A production quality detection and early warning method and device for electric tool gears

Abstract

The application provides a production quality detection and early warning method and device for electric tool gears, comprising: collecting multi-dimensional quality parameters in the gear manufacturing process; performing dynamic load simulation testing on the finished gears, and determining the dynamic performance grade thereof according to the testing data; performing correlation analysis on the multi-dimensional quality parameters of each gear and the corresponding dynamic performance grade thereof, establishing a parameter-performance correlation rule library based on a predefined logic condition combination; acquiring the multi-dimensional quality parameters in real time, and inputting the parameters into the parameter-performance correlation rule library for logic matching; determining the predicted performance grade of the new production gear according to the logic matching result; when the predicted performance grade is lower than the preset qualified grade, triggering an early warning, and outputting the corresponding key defect type and suggested treatment measures according to the triggered specific correlation rule. The application makes the detection of electric tool gears more intelligent and comprehensive.

Description

Technical Field

[0001] This invention relates to the field of gear production inspection technology, and more specifically, to a method and device for early warning of production quality inspection of power tool gears. Background Technology

[0002] The quality of gears in power tools (such as gears in circular saws and angle grinders) directly affects the reliability, lifespan, and safety of the tools. Current production quality inspection methods face the following problems: First, traditional inspections often focus on the conformity checks of static single parameters such as gear dimensions and hardness. These parameters lack a direct and clear correlation with the actual performance of gears under dynamic loads (such as vibration, noise, and fatigue life), leading to the risk of "static conformity, dynamic failure." Second, even when load testing is introduced, it is often only used as a final verification method for random sampling of finished products, failing to achieve real-time early warning and intervention during the production process. Quality problems can only be discovered afterward, resulting in waste of materials and time. Furthermore, some advanced prediction methods rely on complex mathematical models or finite element simulations, which are computationally expensive, slow to respond, and require deep professional knowledge to interpret the results, making it difficult to achieve immediate and universal quality judgment on fast-paced production lines. Summary of the Invention

[0003] Therefore, embodiments of the present invention provide a method and apparatus for early warning of production quality inspection of power tool gears, making the inspection of power tool gears more intelligent and comprehensive.

[0004] To address the aforementioned problems, this invention provides a production quality inspection and early warning method for power tool gears, comprising: collecting multi-dimensional quality parameters during the gear manufacturing process; conducting dynamic load simulation tests on the finished gears and determining their dynamic performance levels based on the test data; performing correlation analysis between the multi-dimensional quality parameters of each gear and their corresponding dynamic performance levels, and establishing a parameter-performance correlation rule library based on predefined logical condition combinations; in subsequent production processes, acquiring the multi-dimensional quality parameters of newly produced gears in real time and inputting the parameters into the parameter-performance correlation rule library for logical matching; determining the predicted performance level of the newly produced gear based on the logical matching results; and triggering an early warning when the predicted performance level is lower than a preset qualified level, and outputting the corresponding key defect type and suggested handling measures based on the specific correlation rules triggered.

[0005] Compared with existing technologies, the technical effects achieved by this solution are as follows: By collecting multi-dimensional process parameters and correlating them with measured dynamic performance levels, a predictive and early warning mechanism based on logical rules is creatively established. This successfully places the conclusions of destructive performance tests, which are retrospective, into the production process, enabling early identification and intervention of quality risks and effectively avoiding the problem of disconnect between traditional static testing and dynamic performance. It abandons the reliance on complex mathematical models and uses logical conditions based on historical data for judgment, making the early warning decision-making process clear, direct, and rapid. It is easy to deploy and apply in industrial settings, significantly improving the real-time performance, foresight, and practicality of quality control. It fundamentally changes the passive inspection model, shifting towards proactive predictive quality control.

[0006] In one embodiment of the present invention, the multi-dimensional quality parameters include at least: internal defect features determined by non-destructive testing, and tooth surface micromorphological features determined by image analysis.

[0007] Compared to existing technologies, this technical solution achieves the following technical effects: it precisely identifies the two key root factors affecting the dynamic service performance of gears. Internal defect characteristics are directly related to the gear's load-bearing capacity and fatigue fracture risk, while the microscopic morphology of the tooth surface directly affects vibration noise and wear efficiency during meshing. Using both as the basic dimensions for quality evaluation ensures the comprehensiveness and relevance of process parameter collection, providing a solid and focused data foundation for establishing accurate performance correlation rules. This avoids prediction biases caused by improper parameter selection or omissions, ensuring that the entire early warning system is built upon the key physical essence affecting performance.

