A method and system for gearbox operating process parameter control

Abstract

The application relates to the technical field of gear box operation process parameter control, and discloses a gear box operation process parameter control method and system, which are characterized in that the small cumulative deviation of robot operation parameters is correlated and analyzed with deep-level vibration and temperature change characteristics of the gear box, and the limitation that traditional methods only rely on single-dimensional information or cannot effectively correlate multidimensional information is overcome. Through comprehensive utilization of different levels of information, the application can earlier and more accurately find potential damage of the gear box, so that timely adjustment of operation control parameters is realized, the damage process is effectively slowed down, sudden failure and production interruption are avoided, and the reliability and economic benefits of the automatic production line are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of gearbox operation process parameter control technology, and more specifically, to a gearbox operation process parameter control method and system. Background Technology

[0002] In modern intelligent manufacturing environments, the operational status of the gearbox—the core transmission component—of automated equipment, especially robots undertaking heavy-duty handling tasks, is crucial to production efficiency and product quality. Traditional control methods primarily rely on real-time monitoring and adjustment of direct operational information such as motor current and speed to cope with known load changes. However, when minute and unidentified changes in operating conditions occur during production, such as subtle differences in workpiece material properties, these traditional methods reveal their limitations. These minute changes in operating conditions can lead to continuous, imperceptible adjustment deviations in the control system, which are insufficient to trigger conventional fault alarms in the short term.

[0003] Over time, these seemingly insignificant deviations accumulate, causing early and insidious uneven wear on critical components inside the gearbox, especially the gears. Signs of this early damage often manifest in deeper operational data such as vibration and temperature. These changes are slow and easily masked by background noise, making it difficult for existing systems to correlate these different levels of information. This hinders the timely detection and handling of potential equipment problems, potentially leading to sudden failures that severely impact production line continuity and economic efficiency.

[0004] In highly automated production lines, when heavy-duty handling robots face unrecognized conditions such as minute changes in the physical properties of workpieces, existing technologies struggle to effectively utilize the continuous minute adjustment deviations in first-level operational information such as motor current and speed accumulated over a long period, combined with the slow evolution trends in second-level operational information such as vibration and temperature, to identify early, hidden, non-uniform wear of gearboxes. Based on this, control strategies can be adjusted in advance to slow down the damage process, thereby avoiding sudden failures and production interruptions.

[0005] Based on the above problems, existing technologies urgently need improvement. Summary of the Invention

[0006] This invention provides a method and system for controlling parameters during gearbox operation. The aim is to address the limitations of existing technologies where, when heavy-duty handling robots face unrecognized conditions such as minute changes in the physical properties of workpieces, it is difficult to effectively utilize the continuous, minute adjustment deviations in first-level operational information (e.g., motor current, speed) accumulated over a long period, combined with the slow evolution trends in second-level operational information (e.g., vibration, temperature), to identify early, hidden, non-uniform wear of the gearbox. Based on this, the control strategy can be adjusted in advance to slow the damage process, thereby avoiding sudden failures and production interruptions.

[0007] The technical solution of this application is as follows: In a first aspect, this application discloses a method for controlling parameters during gearbox operation, comprising the following steps: The operating parameters of the robot's drive components and the physical quantities of the gearbox are collected, including vibration and temperature physical quantities. The complete process of a robot handling one workpiece at a time is defined as a task unit. Based on the task unit, the deviation between the operating parameters of the robot drive component and the preset standard parameters is obtained, and the deviation is accumulated to obtain the cumulative deviation value; Based on the physical quantities of vibration and combined with the instantaneous meshing phase of the gear, the vibration change characteristics corresponding to a specific gear meshing phase are obtained through superposition and averaging. Based on the physical quantity of temperature, obtain the temperature evolution trend inside the gearbox; The cumulative deviation value is correlated with the vibration change characteristics and temperature evolution trend. Based on the correlation results, it is determined whether there is early hidden non-uniform wear in the gearbox. Adjust the gearbox's operating control parameters based on the assessment results.

[0008] This technical solution enables multi-dimensional correlation analysis of the cumulative deviation of robot operating parameters, vibration variation characteristics of gearbox, and temperature evolution trend, thereby effectively identifying early hidden non-uniform wear and adjusting operating control parameters in a timely manner to avoid sudden failures.

