Large casting internal defect intelligent detection shooting path planning method and system

Abstract

The invention belongs to the field of casting product quality detection, and discloses a large casting internal defect intelligent detection shooting path planning method and system.The method comprises the steps that angular points, blind holes and large-wall-thickness areas on a casting model are obtained to serve as quality control points; selecting one quality control point as a starting point, and controlling a ray source to traverse the surface of the casting model according to a preset interval to search supplementary path points between adjacent quality control points: if the interval between the adjacent quality control points meets the preset interval, starting next search by taking the next quality control point as a new starting point; if the distance does not meet the preset distance, searching by taking the previous quality control point as a starting point until the distance between the searched supplementary path point and the quality control point meets the preset distance; repeatedly searching until a stop condition is met; and planning an actual shooting path by using a genetic algorithm. According to the invention, the problems of defects caused by poor shooting angle in actual shooting, poor imaging effects such as mixed casting body structure and easy missing detection can be effectively avoided.

Description

Technical Field

[0001] The present application belongs to the technical field of casting product quality inspection, and more specifically, to a method and system for intelligent inspection of internal defects of large castings and filming path planning. Background Art

[0002] Light alloys such as titanium and aluminum have excellent properties such as low density, high specific strength, and strong corrosion resistance, and are widely used in aerospace, weapons and equipment and other fields. As the requirements for lightweight, integrated, and high-quality parts of advanced core defense equipment continue to increase, the volume of castings such as casings and cabins is getting larger and larger, and their internal structures are becoming more complex, which makes internal quality control more difficult, and it is easy to have internal defects such as loose holes / inclusions, and the requirements for non-destructive testing are rising. In the non-destructive testing of casting defects, X-ray testing is one of the most commonly used technologies. However, the existing shooting technology that selects the shooting area based on manual experience and controls the instrument to project rays has problems such as missed shots, multiple shots, and low efficiency. It cannot adapt to the development trend of casting detection and has become the main bottleneck restricting the urgent realization of intelligent, efficient, and high-quality batch production of high-end equipment. Summary of the invention

[0003] In view of the defects of the prior art, the purpose of this application is to provide a method and system for intelligent detection of internal defects of large castings. The method aims to solve the problems of missed shots, multiple shots and inefficiency in the existing shooting technology of manually selecting the shooting area and performing ray projection.

[0004] To achieve the above objectives, the present application provides a method for intelligent detection of internal defects of large castings and filming path planning, comprising the following steps:

[0005] S1 obtains corner points, blind holes and thick wall areas on the casting model as quality control points;

[0006] S2 selects one of the quality control points as the starting point, and controls the ray source to traverse the surface of the casting model according to the preset spacing to search for supplementary path points between adjacent quality control points: if the spacing between adjacent quality control points meets the preset spacing, the next search is started with the next quality control point as the new starting point; if the spacing between adjacent quality control points does not meet the preset spacing, the supplementary path points are searched in sequence with the previous quality control point as the starting point, until the spacing between the supplementary path point found and the next quality control point meets the preset spacing, and the next search is performed with the next quality control point as the new starting point; the search steps are repeated until the ray mapping areas corresponding to all quality control points and supplementary path points can fully cover the side of the casting model, and the search is stopped to obtain a path point set consisting of quality control points and supplementary path points;

[0007] S3 plans the actual filming path using a genetic algorithm based on the set of path points.

[0008] Furthermore, in step S2, the preset spacing is expressed as:

[0009]

[0010] Where a≤1 is the overlap coefficient between adjacent ray mapping areas; x 相邻 Indicates the distance between adjacent filming points, d 板 represents the side length of the imaging plate; r represents the radius of the bottom of the enveloping cylinder, which is also the distance between the imaging plate and the central axis of the enveloping cylinder; x 源n Indicates the distance from the nth supplementary path point on the casting model to the central axis of the envelope cylinder. is the distance from the ray source to the central axis of the enveloping cylinder.

[0011] Furthermore, in step S2, adjacent supplementary path points must meet the following preset conditions:

[0012] x 源n+1 >x 源n

[0013] Among them, x 源n+1 Indicates the distance from the n+1th supplementary path point on the casting model to the central axis of the enveloping cylinder;

[0014] When the preset conditions are met, the n+1th supplementary path point is used as the starting point for the next search; otherwise, the nth supplementary path point is used as the starting point, and the preset interval is reduced to search again until a new supplementary path point that meets the preset conditions is found, and the next search is started with the new supplementary path point as the starting point.