[0008] In one embodiment of the present invention, the collection of multi-dimensional quality parameters during the gear manufacturing process further includes: scanning the gear blank; if pores are detected and their maximum size exceeds a first threshold, it is determined that there is a major internal defect; if microcracks are detected and their cumulative length exceeds a second threshold, it is determined that there is a potential risk of cracking; after processing, the tooth surface is photographed and analyzed; if the density of vibration marks is identified to exceed a third threshold, it is determined that there is a surface processing defect; if the measured tooth profile error exceeds a fourth threshold, it is determined that there is a tooth profile deviation.

[0009] Compared with existing technologies, the technical effects achieved by this solution are as follows: By setting clear thresholds and making judgments on pore size, microcrack length, vibration density, and tooth profile error, the results of non-destructive testing and image analysis can be automatically and objectively converted into discrete state indicators such as "major internal defects." This not only achieves the digitization and standardization of detection results, avoiding the subjectivity of human interpretation, but more importantly, it transforms continuous measurements into discrete facts suitable for logical rule processing. This lays a direct foundation for efficient matching and reasoning in the subsequent rule base, and is a key step in achieving automated and intelligent early warning for the entire method.

[0010] In one embodiment of the present invention, the step of performing dynamic load simulation testing on the finished gear and determining its dynamic performance level based on the test data further includes: running the gear under simulated hardwood cutting load conditions; if the vibration amplitude exceeds the safety benchmark and is accompanied by periodic impact noise, the dynamic performance is determined to be unqualified; if the vibration amplitude is within the safety benchmark, but there are abnormal high-frequency components in the noise spectrum, the dynamic performance is determined to be critical; if both the vibration and noise indicators are better than the safety benchmark, the dynamic performance is determined to be excellent.

[0011] Compared with existing technologies, the technical effects achieved by adopting this technical solution are as follows: By combining the periodic impact characteristics of vibration amplitude and noise with the logical combination of high-frequency components, it can effectively distinguish different failure modes and health states of gears under simulated real working conditions. It is intuitive and practical, which not only simplifies the performance evaluation system and enables the test results to be directly understood and utilized by the quality management system, but more importantly, the discrete performance level labels it generates are the target variables that are logically associated with manufacturing process parameters, thus building an intuitive bridge between process characteristics and final performance.

[0012] In one embodiment of the present invention, the establishment of the parameter-performance association rule base includes: Rule 1: If a gear's multi-dimensional quality parameters simultaneously satisfy the conditions of having significant internal defects and poor surface processing, its historical dynamic performance level is unqualified; Rule 2: If a gear's multi-dimensional quality parameters satisfy either the condition of having potential cracking risk or tooth profile deviation, its historical dynamic performance level is critical; Rule 3: If a gear's multi-dimensional quality parameters do not satisfy either Rule 1 or Rule 2, its historical dynamic performance level is excellent.

[0013] Compared with existing technologies, the technical effects achieved by adopting this technical solution are as follows: Rule 1 uses "AND" logic to associate the most serious composite defects with non-conforming performance, ensuring the accurate capture of high-risk gears; Rule 2 uses "OR" logic to identify the potential risks, i.e., critical states, caused by a single key defect, improving the sensitivity of the system; and Rule 3 defines the excellent state without the above defects, so that complex quality predictions do not need to be calculated in real time, but can be completed through efficient pattern matching, achieving speed and consistency in early warning decisions.

[0014] In one embodiment of the present invention, the determination results of the internal defect characteristics and tooth surface micromorphology characteristics of the new gear are used as input facts; the input facts are compared with the preconditions of each rule in the parameter-performance association rule base; the rule whose first precondition is fully satisfied is executed, and the conclusion of the rule is used as the predicted performance level of the new gear.

[0015] Compared to existing technologies, the technical effects achieved by this solution are as follows: By using the detection results of the new gear as "input facts" and comparing them one by one with the rule premises, the first rule that is fully satisfied is executed. This process ensures the clarity and exclusivity of the early warning decision. Each gear will inevitably and can only obtain a definite predictive performance level, avoiding ambiguous or conflicting conclusions. It achieves a seamless transformation from concrete data to abstract judgment and is a key execution link in the entire system's function of moving from "learning from history" to "predicting the future," ensuring the stability and reliability of the early warning output.

[0016] In one embodiment of the present invention, if the predicted performance level is unqualified, the highest level warning is triggered, the system outputs a warning message, and automatically locks the gear into the isolation zone; if the predicted performance level is critical, the intermediate warning is triggered, the system outputs a prompt message, and marks a specific identifier on the gear; if the predicted performance level is excellent, no warning is triggered, and the product is released normally.

