Blisk machining quality analysis method based on multi-source three-dimensional measurement information fusion

Abstract

This invention discloses a method for analyzing the processing quality of bladed disks based on the fusion of multi-source three-dimensional measurement information, belonging to the field of multi-source three-dimensional measurement technology. It obtains information about the bladed disks to be processed from a database and constructs a set of blades, collecting initial data of the blades to form an initial processing state data set. The processing process is divided into several processing nodes, and the processing state data of each node is collected in real time to form a processing node data set. The geometric deviation characterization quantity and the change in geometric deviation of the blades are calculated. The processing anomaly response quantity and its change are calculated. The geometric deviation change quantity and the processing anomaly change quantity are weighted to generate a blade processing quality state quantity, which is then summarized into a comprehensive processing state quantity. Combined with a preset threshold, processing quality is judged and anomaly warnings are issued. This method achieves precise monitoring and quantitative analysis of the entire bladed disk processing process, enabling timely detection of processing anomalies and issuing warnings, improving the processing accuracy and consistency of bladed disks, reducing the defect rate, and increasing production efficiency.

Description

Technical Field

[0001] This invention relates to the field of multi-source three-dimensional measurement technology, specifically to a method for analyzing the manufacturing quality of bladed disks based on the fusion of multi-source three-dimensional measurement information. Background Technology

[0002] As aero-engines evolve towards higher thrust-to-weight ratios, higher efficiency, and longer lifespans, bladed disk structures are showing a trend towards integration, thinner walls, and more complex curved surfaces, placing higher demands on manufacturing quality control. Current bladed disk manufacturing processes typically combine CNC machining equipment with various inspection methods for quality control research. Three-dimensional laser scanning technology is used to acquire overall blade profile information, coordinate measuring machine (CMM) technology is used for critical dimension and tolerance detection, visual inspection technology is used to identify surface defects, and the machining equipment itself can collect real-time operating condition information such as speed, load, and temperature. Building on this foundation, academia and engineering have gradually moved from offline quality assessment using single measurement methods to manufacturing process monitoring and quality analysis methods assisted by multi-source measurement data. These methods attempt to achieve quantitative description and anomaly warning of bladed disk manufacturing status by introducing multi-node detection, process data correlation analysis, and quality threshold determination.

[0003] Existing technical solutions still have several shortcomings in the analysis of bladed disk machining quality: Firstly, most methods focus on final inspection or periodic sampling after machining, making it difficult to reflect the dynamic changes in geometric morphology and machining anomalies during machining. This often leads to quality problems being discovered only after they have formed, lacking process foresight. Secondly, the use of multi-source measurement data in existing technologies is mostly limited to parallel use or simple comparison, failing to integrate heterogeneous information such as three-dimensional surfaces, key dimensions, surface images, and machining conditions under a unified analysis model. This makes it difficult to reveal the intrinsic relationship between geometric deviations and machining anomalies. Furthermore, regarding the initial differences between different blades in the bladed disk, most solutions lack a change analysis mechanism based on the initial machining state, easily mixing inherent manufacturing deviations with machining-induced deviations in evaluation, affecting the accuracy of quality judgment. Especially when load fluctuations, abnormal temperature rises, or local surface defects occur during machining, existing methods often cannot simultaneously quantify their impact on geometric evolution, and it is also difficult to form a unified machining quality status quantity for overall assessment and early warning. Summary of the Invention

[0004] The purpose of this invention is to provide a method for analyzing the manufacturing quality of bladed disks based on the fusion of multi-source three-dimensional measurement information, so as to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: A method for analyzing the processing quality of bladed disks based on multi-source 3D measurement information fusion includes the following steps: Step S1: Obtain all information on the bladed disks to be processed from the database and construct a set of blades for the bladed disks to be processed; before processing begins, obtain the initial data of the blades in the bladed disks to be processed; Step S2: Construct an initial processing state data set for the blades; divide the processing process into several processing nodes and construct a processing state data set for the blades under each processing node; Step S3: Based on the initial processing state data set, calculate the initial geometric deviation characterization of the blades before processing; based on the processing state data set, calculate the geometric deviation characterization of the blades under each processing node; Step S4: Based on the initial geometric deviation characterization quantity and the geometric deviation characterization quantity, calculate the geometric deviation change of the blade at the processing node; Step S5: Based on the initial processing state data set of the blade, calculate the initial processing anomaly response quantity of the blade before processing; based on the processing state data set of the blade at the processing node, calculate the processing anomaly response quantity of the blade at the processing node; based on the initial processing anomaly response quantity and the processing anomaly response quantity, calculate the processing anomaly change of the blade at the processing node; Step S6: Based on the geometric deviation change quantity and the processing anomaly change quantity, calculate the processing quality state quantity and the comprehensive processing state quantity of the blade at the processing node, preset thresholds, analyze the processing quality and send an early warning.

[0006] As a preferred embodiment of the bladed disk processing quality analysis method based on multi-source three-dimensional measurement information fusion described in this invention, all bladed disk information to be processed is obtained from the database, and a set of blades for the bladed disk to be processed is constructed, denoted as... ,in, Let J represent the j-th blade in the i-th bladed disk to be processed, and J represent the total number of blades in the i-th bladed disk. A multi-source measurement module is constructed, which includes a laser scanning sensor, a coordinate measuring machine, a vision camera, and a processing condition acquisition device. The laser scanning sensor is used to acquire three-dimensional surface measurement data of the blades in the impeller disk. The coordinate measuring machine is used to acquire key dimension measurement data of the blades in the impeller disk. The vision camera is used to acquire surface image information data of the blades in the impeller disk. The processing condition acquisition device is used to acquire processing condition data of the blades in the impeller disk.

[0007] As a preferred embodiment of the bladed disk processing quality analysis method based on multi-source three-dimensional measurement information fusion described in this invention, the processing condition acquisition device includes a speed acquisition sensor, a load acquisition sensor, and a temperature acquisition sensor. The processing condition data includes speed data, load data, and temperature data. The key dimension measurement data includes blade root diameter data, blade tip diameter data, and blade thickness data. Before processing begins, obtain the j-th blade from the i-th bladed disk to be processed. The initial three-dimensional profile measurement data, initial data of blade root diameter, initial data of blade tip diameter, initial data of blade thickness, initial data of surface image information, initial rotational speed data, initial load data, and initial temperature data.