[0015] Furthermore, when planning the actual filming path, at each quality control point, a ray is made along the normal vector direction of the current quality control point toward the direction of the casting model to determine whether the ray meets the thickness condition and the intersection condition. The thickness condition is that the thickness of the casting model penetrated by the ray is less than or equal to the maximum penetration distance of the actual flaw detection equipment; the intersection condition is that the number of intersections between the ray and the casting model is less than or equal to 3.

[0016] Furthermore, in step S3, if the ray does not meet the thickness condition and / or the intersection condition at the current quality control point, the ray source and the corresponding imaging plate need to be synchronously deflected by a preset angle for shooting, and the ray source and the imaging plate always maintain a facing state;

[0017] The method for searching the preset angle is: deflect the ray source and the imaging plate synchronously, and make a new ray along the normal vector direction of the imaging plate after deflection toward the current quality control point; if the new ray satisfies the thickness condition and the intersection condition, the current deflection angle is used as the preset angle; if the new ray does not satisfy the thickness condition and / or the intersection condition, re-deflect to other angles until the new ray obtained after deflection satisfies the thickness condition and the intersection condition; preferably, the deflection direction is at least one.

[0018] Furthermore, in step S1, the corner points are obtained using a three-dimensional Harris response method.

[0019] Further, in step S1, the large wall thickness area is an area located in the first 10% of the wall thickness on the casting model.

[0020] Furthermore, in step S1, the step of obtaining a blind hole includes: obtaining a boundary ring, dividing the boundary ring into multiple sub-grids, and determining all sub-grids with only one exit; obtaining the ratio of the surface area to the volume of each sub-grid with only one exit, and taking the sub-grid corresponding to the maximum ratio as a blind hole.

[0021] According to a second aspect of the present application, there is also provided a system for implementing the method for intelligent detection of internal defects of large castings as described in any one of the above items, comprising:

[0022] A model building module, used to build a virtual flaw detection device and a casting model, wherein the virtual flaw detection device includes a ray source;

[0023] A supplementary path point acquisition module is used to control the ray source to start from any starting point at one end of the casting model, and to make circular and vertical movements around the central axis of the casting model to collect supplementary path points, wherein: the supplementary path point is the center point of the ray mapping area formed on the casting model by the rays emitted by the ray source, and the adjacent ray mapping areas are intersected, and the movement is stopped when all the ray mapping areas can fully cover the side of the casting model; it is also used to obtain quality control points at corner points, blind holes and large wall thickness of the casting model;

[0024] An additional path point acquisition module, used to acquire quality control points at corner points, blind holes and places with large wall thickness of the casting model;

[0025] The actual filming path output module is used to output the actual filming path based on the supplementary path points, corner points, blind holes and quality control points at locations with large wall thickness.

[0026] According to a third aspect of the present application, a computer-readable storage medium is further provided, wherein the computer-readable storage medium stores a computer program, and when the computer program runs on a processor, the processor executes any of the methods described above.

[0027] According to a fourth aspect of the present application, a computer program is also provided. When the computer program runs on a processor, the processor executes any of the methods described above.

[0028] In general, the above technical solutions conceived by this application have the following beneficial effects compared with the prior art:

[0029] (1) The filming path planning method for intelligent detection of internal defects of large castings provided by the present application finds the quality control points on the casting model, and then traverses and finds all the supplementary path points between adjacent quality control points. The ray mapping areas corresponding to all the quality control points and supplementary path points found can completely cover the side surface of the three-dimensional casting model, and generate a scientific and complete actual filming work path based on the path point set. The actual filming work path helps to achieve all-round detection of actual castings, effectively avoids the problems of missed shots and multiple shots in the actual shooting process, and further effectively reduces the missed detection of casting defects caused by missed shots.

[0030] (2) This application uses a genetic algorithm to generate a scientific, complete and comprehensive shooting path based on quality control points and supplementary path points between adjacent quality control points. Compared with manual image analysis and detection and manual programming, it greatly improves the efficiency of the shooting planning link. In addition, at each quality control point, a method for optimizing the shooting path of the quality control point is proposed to enable shooting at multiple angles of up and down and pitch at a specific quality control point, optimize the shooting angle, improve the effectiveness of shooting and the efficiency of filming, and avoid invalid shooting.