[0017] Compared with existing technologies, the technical effects achieved by this solution are as follows: A graded early warning and response mechanism strictly corresponding to the predicted performance level is established, realizing differentiated risk control. For non-conforming risks, the highest level of isolation and scrapping warnings are implemented to decisively prevent defective products from entering subsequent stages or the market; for critical risks, intermediate warnings are issued with marking and recommendations for downgrading, maximizing the utilization of material value while ensuring safety; and for excellent products, release is carried out without intervention. This graded approach precisely focuses limited quality management resources on risk points of different levels, avoiding the cost waste or risk omission caused by a one-size-fits-all approach. It makes the early warning not just a signal, but a set of operational instructions that can directly drive different responses in the production process, greatly improving the precision and economy of quality control.

[0018] In one embodiment of the present invention, when an alert is triggered, the system automatically retrieves the production process information corresponding to the alert; and associates it with one or more suspicious process steps according to the rule type that triggered the alert.

[0019] Compared to existing technologies, the technical benefits of this solution include: adding crucial process traceability capabilities to the early warning system, enabling reverse tracing from the surface symptoms of quality problems to their root causes in manufacturing. When an alert is triggered, the system automatically links to the specific production process information that generated the defect and maps the rule type to the suspected process step. This function elevates simple quality alarms to a process diagnostic tool, helping production personnel quickly locate the source of problems; for example, a "poor surface finish" alert can be directly directed to the finishing machine tool. This not only accelerates the investigation and correction of quality problems and reduces downtime, but more importantly, it connects discrete quality events with continuous production processes, providing a direct basis for implementing targeted process improvements and promoting continuous optimization of the production process.

[0020] In one embodiment of the present invention, gears that have been warned and dealt with are periodically sampled and subjected to dynamic load simulation tests to obtain their actual performance level; the actual performance level is compared with the predicted performance level determined at the time of the warning; if it is found that the number of cases where the predicted result is inconsistent with the actual result under a certain combination of manufacturing parameters exceeds a preset number, an automatic prompt is made to adjust the corresponding logical conditions or thresholds in the parameter-performance association rule base.

[0021] Compared to existing technologies, the technical benefits of this solution are as follows: the entire early warning system incorporates self-optimization and continuous learning capabilities, ensuring its long-term effectiveness and adaptability. Through periodic sampling verification and comparison of predictive performance with actual performance, the system can autonomously identify prediction deviations in the rule base. When inconsistent cases accumulate to a certain number, the system will proactively prompt adjustments to rules or thresholds. This mechanism enables the system to dynamically evolve in response to changes in production conditions such as raw material fluctuations, equipment wear and tear, and process fine-tuning, avoiding the risk of gradual failure due to the solidification of initial rules.

[0022] This invention also provides a production quality inspection and early warning device for power tool gears. The production quality inspection and early warning device is used to implement a production quality inspection and early warning method. The device includes: a manufacturing parameter acquisition module for acquiring and logically determining the internal defect characteristics and microscopic morphological characteristics of the gear teeth; a load testing and performance determination module for testing the finished gear and logically determining its dynamic performance level; a rule base management module for storing and maintaining the parameter-performance association rule base; a logical reasoning module for matching new gears with the parameter-performance association rule base and outputting the predicted performance level and early warning conclusion; and an early warning execution and feedback module for triggering the corresponding level of early warning action based on the reasoning result and outputting process traceability information.

[0023] The production quality inspection and early warning device is used to implement the production quality inspection and early warning method, and therefore has all the effects of the production quality inspection and early warning method, which will not be elaborated here.

[0024] By adopting the technical solution of the present invention, the following technical effects can be achieved:

[0025] (1) By collecting multi-dimensional process parameters and correlating them with measured dynamic performance levels, a predictive early warning mechanism based on logical rules was creatively established. This successfully placed the conclusions of destructive performance tests in advance during the production process, enabling early identification and intervention of quality risks and effectively avoiding the problem of disconnect between traditional static testing and dynamic performance. It abandons the reliance on complex mathematical models and uses logical conditions based on historical data for judgment, making the early warning decision-making process clear, direct, and fast. It is easy to deploy and apply in industrial settings, significantly improving the real-time performance, foresight, and practicality of quality control. It fundamentally changes the passive inspection mode and shifts to proactive predictive quality control. Attached Figure Description

[0026] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 A flowchart illustrating a production quality inspection and early warning method for power tool gears provided in an embodiment of the present invention;

[0028] Figure 2 This is a schematic diagram of the module structure of a production quality inspection and early warning device for power tool gears provided in an embodiment of the present invention.

