coordinate data of the shape features of the motor stator to be detected;

[0093] S203, based on the acoustic wave technology, sequentially performing internal detection processing on the motor stator to be detected from the detection base point to obtain the position coordinates of each copper wire of the motor stator to be detected;

[0094] The acoustic wave technology here adopts the pulse echo method, and combines the synthetic aperture focusing technology (SAFT) for improving the imaging resolution;

[0095] S204, data extraction processing is performed on the database system to obtain the production process of the motor stator;

[0096] S205, data extraction processing is performed on the production process of the motor stator to determine the winding direction of the copper wire inside the motor stator;

[0097] S206, based on the winding direction of the copper wire inside the motor stator, the coordinates of each position of the copper wire of the motor stator to be detected are processed in series to determine the winding direction of the copper wire of the motor stator to be detected;

[0098] It can be understood that in order to construct a three-dimensional model of the motor stator to be detected, the coordinates of each position of the motor stator outside (i.e. the coordinate data of the shape features of the motor stator to be detected) and the coordinates of each position of the copper wire inside (i.e. the coordinates of each position of the copper wire of the motor stator to be detected) need to be determined first, and then the winding direction of the copper wire needs to be determined to determine the winding direction of the copper wire of the motor stator, and the winding direction of the copper wire will also affect the quality of the motor stator.

[0099] Embodiment 3

[0100] Step S300, model construction processing is performed according to the shape features of the motor stator to be detected and the internal parameters of the motor stator to be detected to determine the three-dimensional model of the motor stator to be detected, which specifically includes the following steps:

[0101] S301, selecting a model construction starting point in the modeling software;

[0102] It is worth noting that the modeling software can be AutoCAD, SolidWorks, Inventor, etc.

[0103] S302, placing a detection base point at the model construction starting point, and constructing an internal structure model of the motor stator to be detected in the modeling software according to the coordinates of each position of the copper wire of the motor stator to be detected and the winding direction of the copper wire of the motor stator to be detected;

[0104] S303, based on the coordinate data of the shape features of the motor stator to be detected, performing secondary construction processing on the internal structure model of the motor stator to be detected to determine the three-dimensional model of the motor stator to be detected.

[0105] Embodiment 4

[0106] Step S400, respectively comparing and analyzing the surface defect data of the motor stator to be detected and the three-dimensional model of the motor stator to be detected to determine the surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected, which specifically includes the following steps:

[0107] S401, judging and processing the surface defect data of the motor stator to be detected and the set surface defect threshold value;

[0108] S402, if the total number of depressions and protrusions on the surface of the motor stator to be detected is greater than or equal to the set surface defect threshold value, outputting that the surface quality of the motor stator to be detected is poor, etc.;

[0109] S403, if the total number of depressions and protrusions on the surface of the motor stator to be detected is less than the set surface defect threshold value, outputting that the surface quality of the motor stator to be detected is excellent;

[0110] S404, model construction processing of the standard motor stator is performed to obtain a three-dimensional model of the standard motor stator;

[0111] S405, based on the three-dimensional model of the standard motor stator, comparative analysis processing is performed on the three-dimensional model of the motor stator to be detected to determine the internal quality score of the motor stator to be detected;

[0112] It can be understood that if the number of surface defects of the motor stator to be detected is too large, it cannot be put into use, because too many defects mean that the service life of the motor stator is too short, which greatly increases the probability of motor failure;

[0113] The step S404 of model construction processing of the standard motor stator to obtain a three-dimensional model of the standard motor stator specifically includes the following steps:

[0114] S4041, image shooting processing of the standard motor stator is performed by an image shooting device to obtain a surface image of the standard motor stator;

[0115] S4042, data extraction processing of the surface image of the standard motor stator is performed to obtain the shape features of the standard motor stator;

[0116] S4043, the surface center of the standard motor stator is placed in a three-dimensional rectangular coordinate system, and the placement direction of the standard motor stator is consistent with the placement direction of the motor stator to be detected, and the coordinate data of the shape features of the standard motor stator are recorded;

[0117] S4044, based on acoustic wave technology, internal detection processing of the standard motor stator is sequentially performed from a detection base point to obtain the position coordinates of the copper wire of the standard motor stator;