[0031] (3) The filming path planning method and system provided in the present application can be applied to non-destructive testing of castings of various shapes and sizes, especially large castings with increasingly larger volumes and increasingly complex internal structures, which can further reduce the difficulty of controlling the internal quality of castings. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a flow chart of a method for intelligent detection of internal defects of large castings and filming path planning provided in an embodiment of the present application;

[0033] Figure 2 It is a schematic diagram of a ray mapping area formed on a casting model by rays emitted by a ray source provided in an embodiment of the present application;

[0034] Figure 3It is a schematic diagram of the intersection of adjacent ray mapping areas formed on a casting model by rays emitted by a ray source provided in an embodiment of the present application;

[0035] Figure 4 It is a schematic diagram of the principle of planning a multi-angle shooting path for a quality control point provided in an embodiment of the present application;

[0036] Figure 5 It is a schematic diagram of the intersection of the ray and the casting model at the quality control point provided in the embodiment of the present application. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0038] The term "and / or" in this article is a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The symbol " / " in this article indicates that the associated objects are in an or relationship, for example, A / B means A or B.

[0039] The terms "first" and "second" in the specification and claims herein are used to distinguish different objects rather than to describe a specific order of the objects. For example, a first response message and a second response message are used to distinguish different response messages rather than to describe a specific order of the response messages.

[0040] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0041] In the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more than two. For example, multiple processing units refer to two or more processing units, etc.; multiple elements refer to two or more elements, etc.

[0042] Next, the technical solutions provided in the embodiments of the present application are introduced.

[0043] This embodiment provides a method for planning the filming path for intelligent detection of internal defects of large castings. In this method, the flaw detection equipment includes a CNC programmable ray source, an imaging plate, and a storage table. The ray source and the imaging plate can move freely in at least the horizontal and vertical directions, and the ray source can emit rays at a certain pitch angle. The imaging plate can adjust its posture to receive the rays following the ray source. In the ray transmission and reception working state, the ray source and the imaging plate keep their centers facing each other. Figure 1 As shown, the filming path planning method for intelligent detection of internal defects of large castings includes the following steps:

[0044] S1 obtains corner points, blind holes and thick wall areas on the casting model as quality control points.

[0045] The definition of quality control points is the area where the quality of the casting body must be paid attention to. That is, quality control points should inevitably appear in the process of shooting path planning, and their shapes should be determined as much as possible through multi-angle shooting. For areas where the quality needs to be checked, it is necessary to achieve full coverage of the mapping range, and calculate the range of the ray mapping area at its center point. If all the key quality attention areas fall within the range of the ray mapping area, the center point of the ray mapping area is determined as a quality control point as a supplementary path point. If there are still areas not included, other quality control points (i.e. supplementary path points) are found at the boundary of the ray mapping area until the corresponding ray mapping areas of all quality control points and supplementary path points completely cover the area.

[0046] In this embodiment, the three-dimensional Harris response method is used to first extract corner points. The principle of corner point extraction is: the Harris corner point detection method measures the degree of change of each point by calculating the autocorrelation matrix of the local area. Corner points are points with significant changes in the local area, and the neighborhood change rate of these points in different directions is large. For three-dimensional data, the Harris response can be expanded by calculating the covariance matrix of the three-dimensional image gradient. The expanded matrix is ​​as follows:

[0047]

[0048] Where M represents the matrix, I x , I y , I z are the gradients of the point in the x, y, and z directions respectively. The gradient is calculated by the Sobel convolution operator. The Harris calculation formula is as follows:

[0049] R = det(M) - k(trace(M)) 2 (2)

[0050] Wherein, R is the corner point response function, k is a constant, usually between 0.04 and 0.06, and a threshold is set to select the corner points. In this embodiment, 10% is selected as the threshold, and points above the threshold are selected and output as the corner points of the image, and the corner points are placed in the quality critical area.