[0009] Furthermore, the cumulative deviation value is correlated with vibration variation characteristics and temperature evolution trends. Based on the correlation results, it is determined whether the gearbox exhibits early, hidden, non-uniform wear, including: Collect real-time motion status information of the robot; Based on the physical quantities of vibration, combined with the instantaneous meshing phase of the gear and the real-time motion state information of the robot, the vibration change characteristics corresponding to a specific gear meshing phase and the structural resonance characteristics corresponding to the robot's body motion phase are obtained through superposition and averaging. Based on vibration variation characteristics and structural resonance characteristics, combined with cumulative deviation values ​​and temperature evolution trends, it is determined whether the gearbox has early, hidden, non-uniform wear.

[0010] This technical solution incorporates real-time motion state information of the robot, enabling more precise separation of gear vibration and body structure resonance, thereby improving the accuracy of identifying early, hidden, non-uniform wear.

[0011] Based on this, the steps of correlating the cumulative deviation value with vibration change characteristics and temperature evolution trend, and determining whether the gearbox has early hidden non-uniform wear based on the correlation results, also include: Obtain information on the robot's current operating status, the expected lifespan parameters of the gearbox, and production requirements; Based on the operating condition information, the judgment thresholds used to determine the cumulative deviation value, vibration change characteristics, and temperature evolution trend are dynamically adjusted. By combining the expected life parameters of the gearbox and the accumulated operating time, the weights of vibration change characteristics and temperature evolution trend in the correlation judgment process are adjusted. Adjust the early warning level and response strategy according to production needs; Based on the adjusted judgment threshold, weight, and warning level, the cumulative deviation value, vibration change characteristics, and temperature evolution trend are correlated and judged to determine whether there is early hidden non-uniform wear in the gearbox.

[0012] This technical solution enables dynamic adjustment of judgment thresholds, weights, and early warning strategies based on actual operating conditions, expected lifespan, and production needs, making wear assessment more flexible and adaptable.

[0013] In some preferred embodiments, the step of obtaining the vibration change characteristics corresponding to a specific gear meshing phase further includes: In the non-critical motion phases set in each task unit of the robot, a small load disturbance with different frequencies and amplitudes is actively applied to the gearbox. The vibration response and ultrasonic emission response of the gearbox to minor load disturbances were collected simultaneously. The collected vibration response was synchronously averaged, and the amplitude and phase changes of the frequency components of the vibration response were analyzed to extract surface wear characteristics. Transient event counting and energy distribution analysis were performed on the acquired ultrasonic acoustic emission response to extract subsurface fatigue characteristics; Based on surface wear characteristics and subsurface fatigue characteristics, the damage trend prediction of the gearbox is revised. The corrected damage trend prediction is used as an auxiliary basis for judgment. Combined with the cumulative deviation value, vibration change characteristics and temperature evolution trend, it is determined whether there is early hidden non-uniform wear in the gearbox.

[0014] This technical solution, by actively applying minute load disturbances and combining data from multiple sensors, can more comprehensively acquire the surface wear and subsurface fatigue characteristics of the gearbox, thereby improving the accuracy of damage trend prediction.

[0015] Furthermore, after correcting the damage trend prediction of the gearbox based on surface wear characteristics and subsurface fatigue characteristics, it also includes: Based on the revised damage trend prediction, the flexible task planning strategy is adaptively adjusted.

[0016] This technical solution enables more accurate damage trend prediction and adaptive adjustment of flexible task planning strategies, achieving more refined preventative maintenance.

[0017] Based on the above, and according to the revised damage trend prediction, the steps for adaptively adjusting the flexible task planning strategy include: Obtain the current operating status information of other key components of the robot; Based on the current operating status information of other key components, assess the current health status and damage sensitivity of other key components; Simulate the damage mitigation effects of different flexible task planning strategies on the gearbox and the impact on the potential increase in load on other key components; Based on the simulation results, the flexible task planning strategy that has the least impact on the load of other key components is selected, and the robot's operating parameters are adjusted according to the selected flexible task planning strategy.