[0029] Explanation of reference numerals in the attached figures:

[0030] 100 is the production quality inspection and early warning device; 110 is the manufacturing parameter acquisition module; 120 is the load testing and performance judgment module; 130 is the rule base management module; 140 is the logic reasoning module; 150 is the early warning execution and feedback module. Detailed Implementation

[0031] To make the above-mentioned objectives, features, and advantages of the present invention more apparent and understandable, the technical solutions in the embodiments of the present invention are clearly and completely described. 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.

[0032] [First Embodiment]

[0033] See Figure 1 This invention provides a method for early warning of production quality inspection of power tool gears, comprising:

[0034] Step S100: Collect multi-dimensional quality parameters during the gear manufacturing process; conduct dynamic load simulation tests on the finished gear products and determine their dynamic performance level based on the test data;

[0035] Step S200: Perform correlation analysis between the multi-dimensional quality parameters of each gear and its corresponding dynamic performance level, and establish a parameter-performance correlation rule library based on predefined logical condition combinations.

[0036] Step S300: In the subsequent production process, the multi-dimensional quality parameters of the newly produced gears are obtained in real time, and the parameters are input into the parameter-performance association rule base for logical matching;

[0037] Step S400: Based on the result of logical matching, determine the predicted performance level of the newly produced gear; when the predicted performance level is lower than the preset qualified level, trigger an early warning, and output the corresponding key defect type and suggested handling measures according to the specific association rules triggered.

[0038] Specifically, the gear production line is equipped with an industrial CT scanner (for internal inspection of raw materials), a high-resolution industrial camera and 3D profilometer (for tooth surface morphology inspection), a dedicated dynamic load test bench (equipped with a vibration acceleration sensor and acoustic microphone), a central server, and various operating terminals.

[0039] Specifically, during system initialization, the first step is to construct an initial parameter-performance association rule base. For example, the following data is collected from 500 gears in a past production batch: CT scan images of each gear blank (for post-production analysis of porosity and cracks) and 3D morphology data of the finished tooth surface; each gear in this batch underwent dynamic load simulation testing on a test bench, and its performance level was determined as excellent, critical, or unqualified. Through manual and statistical analysis of these 500 sets of data, initial logical rules are extracted.

[0040] In one specific embodiment, for each newly produced gear (denoted as gear X), the system executes the following complete process:

[0041] S1: Acquire multi-dimensional quality parameters. Immediately after gear X completes sintering (blank), an industrial CT scan is performed to obtain three-dimensional data of its internal structure. After finishing, an industrial camera and profilometer are used to capture two-dimensional images and three-dimensional contour data of all tooth surfaces.

[0042] S2: Finished Product Performance Assessment (for rule base learning, not real-time warning gear). Gear X, as a finished product, will be sent to a test bench for dynamic load simulation testing. The test bench simulates the working conditions of an electric circular saw cutting hard oak (setting specific speed and load torque). Sensors record the effective value of vibration and noise spectrum during operation. The testing software automatically determines its dynamic performance level according to preset logic, such as "Excellent".

[0043] S3: Association Analysis and Rule Base Establishment (continuously running in the background). The central server binds Gear X's "multi-dimensional quality parameters" to its final "dynamic performance level" and stores them in the historical database. When enough new data is accumulated (e.g., 100 more samples are added), the system will run the association analysis algorithm, check and potentially optimize the established logical rule combinations, and update the parameter-performance association rule base.

[0044] S4: Real-time prediction and early warning (for key steps in subsequent new gears). For the next gear Y that has just been processed in stage S1, the system only executes S1 (collecting its manufacturing parameters), but does not need to wait for it to complete the actual measurement in stage S2. The system immediately sends the manufacturing parameter analysis results of gear Y (i.e., "input facts") to the latest parameter-performance association rule base for logical matching.

[0045] S5: Output and Handling. Based on the matching results (such as matching rule two), the system determines that the predicted performance level of gear Y is critical. Since this level is lower than the preset excellent / qualified level, the system triggers an early warning. The early warning information displays: "Predicted performance: critical; Key defect type: Tooth profile deviation; Recommended handling measures: Mark and downgrade for light-load tools."