[0008] As a preferred embodiment of the bladed disk processing quality analysis method based on multi-source three-dimensional measurement information fusion described in this invention, the blade before processing begins... The initial three-dimensional profile measurement data, initial blade root diameter data, initial blade tip diameter data, initial blade thickness data, initial surface image information data, initial rotational speed data, initial load data, and initial temperature data are respectively denoted as: and and construct blades The initial processing state data set, denoted as ; The processing is divided into several processing nodes, and the blade data acquired in real time at the a-th processing node is... The three-dimensional profile measurement data, blade root diameter data, blade tip diameter data, blade thickness data, surface image information data, rotational speed data, load data, and temperature data are respectively denoted as: and And construct the blade at the a-th processing node. The processing status data set, denoted as .

[0009] As a preferred embodiment of the bladed disk processing quality analysis method based on multi-source three-dimensional measurement information fusion described in this invention, based on the blade... Initial processing state data set Calculate the blade The initial geometric deviation characterization value before processing is calculated using the following formula: ; in, Indicates blade Initial geometric deviation characterization quantity before processing and This represents the preset contribution coefficient of geometric deviation before processing. This indicates the number of measurement points in the initial three-dimensional surface measurement data. This represents the p-th measurement point in the initial three-dimensional surface measurement data. This represents the spatial coordinate data of the p-th measurement point. This represents the spatial center coordinates of all measured points. This represents the initial average thickness of all blades in the i-th bladed disk; Based on the blade at the a-th processing node Processing status data set Calculate the blade at the a-th processing node. The geometric deviation characteristic is calculated using the following formula: ; in, This indicates the blade at the a-th processing node. Geometric deviation characterization quantity, and This represents the preset contribution coefficient for geometric deviation after processing. This represents the number of measurement points in the 3D surface measurement data at the a-th processing node. This represents the q-th measurement point in the 3D surface measurement data at the a-th processing node. This represents the spatial coordinate data of the q-th measurement point under the a-th processing node. This represents the spatial center coordinates of all measurement points under the a-th processing node. This represents the initial average thickness of all blades in the i-th bladed disk at the a-th processing node; Based on blades Initial geometric deviation characterization before processing and the blade at the a-th processing node Geometric deviation characterization quantity, calculate the blade The formula for calculating the geometric deviation change at the a-th processing node is: ,in, Indicates blade The change in geometric deviation at the a-th processing node.

[0010] As a preferred embodiment of the bladed disk processing quality analysis method based on multi-source three-dimensional measurement information fusion described in this invention, based on the blade... Initial processing state data set Calculate the blade The initial processing anomaly response quantity before processing is calculated using the following formula: ; in, Indicates blade Initial processing anomaly response quantity before processing and This represents the preset abnormal sensitivity weighting coefficient before processing. This represents the average of the initial data for the root diameter of all blades in the i-th bladed disk. This represents the average of the initial data for all blade tip diameters in the i-th bladed disk. This indicates the number of local image regions in the initial data of the surface image information. The local image region number represents the initial data of the surface image information. This represents the nth local image region in the initial data of the surface image information. Indicates the significance of image anomalies; Based on the blade at the a-th processing node Processing status data set Calculate the blade at the a-th processing node. The processing anomaly response quantity is calculated using the following formula: ,in, This indicates the blade at the a-th processing node. The amount of abnormal processing response, and This represents the preset abnormal sensitivity weighting coefficient after processing. This represents the average value of the blade root diameter data in the i-th bladed disk at the a-th processing node. This represents the average value of the blade tip diameter data in the i-th bladed disk at the a-th processing node. This represents the number of local image regions in the lower surface image information data of the a-th processing node. This represents the local image region index of the lower surface image information data of the a-th processing node. This represents the z-th local image region in the image information data of the lower surface of the a-th processing node. This represents the metric for image anomaly significance at the a-th processing node; Based on blades Initial processing anomaly response quantity before processing and the blade at the a-th processing node Processing abnormal response quantity Calculate the blade The formula for calculating the abnormal change in processing at the a-th processing node is: ,in, Indicates blade The amount of abnormal changes in processing at the a-th processing node.

[0011] As a preferred embodiment of the bladed disk processing quality analysis method based on multi-source three-dimensional measurement information fusion described in this invention, based on the blade... Geometric deviation change at the a-th processing node and leaves The amount of abnormal changes in processing at the a-th processing node Calculate the blade The processing quality status quantity at the a-th processing node is calculated using the following formula: ,in, Indicates blade The processing quality status quantity at the a-th processing node, This represents the influence coefficient of the preset geometric deviation change. This represents the preset influence coefficient of abnormal processing variation. Based on blades Machining quality status quantity at the a-th machining node Calculate the comprehensive machining state quantity of all blades at the a-th machining node. The calculation formula is as follows: ,in, This represents the overall processing status quantity of all blades at the a-th processing node; The preset threshold for the overall processing state quantity is: if the overall processing state quantity of all blades at the a-th processing node is... If the value is greater than or equal to the threshold value of the comprehensive processing state quantity, it is determined that the processing quality of the impeller is poor at the a-th processing node, and an early warning is issued to the relevant personnel.

[0012] The bladed disk processing quality analysis system based on multi-source three-dimensional measurement information fusion includes: a set construction and initial data acquisition module, a processing state set construction module, a characterization quantity calculation and change quantity calculation module, a response quantity calculation and change quantity calculation module, and a state quantity calculation and analysis module. The set construction and initial data acquisition module: retrieves all information on the bladed disks to be processed from the database and constructs a set of blades for the bladed disks to be processed; before processing begins, it acquires the initial data of the blades in the bladed disks to be processed. The processing state set construction module: constructs the initial processing state data set of the blade; divides the processing process into several processing nodes, and constructs the processing state data set of the blade under the processing node; The module for calculating the characteristic quantity and the change quantity: calculates the initial geometric deviation characteristic quantity of the blade before processing based on the initial processing state data set; calculates the geometric deviation characteristic quantity of the blade at the processing node based on the processing state data set; and calculates the change quantity of geometric deviation of the blade at the processing node based on the initial geometric deviation characteristic quantity and the geometric deviation characteristic quantity. The response and change calculation module calculates the initial processing anomaly response of the blade before processing based on the initial processing state data set of the blade; it calculates the processing anomaly response of the blade at the processing node based on the processing state data set of the blade at the processing node; and it calculates the processing anomaly change of the blade at the processing node based on the initial processing anomaly response and the processing anomaly response. The state quantity calculation and analysis module calculates the processing quality state quantity and comprehensive processing state quantity of the blade at the processing node based on the geometric deviation change and the processing anomaly change, presets a threshold, analyzes the processing quality, and sends an early warning.