[0018] This technical solution enables the comprehensive consideration of the health status of the gearbox and other key components when adjusting flexible task planning strategies, avoiding the problem of "robbing Peter to pay Paul" and achieving global optimization.

[0019] Preferably, adjusting the robot's operating parameters according to the selected flexible task planning strategy includes: Based on the selected flexible task planning strategy, adjust the operating speed, acceleration, or torque limits of specific motion axes of the robot to reduce the instantaneous load or impact on the gearbox. And / or, based on the selected flexible task planning strategy, optimize the robot's motion trajectory or posture to avoid the gearbox being subjected to high stress or resonant frequency for a long time under a specific motion phase; And / or, depending on the selected flexible task planning strategy, adjust the cycle time or task sequence of the robot's task units to increase the gearbox's cooling or recovery time.

[0020] This technical solution provides a variety of specific ways to adjust operating parameters, allowing for flexible selection of the most suitable strategy based on actual conditions to achieve the best damage mitigation effect.

[0021] In one implementation, the effects of different flexible task planning strategies on damage mitigation of the gearbox and the potential increase in load on other critical components are simulated, including: Acquire real-time operational data of the robot, gearbox, and other key components; By comparing real-time operating data with corresponding parameters in the simulation environment, the deviation between the simulation environment and the actual operating environment can be identified and quantified. Based on the identified and quantified deviations, the component damage evolution parameters in the simulation environment are adjusted. These component damage evolution parameters include the material fatigue coefficient, wear rate constant, and thermal conductivity coefficient. In the adjusted simulation environment, different flexible task planning strategies were implemented, and the damage mitigation effect of the gearbox and the impact on the potential increase in load on other critical components were calculated.

[0022] This technical solution improves the accuracy and reliability of simulation results by calibrating the simulation environment with real-time data, making the evaluation of flexible task planning strategies closer to reality.

[0023] In another implementation, the step of simulating the damage mitigation effect of different flexible task planning strategies on the gearbox and the impact on the potential increase in load on other critical components further includes: Obtain external environmental information about the robot's operating environment; Based on external environmental information, adjust the environmental impact parameters in the damage trend prediction of the gearbox, and adjust the environmental impact parameters in the damage sensitivity of other key components. Based on the adjusted environmental impact parameters, the evaluation results of the flexible mission planning strategy were revised. The evaluation results include the assessment of the damage mitigation effect on the gearbox and the assessment of the potential increase in load on other critical components. Based on the revised evaluation results, the flexible task planning strategy that best mitigates gearbox damage and has the least impact on the load of other critical components is selected.

[0024] By incorporating external environmental information through this technical solution, the impact of environmental factors on damage evolution and strategy assessment can be considered more comprehensively, thereby further optimizing flexible mission planning.

[0025] Secondly, this application also discloses a gearbox operating process parameter control system, the system comprising: The data acquisition module is used to collect the operating parameters of the robot's drive components and the physical quantities of the gearbox, including vibration and temperature physical quantities. The task unit definition module is used to define the complete process of the robot handling one workpiece at a time as a task unit. The deviation accumulation module is used to obtain the deviation between the operating parameters of the robot drive component and the preset standard parameters according to the task unit, and accumulate the deviation to obtain the deviation accumulation value. The vibration feature acquisition module is used to acquire the vibration change features corresponding to a specific gear meshing phase by superimposing and averaging the collected vibration physical quantities of the gearbox and combining them with the instantaneous meshing phase of the gear. The temperature trend acquisition module is used to obtain the temperature evolution trend inside the gearbox based on the collected temperature physical quantities of the gearbox. The correlation judgment module is used to correlate the cumulative deviation value with the vibration change characteristics and temperature evolution trend. Based on the correlation results, it determines whether there is early hidden non-uniform wear in the gearbox and outputs instructions to adjust the gearbox operation control parameters based on the judgment results.

[0026] This technical solution provides a system for controlling the parameters of gearbox operation. Through modular design, it can efficiently integrate various functions and effectively monitor and control the early, hidden, non-uniform wear of the gearbox.

