CN102785129B - The online test method of the surface machining accuracy of complex parts - Google Patents
The online test method of the surface machining accuracy of complex parts Download PDFInfo
- Publication number
- CN102785129B CN102785129B CN201210266355.6A CN201210266355A CN102785129B CN 102785129 B CN102785129 B CN 102785129B CN 201210266355 A CN201210266355 A CN 201210266355A CN 102785129 B CN102785129 B CN 102785129B
- Authority
- CN
- China
- Prior art keywords
- rbf
- network
- error
- ball
- curved surface
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000003754 machining Methods 0.000 title claims abstract description 25
- 238000010998 test method Methods 0.000 title claims 5
- 238000001514 detection method Methods 0.000 claims abstract description 57
- 238000003801 milling Methods 0.000 claims abstract description 28
- 238000000034 method Methods 0.000 claims abstract description 25
- 230000008569 process Effects 0.000 claims abstract description 8
- 238000012549 training Methods 0.000 claims description 13
- 210000002569 neuron Anatomy 0.000 claims description 9
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 238000004519 manufacturing process Methods 0.000 claims description 7
- 230000004913 activation Effects 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000013459 approach Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000000946 synaptic effect Effects 0.000 claims description 3
- 230000007935 neutral effect Effects 0.000 claims 7
- 238000000205 computational method Methods 0.000 claims 2
- 239000000523 sample Substances 0.000 abstract description 44
- 238000012545 processing Methods 0.000 abstract description 14
- 238000005259 measurement Methods 0.000 abstract description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Landscapes
- Numerical Control (AREA)
- Machine Tool Sensing Apparatuses (AREA)
Abstract
本发明是一种复杂零件的曲面加工精度的在线检测方法。包括如下步骤：1）将接触式触发测头安装在数控铣床的主轴上；2）对被测曲面进行检测路径规划；3）驱动数控铣床，逐一获取测头与待测曲面接触点处的测球中心的坐标；测球是接触式触发测头的一个部件，是一个具有高精度和高硬度的圆球，安装在测头主体上并与被测曲面直接接触；4)对上述测球中心的坐标进行测球半径补偿、测头预行程误差补偿和机床误差补偿，从而获得测点高精度的检测结果；5)将检测结果与零件的理想CAD模型进行对比分析，找到各个测点与CAD模型对应的偏差，从而获得待测曲面的加工精度。本发明使高精度的加工精度检测过程直接在数控铣床上进行，避免了零件多次装夹所带来的定位误差。
The invention is an on-line detection method for the processing precision of the curved surface of complex parts. It includes the following steps: 1) Install the touch trigger probe on the spindle of the CNC milling machine; 2) Plan the detection path for the surface to be measured; 3) Drive the CNC milling machine to obtain the measurement points at the contact points between the probe and the surface to be measured one by one. The coordinates of the ball center; the measuring ball is a part of the contact trigger probe, which is a ball with high precision and high hardness, installed on the main body of the probe and directly in contact with the measured surface; 4) the center of the above measuring ball The coordinates of the measuring ball radius compensation, the probe pre-travel error compensation and the machine tool error compensation, so as to obtain the high-precision detection results of the measuring points; The deviation corresponding to the model, so as to obtain the machining accuracy of the surface to be measured. The invention enables the high-precision machining accuracy detection process to be directly carried out on the numerical control milling machine, thereby avoiding the positioning error caused by multiple clamping of parts.
Description
技术领域technical field
本发明是一种复杂零件的曲面加工精度的在线检测方法，特别是一种在数控铣床上对复杂曲面零件进行加工后、直接在该机床上对该零件进行加工精度的在线检测的复杂零件的曲面加工精度的在线检测方法，属于复杂零件的曲面加工精度的在线检测方法的改造技术。The present invention is an on-line detection method of the machining accuracy of the curved surface of complex parts, in particular to a method for processing complex curved surface parts on a CNC milling machine, and then directly performing online detection of the processing accuracy of the complex parts on the machine tool. The invention relates to an online detection method of curved surface machining accuracy, which belongs to the transformation technology of the online detection method of curved surface processing accuracy of complex parts.
背景技术Background technique
随着制造业技术和装备的不断进步，对复杂零件/产品的精度、效率、质量和外观要求愈来愈高。在复杂曲面零件的生产过程中，需要用相应的检测技术对其加工精度进行检测和控制。基于三坐标测量机（CMM）的检测技术常用于精密零件的形位精度检测，但存在工件二次装夹定位误差问题及大型零件测量的局限性问题。在数控铣床上直接进行加工精度的在线检测，形成“加工－测量－补偿”的闭环加工检测系统，具有十分重要的意义。With the continuous advancement of manufacturing technology and equipment, the requirements for precision, efficiency, quality and appearance of complex parts/products are getting higher and higher. In the production process of complex curved surface parts, it is necessary to use corresponding detection technology to detect and control its machining accuracy. The detection technology based on three-coordinate measuring machine (CMM) is often used in the shape and position accuracy detection of precision parts, but there are problems of positioning errors in the secondary clamping of workpieces and limitations in the measurement of large parts. It is of great significance to directly conduct on-line detection of machining accuracy on the CNC milling machine and form a closed-loop processing detection system of "processing-measurement-compensation".