[0118] S4045, based on the winding direction of the internal copper wire of the motor stator, the position coordinates of the copper wire of the standard motor stator are serially processed to determine the winding direction of the copper wire of the standard motor stator;

[0119] S4046, in the modeling software, constructing a model according to the winding direction of the copper wire of the standard motor stator, the position coordinates of each copper wire of the standard motor stator, and the coordinate data recording the shape characteristics of the standard motor stator, and obtaining a three-dimensional model of the standard motor stator;

[0120] It can be understood that in order to determine whether there is a defect inside the motor stator, a control group needs to be set, and the corresponding model constructed by the parameters of the standard motor stator is the most perfect control group, therefore, the three-dimensional model of the standard motor stator is constructed to determine whether there is a defect inside the motor stator to be detected.

[0121] The step S405, based on the three-dimensional model of the standard motor stator, the three-dimensional model of the motor stator to be detected is compared and analyzed, and the internal quality score of the motor stator to be detected is determined, which specifically includes the following steps:

[0122] S4051, the three-dimensional model of the standard motor stator and the three-dimensional model of the motor stator to be detected are analyzed and processed;

[0123] S4052, if the three-dimensional model of the motor stator to be detected and the three-dimensional model of the standard motor stator are completely coincident, the internal quality of the motor stator to be detected is output as excellent;

[0124] S4053, if the three-dimensional model of the motor stator to be detected and the three-dimensional model of the standard motor stator are not completely coincident, the internal quality of the motor stator to be detected is output as poor.

[0125] Embodiment 5

[0126] Step S500, the surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected are comprehensively evaluated, and the quality of the motor stator to be detected is determined, which specifically includes the following steps:

[0127] S501, the surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected are comprehensively evaluated;

[0128] S502, if the surface quality of the motor stator to be detected is poor, the quality of the motor stator to be detected is output as poor;

[0129] S503, if the internal quality of the motor stator to be detected is poor, the quality of the motor stator to be detected is output as poor;

[0130] S504, if the surface quality of the motor stator to be detected is excellent, and the internal quality of the motor stator to be detected is excellent, the quality of the motor stator to be detected is output as excellent;

[0131] It can be understood that as long as the surface quality of the motor stator is poor or the internal quality of the motor stator is poor, the motor stator cannot be put into use, and if used, the probability of motor failure will be increased.

[0132] Referring to Figure 2 The motor stator defect detection system based on image analysis for realizing the motor stator defect detection method based on image analysis comprises:

[0133] The intelligent analysis terminal is used for controlling each module to evaluate and process the surface quality and internal quality of the motor stator to be detected, and determine the quality of the motor stator to be detected.

[0134] The database system is used for storing the production process of the motor stator.

[0135] The image shooting device is used for image shooting of the motor stator to be detected and the standard motor stator, and obtaining the surface image of the motor stator to be detected and the surface image of the standard motor stator.

[0136] The image processing module is used for feature extraction of the surface image of the motor stator to be detected and the surface image of the standard motor stator, and obtaining the shape features of the motor stator to be detected and the shape features of the standard motor stator.

[0137] The grid analysis module is used for defect detection of the grid to be analyzed, and determining the surface defect data of the motor stator to be detected.

[0138] The coordinate determination module is used for coordinate analysis of the motor stator to be detected and the standard motor stator, and determining the position coordinates of the copper wire of the motor stator to be detected and the position coordinates of the copper wire of the standard motor stator.

[0139] The model construction module is used for constructing the three-dimensional model of the standard motor stator and the three-dimensional model of the motor stator to be detected.

[0140] The quality evaluation module is used for comprehensive evaluation of the surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected, and determining the quality of the motor stator to be detected.