[0027] Beneficial Effects: The gearbox operation process parameter control method and system disclosed in this application defines a complete process of the robot handling one workpiece at a time as a task unit by collecting the operating parameters of the robot drive components and the physical quantities of the gearbox (including vibration and temperature). Based on this, the method can obtain and accumulate the deviation between the robot drive component operating parameters and preset standard parameters according to the task unit, thus obtaining a cumulative deviation value. Simultaneously, it combines the instantaneous meshing phase of the gears to perform superposition and averaging of the vibration physical quantities, obtaining the vibration change characteristics corresponding to a specific gear meshing phase; and it obtains the temperature evolution trend inside the gearbox based on the temperature physical quantity. Finally, it correlates the cumulative deviation value with the vibration change characteristics and the temperature evolution trend in multiple dimensions, determines whether the gearbox has early, hidden, non-uniform wear based on the correlation results, and adjusts the gearbox's operation control parameters accordingly.

[0028] Through the above technical solution, this application effectively solves the problem of difficulty in identifying early, hidden, non-uniform wear of gearboxes in heavy-duty handling robots in existing technologies. Specifically, this method correlates the small cumulative deviations of robot operating parameters with the deep-seated vibration and temperature change characteristics of the gearbox, overcoming the limitations of traditional methods that rely solely on single-dimensional information or cannot effectively correlate multi-dimensional information. By comprehensively utilizing information at different levels, this application can detect potential damage to the gearbox earlier and more accurately, thereby enabling timely adjustments to operating control parameters, effectively slowing down the damage process, avoiding sudden failures and production interruptions, and significantly improving the reliability and economic benefits of automated production lines. Attached Figure Description

[0029] Figure 1 This is a flowchart illustrating a method for controlling parameters during gearbox operation according to an embodiment of the present invention. Figure 2 This is a flowchart of a method for determining whether a gearbox has early, hidden, non-uniform wear based on correlation results, provided by an embodiment of the present invention. Figure 3 This is a schematic diagram of a gearbox operation process parameter control system provided in an embodiment of the present invention. Detailed Implementation

[0030] 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.

[0031] Reference Figure 1 , Figure 1 This is a flowchart illustrating a method for controlling parameters during gearbox operation according to an embodiment of the present invention, including: S11, Collect the operating parameters of the robot drive components and the physical quantities of the gearbox, including vibration physical quantities and temperature physical quantities; S12 defines the complete process of the robot handling one workpiece at a time as a task unit; S13, according to the task unit, obtain the deviation between the operating parameters of the robot drive component and the preset standard parameters, and accumulate the deviation to obtain the cumulative deviation value; S14. Based on the vibration physical quantity and combined with the instantaneous meshing phase of the gear, the vibration change characteristics corresponding to a specific gear meshing phase are obtained through superposition averaging. S15, Based on the temperature physical quantity, obtain the temperature evolution trend inside the gearbox; S16, correlate the cumulative deviation value with the vibration change characteristics and the temperature evolution trend, and determine whether the gearbox has early hidden non-uniform wear based on the correlation result; S17, Adjust the operating control parameters of the gearbox according to the judgment result.

[0032] This application, through comprehensive analysis of the deviation of operating parameters of robot drive components, vibration variation characteristics of gearbox, and temperature evolution trend, can effectively identify early hidden non-uniform wear of gearbox and adjust operating control parameters accordingly, thereby avoiding sudden failures and improving the stability and reliability of the production line.

[0033] To better understand this application, some key terms involved will be explained first.

[0034] "Operating parameters of robot drive components" usually refer to real-time data such as motor current, speed, torque, and position during robot movement. These parameters directly reflect the robot's working status and load conditions.

[0035] "Physical quantities of the gearbox" refer to physical indicators used to characterize the health status of the gearbox, mainly including "vibration physical quantities" and "temperature physical quantities". Among them, "vibration physical quantities" can be collected by acceleration sensors or vibration sensors installed on the gearbox housing, reflecting dynamic characteristics such as gear meshing and bearing operation; "temperature physical quantities" can be obtained by thermocouples or infrared thermometers, reflecting the thermodynamic state of the gearbox, such as friction and heat dissipation.

[0036] A "task unit" refers to the smallest work cycle in which a robot completes a full workpiece handling process, such as the entire process from picking up the workpiece to placing it. Dividing the continuous operation of a robot into discrete task units helps to standardize and periodically analyze operating parameters.