在线检测技术即是在数控铣床上直接对数控加工后的零件进行检测，通过对检测数据进行分析得到零件的加工精度。该技术适合于各类尺寸大小、复杂曲面零件的加工精度检测，能有效提高数控铣床的零件加工精度，已得到学术界和工业界的广泛关注。数控铣床零件加工精度的在线检测方法，具有广泛的应用前景。On-line detection technology is to directly detect the CNC-machined parts on the CNC milling machine, and obtain the machining accuracy of the parts by analyzing the detection data. This technology is suitable for the machining accuracy detection of various sizes and complex curved surface parts, and can effectively improve the machining accuracy of CNC milling machines. It has attracted extensive attention from academia and industry. The on-line detection method of machining accuracy of CNC milling machine parts has a wide application prospect.
目前国内所开发的在线检测系统，检测功能较为薄弱，基本上停留于数控系统本身所提供的一些功能，检测对象大多数是针对简单规则形体（如平面、圆、圆柱和凸台等），而针对复杂曲面的在线检测研究较少。At present, the online detection system developed in China has relatively weak detection functions, and basically stays at some functions provided by the CNC system itself. Most of the detection objects are for simple regular shapes (such as planes, circles, cylinders and bosses, etc.), while There are few researches on online detection for complex surfaces.
数控铣床测量环境复杂，误差影响因素多，数控铣床在线检测与高精度的CMM相比仍然存在较大差距，难以获得满意的实际测量精度。The measurement environment of CNC milling machine is complex, and there are many factors affecting the error. Compared with the high-precision CMM, there is still a big gap between the online detection of CNC milling machine and it is difficult to obtain satisfactory actual measurement accuracy.
发明内容Contents of the invention
本发明的目的在于考虑上述问题而提供一种可以获得高精度的检测结果，进而得到被加工曲面的加工精度的复杂零件的曲面加工精度的在线检测方法。本发明克服了复杂曲面需要移动到CMM上才能获得高精度检测结果的不足之处，能直接在数控铣床上对具有复杂曲面特征的零件进行形面精度的在线检测，通过误差补偿获得高精度的检测结果，最终得到零件的形面加工精度，有效地提高了生产效率。The object of the present invention is to consider the above problems and provide an online detection method of the curved surface processing accuracy of complex parts that can obtain high-precision detection results and further obtain the processing accuracy of the processed curved surface. The invention overcomes the disadvantage that complex curved surfaces need to be moved to the CMM to obtain high-precision detection results, and can directly perform online detection of shape and surface accuracy on parts with complex curved surface features on a CNC milling machine, and obtain high-precision results through error compensation The detection results can finally obtain the machining accuracy of the shape and surface of the parts, which effectively improves the production efficiency.
本发明的技术方案是：本发明的复杂零件的曲面加工精度的在线检测方法，包括有如下步骤：The technical solution of the present invention is: the online detection method of the curved surface machining accuracy of the complex part of the present invention, comprises the following steps:
1）将接触式触发测头安装在数控铣床的主轴上；1) Install the touch trigger probe on the spindle of the CNC milling machine;
2）对被测曲面进行检测路径规划；2) Carry out inspection path planning on the surface to be tested;
3）驱动数控铣床，逐一获取测头与待测曲面接触点处的测球中心（3）的坐标；测球是接触式触发测头的一个部件，其形状是一个具有高制造精度和高硬度的圆球，安装在测头主体上并与被测曲面直接接触；3) Drive the CNC milling machine to obtain the coordinates of the ball center (3) at the contact point between the probe and the surface to be measured one by one; the ball is a part of the contact trigger probe, and its shape is a high manufacturing precision and high hardness The ball is installed on the main body of the probe and is in direct contact with the measured surface;
4)对上述测球中心的坐标进行测球半径补偿、测头预行程误差补偿和机床误差补偿，从而获得测点高精度的检测结果；其中机床误差需采用机床误差检测仪器得到；4) Perform ball radius compensation, probe pre-travel error compensation and machine tool error compensation on the coordinates of the above-mentioned measuring ball center, so as to obtain a high-precision detection result of the measuring point; the machine tool error needs to be obtained by using a machine tool error detection instrument;
5)将检测结果与零件的理想CAD模型进行对比分析，找到各个测点与CAD模型对应的偏差，从而获得待测曲面的加工精度。5) Compare and analyze the test results with the ideal CAD model of the part, and find the deviation between each measuring point and the CAD model, so as to obtain the machining accuracy of the surface to be tested.
上述测球半径补偿、测头预行程误差补偿需要计算出曲面测点的法矢方向，对于测头预行程误差补偿，还需要计算出测头在各个测点法矢方向的预行程误差，补偿方法是采用基于径向基函数（RBF）的神经网络进行预测。The above-mentioned probe ball radius compensation and probe pre-travel error compensation need to calculate the normal vector direction of the surface measuring point. For the probe pre-travel error compensation, it is also necessary to calculate the pre-travel error of the probe in the normal vector direction of each measuring point, and the compensation The method is to use a neural network based on radial basis function (RBF) for prediction.
