阅读说明：本技术 一种大批量航空零件质检方法 (Quality inspection method for large-batch aviation parts ) 是由 汪俊 吴波 李红卫 张沅 于 2020-05-19 设计创作，主要内容包括：一种大批量航空零件质检方法,包括以下步骤：识别零件类型,得到零件属性信息；根据零件属性信息规划扫描路径,根据扫描路径对航空零件进行扫描,得到各帧点云数据；通过针对各帧点云数据的配准拼接得到零件的实测点云模型,并建立实测点云模型和理论模型之间的转换矩阵；针对实测点云模型进行零件的特征提取,得到零件特征数据,并将零件特征数据与理论模型指标数据进行对比衡量,得到误差指标评估结果,更新零件检测报告数据,完成单个零件质检；重复上述步骤,直到所有零件质检完成,生成最终检测报告。本发明提高了航空零件检测效率与自动化水平,整个检测过程可以无需人工干预。(A quality inspection method for large-batch aviation parts comprises the following steps: identifying the type of the part to obtain part attribute information; planning a scanning path according to the attribute information of the part, and scanning the aviation part according to the scanning path to obtain point cloud data of each frame; obtaining an actual measurement point cloud model of the part by registering and splicing point cloud data of each frame, and establishing a conversion matrix between the actual measurement point cloud model and a theoretical model; extracting the characteristics of the parts aiming at the actually measured point cloud model to obtain the characteristic data of the parts, comparing and measuring the characteristic data of the parts with the index data of the theoretical model to obtain an error index evaluation result, updating the detection report data of the parts and completing the quality inspection of the single parts; and repeating the steps until the quality inspection of all the parts is finished, and generating a final inspection report. The invention improves the detection efficiency and automation level of aviation parts, and the whole detection process can be free from manual intervention.)

1. A quality inspection method for large-batch aviation parts is characterized by comprising the following steps:

identifying the type of the part to obtain part attribute information;

planning a scanning path according to the attribute information of the part, and scanning the aviation part according to the scanning path to obtain point cloud data of each frame;

obtaining an actual measurement point cloud model of the part by registering and splicing point cloud data of each frame, and establishing a conversion matrix between the actual measurement point cloud model and a theoretical model;

extracting the characteristics of the parts aiming at the actually measured point cloud model to obtain the characteristic data of the parts, comparing and measuring the characteristic data of the parts with the index data of the theoretical model to obtain an error index evaluation result, updating the detection report data of the parts and completing the quality inspection of the single parts;

and repeating the steps until the quality inspection of all the parts is finished, and generating a final inspection report.

2. The method for quality inspection of bulk aviation parts according to claim 1, wherein the identifying part types and obtaining part attribute information specifically comprises the steps of:

the infrared sensor detects a part on the production line, and when a new part is detected, the system retrieves attribute information of the part prestored in the system through a two-dimensional code on the part, wherein the attribute information comprises theoretical model information of the part.

3. The method for quality inspection of high volume aerospace parts according to claim 1, wherein the step of planning the scan path based on part attribute information comprises the steps of:

calculating theoretical digital-analog characteristics of the part according to the part attribute information, wherein the theoretical digital-analog characteristics comprise the outline curved surface characteristics, the hole characteristics and the boundary characteristics of the part;

calculating to obtain a preliminary scanning path according to theoretical digital-analog characteristics of the part;

discretizing the primary scanning path, then generating by a scanning normal method and calculating a scanning track to obtain a part scanning path, and controlling the mechanical arm to drive the three-dimensional laser scanner to scan the part from different angles according to the part scanning path.

4. The method for quality inspection of high-volume aerospace parts according to claim 3, wherein the scanning angle of the three-dimensional laser scanner is calculated and obtained according to part attribute information.