[0013] Furthermore, the characterization quantity calculation and change quantity calculation module includes a characterization quantity calculation unit and a change quantity calculation unit; The characterization calculation unit calculates the initial geometric deviation characterization of the blade before processing based on the initial processing state data set of the blade, and calculates the geometric deviation characterization of the blade at the a-th processing node based on the processing state data set of the blade at the a-th processing node. The change calculation unit calculates the change in geometric deviation of the blade at the a-th processing node based on the initial geometric deviation characterization of the blade before processing and the geometric deviation characterization of the blade at the a-th processing node.

[0014] Furthermore, the response quantity calculation and change quantity calculation module includes a response quantity calculation unit and a change quantity calculation unit; The response quantity calculation unit: calculates the initial processing abnormal response quantity of the blade before processing based on the initial processing state data set of the blade, and calculates the processing abnormal response quantity of the blade at the a-th processing node based on the processing state data set of the blade at the a-th processing node. The change calculation unit calculates the change in processing anomaly of the blade at the a-th processing node based on the initial processing anomaly response of the blade before processing and the processing anomaly response of the blade at the a-th processing node.

[0015] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: The bladed disk processing quality analysis method based on multi-source three-dimensional measurement information fusion provided by this invention obtains all information of the bladed disks to be processed from the database and constructs a set of blades. It collects the initial three-dimensional profile, key dimensions, surface image information, and processing condition data of the blades, realizing comprehensive digital modeling of the processed object and providing reliable benchmark data for subsequent processing status analysis. The processing process is divided into several nodes, and an initial processing state and node processing state data set are constructed, enabling continuous tracking and refined management of the processing process, providing data support for phased quality assessment. Based on the initial processing state and processing node states, the geometric deviation characterization quantity and its change quantity of the blades are calculated, realizing the blade... Quantitative assessment of blade shape and size variations helps to precisely control machining accuracy. Calculating machining anomaly response quantities and their changes based on initial and node machining states enables early identification and quantitative analysis of machining anomalies such as surface defects and operating condition fluctuations, thereby reducing machining risks. Furthermore, weighting geometric deviation changes and machining anomaly changes to calculate blade machining quality status quantities, and summarizing them into a comprehensive machining status quantity for threshold analysis, enables quantitative judgment and anomaly warning of the overall machining quality of the bladed disk. This effectively improves the machining accuracy and consistency of the bladed disk, reduces the defect rate, and increases production efficiency, forming a complete closed loop from initial modeling and status monitoring to quality judgment and early warning. This provides a scientific, systematic, and quantifiable technical guarantee for the reliable manufacturing of high-performance bladed disks. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0017] Figure 1 This is a schematic diagram of the steps of the bladed disk processing quality analysis method based on multi-source three-dimensional measurement information fusion of the present invention; Figure 2 This is a schematic diagram of the structure of the bladed disk processing quality analysis system based on the fusion of multi-source three-dimensional measurement information of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 In this first embodiment: a method for analyzing the manufacturing quality of bladed disks based on the fusion of multi-source three-dimensional measurement information is provided. This method includes the following steps: Step S1: Obtain all information on the bladed disks to be processed from the database and construct a set of blades for the bladed disks to be processed; before processing begins, obtain the initial data of the blades in the bladed disks to be processed.

[0020] Specifically, all information on the impellers to be processed is retrieved from the database, and a set of blades for the impellers to be processed is constructed, denoted as . ,in, Let J represent the j-th blade in the i-th bladed disk to be processed, and J represent the total number of blades in the i-th bladed disk. A multi-source measurement module is constructed, which includes a laser scanning sensor, a coordinate measuring machine, a vision camera, and a processing condition acquisition device. The laser scanning sensor is used to acquire three-dimensional surface measurement data of the blades in the impeller disk. The coordinate measuring machine is used to acquire key dimension measurement data of the blades in the impeller disk. The vision camera is used to acquire surface image information data of the blades in the impeller disk. The processing condition acquisition device is used to acquire processing condition data of the blades in the impeller disk.

[0021] Furthermore, the processing condition acquisition device includes a speed acquisition sensor, a load acquisition sensor, and a temperature acquisition sensor. The processing condition data includes speed data, load data, and temperature data. The key dimension measurement data includes blade root diameter data, blade tip diameter data, and blade thickness data. Before processing begins, obtain the j-th blade from the i-th bladed disk to be processed. The initial three-dimensional profile measurement data, initial data of blade root diameter, initial data of blade tip diameter, initial data of blade thickness, initial data of surface image information, initial rotational speed data, initial load data, and initial temperature data.

[0022] In this invention, by retrieving all information about the bladed disk to be processed from a database and constructing a blade set, the initial three-dimensional profile, key dimensions, surface image information, and processing condition data of each blade are obtained, achieving comprehensive digital modeling of the processing object. This process ensures that the initial state of each blade is known before processing begins, providing accurate benchmark data for subsequent processing status monitoring and deviation analysis. This effectively avoids deviations in processing quality judgment caused by missing or incorrect initial data, ultimately achieving the beneficial effect of providing reliable basic data for processing quality analysis.

[0023] It should be noted that the bladed disk is a core component of aero-engines and turbines, and its geometric accuracy directly determines aerodynamic efficiency, rotational balance, and structural strength. The geometric parameters in this invention include the three-dimensional profile, root diameter, tip diameter, and blade thickness. Specifically, the three-dimensional profile affects airflow efficiency; deviations in the profile can lead to decreased thrust and increased energy consumption. The root diameter is a critical part connecting the blade to the main shaft; diameter deviations affect connection strength and stress uniformity. The tip diameter determines the clearance between the blade and the casing; excessive deviations can lead to airflow leakage, while insufficient deviations can cause friction. Blade thickness affects structural stiffness and aerodynamic load distribution; uneven thickness can cause rotational vibration. Existing technologies often confuse initial geometric deviations with processing-induced deviations. The patent calculates these parameters by adding initial values, node values, and changes, accurately separating inherent deviations (such as uneven raw material thickness) from processing deviations (such as surface deformation caused by tool wear), making quality judgment more objective.