上述机床误差补偿所用的机床误差检测仪器为激光干涉仪。The machine tool error detection instrument used for the above machine tool error compensation is a laser interferometer.
上述测点与CAD模型对应的偏差采用点到曲面的最小距离的计算方法求解。The deviation between the above measuring points and the CAD model is solved by calculating the minimum distance from the point to the surface.
上述基于径向基函数RBF的神经网络进行预测的方法如下：The above-mentioned neural network based on radial basis function RBF predicts the method as follows:
正则化RBF网络是一种具有单隐含层的三层前馈局部逼近网络，已经证明，它与BP网络都能以任意精度逼近任意连续函数；并且，相比BP网络，其训练时间更短,并且它同时满足对样本的逼近误差和逼近曲线的平滑性,在实践中，网络的监督训练可以看成是一种曲线拟合的过程,利用正则化RBF算法，通过对网络的训练，实现输入和输出空间之间的非线性映射；The regularized RBF network is a three-layer feed-forward local approximation network with a single hidden layer. It has been proved that it and the BP network can approximate any continuous function with arbitrary precision; and, compared with the BP network, its training time is shorter , and it satisfies the approximation error of the sample and the smoothness of the approximation curve at the same time. In practice, the supervised training of the network can be regarded as a curve fitting process, and the regularized RBF algorithm is used to train the network to achieve non-linear mapping between input and output spaces;
正则化RBF网络的拓扑结构由一个径向基神经元的隐层、一个线性神经元的输出层组成，网络的输入点数量为N，隐节点数量为P个，输出节点数量为l个；网络的隐节点数等于输入样本数，并将所有输入样本设为径向基函数的中心，各径向基函数取统一的扩展常数；The topology of the regularized RBF network consists of a hidden layer of radial basis neurons and an output layer of linear neurons. The number of input points of the network is N, the number of hidden nodes is P, and the number of output nodes is l; the network The number of hidden nodes is equal to the number of input samples, and all input samples are set as the center of the radial basis function, and each radial basis function takes a uniform expansion constant;
RBF实现由输入到输出的映射，()采用径向基函数为任一隐节点的激活函数，选用Gauss函数作为径向基函数；W为输出权矩阵，其中（；）为隐层第个节点到输出层第个节点间的突触权值；采用线性激活函数作为输出层神经元；RBF is realized by inputting to output the mapping, ( ) uses the radial basis function as the activation function of any hidden node, and uses the Gauss function as the radial basis function; W is the output weight matrix, where ( ; ) is the hidden layer node to the output layer Synaptic weights between nodes; using a linear activation function as the output layer neurons;
根据正则化RBF网络原理，RBF的训练过程为：（1）确定RBF神经网络输入及输出变量，即以检测方向作为网络的输入节点，相应的预行程误差为网络的输出节点；（2）组成训练集对网络进行训练，即从检测到的预行程误差数据中随机选取若干组作为网络的教师数据；（3）输入预测样本，用训练好的网络预测任意检测方向的预行程误差，用剩下的测点数据作为预测样本。According to the principle of regularized RBF network, the training process of RBF is: (1) Determine the input and output variables of the RBF neural network, that is, the detection direction is used as the input node of the network, and the corresponding pre-travel error is the output node of the network; (2) Composition The training set trains the network, that is, randomly selects several groups from the detected pre-travel error data as the teacher data of the network; (3) input prediction samples, use the trained network to predict the pre-travel error of any detection direction, and use the remaining The measured point data below are used as forecast samples.
上述点到曲面的最小距离的计算方法求解的方法如下：将曲面分割为足够小的网格，计算测点(xm,ym,zm)到全部网格节点的距离，所有这些距离的最小值就是点到曲面的最小距离。The method of calculating the minimum distance from the above point to the surface is as follows: Divide the surface into small enough grids, calculate the distances from the measurement points (x m , y m , z m ) to all grid nodes, and all these distances The minimum value is the minimum distance from the point to the surface.
本发明由于采用在待检测的复杂曲面加工完成后，在数控铣床的工作台上直接进行检测的方法，对获取的数据进行误差补偿从而获得高精度的检测结果，对此检测结果进行分析进而获得被测曲面的加工精度。本发明的优点是：本发明的方法可以在数控铣床上直接对加工完成后的复杂曲面进行在线检测，通过对获取的数据进行测球半径补偿、预行程误差补偿和机床误差补偿，获得高精度的检测结果，进而得到被加工曲面的加工精度，克服了复杂曲面需要移动到CMM上才能获得高精度检测结果的不足之处，有效地提高了生产效率，本发明具有显著的经济效益、社会效益。本发明是一种设计巧妙，性能优良，方便实用的复杂零件的曲面加工精度的在线检测方法。The present invention adopts the method of directly detecting on the workbench of the CNC milling machine after the processing of the complex curved surface to be detected is completed, and performs error compensation on the acquired data to obtain a high-precision detection result, and then analyzes the detection result to obtain The machining accuracy of the measured surface. The advantage of the present invention is: the method of the present invention can directly detect the complex curved surface after processing on the CNC milling machine online, and obtain high precision by performing ball radius compensation, pre-travel error compensation and machine tool error compensation on the acquired data. The detection results can be obtained, and then the machining accuracy of the processed curved surface can be obtained, which overcomes the disadvantage that the complex curved surface needs to be moved to the CMM to obtain high-precision detection results, and effectively improves the production efficiency. The present invention has significant economic and social benefits . The invention is a convenient and practical on-line detection method for the processing precision of the curved surface of complex parts with ingenious design, excellent performance.