[0141] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above-mentioned embodiments, and the above-mentioned embodiments and descriptions in the specification are only the principles of the present application. Various changes and improvements can be made without departing from the spirit and scope of the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for detecting defects in a stator of an electrical machine based on image analysis, characterized in that, The method comprises the following steps: acquiring a surface image of a motor stator to be detected, performing feature analysis on the surface image of the motor stator to be detected, determining surface defect data of the motor stator to be detected and an external feature of the motor stator to be detected; determining a detection base point, performing internal detection on the motor stator to be detected according to the detection base point, and determining internal parameters of the motor stator to be detected; wherein the internal parameters of the motor stator to be detected include position coordinates of each copper wire of the motor stator to be detected and a winding direction of the copper wire of the motor stator to be detected; performing model construction on the external feature of the motor stator to be detected and the internal parameters of the motor stator to be detected to determine a three-dimensional model of the motor stator to be detected; performing comparative analysis on the surface defect data of the motor stator to be detected and the three-dimensional model of the motor stator to be detected to determine a surface quality score of the motor stator to be detected and an internal quality score of the motor stator to be detected; performing comprehensive evaluation on the surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected to determine a quality of the motor stator to be detected; the step of determining the detection base point, performing internal detection on the motor stator to be detected according to the detection base point, and determining the internal parameters of the motor stator to be detected specifically comprises the following steps: setting a surface center point of the motor stator to be detected as the detection base point, and constructing a three-dimensional rectangular coordinate system with the detection base point as the coordinate origin; placing the motor stator to be detected in the three-dimensional rectangular coordinate system and recording coordinate data of the external feature of the motor stator to be detected; based on acoustic wave technology, sequentially performing internal detection on the motor stator to be detected from the detection base point to obtain position coordinates of each copper wire of the motor stator to be detected; performing data extraction on a database system to obtain a production process of the motor stator; performing data extraction on the production process of the motor stator to determine an internal copper wire winding direction of the motor stator; based on the internal copper wire winding direction of the motor stator, serially processing the position coordinates of each copper wire of the motor stator to be detected to determine the winding direction of the copper wire of the motor stator to be detected.

2. The method for detecting defects in a stator of an electrical machine based on image analysis according to claim 1, characterized in that, the step of acquiring the surface image of the motor stator to be detected, performing feature analysis on the surface image of the motor stator, determining surface defect data of the motor stator to be detected and an external feature of the motor stator to be detected specifically comprises the following steps: performing image shooting on the motor stator to be detected by an image shooting device to obtain the surface image of the motor stator to be detected; performing region analysis on the surface image of the motor stator to be detected to determine the surface defect data of the motor stator to be detected; performing data extraction on the surface image of the motor stator to be detected to obtain the external feature of the motor stator to be detected.

3. A method of detecting defects in a stator of an electrical machine based on image analysis according to claim 2, characterized in that, the step of performing region analysis on the surface image of the motor stator to be detected to determine the surface defect data of the motor stator to be detected specifically comprises the following steps: performing grid division on the surface image of the motor stator to be detected to obtain a grid of an analysis unit; performing grid content analysis on the grid of the analysis unit; if there are redundant lines in the grid and the lines intersect at a point and extend outward, output grid protrusion information and record the protrusion position and the number of protrusions; If there are redundant lines in the grid, and the lines intersect at a point and extend inward, output the grid area recess information, record the recess position and recess number; If there are no redundant lines in the grid, output the grid normal information.

4. The method for detecting defects in a stator of an electrical machine based on image analysis according to claim 1, characterized in that, The model construction processing according to the shape characteristics of the motor stator to be detected and the internal parameters of the motor stator to be detected, to determine the three-dimensional model of the motor stator to be detected, specifically includes the following steps: Select the model construction starting point in the modeling software; Place the detection base point at the model construction starting point, and construct the internal structure model of the motor stator to be detected in the modeling software according to the coordinates of each position of the copper wire of the motor stator to be detected and the winding direction of the copper wire of the motor stator to be detected; Based on the coordinate data of the shape characteristics of the motor stator to be detected, the internal structure model of the motor stator to be detected is constructed again, and the three-dimensional model of the motor stator to be detected is determined.

5. A method of detecting defects in a stator of an electrical machine based on image analysis according to claim 4, characterized in that, The comparison and analysis of the surface defect data of the motor stator to be detected and the three-dimensional model of the motor stator to be detected, to determine the surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected, specifically includes the following steps: Judge the surface defect data of the motor stator to be detected and the set surface defect threshold value; If the total number of recesses and protrusions on the surface of the motor stator to be detected is greater than or equal to the set surface defect threshold value, output the surface quality of the motor stator to be detected as poor, etc.; If the total number of recesses and protrusions on the surface of the motor stator to be detected is less than the set surface defect threshold value, output the surface quality of the motor stator to be detected as excellent; Model construction processing is performed on the standard motor stator to obtain a three-dimensional model of the standard motor stator; Based on the three-dimensional model of the standard motor stator, the three-dimensional model of the motor stator to be detected is compared and analyzed to determine the internal quality score of the motor stator to be detected.