[0024] This invention also includes abnormal parameters, namely surface image, rotational speed, load, and temperature. Among them, the surface image directly identifies defects such as scratches and dents, which cannot be reflected by geometric dimensions but will seriously affect the fatigue life of the blades. Rotational speed and load are related to the cutting state of the tool (e.g., sudden drop in rotational speed → insufficient cutting force → residual material on the surface; sudden change in load → tool wear → dimensional deviation). Temperature is related to material deformation (e.g., excessively high temperature → thermal expansion and contraction of material → geometric deviation after machining). Bladed disk machining is a thin-walled, complex curved surface, and high-precision machining process, which is extremely sensitive to fluctuations in operating conditions (e.g., thin-walled parts are prone to deformation due to uneven temperature). Adding rotational speed, load, and temperature parameters can accurately monitor the stability of the machining process and avoid batch defects caused by uncontrolled operating conditions.

[0025] Step S2: Construct the initial processing state data set of the blade; divide the processing process into several processing nodes and construct the processing state data set of the blade under the processing node.

[0026] Specifically, the blades will be processed before the process begins. The initial three-dimensional profile measurement data, initial blade root diameter data, initial blade tip diameter data, initial blade thickness data, initial surface image information data, initial rotational speed data, initial load data, and initial temperature data are respectively denoted as: and and construct blades The initial processing state data set, denoted as ; The processing is divided into several processing nodes, and the blade data acquired in real time at the a-th processing node is... The three-dimensional profile measurement data, blade root diameter data, blade tip diameter data, blade thickness data, surface image information data, rotational speed data, load data, and temperature data are respectively denoted as: and And construct the blade at the a-th processing node. The processing status data set, denoted as .

[0027] In this invention, an initial processing state data set is formed by combining the initial three-dimensional profile, key dimensions, surface image information, and processing condition data of the blade. The processing process is then divided into several nodes, and data from these nodes is collected in real time to form a processing state data set. This achieves continuous state tracking and refined management of the processing process. This step systematically correlates the static initial state with the dynamic processing state, providing a clear data structure for subsequent calculations of geometric deviations and abnormal response quantities. It also facilitates phased quality assessments, thereby achieving the beneficial effects of real-time monitoring and traceable management of the processing process.

[0028] Step S3: Based on the initial processing state data set, calculate the initial geometric deviation characterization of the blade before processing; based on the processing state data set, calculate the geometric deviation characterization of the blade at the processing node; based on the initial geometric deviation characterization and the geometric deviation characterization, calculate the geometric deviation change of the blade at the processing node.

[0029] Specifically, based on the blade Initial processing state data set Calculate the blade The initial geometric deviation characterization value before processing is calculated using the following formula: ; in, Indicates blade Initial geometric deviation characterization quantity before processing and This represents the preset contribution coefficient of geometric deviation before processing. This indicates the number of measurement points in the initial three-dimensional surface measurement data. This represents the p-th measurement point in the initial three-dimensional surface measurement data. This represents the spatial coordinate data of the p-th measurement point. This represents the spatial center coordinates of all measured points. This represents the initial average thickness of all blades in the i-th bladed disk; It should be noted that the core of the formula is multi-dimensional geometric bias weighted fusion, which aligns with the core idea of ​​multi-source data fusion. It is the three-dimensional surface measurement data of the blade before processing (acquired by laser scanning sensor). This is the total number of measurement points on the profile. These are the spatial coordinates of the p-th measurement point. This refers to the spatial center coordinates of all measurement points (i.e., the center of the ideal profile). This item quantifies the degree of deviation of the overall blade profile (e.g., whether the blade is warped or deformed) by calculating the average distance from all profile points to the ideal center. Quantifying the diameter difference between the blade root and the blade tip—in actual processing, if this difference is too large, the blade will be unbalanced when rotating, causing engine vibration. Therefore, it is a core geometric indicator. The formula quantifies the consistency of the thickness of a single blade with that of the entire bladed disk (inconsistent blade thickness will affect aerodynamic efficiency). This formula establishes a geometric benchmark before processing, and all deviations in subsequent processing are referenced to this benchmark. This avoids misjudging inherent manufacturing deviations of the blades (such as slight deformation of raw materials) as processing defects. It quantifies the initial geometric state of each blade and provides a basis for subsequent personalized quality assessment (different blades have different initial deviations and cannot be judged by a uniform standard).

[0030] Based on the blade at the a-th processing node Processing status data set Calculate the blade at the a-th processing node. The geometric deviation characteristic is calculated using the following formula: ; in, This indicates the blade at the a-th processing node. Geometric deviation characterization quantity, and This represents the preset contribution coefficient for geometric deviation after processing. This represents the number of measurement points in the 3D surface measurement data at the a-th processing node. This represents the q-th measurement point in the 3D surface measurement data at the a-th processing node. This represents the spatial coordinate data of the q-th measurement point under the a-th processing node. This represents the spatial center coordinates of all measurement points under the a-th processing node. This represents the initial average thickness of all blades in the i-th bladed disk at the a-th processing node; Based on blades Initial geometric deviation characterization before processing and the blade at the a-th processing node Geometric deviation characterization quantity, calculate the blade The formula for calculating the geometric deviation change at the a-th processing node is: ,in, Indicates blade The change in geometric deviation at the a-th processing node.

[0031] In this invention, the initial geometric deviation characterization and nodal geometric deviation characterization of the blade are calculated based on the initial processing state and the processing node state, and the geometric deviation change is further calculated, thereby realizing a quantitative assessment of the geometric shape change of the blade during processing. This step accurately reflects the dimensional changes and shape shifts of the blade during processing, helps to promptly detect abnormal processing trends, and provides a quantitative basis for geometric error control and accuracy optimization. This achieves the beneficial effect of scientifically monitoring processing accuracy and improving the consistency of finished products.

[0032] Step S4: Based on the initial processing state data set of the blade, calculate the initial processing anomaly response amount of the blade before processing; based on the processing state data set of the blade under the processing node, calculate the processing anomaly response amount of the blade under the processing node; based on the initial processing anomaly response amount and the processing anomaly response amount, calculate the processing anomaly change amount of the blade at the processing node.