附图说明Description of drawings
图1为本发明测球半径补偿原理图；Fig. 1 is the schematic diagram of measuring ball radius compensation of the present invention;
图2为本发明预行程误差图；Fig. 2 is the pre-travel error figure of the present invention;
图3为本发明的方法流程图；Fig. 3 is a flow chart of the method of the present invention;
图4为正则化RBF网络的示意图；Figure 4 is a schematic diagram of a regularized RBF network;
图5为点到曲面的最小距离的示意图。Fig. 5 is a schematic diagram of the minimum distance from a point to a curved surface.
图中：1-法矢方向，2-被测曲面，3-测球中心，4-接触式触发测头，5-接触点，6-检测方向，7-预行程误差，8-测头高速定位移动方向，9-测头低速接近方向，10-测头的预接触距离，11-工件。In the figure: 1-normal vector direction, 2-surface to be tested, 3-ball center, 4-contact trigger probe, 5-contact point, 6-detection direction, 7-pre-travel error, 8-probe high speed Positioning movement direction, 9-probe low-speed approach direction, 10-pre-contact distance of the probe, 11-workpiece.
具体实施方式detailed description
实施例：Example:
本发明的复杂零件的曲面加工精度的在线检测方法，The online detection method of the curved surface machining accuracy of complex parts of the present invention,
本发明的复杂零件的曲面加工精度的在线检测方法，包括有如下步骤：The online detection method of the curved surface machining accuracy of complex parts of the present invention comprises the following steps:
1）将接触式触发测头4安装在数控铣床的主轴上；1) Install the touch trigger probe 4 on the spindle of the CNC milling machine;
2）对被测曲面2进行检测路径规划；2) Carry out detection path planning on the measured surface 2;
3）驱动数控铣床，逐一获取测头与待测曲面接触点处的测球中心3的坐标；测球是接触式触发测头的一个部件，其形状是一个具有高制造精度和高硬度的圆球，安装在测头主体上并与被测曲面直接接触；3) Drive the CNC milling machine to obtain the coordinates of the ball center 3 at the contact point between the probe and the surface to be measured one by one; the ball is a part of the contact trigger probe, and its shape is a circle with high manufacturing accuracy and high hardness Ball, mounted on the probe body and in direct contact with the measured surface;
4)对上述测球中心3的坐标进行测球半径补偿、测头预行程误差补偿和机床误差补偿，从而获得测点高精度的检测结果；其中机床误差需采用机床误差检测仪器得到；4) Perform ball radius compensation, probe pre-travel error compensation and machine tool error compensation on the coordinates of the above-mentioned measuring ball center 3, so as to obtain high-precision detection results of the measuring point; wherein the machine tool error needs to be obtained by using a machine tool error detection instrument;
5)将检测结果与零件的理想CAD模型进行对比分析，找到各个测点与CAD模型对应的偏差，从而获得待测曲面的加工精度。5) Compare and analyze the test results with the ideal CAD model of the part, and find the deviation between each measuring point and the CAD model, so as to obtain the machining accuracy of the surface to be tested.
上述测球半径补偿、测头预行程误差补偿需要计算出曲面测点的法矢方向，对于测头预行程误差补偿，还需要计算出测头在各个测点法矢方向的预行程误差，补偿方法是采用基于径向基函数（RBF）的神经网络进行预测。The above-mentioned probe ball radius compensation and probe pre-travel error compensation need to calculate the normal vector direction of the surface measuring point. For the probe pre-travel error compensation, it is also necessary to calculate the pre-travel error of the probe in the normal vector direction of each measuring point, and the compensation The method is to use a neural network based on radial basis function (RBF) for prediction.
上述机床误差补偿所用的机床误差检测仪器为激光干涉仪。The machine tool error detection instrument used for the above machine tool error compensation is a laser interferometer.
上述测点与CAD模型对应的偏差采用点到曲面的最小距离的计算方法求解。The deviation between the above measuring points and the CAD model is solved by calculating the minimum distance from the point to the surface.