6. A method of detecting defects in a stator of an electrical machine based on image analysis according to claim 5, characterized in that, The model construction processing of the standard motor stator to obtain the three-dimensional model of the standard motor stator specifically includes the following steps: Image shooting of the standard motor stator is performed by an image shooting device to obtain a surface image of the standard motor stator; Data extraction processing is performed on the surface image of the standard motor stator to obtain the shape characteristics of the standard motor stator; The surface center of the standard motor stator is placed in a three-dimensional rectangular coordinate system, and the placement direction of the standard motor stator is consistent with the placement direction of the motor stator to be detected, and the coordinate data of the shape characteristics of the standard motor stator is recorded; Based on acoustic technology, internal detection processing is sequentially performed on the standard motor stator from the detection base point to obtain the coordinates of each position of the copper wire of the standard motor stator; Based on the winding direction of the internal copper wire of the motor stator, the coordinates of each position of the copper wire of the standard motor stator are serially processed to determine the winding direction of the copper wire of the standard motor stator; In the modeling software, a model is constructed according to the winding direction of the copper wire of the standard motor stator, the coordinates of each position of the copper wire of the standard motor stator, and the recorded coordinate data of the shape characteristics of the standard motor stator to obtain the three-dimensional model of the standard motor stator.

7. A method of detecting defects in a stator of an electrical machine based on image analysis according to claim 6, characterized in that, The three-dimensional model based on the standard motor stator is compared and analyzed with the three-dimensional model of the motor stator to be detected to determine the internal quality score of the motor stator to be detected, which specifically includes the following steps: The three-dimensional model of the standard motor stator and the three-dimensional model of the motor stator to be detected are analyzed and processed for coincidence degree; If the three-dimensional model of the motor stator to be detected and the three-dimensional model of the standard motor stator are completely coincident, the internal quality of the motor stator to be detected is output as excellent; If the three-dimensional model of the motor stator to be detected and the three-dimensional model of the standard motor stator are not completely coincident, the internal quality of the motor stator to be detected is output as poor.

8. A method of detecting defects in a stator of an electrical machine based on image analysis according to claim 7, characterized in that, The surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected are comprehensively evaluated to determine the quality of the motor stator to be detected, which specifically includes the following steps: The surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected are comprehensively evaluated; If the surface quality of the motor stator to be detected is poor, the quality of the motor stator to be detected is output as poor; If the internal quality of the motor stator to be detected is poor, the quality of the motor stator to be detected is output as poor; If the surface quality of the motor stator to be detected is excellent and the internal quality of the motor stator to be detected is excellent, the quality of the motor stator to be detected is output as excellent.

9. An image analysis based motor stator defect detection system for implementing an image analysis based motor stator defect detection method as claimed in any one of claims 1 to 8, characterized by It includes: An intelligent analysis terminal for controlling each module to evaluate the surface quality and internal quality of the motor stator to be detected and determine the quality of the motor stator to be detected, and for controlling data transmission and information interaction between each module; A database system for storing the production process of the motor stator; An image capturing device for capturing images of the motor stator to be detected and the standard motor stator to obtain surface images of the motor stator to be detected and the standard motor stator; An image processing module for feature extraction processing of the surface images of the motor stator to be detected and the standard motor stator to obtain the shape features of the motor stator to be detected and the standard motor stator; A grid analysis module for defect detection processing of the grid of the unit to be analyzed to determine the surface defect data of the motor stator to be detected; A coordinate determination module for coordinate analysis processing of the motor stator to be detected and the standard motor stator to determine the coordinates of each position of the copper wire of the motor stator to be detected and the coordinates of each position of the copper wire of the standard motor stator; A model construction module for constructing the three-dimensional model of the standard motor stator and the three-dimensional model of the motor stator to be detected; A quality evaluation module for comprehensive evaluation of the surface quality score of the motor stator to be detected and the internal quality score of the motor stator to be detected to determine the quality of the motor stator to be detected.

Images