5. The method for quality inspection of large-batch aviation parts according to claim 1, wherein the step of obtaining an actually measured point cloud model of the part by registration and splicing of each frame of point cloud data and establishing a transformation matrix between the actually measured point cloud model and a theoretical model comprises the following steps:

establishing an accurate corresponding point set of each frame of point cloud data through an ICP (inductively coupled plasma) algorithm;

integrating the parameterized color and geometric information of corresponding points in the accurate corresponding point set to obtain an optimization target;

minimizing the optimization target to obtain a fine registration transformation matrix;

registering and splicing the point cloud data of each frame according to the fine registration transformation matrix to obtain an actually measured point cloud model of the part;

extracting characteristic points aiming at the actually measured point cloud model, matching the extracted characteristic points with the characteristic points in the theoretical model of the part, solving to obtain a conversion matrix, and establishing a global coordinate system for the conversion of the actually measured point cloud model so that the actually measured point cloud model and the theoretical model are in the same global coordinate system.

6. The method for quality inspection of high volume aerospace parts according to claim 1, wherein the part feature data includes part contour features, hole features and boundary features.

Technical Field

The invention belongs to the field of part quality inspection, and particularly relates to a mass aviation part quality inspection method.

Background

The aircraft is assembled by huge part, because the aircraft requires very high to the security, so also very high to part quality requirement, nevertheless to such a large amount of and a great variety of aircraft parts if all by artifical detection not only waste time and energy, it is with high costs, detection effect also depends on personnel's experience technique and input state, and can't make the testing result obtain guaranteeing.

Disclosure of Invention

Aiming at the defects in the prior art, the invention provides a quality inspection method for large-batch aviation parts, which can quickly evaluate the quality of the large-batch aviation parts.

In order to achieve the purpose, the invention adopts the following technical scheme:

a quality inspection method for large-batch aviation parts comprises the following steps:

identifying the type of the part to obtain part attribute information;

planning a scanning path according to the attribute information of the part, and scanning the aviation part according to the scanning path to obtain point cloud data of each frame;

obtaining an actual measurement point cloud model of the part by registering and splicing point cloud data of each frame, and establishing a conversion matrix between the actual measurement point cloud model and a theoretical model;

extracting the characteristics of the parts aiming at the actually measured point cloud model to obtain the characteristic data of the parts, comparing and measuring the characteristic data of the parts with the index data of the theoretical model to obtain an error index evaluation result, updating the detection report data of the parts and completing the quality inspection of the single parts;

and repeating the steps until the quality inspection of all the parts is finished, and generating a final inspection report.

In order to optimize the technical scheme, the specific measures adopted further comprise:

further, the identifying the type of the part and obtaining the part attribute information specifically includes the following steps:

the infrared sensor detects a part on the production line, and when a new part is detected, the system retrieves attribute information of the part prestored in the system through a two-dimensional code on the part, wherein the attribute information comprises theoretical model information of the part.

Further, the planning of the scanning path according to the part attribute information includes the following steps:

calculating theoretical digital-analog characteristics of the part according to the part attribute information, wherein the theoretical digital-analog characteristics comprise the outline curved surface characteristics, the hole characteristics and the boundary characteristics of the part;

calculating to obtain a preliminary scanning path according to theoretical digital-analog characteristics of the part;

discretizing the primary scanning path, then generating by a scanning normal method and calculating a scanning track to obtain a part scanning path, and controlling the mechanical arm to drive the three-dimensional laser scanner to scan the part from different angles according to the part scanning path.

Further, the scanning angle of the three-dimensional laser scanner is calculated and obtained according to the part attribute information.

Further, the step of obtaining the actually measured point cloud model of the part by registering and splicing the point cloud data of each frame and establishing the conversion matrix between the actually measured point cloud model and the theoretical model comprises the following steps:

establishing an accurate corresponding point set of each frame of point cloud data through an ICP (inductively coupled plasma) algorithm;

integrating the parameterized color and geometric information of corresponding points in the accurate corresponding point set to obtain an optimization target;

minimizing the optimization target to obtain a fine registration transformation matrix;

registering and splicing the point cloud data of each frame according to the fine registration transformation matrix to obtain an actually measured point cloud model of the part;

extracting characteristic points aiming at the actually measured point cloud model, matching the extracted characteristic points with the characteristic points in the theoretical model of the part, solving to obtain a conversion matrix, and establishing a global coordinate system for the conversion of the actually measured point cloud model so that the actually measured point cloud model and the theoretical model are in the same global coordinate system.