[0046] Specifically, for internal defect assessment, the system processes the CT scan image of gear X. The software algorithm identifies all suspected pore areas and calculates their equivalent diameter. If the equivalent diameter of any pore exceeds a first threshold (e.g., set to Φ0.8mm), the system automatically marks it as "Major internal defect exists" = Yes. Simultaneously, the algorithm detects linear defects (cracks) and accumulates the length of all microcracks. If the accumulated length exceeds a second threshold (e.g., set to 2mm), it is marked as "Potential cracking risk exists" = Yes. For tooth surface morphology assessment: the system analyzes the tooth surface image of gear X. A texture analysis algorithm calculates the number of "striates" (periodic textures) per unit area. If this density exceeds a third threshold (e.g., set to 5 stripes per square millimeter), it is marked as "Poor surface finish" = Yes. A 3D profilometer measures the tooth profile curve and compares it with the standard theoretical tooth profile, calculating the maximum error. If this error exceeds a fourth threshold (e.g., set to 0.02mm), it is marked as "Tooth profile out of tolerance" = Yes.

[0047] Specifically, in the dynamic performance level determination, gear X operates under rated conditions on the test bench, and the system collects: vibration signal: calculates its effective value of vibration velocity (RMS), and noise signal: performs spectrum analysis.

[0048] Unacceptable: If the vibration RMS value exceeds the safety benchmark value (e.g., 1.5 mm / s) and there are obvious periodic impact spikes (with energy significantly higher than the background noise) at the meshing frequency and its harmonics in the noise spectrum, it is deemed unacceptable.

[0049] Critical: If the vibration RMS value does not exceed the above safety benchmark, but there are abnormal, continuous high-frequency components in the high-frequency band of the noise spectrum (e.g., 4-8kHz) (the amplitude of which is higher than the typical spectrum of a qualified gear), it is determined to be critical.

[0050] Excellent: If the vibration RMS value is better than (lower than) the safety benchmark value, and the noise spectrum does not have the above-mentioned abnormal periodic impacts or abnormal high-frequency components, it is judged as excellent.

[0051] Furthermore, for gears predicted to be defective: the system triggers the highest-level warning (red alarm light flashes, terminal screen pops up). The control program automatically locks the conveyor belt and sends the gear into the isolation material box. Simultaneously, the system marks its status as locked-to-be-scrapped in the database and generates a report. For gears predicted to be critical: the system triggers a medium-level warning (yellow alarm light stays on). The system controls a marking machine to laser-engrave a "△" symbol on the non-working surface of the gear (this can be changed according to actual conditions). The terminal screen prompts the operator: "Gear XXX, predicted performance is critical (tooth profile out of tolerance), recommended to be assigned to a low-load product line." For gears predicted to be "excellent": the system does not trigger any audible or visual alarms, and the gear normally flows into the next packaging process.

[0052] Furthermore, when gear Y triggers a critical warning (due to tooth profile deviation), the warning execution module immediately sends a query request to the Manufacturing Execution System (MES). Based on gear Y's unique QR code, the MES provides its complete production history, including: "Finishing – Machine Tool Number: xx; Operating Team: xx; Processing Time: xx". The system associates this information with the warning conclusion, recording in the warning log: "Suspicious Process Step: Finishing (Machine Tool xx)", providing maintenance personnel with a precise starting point for investigation.

[0053] Furthermore, at the end of each month, quality engineers randomly sample 5% of all gears with critical warnings for actual dynamic load testing. For example, a gear previously predicted to be critical (due to "tooth profile deviation") might be found to have excellent performance in actual testing. The system records this case. After one quarter, the system statistics show that among gears predicted to be critical due to tooth profile deviation alone, the proportion with excellent performance in actual testing reached 15% (exceeding the preset 10% inconsistency threshold). The system automatically sends a notification to the administrator: "The confidence level of the correlation between tooth profile deviation and critical performance in Rule 2 has decreased. It is recommended to review and adjust the fourth threshold or modify the rule logic."

[0054] Preferably, by collecting multi-dimensional process parameters and correlating them with measured dynamic performance levels, a predictive and early warning mechanism based on logical rules has been creatively established. This successfully places the conclusions of destructive performance tests, which are retrospective, into the production process, enabling early identification and intervention of quality risks and effectively avoiding the problem of disconnect between traditional static testing and dynamic performance. It abandons the reliance on complex mathematical models and uses logical conditions based on historical data for judgment, making the early warning decision-making process clear, direct, and rapid. It is easy to deploy and apply in industrial settings, significantly improving the real-time performance, foresight, and practicality of quality control. It fundamentally changes the passive inspection model, shifting towards proactive predictive quality control.