[0033] Specifically, based on the blade Initial processing state data set Calculate the blade The initial processing anomaly response quantity before processing is calculated using the following formula: ; in, Indicates blade Initial processing anomaly response quantity before processing and This represents the preset abnormal sensitivity weighting coefficient before processing. This represents the average of the initial data for the root diameter of all blades in the i-th bladed disk. This represents the average of the initial data for all blade tip diameters in the i-th bladed disk. This indicates the number of local image regions in the initial data of the surface image information. The local image region number represents the initial data of the surface image information. This represents the nth local image region in the initial data of the surface image information. Indicates the significance of image anomalies; It should be noted that the core of the formula is the quantification of anomaly sensitivity. It is the average diameter at the base of all blades on the same leaf disc. This is the average leaf tip diameter. Unlike the absolute deviation in the geometric deviation formula, this term uses relative deviation (deviation / mean) to quantify dimensional consistency. In practice, the absolute deviation of large-sized leaves is more permissible, and the relative deviation is more reflective of anomalies (e.g., for a root diameter of 100mm, a deviation of 1mm is 1%; for a diameter of 10mm, a deviation of 1mm is 10%, the latter being a more serious anomaly). It is image data of the blade surface before processing (acquired by a vision camera). It is the number of local image regions on the surface. This is a saliency measure of image anomalies in the nth region (e.g., the area / depth quantification value of scratches or dents; a larger value indicates a more severe defect). This term quantifies the initial surface quality of the blade. This formula establishes an anomaly baseline before processing, distinguishing between initial defects and processing-induced defects—for example, a blade may have slight scratches on its surface before processing (…). If the scratches do not expand after processing (including the defect), then there are no new abnormalities in the processing.

[0034] Furthermore, the image anomaly saliency metric is a core indicator for converting qualitative defects on the leaf surface (such as scratches, dents, and burrs) into quantitative data. This is achieved by capturing high-resolution grayscale or color images of the leaf surface using a vision camera (industrial CCD / CMOS camera) (grayscale images are preferred to reduce computational load). During shooting, fixed lighting conditions are necessary (avoiding glare and shadow interference), and the lens should be aimed at key areas of the leaf (such as the leaf tip, root, pressure surface / suction surface). The entire leaf surface image is divided into several non-overlapping local image regions (e.g., (This indicates that the area is divided into 20 regions). The principle of division is that small regions cover areas prone to critical defects, and large regions cover flat areas. For example, the leaf tip is prone to scratches and can be divided into 10 small regions; the middle of the leaf is flat and can be divided into 5 large regions. Image preprocessing is performed. Based on the preprocessed image, an algorithm identifies abnormal regions (parts that are inconsistent with the normal surface). The core method aligns with the needs of patented processing anomaly identification and can be referenced below: Set a grayscale threshold (e.g., the grayscale value of a normal leaf surface is concentrated between 150-200), and areas below / above this threshold are judged as abnormal (e.g., the grayscale value of a scratched area is <120, and the grayscale value of a dented area is <100). Perform basic quantization dimensions: calculate the proportion of pixels in the abnormal area to the total number of pixels in the local image area, calculate the average gray level difference between the abnormal area and the normal area, and directly measure the actual depth of the depression and burrs (unit μm), such as a depression depth of 15μm. Map the basic quantization results to the [0,1] interval (to avoid numerical differences caused by different defect types and different area sizes); if there are multiple defects in the same local area (such as scratches + dents), take the maximum value of each defect quantization value.

[0035] Based on the blade at the a-th processing node Processing status data set Calculate the blade at the a-th processing node. The processing anomaly response quantity is calculated using the following formula: ,in, This indicates the blade at the a-th processing node. The amount of abnormal processing response, and This represents the preset abnormal sensitivity weighting coefficient after processing. This represents the average value of the blade root diameter data in the i-th bladed disk at the a-th processing node. This represents the average value of the blade tip diameter data in the i-th bladed disk at the a-th processing node. This represents the number of local image regions in the lower surface image information data of the a-th processing node. This represents the local image region index of the lower surface image information data of the a-th processing node. This represents the z-th local image region in the image information data of the lower surface of the a-th processing node. This represents the metric for image anomaly significance at the a-th processing node; Based on blades Initial processing anomaly response quantity before processing and the blade at the a-th processing node Processing abnormal response quantity Calculate the blade The formula for calculating the abnormal change in processing at the a-th processing node is: ,in, Indicates blade The amount of abnormal changes in processing at the a-th processing node.

[0036] In this invention, the initial processing anomaly response and node processing anomaly response of the blade are calculated based on the initial processing state and processing node state, and the anomaly change is further calculated, thereby achieving a quantitative characterization of processing anomalies such as blade surface defects and operating condition fluctuations. The purpose of this step is to comprehensively transform operating condition data, surface images, and key dimensional deviations into processing anomaly indicators, enabling early identification and quantification of processing anomalies. This provides a basis for timely adjustment of processing parameters and avoidance of batch defects, thereby achieving the beneficial effects of reducing processing risks and improving processing reliability.

[0037] Step S5: Based on the geometric deviation change and the machining anomaly change, calculate the machining quality status and comprehensive machining status of the blade at the machining node, preset the threshold, analyze the machining quality and send an early warning.

[0038] Specifically, based on the blade Geometric deviation change at the a-th processing node and leaves The amount of abnormal changes in processing at the a-th processing node Calculate the blade The processing quality status quantity at the a-th processing node is calculated using the following formula: ,in, Indicates blade The processing quality status quantity at the a-th processing node, This represents the influence coefficient of the preset geometric deviation change. This represents the preset influence coefficient of abnormal processing variation. Based on blades Machining quality status quantity at the a-th machining node Calculate the comprehensive machining state quantity of all blades at the a-th machining node. The calculation formula is as follows: ,in, This represents the overall processing status quantity of all blades at the a-th processing node; The preset threshold for the overall processing state quantity is: if the overall processing state quantity of all blades at the a-th processing node is... If the value is greater than or equal to the threshold value of the comprehensive processing state quantity, it is determined that the processing quality of the impeller is poor at the a-th processing node, and an early warning is issued to the relevant personnel.