上述基于径向基函数（RBF）的神经网络进行预测的方法如下：The prediction method of the above-mentioned neural network based on radial basis function (RBF) is as follows:
正则化RBF网络是一种具有单隐含层的三层前馈局部逼近网络，已经证明，它与BP网络都能以任意精度逼近任意连续函数；并且，相比BP网络，其训练时间更短,并且它同时满足对样本的逼近误差和逼近曲线的平滑性,在实践中，网络的监督训练可以看成是一种曲线拟合的过程,利用正则化RBF算法，通过对网络的训练，实现输入和输出空间之间的非线性映射；The regularized RBF network is a three-layer feed-forward local approximation network with a single hidden layer. It has been proved that it and the BP network can approximate any continuous function with arbitrary precision; and, compared with the BP network, its training time is shorter , and it satisfies the approximation error of the sample and the smoothness of the approximation curve at the same time. In practice, the supervised training of the network can be regarded as a curve fitting process, and the regularized RBF algorithm is used to train the network to achieve non-linear mapping between input and output spaces;
正则化RBF网络的拓扑结构由一个径向基神经元的隐层、一个线性神经元的输出层组成，网络的输入点数量为N，隐节点数量为P个，输出节点数量为l个；网络的隐节点数等于输入样本数，并将所有输入样本设为径向基函数的中心，各径向基函数取统一的扩展常数；The topology of the regularized RBF network consists of a hidden layer of radial basis neurons and an output layer of linear neurons. The number of input points of the network is N, the number of hidden nodes is P, and the number of output nodes is l; the network The number of hidden nodes is equal to the number of input samples, and all input samples are set as the center of the radial basis function, and each radial basis function takes a uniform expansion constant;
RBF实现由输入到输出的映射，()采用径向基函数为任一隐节点的激活函数，选用Gauss函数作为径向基函数；W为输出权矩阵，其中（；）为隐层第个节点到输出层第个节点间的突触权值；采用线性激活函数作为输出层神经元；RBF is realized by inputting to output the mapping, ( ) uses the radial basis function as the activation function of any hidden node, and uses the Gauss function as the radial basis function; W is the output weight matrix, where ( ; ) is the hidden layer node to the output layer Synaptic weights between nodes; using a linear activation function as the output layer neurons;
根据正则化RBF网络原理，RBF的训练过程为：（1）确定RBF神经网络输入及输出变量，即以检测方向作为网络的输入节点，相应的预行程误差为网络的输出节点；（2）组成训练集对网络进行训练，即从检测到的预行程误差数据中随机选取若干组作为网络的教师数据；（3）输入预测样本，用训练好的网络预测任意检测方向的预行程误差，用剩下的测点数据作为预测样本。According to the principle of regularized RBF network, the training process of RBF is: (1) Determine the input and output variables of the RBF neural network, that is, the detection direction is used as the input node of the network, and the corresponding pre-travel error is the output node of the network; (2) Composition The training set trains the network, that is, randomly selects several groups from the detected pre-travel error data as the teacher data of the network; (3) input prediction samples, use the trained network to predict the pre-travel error of any detection direction, and use the remaining The measured point data below are used as forecast samples.
上述点到曲面的最小距离的计算方法求解的方法如下：将曲面分割为足够小的网格，计算测点(xm,ym,zm)到全部网格节点的距离，所有这些距离的最小值就是点到曲面的最小距离。The method of calculating the minimum distance from the above point to the surface is as follows: Divide the surface into small enough grids, calculate the distances from the measurement points (x m , y m , z m ) to all grid nodes, and all these distances The minimum value is the minimum distance from the point to the surface.
本发明实施例是利用数控铣床对复杂曲面工件进行高精度的检测方法，该方法适应于当数控铣床对工件11进行加工，本发明实施例的工件11是复杂曲面零件，工件11完成一个加工工序后，直接在数控铣床的工作台上对工件11进行检测，实现加工和检测都在数控铣床上进行，可以避免将工件11移动到其他检测设备（如三坐标测量机）上检测带来的二次定位误差，也可避免对尺寸和重量大的工件11进行搬运所带来的不便。本实施例的方法中被检测的工件11在加工完成后，在数控铣床的工作台上直接检测，包括以下步骤：The embodiment of the present invention uses a numerical control milling machine to perform a high-precision detection method on a workpiece with a complex curved surface. This method is suitable for processing the workpiece 11 with a numerical control milling machine. The workpiece 11 in the embodiment of the present invention is a complex curved surface part, and the workpiece 11 completes a processing procedure. Finally, the workpiece 11 is detected directly on the workbench of the CNC milling machine, so that both processing and detection are carried out on the CNC milling machine, which can avoid secondary problems caused by moving the workpiece 11 to other detection equipment (such as a three-coordinate measuring machine) for detection. The secondary positioning error can also avoid the inconvenience caused by handling the workpiece 11 with large size and weight. The detected workpiece 11 in the method of the present embodiment is directly detected on the workbench of the CNC milling machine after the processing is completed, comprising the following steps:
步骤一：将接触式触发测头4安装在数控铣床的主轴上。主轴带动接触式触发测头4运动，接触式触发测头4实施对被测曲面2的坐标检测，检测结果记录在检测软件中。Step 1: Install the touch trigger probe 4 on the spindle of the CNC milling machine. The main shaft drives the touch-trigger probe 4 to move, and the touch-trigger probe 4 implements the coordinate detection of the measured curved surface 2, and the detection result is recorded in the detection software.