[0055] Specifically, the multi-dimensional quality parameters include at least: internal defect features determined by non-destructive testing, and microscopic morphological features of the tooth surface determined by image analysis.

[0056] The optimal approach precisely identifies the two key fundamental factors influencing the dynamic service performance of gears. Internal defects directly relate to the gear's load-bearing capacity and fatigue fracture risk, while the microscopic morphology of the tooth surface directly affects vibration, noise, and wear efficiency during meshing. Using both as the foundational dimensions for quality evaluation ensures the comprehensiveness and relevance of process parameter collection, providing a solid and focused data basis for establishing accurate performance correlation rules. This avoids prediction biases caused by improper parameter selection or omissions, ensuring the entire early warning system is built upon the key physical essence affecting performance.

[0057] Specifically, the collection of multi-dimensional quality parameters during the gear manufacturing process also includes: scanning the gear blank; if porosity is detected and its maximum size exceeds the first threshold, it is determined that there is a major internal defect; if microcracks are detected and their cumulative length exceeds the second threshold, it is determined that there is a potential risk of cracking; after processing, the tooth surface is photographed and analyzed; if the density of vibration marks is identified to exceed the third threshold, it is determined that there is a surface finishing defect; if the measured tooth profile error exceeds the fourth threshold, it is determined that there is a tooth profile deviation.

[0058] Preferably, by setting clear thresholds for pore size, microcrack length, vibration density, and tooth profile error, and then making judgments, the results of non-destructive testing and image analysis can be automatically and objectively converted into discrete state indicators such as "major internal defects." This not only achieves the digitization and standardization of detection results, avoiding the subjectivity of human interpretation, but more importantly, it transforms continuous measurement values ​​into discrete facts suitable for logical rule processing. This lays a direct foundation for efficient matching and reasoning in the subsequent rule base, and is a key step in achieving automated and intelligent early warning for the entire method.

[0059] Specifically, dynamic load simulation tests are conducted on the finished gears that have been manufactured, and the dynamic performance level is determined based on the test data. This includes: running the gears under load conditions simulating the cutting of hardwood; if the vibration amplitude exceeds the safety benchmark and is accompanied by periodic impact noise, the dynamic performance is deemed unqualified; if the vibration amplitude is within the safety benchmark, but there are abnormal high-frequency components in the noise spectrum, the dynamic performance is deemed critical; if both the vibration and noise indicators are better than the safety benchmark, the dynamic performance is deemed excellent.

[0060] Preferably, by combining the periodic impact characteristics of vibration amplitude and noise with the logical combination of high-frequency components, it is possible to effectively distinguish different failure modes and health states of gears under simulated real working conditions. This is intuitive and practical, not only simplifying the performance evaluation system and enabling the test results to be directly understood and utilized by the quality management system, but more importantly, the discrete performance level labels it generates are the target variables that are logically associated with manufacturing process parameters, thus building an intuitive bridge between process characteristics and final performance.

[0061] Specifically, the parameter-performance correlation rule base includes: Rule 1: If a gear's multi-dimensional quality parameters simultaneously satisfy the conditions of having significant internal defects and poor surface processing, its historical dynamic performance level is unqualified; Rule 2: If a gear's multi-dimensional quality parameters satisfy either the condition of having potential cracking risk or tooth profile deviation, its historical dynamic performance level is critical; Rule 3: If a gear's multi-dimensional quality parameters do not satisfy any of the conditions in Rule 1 and Rule 2, its historical dynamic performance level is excellent.

[0062] Preferably, Rule 1 uses AND logic to associate the most severe composite defects with non-conforming performance, ensuring the accurate capture of high-risk gears; Rule 2 uses OR logic to identify the potential risks, i.e., critical states, caused by a single key defect, improving the sensitivity of the system; and Rule 3 defines the excellent state without the above defects, so that complex quality predictions do not need to be calculated in real time, but can be completed through efficient pattern matching, achieving speed and consistency in early warning decisions.

[0063] Specifically, the determination results of the internal defect characteristics and tooth surface micromorphology characteristics of the new gear are used as input facts; the input facts are compared with the preconditions of each rule in the parameter-performance association rule base; the rule whose first precondition is fully satisfied is executed, and the conclusion of the rule is used as the predicted performance level of the new gear.

[0064] Preferably, by using the detection results of the new gear as "input facts" and comparing them one by one with the rule premises, the first rule that is fully satisfied is executed. This process ensures the clarity and exclusivity of the early warning decision. Each gear will inevitably and can only obtain a definite predictive performance level, avoiding ambiguous or conflicting conclusions. It achieves a seamless transformation from concrete data to abstract judgment and is a key execution link in the entire system's function of moving from "learning from history" to "predicting the future," ensuring the stability and reliability of the early warning output.