[0051] Get the quality control points at the first 10% of the wall thickness of the casting model. Defects are very likely to occur in thick parts of the casting due to slow filling and other reasons, and an algorithm needs to be used to identify these areas. Because the normal vectors of the triangular facets are stored in the original STL format file, this embodiment uses the normal vector method to analyze the wall thickness of the entire casting model. Rays are extended from the three points of the triangular facet in the opposite direction of their normals, and the distance to the first triangular facet contacted by the rays is recorded. The average value of the wall thickness of the three points is taken as the wall thickness of the entire triangular facet, and then a certain weight of the wall thickness of the three adjacent triangles is taken and combined with the triangle to obtain the wall thickness of the entire triangular facet. Similarly, this embodiment selects areas with larger wall thickness of the casting (usually the first 10%) for recording, and determines these areas as quality control points.

[0052] During the pouring process, air easily accumulates at the bottom and side walls of the blind hole, especially during the pouring process, the air cannot be discharged smoothly, forming bubbles, which eventually solidify into pores. At the same time, due to the slow heat dissipation, shrinkage cavities and shrinkage porosity are easily formed, so it is necessary to analyze and search for the blind hole area. In step S2, after collecting the supplementary path points and the quality control points in the above embodiment, the following steps are also performed to obtain these blind holes:

[0053] (1) Obtain a boundary ring, divide the boundary ring into multiple sub-grids, and determine all sub-grids that have only one exit.

[0054] (2) Obtain the ratio of the surface area to the volume of each sub-grid with only one exit, and take the sub-grid corresponding to the maximum ratio as a blind hole. In step S3, these blind holes, the quality control points at the thickest wall, and the supplementary path points determined in step S2 are taken as a set to output the actual filming path.

[0055] Specifically, read the surface topology information of the casting model and identify the boundary ring on the model. The boundary ring described in this embodiment is composed of the boundary edges in the model. When an edge on a triangular facet is shared by only one face (generally, an edge is shared by two faces), and several boundary edges are surrounded by a ring, the boundary ring is found. Through the mesh segmentation method, the sub-grid at the boundary ring is analyzed, and the connectivity analysis is performed on it. If the sub-grid has only one outlet, the sub-grid may be a blind hole, and all sub-grids with only one outlet are found. Then calculate the ratio of the surface area to the volume of all sub-grids with only one outlet, select the sub-grid corresponding to the maximum ratio, and the corresponding structure of the sub-grid is considered to be a blind hole.

[0056] S2 selects a quality control point obtained in step S1 as the starting point, controls the ray source according to the preset spacing to traverse the surface of the casting model to find the supplementary path points between adjacent quality control points, and judges during the search process: if the spacing between adjacent quality control points meets the preset spacing, then the next search is started with the next quality control point as the new starting point; if the spacing between adjacent quality control points does not meet the preset spacing, then the supplementary path points are searched in sequence with the previous quality control point as the starting point, until the spacing between the supplementary path point found and the next quality control point meets the preset spacing, then the next search is performed with the next quality control point as the new starting point; repeat the aforementioned search steps until the ray mapping areas corresponding to all quality control points and supplementary path points can completely cover the side of the casting model, and stop the search to obtain a path point set consisting of quality control points and supplementary path points between adjacent quality control points;

[0057] Specifically, the ray source can be controlled to start from any quality control point of the casting model and make circular and vertical movements around the central axis of the casting model to collect additional path points. Figure 2 As shown, the supplementary path point is the center point of the ray mapping area (rectangular box in the figure) formed by the rays emitted by the ray source on the casting model. The ray source is in an unfixed area far away from the enveloping cylinder, and the imaging plate is located on the enveloping cylinder for receiving rays. Figure 3 As shown, adjacent ray mapping areas are intersected, and the search movement is stopped when all ray mapping areas can completely cover the side surfaces of the casting model.

[0058] More specifically, in order to ensure that no areas are missed, it is necessary to ensure that all areas of the casting can be inspected by the flaw detection equipment. The radiation source of the casting flaw detection equipment is a small circular plane, and the emitted radiation is Figure 2 As shown in the figure, the conical scattering penetrates the casting and is received by the rectangular plane imaging plate. In general, the geometric centers of the ray source and the imaging plate are in the normal direction of each other, that is, they are perpendicular to each other. The range of the final conical ray mapping area is the position of the imaging plate, that is, the effective range of ray scattering. This kind of scattering has an edge effect, that is, the rays are dense in the center and sparse around, so it is necessary to ensure that the key area is within the effective range of ray scattering and as close to the center as possible. Most casting flaw detection equipment generally uses an imaging plate size of 430mm×430mm. Therefore, in a normally placed casting, the closer the envelope surface is to the casting, the shorter the distance between its supplementary path points, and the farther the envelope surface is from the casting, the greater the distance between adjacent supplementary path points.