[0039] In this invention, the blade processing quality status is calculated by weighting the changes in geometric deviation and processing anomalies, and then summarizing these to calculate a comprehensive processing status. Combined with preset thresholds, this enables processing quality analysis and early warning functions, achieving a quantitative assessment and anomaly warning of the overall processing quality of the bladed disk. This step maps local blade deviations and anomalies to the overall bladed disk quality status, directly reflecting the quality level of processing nodes. The early warning mechanism prompts timely intervention by staff to prevent defects from escalating, thereby improving the overall pass rate of bladed disk processing and ensuring production safety and processing efficiency.

[0040] In this embodiment, fifteen preset coefficients are involved, including: preset geometric deviation contribution coefficient before processing. and Preset contribution coefficient of geometric deviation after processing and Preset abnormality sensitivity weighting coefficients before processing and Preset abnormal sensitivity weighting coefficients after processing and Preset influence coefficient of geometric deviation variation Preset influence coefficient of abnormal processing variation ,in: This reflects the importance of the initial deformation of the mold surface caused by raw materials or previous processes. If the impeller is a precision casting, the initial accuracy of the mold surface is required to be high. A larger value can be taken (e.g., 0.4~0.6). If it is a rough-machined blank, it can be appropriately reduced. This reflects the initial state of the blade's rotational balance. For high-speed rotating components, this difference affects dynamic balance. It is usually assigned a medium or high weight (e.g., 0.3 to 0.5). The initial difference in thickness uniformity between blades reflects the aerodynamic design's sensitivity to thickness distribution. A higher value can be taken; conversely, a lower value can be taken.

[0041] Based on statistical analysis of historical qualified blank data, the standard deviation of each dimension of data is calculated, normalized, and used as the initial value of the coefficients. Alternatively, the coefficients can be directly allocated by process experts according to the importance of the design tolerances, which usually meet the following requirements. ; If the machining process (such as milling) easily causes blade chatter and tool deflection, resulting in surface distortion, then A larger value should be selected (e.g., 0.5~0.7); The root tip diameter is usually guaranteed during the finishing stage. If this process is prone to exceeding tolerances, then... The adjustment needs to be increased at the corresponding nodes; It is strongly related to tool wear and clamping stability, especially during long-term machining or in thin-walled areas. The weight needs to be increased; Furthermore, during the machining process, the sensitivity of each geometric dimension to process fluctuations differs from that in the initial state; for example, cutting forces can easily lead to surface deformation. Increased sensitivity means tool wear may have a greater impact on thickness control. (Weight changes), therefore, it is necessary to set the geometric deviation contribution coefficient before processing and the geometric deviation contribution coefficient after processing; and By simulating the machining process or conducting trial cutting experiments, the degree of deterioration of data in each geometric dimension under different typical process failures (such as tool wear and vibration) can be observed, and adjustments can be made accordingly. The coefficients can also be obtained through reverse optimization learning based on a historical failure case library, and typically satisfy the following conditions: ; The initial dispersion of dimensions may indicate material or fixture problems. The weight is set according to the width of the dimensional tolerance band; the tighter the tolerance, the higher the weight. Initial surface defects (scratches, corrosion) have a significant impact on fatigue life. Typically, based on visual inspection standards and a defect sample library, the severity level of the defect is mapped to a weight value. and The settings are typically combined with incoming material inspection standards and the tolerance for defects in downstream processes, and are set by the quality engineer. For example, if there is a subsequent polishing process, then... It can be appropriately reduced, usually meeting the requirements. ; :and Similarly, but with a greater focus on processing-induced dimensional dispersion. The weights can be adjusted by inversely based on the dimensional capability index of the process.

[0042] : Monitor new defects generated during processing (such as vibration marks, burns), the weight of which depends on the level of impact of the defect on product performance; : Control the weight of the rate of change of speed, load, and temperature relative to the initial value in abnormal quantities; determine the correlation strength between operating condition fluctuations and quality results through correlation analysis (such as correlation and regression models), and allocate weights accordingly.

[0043] Furthermore, during processing, dimensional and surface defects are consequential anomalies, while fluctuations in operating conditions are precursory anomalies. The coefficient system must be able to respond to both types of signals simultaneously, with a greater focus on dynamic signals. and The core objective is to establish a baseline, eliminate inherent disturbances, and monitor the inherent state of the blades. and The core objective is to detect process anomalies and provide early warnings of new risks. Since the monitoring focuses on the processing process itself, it's necessary to set pre-processing and post-processing anomaly sensitivity weight coefficients. These weightings can be assigned by establishing a correlation model (such as correlation analysis or regression model) between operating condition fluctuation patterns and final quality defects through fault injection experiments or analysis of historical downtime data. This typically satisfies... ; A larger value indicates that the process or product focuses more on achieving dimensional and shape accuracy. This applies to final finishing and forming processes. A larger value indicates that the process prioritizes stability and defect-free operation. This is suitable for roughing, semi-finishing, or processes with extremely high surface integrity requirements. and Generally satisfies The ratio between the two reflects the trade-off between quality and cost. For example, during the trial production stage, to ensure safety, a certain percentage can be set... To achieve precision during the stable mass production stage, it is possible to set... The decision can be made by combining the opinions of the design department (emphasizing geometric accuracy), the process department (emphasizing process stability), and the quality department (emphasizing risk control), along with the functional positioning of the impeller and historical quality loss data.

[0044] Please see Figure 2 In this second embodiment: a bladed disk processing quality analysis system based on multi-source three-dimensional measurement information fusion is provided. The system includes: a set construction and initial data acquisition module, a processing state set construction module, a characterization quantity calculation and change quantity calculation module, a response quantity calculation and change quantity calculation module, and a state quantity calculation and analysis module. The set construction and initial data acquisition module: retrieves all information on the bladed disks to be processed from the database and constructs a set of blades for the bladed disks to be processed; before processing begins, it acquires the initial data of the blades in the bladed disks to be processed. The processing state set construction module: constructs the initial processing state data set of the blade; divides the processing process into several processing nodes, and constructs the processing state data set of the blade under the processing node; The module for calculating the characteristic quantity and the change quantity: calculates the initial geometric deviation characteristic quantity of the blade before processing based on the initial processing state data set; calculates the geometric deviation characteristic quantity of the blade at the processing node based on the processing state data set; and calculates the change quantity of geometric deviation of the blade at the processing node based on the initial geometric deviation characteristic quantity and the geometric deviation characteristic quantity. The response and change calculation module calculates the initial processing anomaly response of the blade before processing based on the initial processing state data set of the blade; it calculates the processing anomaly response of the blade at the processing node based on the processing state data set of the blade at the processing node; and it calculates the processing anomaly change of the blade at the processing node based on the initial processing anomaly response and the processing anomaly response. The state quantity calculation and analysis module calculates the processing quality state quantity and comprehensive processing state quantity of the blade at the processing node based on the geometric deviation change and the processing anomaly change, presets a threshold, analyzes the processing quality, and sends an early warning.