[0065] Specifically, if the predicted performance level is unqualified, the highest level warning is triggered, the system outputs a warning message, and automatically locks the gear into the isolation zone; if the predicted performance level is critical, the intermediate warning is triggered, the system outputs a prompt message, and marks a specific mark on the gear; if the predicted performance level is excellent, no warning is triggered, and the product is released normally.

[0066] The preferred approach establishes a tiered early warning and response mechanism that strictly corresponds to the predicted performance level, enabling differentiated risk management. For non-conforming risks, the highest level of isolation and scrapping warnings are implemented to decisively prevent defective products from entering subsequent stages or the market. For critical risks, intermediate warnings are issued with marking and recommendations for downgrading, maximizing the utilization of material value while ensuring safety. For excellent products, release without intervention. This tiered approach precisely focuses limited quality management resources on risk points of different levels, avoiding the cost waste or risk omission caused by a one-size-fits-all approach. It makes the early warning not just a signal, but a set of operational instructions that can directly drive different responses in the production process, greatly improving the precision and economy of quality control.

[0067] Specifically, when an alert is triggered, the system automatically retrieves the production process information corresponding to the alert; and based on the rule type that triggered the alert, it associates it with one or more suspicious process steps.

[0068] Preferably, this adds crucial process traceability capabilities to the early warning system, enabling reverse tracing from the surface symptoms of quality problems to their root causes in production. When an early warning is triggered, the system automatically links to the specific production process information that generated the defect and maps the rule type to the suspected process step. This function elevates a simple quality alarm to a process diagnostic tool, helping production personnel quickly locate the source of the problem; for example, it can directly point an early warning for "poor surface finish" to the finishing machine tool. This not only accelerates the investigation and correction of quality problems and reduces downtime, but more importantly, it links discrete quality events to continuous production processes, providing a direct basis for implementing targeted process improvements and promoting continuous optimization of the production process.

[0069] Specifically, gears that have been warned and dealt with are sampled periodically and subjected to dynamic load simulation tests to obtain their actual performance level. The actual performance level is compared with the predicted performance level determined at the time of the warning. If more than a preset number of cases are found where the predicted results are inconsistent with the actual results under a certain combination of manufacturing parameters, the system will automatically prompt that the corresponding logical conditions or thresholds in the parameter-performance association rule base need to be adjusted.

[0070] Preferably, the entire early warning system incorporates self-optimization and continuous learning capabilities, ensuring its long-term effectiveness and adaptability. Through periodic sampling verification and comparison of predictive performance with actual performance, the system can autonomously identify prediction deviations in the rule base. When inconsistent cases accumulate to a certain number, the system will proactively prompt adjustments to rules or thresholds. This mechanism enables the system to dynamically evolve in response to changes in production conditions such as raw material fluctuations, equipment wear and tear, and process fine-tuning, avoiding the risk of gradually becoming ineffective due to the solidification of initial rules.

[0071] See Figure 2 The present invention also provides a production quality inspection and early warning device 100 for power tool gears. The production quality inspection and early warning device 100 is used to implement a production quality inspection and early warning method. The production quality inspection and early warning device 100 includes: a manufacturing parameter acquisition module 110, used to acquire and logically determine the internal defect characteristics and microscopic morphological characteristics of the gear teeth; a load testing and performance determination module 120, used to test the finished gear and logically determine its dynamic performance level; a rule base management module 130, used to store and maintain a parameter-performance association rule base; a logical reasoning module 140, used to match new gears with the parameter-performance association rule base and output predicted performance levels and early warning conclusions; and an early warning execution and feedback module 150, used to trigger corresponding level early warning actions based on the reasoning results and output process traceability information.