[0059] The aforementioned preset spacing is limited to:

[0060]

[0061] Where a≤1 is the overlap coefficient between adjacent ray mapping areas; x 相邻 Indicates the distance between adjacent filming points, d 板 represents the side length of the imaging plate; r represents the radius of the bottom of the enveloping cylinder, which is also the distance between the imaging plate and the central axis of the enveloping cylinder; x 源n Indicates the distance from the nth supplementary path point on the casting model to the central axis of the envelope cylinder. is the distance from the ray source to the central axis of the enveloping cylinder.

[0062] Each adjacent supplementary path point found must meet the following preset conditions:

[0063] x 源n+1 >x 源n (4)

[0064] Among them, x 源n+1 Indicates the distance from the n+1th supplementary path point on the casting model to the central axis of the enveloping cylinder;

[0065] If the preset condition is met, the n+1th supplementary path point is used as the starting point for the next search; otherwise, the nth supplementary path point is used as the starting point, and the preset distance is reduced to search again, such as by reducing the x in the preset distance formula. 源n , to reduce the calculated x 相邻 , until a new supplementary path point that meets the preset conditions is found, and then the next search begins with the new supplementary path point as the starting point.

[0066] S3 outputs the actual filming path based on the quality control points and supplementary path points found in step S1 and step S2.

[0067] Specifically, based on the path point set, a genetic algorithm (GA) is used to generate the actual filming path. During the actual filming path generation process, the properties of each path point are checked. If the path point is a quality control point found in step S1, such as Figure 4 As shown, the wall thickness check and occlusion check are performed on the points on the original model of the quality control point, and each quality control point in the initial path is optimized according to the following steps:

[0068] A ray is drawn along the normal vector direction of the current quality control point toward the casting model to determine whether the ray meets the thickness condition and the intersection condition. The thickness condition is: the thickness of the ray penetrating the casting model is less than or equal to the maximum penetration distance of the actual flaw detection equipment; the intersection condition is: the number of intersections between the ray and the casting model is less than or equal to 3.

[0069] If at the current quality control point, the ray along the normal vector direction of the current quality control point toward the casting model does not meet the thickness condition and the intersection condition, or does not meet one of the thickness condition and the intersection condition, such as Figure 5 As shown in the figure, when there are four intersections, h1 is the wall thickness between intersections 1 and 2, and h2 is the wall thickness between intersections 3 and 4. Then the ray source and the corresponding imaging plate need to be synchronously deflected by a preset angle to shoot the quality control point, and the ray source and the imaging plate should always be kept facing each other before and after rotation. The method for finding the preset angle is:

[0070] The ray source and the imaging plate are deflected synchronously. In this embodiment, the first deflection angle is A new ray is made toward the current quality control point along the normal vector direction of the deflected imaging plate. If the new ray meets the thickness condition and the intersection condition, the current deflection angle is used as the preset angle to be deflected during actual filming. If the new ray does not meet the thickness condition and the intersection condition, or the new ray does not meet any of the thickness condition and the intersection condition, the ray source and the imaging plate are synchronously deflected to other angles again, and it is determined whether the new ray obtained at this angle meets the thickness condition and the intersection condition. If so, the angle is used as the preset angle. If not, the deflection angle is readjusted until the new ray obtained at any deflection angle can simultaneously meet the thickness condition and the intersection condition and the search is stopped.

[0071] In a preferred embodiment, the deflection direction of the preset angle can be multiple, such as the ray source and the imaging plate can be deflected by preset angles from the left, right, pitch and elevation directions for multiple shots, or they can be deflected by preset angles in any other direction toward the casting for shooting.