步骤二：对被测曲面2进行检测路径规划。利用检测软件对被测曲面进行测点规划和检测路径规划。Step 2: Carry out detection path planning on the measured surface 2 . Use the detection software to plan the measurement point and detection path of the measured surface.
步骤三：驱动数控铣床，逐一获取接触式触发测头4与被测曲面接触点5处的测球中心3的坐标；Step 3: Drive the CNC milling machine to obtain the coordinates of the measuring ball center 3 at the contact point 5 between the touch trigger probe 4 and the measured curved surface one by one;
步骤四：对上述的测球中心3坐标进行测球半径补偿、接触式触发测头4预行程误差补偿和机床误差补偿，从而获得测点高精度的检测结果；Step 4: Perform ball radius compensation, contact trigger probe 4 pre-travel error compensation and machine tool error compensation for the above-mentioned 3 coordinates of the measuring ball center, so as to obtain high-precision testing results of the measuring point;
如图1所示，测球半径补偿的关键是求出被测点的法矢方向1，然后利用公式进行测球半径补偿As shown in Figure 1, the key to compensation of the radius of the measuring ball is to find the normal vector direction 1 of the measured point, and then use the formula to compensate the radius of the measuring ball
(1) (1)
上式中，(X,Y,Z)是接触点A的坐标，(x,y,z)是测球中心B的坐标，r为测球半径，n为测点单位法矢量。In the above formula, (X, Y, Z) are the coordinates of the contact point A, (x, y, z) are the coordinates of the center B of the measuring ball, r is the radius of the measuring ball, and n is the unit normal vector of the measuring point.
测头预行程误差7的补偿方面，如图2所示，从接触式触发测头4接触工件表面到触发信号产生的这段时间内，接触式触发测头4额外运动微小距离，一般称由于这段微小距离所引起的误差值为测头预行程误差。需利用标准球检测出测头在各个方向上的预行程误差，并利用径向基函数（RBF）算法预测出任意法矢方向1上的预行程误差7。In terms of compensation for the probe pre-travel error 7, as shown in Figure 2, during the period from when the touch trigger probe 4 touches the workpiece surface to when the trigger signal is generated, the touch trigger probe 4 moves a small extra distance, which is generally called due to The error caused by this tiny distance is the probe pre-travel error. It is necessary to use the standard ball to detect the pre-travel error of the probe in all directions, and use the radial basis function (RBF) algorithm to predict the pre-travel error 7 in any normal vector direction 1.
在机床误差补偿方面，根据运动学原理，物体沿某一直线运动具有六个自由度，即三个平移自由度和三个回转自由度，也就是具有六个几何误差分量，即沿三个相互垂直方向的直线度误差和对三个相互垂直方向的转动误差。三轴数控铣床一般都具有（X，Y，Z）三个相互垂直的直线运动轴。因此对于三轴数控铣床来说，沿三个轴运动共存在18项误差分量，再加上三个轴之间还存在垂直度误差，总共有21项误差分量。本实施例利用激光干涉仪测出这21项误差分量。In terms of machine tool error compensation, according to the principle of kinematics, the movement of an object along a certain line has six degrees of freedom, that is, three translational degrees of freedom and three rotational degrees of freedom, that is, six geometric error components, that is, along three mutual The straightness error in the vertical direction and the rotation error for three mutually perpendicular directions. Three-axis CNC milling machines generally have (X, Y, Z) three mutually perpendicular linear motion axes. Therefore, for a three-axis CNC milling machine, there are 18 error components along the three axes, plus there are verticality errors between the three axes, and there are 21 error components in total. In this embodiment, the 21 error components are measured by a laser interferometer.
将上述获得的测点法矢、预行程误差和机床误差输入到检测软件系统中，并对数控系统反馈回的测球中心坐标进行补偿。The normal vector of the measuring point, pre-travel error and machine tool error obtained above are input into the detection software system, and the center coordinates of the measuring ball fed back by the numerical control system are compensated.
本实施例使用TP6L测头系统。This embodiment uses the TP6L probe system.
步骤五：将检测结果与零件的理想CAD模型进行对比分析，找到各个测点与CAD模型对应的偏差，从而获得待测曲面的加工精度。Step 5: Compare and analyze the test results with the ideal CAD model of the part, find the deviation between each measuring point and the CAD model, and obtain the machining accuracy of the surface to be tested.
上述步骤二中需进行测点规划（即测点在曲面上的排列方式），并计算出各个测点的法矢方向1，然后进行检测路径规划。所述法矢为测点所在曲面上并通过所述待测点的法矢。In the above step 2, it is necessary to plan the measuring points (that is, the arrangement of the measuring points on the surface), and calculate the normal vector direction 1 of each measuring point, and then plan the detection path. The normal vector is the normal vector on the curved surface where the measuring point is located and passes through the point to be measured.
上述步骤三中接触式触发测头4沿法矢方向1匀速接触被测曲面上的测点，在该过程中，接触式触发测头4沿检测方向6从测头高速定位移动方向8向测头低速接近方向9运动,图中测头的预接触距离10，在接触过程中利用软件系统接收接触式触发测头4的中心坐标信息。In the above step 3, the touch trigger probe 4 contacts the measuring point on the measured surface at a constant speed along the normal vector direction 1. The head moves in the approach direction 9 at a low speed. The pre-contact distance of the probe in the figure is 10. During the contact process, the software system is used to receive the center coordinate information of the touch trigger probe 4 .