[0072] The production quality inspection and early warning device 100 is used to implement the production quality inspection and early warning method, and therefore has all the effects of the production quality inspection and early warning method, which will not be elaborated here.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for early warning of production quality inspection of power tool gears, characterized in that, include: Collect multi-dimensional quality parameters during the gear manufacturing process; Dynamic load simulation tests are conducted on the finished gears that have been manufactured, and their dynamic performance level is determined based on the test data. The multi-dimensional quality parameters of each gear are correlated with their corresponding dynamic performance levels. Based on predefined combinations of logical conditions, a parameter-performance correlation rule library is established. In the subsequent production process, the multi-dimensional quality parameters of the newly produced gears are acquired in real time, and the parameters are input into the parameter-performance association rule library for logical matching. Based on the results of logical matching, the predicted performance level of the newly produced gear is determined; When the predicted performance level is lower than the preset qualified level, an early warning is triggered, and the corresponding key defect type and suggested handling measures are output according to the specific association rule triggered. The establishment of the parameter-performance association rule base includes: Rule 1: If a gear's multi-dimensional quality parameters simultaneously satisfy the conditions of having significant internal defects and poor surface finish, its historical dynamic performance level is unqualified. Rule 2: If a gear's multi-dimensional quality parameters satisfy any of the conditions of potential cracking risk or tooth profile deviation, then its historical dynamic performance level is critical. Rule 3: If a gear's multi-dimensional quality parameters do not meet any of the conditions in Rule 1 and Rule 2, then its historical dynamic performance level is excellent. The multi-dimensional quality parameters include at least: internal defect features determined by non-destructive testing, and microscopic morphological features of the tooth surface determined by image analysis.

2. The method for early warning of production quality inspection of power tool gears according to claim 1, characterized in that, The multi-dimensional quality parameters collected during the gear manufacturing process also include: If a pore is detected when scanning the gear blank and its maximum size exceeds the first threshold, it is determined that there is a major internal defect. If a microcrack is detected and its cumulative length exceeds the second threshold, it is determined that there is a potential risk of cracking. After processing, the tooth surface is photographed and analyzed. If the density of vibration marks exceeds the third threshold, it is determined that there is a surface processing defect. If the measured tooth profile error exceeds the fourth threshold, it is determined that there is a tooth profile deviation.

3. The method for early warning of production quality inspection of power tool gears according to claim 2, characterized in that, The process of conducting dynamic load simulation tests on the manufactured gears and determining their dynamic performance level based on the test data also includes: If the vibration amplitude of the gear exceeds the safety benchmark and is accompanied by periodic impact noise when the gear is running under load conditions simulating hardwood cutting, the dynamic performance is deemed unqualified. If the vibration amplitude is within the safety range, but there are abnormal high-frequency components in the noise spectrum, the dynamic performance is judged to be critical. If both vibration and noise indicators are better than the safety benchmark, the dynamic performance is judged to be excellent.

4. The method for early warning of production quality inspection of power tool gears according to claim 1, characterized in that, The results of judging the internal defect characteristics and the micro-morphological characteristics of the tooth surface of the new gear are used as input facts; The input facts are compared one by one with the preconditions of each rule in the parameter-performance association rule base; The rule that the first precondition is fully met is executed, and the conclusion of that rule is used as the predicted performance level of the new gear.

5. The method for early warning of production quality inspection of power tool gears according to claim 4, characterized in that, If the predicted performance level is unqualified, the highest level warning will be triggered, the system will output a warning message, and automatically lock the gear into the isolation zone; If the predicted performance level is critical, a medium-level warning will be triggered, the system will output a prompt message, and mark a specific identifier on the gear. If the predicted performance level is excellent, no warning will be triggered, and the product will be released normally.

6. The method for early warning of production quality inspection of power tool gears according to claim 1, characterized in that, When an alert is triggered, the system automatically retrieves the production process information corresponding to the alert. Based on the rule type that triggers the alert, it is associated with one or more suspicious process steps.

7. The method for early warning of production quality inspection of power tool gears according to claim 1, characterized in that, Regularly sample gears that have been warned and dealt with, and conduct dynamic load simulation tests to obtain their actual performance level; Compare the actual performance level with the predicted performance level determined at the time of the warning; If the number of cases where the predicted results do not match the actual results under a specific combination of manufacturing parameters exceeds the preset number, the system will automatically prompt that the corresponding logical conditions or thresholds in the parameter-performance association rule base need to be adjusted.

8. A production quality inspection and early warning device for power tool gears, wherein the production quality inspection and early warning device is used to implement the production quality inspection and early warning method as described in any one of claims 1-7, characterized in that, The production quality detection and early warning device includes: The manufacturing parameter acquisition module is used to acquire and logically determine the internal defect characteristics and microscopic morphology characteristics of the gear teeth. The load testing and performance assessment module is used to test the finished gears and logically determine their dynamic performance level. The rule base management module is used to store and maintain the parameter-performance association rule base; The logic reasoning module is used to match the new gear with the parameter-performance association rule base and output the predicted performance level and warning conclusion. The early warning execution and feedback module is used to trigger early warning actions of the corresponding level based on the reasoning results and output process traceability information.

Images