[0072] The actual filming path generated by the genetic algorithm is stored in the computer in the form of point coordinates to establish a cylindrical coordinate system, in which the change of θ in the cylindrical coordinates is achieved by rotating the table by controlling the rotation axis of the table, the change of the z-axis is achieved by controlling the robotic arm to move the radiation source and the imaging plate in the vertical direction, and the deflection operation is completed by rotating the imaging plate. At the same time, the radiation source needs to cooperate with the movement of the imaging plate to keep it centered. Based on this relationship, a CNC coordinate code is generated and transmitted to the machine to control the machine movement.

[0073] In other preferred embodiments, a system for implementing the method for intelligent detection of internal defects of large castings and filming path planning as described in any of the above embodiments is also provided, comprising:

[0074] Quality control point acquisition module, used to acquire corner points, blind holes and thick wall areas on the casting model as quality control points;

[0075] A path point set acquisition module is used to select a quality control point as a starting point, and control the ray source to traverse the surface of the casting model according to a preset spacing to search for supplementary path points between adjacent quality control points: if the spacing between adjacent quality control points meets the preset spacing, the next search is started with the next quality control point as a new starting point; if the spacing between adjacent quality control points does not meet the preset spacing, the supplementary path points are searched in sequence with the previous quality control point as the starting point, until the spacing between the supplementary path point found and the next quality control point meets the preset spacing, and the next search is performed with the next quality control point as the new starting point; the search is stopped until the ray mapping areas corresponding to all quality control points and supplementary path points can fully cover the side of the casting model, and a path point set consisting of quality control points and supplementary path points is obtained;

[0076] The actual filming path planning module is used to plan the actual filming path based on the set of path points.

[0077] It can be understood that the detailed functional implementation of each of the above-mentioned units / modules can be found in the introduction of the aforementioned method embodiment, and will not be repeated here.

[0078] The above embodiment has the following advantages:

[0079] 1. A search method for all quality control points of castings (supplementary path points, corner points, blind holes, and quality control points at thick walls) is proposed, which automatically identifies areas of castings that are more prone to defects from multiple dimensions, greatly simplifying the workload of manual screening.

[0080] 2. A method for planning a detection path without blind spots based on quality control points and flaw detection models is proposed. This method not only realizes the planning of non-destructive testing paths for castings without blind spots, but also conducts multi-angle shooting and detection on quality control points, effectively preventing problems such as defects in defect-prone areas being easily obscured by the casting body structure and a single defect assessment angle, and is conducive to subsequent archiving and investigation.

[0081] It should be understood that the above-mentioned device is used to execute the method in the above-mentioned embodiment. The implementation principle and technical effect of the corresponding program module in the device are similar to those described in the above-mentioned method. The working process of the device can refer to the corresponding process in the above-mentioned method, which will not be repeated here.

[0082] Based on the method in the above embodiment, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method in the above embodiment.

[0083] Based on the method in the above embodiment, an embodiment of the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the method in the above embodiment.

[0084] It is understandable that the processor in the embodiment of the present application can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor can be a microprocessor or any conventional processor.

[0085] The method steps in the embodiments of the present application can be implemented by hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, mobile hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC.

[0086] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions may be transmitted from a website site, a computer, a server or a data center to another website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.

[0087] It should be understood that the various numerical numbers involved in the embodiments of the present application are only used for the convenience of description and are not used to limit the scope of the embodiments of the present application.

[0088] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for intelligent detection of internal defects of large castings and filming path planning, characterized in that: The following steps are involved: S1 obtains corner points, blind holes and thick wall areas on the casting model as quality control points; S2 selects one of the quality control points as the starting point, controls the ray source to traverse the casting model surface according to the preset spacing to search for supplementary path points between adjacent quality control points: if the spacing between adjacent quality control points meets the preset spacing, starts the next search with the next quality control point as the new starting point; If the distance between adjacent quality control points does not meet the preset distance, the previous quality control point is used as the starting point to search for supplementary path points in sequence until the distance between the found supplementary path point and the next quality control point meets the preset distance, and then the next search is performed with the next quality control point as the new starting point; Repeat the searching step until the ray mapping areas corresponding to all the quality control points and the supplementary path points can completely cover the side of the casting model, and then stop searching to obtain a path point set consisting of the quality control points and the supplementary path points; S3 plans the actual filming path using a genetic algorithm based on the set of path points.