上述步骤四中，需利用标准球检测出测头在各个方向上的预行程误差，并利用径向基函数（RBF）算法预测出任意法矢方向上的预行程误差。利用激光干涉仪测量得到机床的几何误差。In the above step four, it is necessary to use the standard sphere to detect the pre-travel error of the probe in all directions, and use the radial basis function (RBF) algorithm to predict the pre-travel error in any normal vector direction. The geometric error of the machine tool is measured by laser interferometer.
Claims (4)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210266355.6A CN102785129B (en) | 2012-07-30 | 2012-07-30 | The online test method of the surface machining accuracy of complex parts |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210266355.6A CN102785129B (en) | 2012-07-30 | 2012-07-30 | The online test method of the surface machining accuracy of complex parts |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102785129A CN102785129A (en) | 2012-11-21 |
| CN102785129B true CN102785129B (en) | 2016-08-03 |
Family
ID=47150819
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210266355.6A Active CN102785129B (en) | 2012-07-30 | 2012-07-30 | The online test method of the surface machining accuracy of complex parts |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102785129B (en) |
Families Citing this family (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103264318B (en) * | 2013-04-19 | 2015-11-18 | 湖北三江航天险峰电子信息有限公司 | A kind of online test method of three-dimensional profile |
| CN103777570B (en) * | 2014-01-07 | 2017-03-01 | 浙江大学 | Mismachining tolerance quick detection compensation method based on nurbs surface |
| CN103831669A (en) * | 2014-03-20 | 2014-06-04 | 蒋峰 | Circular degree error online measurement system and measurement method |
| CN104002174B (en) * | 2014-06-13 | 2016-06-01 | 沈阳飞机工业（集团）有限公司 | A kind of method simplifying use sphere point location |
| CN104002197A (en) * | 2014-06-13 | 2014-08-27 | 沈阳飞机工业（集团）有限公司 | Normal direction seeking device adopted in automatic hole manufacturing |
| CN104698964B (en) * | 2014-10-27 | 2017-05-03 | 大连理工大学 | Complex surface numerical control machining motion analyzing method based on mapping |
| CN104625876B (en) * | 2015-02-17 | 2018-02-09 | 中国船舶重工集团公司第七一一研究所 | Supercharger impeller blade machining process based on on-machine measurement |
| CN104750914A (en) * | 2015-03-06 | 2015-07-01 | 广西科技大学 | Unknown free-form curved surface modeling method |
| CN105127492B (en) * | 2015-09-07 | 2017-11-14 | 上海交通大学 | The method of straight engine the combustion chamber online compensation processing |
| CN107121113A (en) * | 2017-04-24 | 2017-09-01 | 上海现代先进超精密制造中心有限公司 | The detection method of heavy caliber based on three coordinates, complex free curved surface element |
| CN107480377B (en) * | 2017-05-16 | 2018-12-14 | 安徽工业大学 | Three coordinate measuring machine gauge head pretravel error prediction method based on hybrid modeling |
| CN107238364B (en) * | 2017-06-30 | 2019-07-12 | 四川大学 | Precision Compensation Method for Ball Nose Radius of Contact Measuring Stylus |
| CN108106522A (en) * | 2017-11-29 | 2018-06-01 | 中国航发沈阳黎明航空发动机有限责任公司 | A kind of method for three-dimensional measurement of irregular surface |
| CN108050981B (en) * | 2017-12-28 | 2019-11-26 | 上海交通大学 | A kind of three coordinate measuring engine measurement method of complex part surface planarity measurement |
| CN109202539B (en) * | 2018-08-23 | 2020-10-30 | 北京动力机械研究所 | Online detection method for composite material weak-rigidity special-shaped structure |
| CN109407616B (en) * | 2018-09-29 | 2020-11-27 | 广东科杰机械自动化有限公司 | Method for realizing real-time track compensation based on measured data |
| CN109341634B (en) * | 2018-11-29 | 2021-01-01 | 中国航发南方工业有限公司 | Method for measuring profile size of precision casting turbine blade |
| CN111950189B (en) * | 2019-05-14 | 2024-09-27 | 华中科技大学 | A tool machining path planning method suitable for curved surfaces |
| CN110186405B (en) * | 2019-05-30 | 2021-02-02 | 华中科技大学无锡研究院 | 3D Radius Compensation and Cross Compensation Point Correction Method for Probe Ball in Contact Scanning Probe of Blade Section |
| CN111336962B (en) * | 2020-02-25 | 2021-11-12 | 深圳星友方科技有限公司 | Method and system for online measuring workpiece by spark machine |
| CN111504227B (en) * | 2020-06-17 | 2021-06-01 | 北京理工大学 | A confocal axial monitoring method of femtosecond laser processing parameters based on deep learning |
| CN111857069A (en) * | 2020-07-09 | 2020-10-30 | 深圳先进技术研究院 | Control system and method for CNC machining and inspection, and CNC machining and inspection system |
| CN112461175A (en) * | 2020-10-15 | 2021-03-09 | 中国航发沈阳黎明航空发动机有限责任公司 | Method for measuring wide chord and large torsion angle blade profile of fan blisk |
| CN112917241B (en) * | 2021-03-02 | 2022-02-11 | 清华大学深圳国际研究生院 | Hole series form and position error correction method |
| US12135544B2 (en) | 2021-10-14 | 2024-11-05 | Worldwide Superabrasives, LLC | CNC add-on sensor system and method for real-time detection of tool anomalies |