Further, the part feature data includes a contour feature, a hole feature, and a boundary feature of the part.

The invention has the beneficial effects that:

the quality inspection method for the large-batch aviation parts provided by the invention uses a flow line mode to detect the large-batch aviation parts, adopts a mechanical arm to carry a three-dimensional laser scanner to realize the acquisition of point cloud data of the parts, then reconstructs point cloud of the aviation parts in a three-dimensional reconstruction mode, and compares actual characteristic indexes of the parts with digital-analog indexes of part standards by calculation, so that the quality of the aviation parts is rapidly evaluated in a large batch, a detection report is automatically generated, the detection efficiency and the automation level of the aviation parts are improved, and the whole detection process can be free from manual intervention.

Drawings

FIG. 1 is a schematic flow chart of the method of the present invention.

Detailed Description

The present invention will now be described in further detail with reference to the accompanying drawings.

It should be noted that the terms "upper", "lower", "left", "right", "front", "back", etc. used in the present invention are for clarity of description only, and are not intended to limit the scope of the present invention, and the relative relationship between the terms and the terms is not limited by the technical contents of the essential changes.

In one embodiment of the present invention, as shown in fig. 1, a method for quality inspection of a large batch of aviation parts comprises the following steps:

step one, identifying the type of the part

When the system is started, the aviation parts on the assembly line are firstly detected by the infrared sensor, the system senses that a new part comes at the moment, then the system reads the two-dimensional code attached to the part, the system can retrieve the attribute information of each part pre-stored in the system through the two-dimensional code on the part, and the attribute information mainly comprises theoretical digital-analog information of the part;

step two, scanning path planning

And C, calculating theoretical digital-analog characteristics of the part according to the theoretical digital-analog information of the part obtained in the step I. Extracting the appearance curved surface characteristics, the hole characteristics, the boundary characteristics and the like of the part according to a theoretical digital model of the part, calculating according to the characteristics to obtain a scanning path, discretizing the scanning path, generating a scanning normal direction, calculating a scanning track to obtain a final scanning path of the part, and controlling a mechanical arm to drive a three-dimensional laser scanner to scan the part from different angles according to the path;

step three, acquiring point cloud data

Guiding the mechanical arm to scan the part by using a three-dimensional laser scanner according to the scanning path obtained in the step two, and scanning the part from multiple angles to obtain each frame of point cloud data;

step four, registration splicing and model conversion of point cloud data

And C, establishing an actual measurement model of the part after registration and splicing of the point cloud data of each frame of multiple angles obtained in the step three, namely finding out common points among the point cloud data of each frame, covering the common points to obtain a complete actual measurement model of the part, establishing an accurate corresponding point set in the point cloud to be matched by adopting an ICP (inductively coupled plasma) algorithm, integrating colors and geometric information of parameterized corresponding points to form an optimization target, solving the accurate registration transformation matrix by minimizing the target, and further realizing registration and splicing of the point cloud data to obtain the actual measurement aviation part point cloud model. Extracting characteristic points of the actually measured aviation part point cloud model, matching the actually measured characteristic points with the theoretical model characteristic points of the part, solving a conversion matrix, and establishing a global coordinate system for the actually measured model conversion so that the theoretical model and the actually measured model of the part can be compared under the same global coordinate system;

step five, processing the characteristics of the parts

Extracting the characteristics of the part from the part actual measurement model obtained in the step four, wherein the characteristic data of the part such as hole characteristics, shaft characteristics, surface characteristics, boundary characteristics and the like are mainly obtained through calculation;

step six, evaluating part indexes

And C, performing corresponding index calculation on the part according to the part characteristic data obtained by calculation in the step five, and comparing and measuring the index data of the actually measured part model with the index data of the theoretical part model to obtain an error index evaluation result. The theoretical digital model is established based on an accurate aviation part model;

seventhly, updating the data of the part detection report

And updating the part detection report data according to the error index evaluation result of the aviation parts obtained in the sixth step, so that the complete aviation part detection process is completed, then conveying the next aviation part to a specified position by a production line conveyor belt for detection, and automatically generating a final detection report until all parts are detected.

The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above-mentioned embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.