2. The method for intelligent detection of internal defects of large castings according to claim 1, characterized in that: In step S2, the preset spacing is expressed as: Where a≤1 is the overlap coefficient between adjacent ray mapping areas; x 相邻 Indicates the distance between adjacent filming points, d 板 represents the side length of the imaging plate; r represents the radius of the bottom of the enveloping cylinder, which is also the distance between the imaging plate and the central axis of the enveloping cylinder; x 源n Indicates the distance from the nth supplementary path point on the casting model to the central axis of the envelope cylinder. is the distance from the ray source to the central axis of the enveloping cylinder.

3. The method for intelligent detection of internal defects of large castings according to claim 2, characterized in that: In step S2, adjacent supplementary path points must meet the following preset conditions: x 源n+1 >x 源n Among them, x 源n+1 Indicates the distance from the n+1th supplementary path point on the casting model to the central axis of the enveloping cylinder; When the preset conditions are met, the n+1th supplementary path point is used as the starting point for the next search; otherwise, the nth supplementary path point is used as the starting point, and the preset interval is reduced to search again until a new supplementary path point that meets the preset conditions is found, and the next search is started with the new supplementary path point as the starting point.

4. The method for intelligent detection of internal defects of large castings according to claim 1, characterized in that: In step S3, when planning the actual filming path, at each quality control point, a ray is made along the normal vector direction of the current quality control point toward the direction of the casting model to determine whether the ray meets the thickness condition and the intersection condition. The thickness condition is that the thickness of the casting model penetrated by the ray is less than or equal to the maximum penetration distance of the actual flaw detection equipment; the intersection condition is that the number of intersections between the ray and the casting model is less than or equal to 3.

5. The method for intelligent detection of internal defects of large castings according to claim 4, characterized in that: In step S3, if the ray does not meet the thickness condition and / or the intersection condition at the current quality control point, the ray source and the corresponding imaging plate need to be synchronously deflected by a preset angle for shooting, and the ray source and the imaging plate always remain in a facing state; the method for finding the preset angle is: Deflect the ray source and the imaging plate synchronously, and make a new ray toward the current quality control point along the normal vector direction of the deflected imaging plate, and if the new ray satisfies the thickness condition and the intersection condition, the current deflection angle is used as the preset angle; If the new ray does not satisfy the thickness condition and / or the intersection condition, it is re-deflected at other angles until the new ray obtained after the deflection satisfies the thickness condition and the intersection condition and the search is stopped; preferably, the deflection direction is at least one.

6. The method for intelligent detection of internal defects of large castings according to claim 1, characterized in that: In step S1, the corner point is obtained by using a three-dimensional Harris response method; the large wall thickness area is an area located in the first 10% of the wall thickness on the casting model.

7. The method for intelligent detection of internal defects of large castings according to claim 1, characterized in that: In step S1, the step of obtaining the blind hole includes: A boundary ring is obtained, and the boundary ring is divided into a plurality of sub-grids, and all sub-grids having only one exit are determined; a ratio of the surface area to the volume of each sub-grid having only one exit is obtained, and a sub-grid corresponding to the maximum ratio is used as a blind hole.

8. A system for implementing the method for intelligent detection of internal defects of large castings according to any one of claims 1 to 7, characterized in that: include: Quality control point acquisition module, used to acquire corner points, blind holes and thick wall areas on the casting model as quality control points; A path point set acquisition module is used to select one of the quality control points as a starting point, and control the ray source to traverse the casting model surface according to a preset spacing to search for supplementary path points between adjacent quality control points: if the spacing between adjacent quality control points meets the preset spacing, the next search is started with the next quality control point as a new starting point; If the distance between adjacent quality control points does not meet the preset distance, the supplementary path points are searched in sequence starting from the previous quality control point until the distance between the supplementary path point found and the next quality control point meets the preset distance, and the next search is performed with the next quality control point as the new starting point; the search is stopped until the ray mapping areas corresponding to all quality control points and supplementary path points can fully cover the side of the casting model, and a path point set consisting of quality control points and supplementary path points is obtained; The actual filming path planning module is used to plan the actual filming path based on the path point set using a genetic algorithm.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program runs on a processor, the processor is caused to execute the method according to any one of claims 1 to 7.

10. A computer program, characterized in that When the computer program runs on a processor, the processor is caused to execute the method according to any one of claims 1 to 7.

Images