[0045] Furthermore, the characterization quantity calculation and change quantity calculation module includes a characterization quantity calculation unit and a change quantity calculation unit; The characterization calculation unit calculates the initial geometric deviation characterization of the blade before processing based on the initial processing state data set of the blade, and calculates the geometric deviation characterization of the blade at the a-th processing node based on the processing state data set of the blade at the a-th processing node. The change calculation unit calculates the change in geometric deviation of the blade at the a-th processing node based on the initial geometric deviation characterization of the blade before processing and the geometric deviation characterization of the blade at the a-th processing node.

[0046] Furthermore, the response quantity calculation and change quantity calculation module includes a response quantity calculation unit and a change quantity calculation unit; The response quantity calculation unit: calculates the initial processing abnormal response quantity of the blade before processing based on the initial processing state data set of the blade, and calculates the processing abnormal response quantity of the blade at the a-th processing node based on the processing state data set of the blade at the a-th processing node. The change calculation unit calculates the change in processing anomaly of the blade at the a-th processing node based on the initial processing anomaly response of the blade before processing and the processing anomaly response of the blade at the a-th processing node.

[0047] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0048] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A blisk machining quality analysis method based on multi-source three-dimensional measurement information fusion, characterized in that, The method includes the following steps: Step S1: Retrieve all information on the bladed disks to be processed from the database and construct a set of blades for the bladed disks to be processed; before processing begins, obtain the initial data of the blades in the bladed disks to be processed. Step S2: Construct the initial processing state data set of the blade; divide the processing process into several processing nodes and construct the processing state data set of the blade under each processing node; Step S3: Based on the initial processing state data set, calculate the initial geometric deviation characterization of the blade before processing; based on the processing state data set, calculate the geometric deviation characterization of the blade at the processing node; based on the initial geometric deviation characterization and the geometric deviation characterization, calculate the geometric deviation change of the blade at the processing node. Step S4: Based on the initial processing state data set of the blade, calculate the initial processing anomaly response amount of the blade before processing; based on the processing state data set of the blade under the processing node, calculate the processing anomaly response amount of the blade under the processing node; based on the initial processing anomaly response amount and the processing anomaly response amount, calculate the processing anomaly change amount of the blade at the processing node. Step S5: Based on the geometric deviation change and the machining anomaly change, calculate the machining quality status and comprehensive machining status of the blade at the machining node, preset the threshold, analyze the machining quality and send an early warning.

2. The blisk machining quality analysis method based on multi-source three-dimensional measurement information fusion according to claim 1, characterized in that, The specific implementation process of step S1 includes: Obtain all the information of the to-be-processed blisks from the database, and construct a blade set of the to-be-processed blisk, denoted as wherein, denotes the jth blade in the ith to-be-processed blisk, and J denotes the total number of blades in the ith blisk. A multi-source measurement module is constructed, which includes a laser scanning sensor, a coordinate measuring machine, a vision camera, and a processing condition acquisition device. The laser scanning sensor is used to acquire three-dimensional surface measurement data of the blades in the impeller disk. The coordinate measuring machine is used to acquire key dimension measurement data of the blades in the impeller disk. The vision camera is used to acquire surface image information data of the blades in the impeller disk. The processing condition acquisition device is used to acquire processing condition data of the blades in the impeller disk.

3. The method for analyzing the manufacturing quality of bladed disks based on multi-source three-dimensional measurement information fusion according to claim 2, characterized in that, The specific implementation process of step S1 also includes: The processing condition acquisition device includes a speed acquisition sensor, a load acquisition sensor, and a temperature acquisition sensor. The processing condition data includes speed data, load data, and temperature data. The key dimension measurement data includes blade root diameter data, blade tip diameter data, and blade thickness data. Before processing begins, obtain the j-th blade from the i-th bladed disk to be processed. The initial three-dimensional profile measurement data, initial data of blade root diameter, initial data of blade tip diameter, initial data of blade thickness, initial data of surface image information, initial rotational speed data, initial load data, and initial temperature data.

4. The method for analyzing the manufacturing quality of bladed disks based on multi-source three-dimensional measurement information fusion according to claim 3, characterized in that, The specific implementation process of step S2 includes: Before processing begins, the blades The initial three-dimensional profile measurement data, initial blade root diameter data, initial blade tip diameter data, initial blade thickness data, initial surface image information data, initial rotational speed data, initial load data, and initial temperature data are respectively denoted as: and and construct blades The initial processing state data set, denoted as ; The processing is divided into several processing nodes, and the blade data acquired in real time at the a-th processing node is... The three-dimensional profile measurement data, blade root diameter data, blade tip diameter data, blade thickness data, surface image information data, rotational speed data, load data, and temperature data are respectively denoted as: and And construct the blade at the a-th processing node. The processing status data set, denoted as .

5. The method for analyzing the manufacturing quality of bladed disks based on multi-source three-dimensional measurement information fusion according to claim 4, characterized in that, The specific implementation process of step S3 includes: Based on blades Initial processing state data set Calculate the blade The initial geometric deviation characterization value before processing is calculated using the following formula: ； in, Indicates blade Initial geometric deviation characterization quantity before processing and This represents the preset contribution coefficient of geometric deviation before processing. This indicates the number of measurement points in the initial three-dimensional surface measurement data. This represents the p-th measurement point in the initial three-dimensional surface measurement data. This represents the spatial coordinate data of the p-th measurement point. This represents the spatial center coordinates of all measured points. This represents the initial average thickness of all blades in the i-th bladed disk; Based on the blade at the a-th processing node Processing status data set Calculate the blade at the a-th processing node. The geometric deviation characteristic is calculated using the following formula: ； in, This indicates the blade at the a-th processing node. Geometric deviation characterization quantity, and This represents the preset contribution coefficient for geometric deviation after processing. This represents the number of measurement points in the 3D surface measurement data at the a-th processing node. This represents the q-th measurement point in the 3D surface measurement data at the a-th processing node. This represents the spatial coordinate data of the q-th measurement point under the a-th processing node. This represents the spatial center coordinates of all measurement points under the a-th processing node. This represents the initial average thickness of all blades in the i-th bladed disk at the a-th processing node; Based on blades Initial geometric deviation characterization before processing and the blade at the a-th processing node Geometric deviation characterization quantity, calculate the blade The formula for calculating the geometric deviation change at the a-th processing node is: ,in, Indicates blade The change in geometric deviation at the a-th processing node.