| CN114777670A (en) * | 2022-04-21 | 2022-07-22 | 西安交通大学 | Curved surface on-machine measuring method based on contact type measuring head |
| CN117372554B (en) * | 2023-09-14 | 2024-06-04 | 华中科技大学 | Three-coordinate blade section reconstruction method based on radial basis function |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101745845A (en) * | 2009-12-07 | 2010-06-23 | 哈尔滨工业大学 | Measuring method of outer contour shape of metal part and detecting method of processing precision |
| CN103142664A (en) * | 2013-03-29 | 2013-06-12 | 厦门大学 | Method for extracting saikosaponin from bupleurum chinense by using two-level foam separation process |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06170700A (en) * | 1992-12-03 | 1994-06-21 | Yaskawa Electric Corp | Milling tool having odd number of cutting edges and machining error measuring method/device for machining using this tool |
-
2012
- 2012-07-30 CN CN201210266355.6A patent/CN102785129B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101745845A (en) * | 2009-12-07 | 2010-06-23 | 哈尔滨工业大学 | Measuring method of outer contour shape of metal part and detecting method of processing precision |
| CN103142664A (en) * | 2013-03-29 | 2013-06-12 | 厦门大学 | Method for extracting saikosaponin from bupleurum chinense by using two-level foam separation process |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102785129A (en) | 2012-11-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102785129B (en) | The online test method of the surface machining accuracy of complex parts | |
| CN102785128B (en) | The part processing precision on-line detecting system of NC Machine lathe and detection method | |
| CN102699761B (en) | Error identification method of five-axis numerically controlled machine tool based on S-shaped test specimen | |
| CN102501136B (en) | On-machine detection measuring head and detection system for numerical control machine tool | |
| CN107315391B (en) | A pre-travel error compensation method for online detection of CNC machine tools | |
| CN109648399B (en) | Five-axis linkage machine tools dynamic error and quiescent error method for comprehensive detection | |
| Li et al. | Enhancement and evaluation in path accuracy of industrial robot for complex surface grinding | |
| CN103034166A (en) | Recognition method of critical geometrical error source of machine tool | |
| CN102001024A (en) | Measuring method for in-site measurement of free-form curved surface based on machining machine tool | |
| CN110181334B (en) | On-machine detection device and detection method for surface error of free-form surface optical element based on white light confocal principle | |
| CN102866672A (en) | Online detecting method of numerical control machining middle state of plane structural member | |
| CN112008492A (en) | Method for identifying perpendicularity error of translational shaft of gantry numerical control machine tool | |
| CN108803487A (en) | A kind of point profile errors prediction technique on part side milling surface | |
| Aliakbari et al. | An adaptive computer-aided path planning to eliminate errors of contact probes on free-form surfaces using a 4-DOF parallel robot CMM and a turn-table | |
| Slamani et al. | Dynamic and geometric error assessment of an XYC axis subset on five-axis high-speed machine tools using programmed end point constraint measurements | |
| CN112069612A (en) | A Method for Evaluating Measurement Uncertainty of Gear Measuring Center | |
| CN116909209A (en) | A CNC machine tool error modeling and prediction method considering dynamic thermal errors | |
| Han et al. | A review of geometric error modeling and error detection for CNC machine tool | |
| Cappetti et al. | Fuzzy approach to measures correction on Coordinate Measuring Machines: The case of hole-diameter verification | |
| Xing et al. | Comparison of direct and indirect methods for five-axis machine tools geometric error measurement | |
| Lin et al. | Probe radius compensated by the multi-cross product method in freeform surface measurement with touch trigger probe CMM | |
| CN109202539B (en) | Online detection method for composite material weak-rigidity special-shaped structure | |
| CN110531709A (en) | A kind of method of analytic surface part's machining errors and feed rate relationship | |
| CN113868890A (en) | An automatic three-coordinate measurement simulation system suitable for thin plate parts | |
| CN202656009U (en) | Part processing accuracy online detection system facing numerically controlled lathe |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240425 Address after: 230000 building 6, alumni Innovation Park, China University of science and technology, Tianshui Road, Luyang District, Hefei City, Anhui Province Patentee after: Anhui Huachuang Hongdu Photoelectric Technology Co.,Ltd. Country or region after: China Address before: 510006 No. 100 West Ring Road, Guangzhou University, Guangzhou, Guangdong, Panyu District Patentee before: GUANGDONG University OF TECHNOLOGY Country or region before: China |