6. The method for analyzing the manufacturing quality of bladed disks based on multi-source three-dimensional measurement information fusion according to claim 5, characterized in that, The specific implementation process of step S4 includes: Based on blades Initial processing state data set Calculate the blade The initial processing anomaly response quantity before processing is calculated using the following formula: ； in, Indicates blade Initial processing anomaly response quantity before processing and This represents the preset abnormal sensitivity weighting coefficient before processing. This represents the average of the initial data for the root diameter of all blades in the i-th bladed disk. This represents the average of the initial data for all blade tip diameters in the i-th bladed disk. This indicates the number of local image regions in the initial data of the surface image information. The local image region number represents the initial data of the surface image information. This represents the nth local image region in the initial data of the surface image information. Indicates the significance of image anomalies; Based on the blade at the a-th processing node Processing status data set Calculate the blade at the a-th processing node. The processing anomaly response quantity is calculated using the following formula: ,in, This indicates the blade at the a-th processing node. The amount of abnormal processing response, and This represents the preset abnormal sensitivity weighting coefficient after processing. This represents the average value of the blade root diameter data in the i-th bladed disk at the a-th processing node. This represents the average value of the blade tip diameter data in the i-th bladed disk at the a-th processing node. This represents the number of local image regions in the lower surface image information data of the a-th processing node. This represents the local image region index of the lower surface image information data of the a-th processing node. This represents the z-th local image region in the image information data of the lower surface of the a-th processing node. This represents the metric for image anomaly significance at the a-th processing node; Based on blades Initial processing anomaly response quantity before processing and the blade at the a-th processing node Processing abnormal response quantity Calculate the blade The formula for calculating the abnormal change in processing at the a-th processing node is: ,in, Indicates blade The amount of abnormal changes in processing at the a-th processing node.

7. The method for analyzing the manufacturing quality of bladed disks based on multi-source three-dimensional measurement information fusion according to claim 6, characterized in that, The specific implementation process of step S5 includes: Based on blades Geometric deviation change at the a-th processing node and leaves The amount of abnormal changes in processing at the a-th processing node Calculate the blade The processing quality status quantity at the a-th processing node is calculated using the following formula: ,in, Indicates blade The processing quality status quantity at the a-th processing node, This represents the influence coefficient of the preset geometric deviation change. This represents the preset influence coefficient of abnormal processing variation. Based on blades Machining quality status quantity at the a-th machining node Calculate the comprehensive machining state quantity of all blades at the a-th machining node. The calculation formula is as follows: ,in, This represents the overall processing status quantity of all blades at the a-th processing node; The preset threshold for the overall processing state quantity is: if the overall processing state quantity of all blades at the a-th processing node is... If the value is greater than or equal to the threshold value of the comprehensive processing state quantity, it is determined that the processing quality of the impeller is poor at the a-th processing node, and an early warning is issued to the relevant personnel.

8. A bladed disk manufacturing quality analysis system based on multi-source three-dimensional measurement information fusion, comprising the bladed disk manufacturing quality analysis method based on multi-source three-dimensional measurement information fusion as described in any one of claims 1-7, characterized in that, The system includes: a set construction and initial data acquisition module, a processing state set construction module, a characterization quantity calculation and change quantity calculation module, a response quantity calculation and change quantity calculation module, and a state quantity calculation and analysis module; The set construction and initial data acquisition module: retrieves all information on the bladed disks to be processed from the database and constructs a set of blades for the bladed disks to be processed; before processing begins, it acquires the initial data of the blades in the bladed disks to be processed. The processing state set construction module: constructs the initial processing state data set of the blade; divides the processing process into several processing nodes, and constructs the processing state data set of the blade under the processing node; The module for calculating the characteristic quantity and the change quantity: calculates the initial geometric deviation characteristic quantity of the blade before processing based on the initial processing state data set; calculates the geometric deviation characteristic quantity of the blade at the processing node based on the processing state data set; and calculates the change quantity of geometric deviation of the blade at the processing node based on the initial geometric deviation characteristic quantity and the geometric deviation characteristic quantity. The response and change calculation module calculates the initial processing anomaly response of the blade before processing based on the initial processing state data set of the blade; it calculates the processing anomaly response of the blade at the processing node based on the processing state data set of the blade at the processing node; and it calculates the processing anomaly change of the blade at the processing node based on the initial processing anomaly response and the processing anomaly response. The state quantity calculation and analysis module calculates the processing quality state quantity and comprehensive processing state quantity of the blade at the processing node based on the geometric deviation change and the processing anomaly change, presets a threshold, analyzes the processing quality, and sends an early warning.

9. The bladed disk machining quality analysis system according to claim 8, characterized in that: The module for calculating the characteristic quantity and the change quantity includes a characteristic quantity calculation unit and a change quantity calculation unit; The characterization calculation unit calculates the initial geometric deviation characterization of the blade before processing based on the initial processing state data set of the blade, and calculates the geometric deviation characterization of the blade at the a-th processing node based on the processing state data set of the blade at the a-th processing node. The change calculation unit calculates the change in geometric deviation of the blade at the a-th processing node based on the initial geometric deviation characterization of the blade before processing and the geometric deviation characterization of the blade at the a-th processing node.

10. The bladed disk machining quality analysis system according to claim 9, characterized in that: The response quantity calculation and change quantity calculation module includes a response quantity calculation unit and a change quantity calculation unit; The response quantity calculation unit: calculates the initial processing abnormal response quantity of the blade before processing based on the initial processing state data set of the blade, and calculates the processing abnormal response quantity of the blade at the a-th processing node based on the processing state data set of the blade at the a-th processing node. The change calculation unit calculates the change in processing anomaly of the blade at the a-th processing node based on the initial processing anomaly response of the blade before processing and the processing anomaly response of the blade at the a-th processing node.

Images