[0037] "Cumulative deviation value" refers to the cumulative amount of difference between the actual operating parameters of the robot's drive components and the preset standard parameters across multiple task units. This accumulation can amplify small, persistent operational anomalies, making them easier to detect.

[0038] The "instantaneous meshing phase of a gear" refers to the precise moment and position at which the tooth surfaces of a gear come into contact with each other during rotation. By analyzing vibration in conjunction with the meshing phase, it is possible to more accurately pinpoint which specific meshing state of the gear pair is causing the vibration.

[0039] "Superposition averaging" is a signal processing technique that can effectively suppress random noise and highlight periodic features related to specific events (such as gear meshing) by synchronously superimposing multiple periodic signals.

[0040] "Vibration change characteristics" refers to the changes in the amplitude, frequency, phase, and other characteristics of vibration signals related to the meshing phase of a specific gear after superposition and averaging. These changes are often a direct manifestation of gear wear and damage.

[0041] "Temperature evolution trend" refers to the pattern of temperature change inside the gearbox over time, including the rate of temperature increase, stabilization, decrease, and the presence of abnormal fluctuations. These trends can reflect changes in the internal friction state and heat dissipation performance of the gearbox.

[0042] "Early-stage hidden non-uniform wear" refers to the uneven wear on the surface or subsurface of the gearbox that occurs in the early stages of operation due to local stress concentration, poor lubrication, or minor changes in operating conditions. This wear is difficult to observe directly through conventional means. In the early stages of development, this wear often does not affect the normal operation of the equipment, but it will gradually deteriorate and eventually lead to failure.

[0043] "Operation control parameters" refer to parameters used to adjust the working state of the robot and gearbox, such as the robot's movement speed, acceleration, torque limit, and task cycle time.

[0044] In the embodiments of this application, it is first necessary to collect the operating parameters of the robot drive components and the physical quantities of the gearbox. The operating parameters of the robot drive components can be directly obtained through the robot controller or motor driver; for example, data from a motor current sensor, encoder, or torque sensor can be read in real time. The physical quantities of the gearbox, including vibration and temperature quantities, can be collected by installing appropriate sensors on the gearbox housing. For example, a piezoelectric accelerometer can be installed to acquire vibration quantities, and a thermocouple or infrared temperature sensor can be installed to acquire temperature quantities. These sensors should collect data at a sufficiently high sampling frequency to ensure that subtle changes in the operating state can be captured.

[0045] Next, the complete process of the robot handling one workpiece at a time is defined as a task unit. For example, on an automobile production line, a robot picks up a car door at point A, moves it to point B for installation, and then returns to point A to pick up the next car door. This entire cycle can be defined as a task unit. This definition method makes the analysis of robot operating parameters periodic and comparable.

[0046] Then, based on the aforementioned task units, the deviation between the operating parameters of the robot's drive components and preset standard parameters is obtained, and this deviation is accumulated to obtain a cumulative deviation value. For example, the motor current and speed curves when the robot is handling a standard workpiece can be set as preset standard parameters. In each task unit, the actual collected motor current and speed curves are compared with the standard curves, and their instantaneous or average deviations are calculated. These deviations can be positive (indicating that the actual value is higher than the standard value) or negative (indicating that the actual value is lower than the standard value). Subsequently, these deviations are accumulated across multiple task units, for example, by summation or integration, to obtain a cumulative deviation value that reflects the degree of deviation from the long-term operating state.

[0047] Simultaneously, based on the collected vibration physical quantities and combined with the instantaneous meshing phase of the gears, the vibration change characteristics corresponding to a specific gear meshing phase are obtained through superposition and averaging. For example, the instantaneous meshing phase information of the gears can be accurately obtained by installing a rotary encoder inside the gearbox. After the vibration signal is collected, the vibration signals of multiple meshing cycles are superimposed and averaged based on the gear meshing cycle. This processing method can effectively filter out random noise and non-periodic vibration components, highlighting the periodic vibration characteristics related to gear meshing. By analyzing the amplitude, frequency components, or waveform changes of the superimposed and averaged vibration signal, the vibration change characteristics corresponding to a specific gear meshing phase can be extracted. For example, whether the impact vibration energy of a specific gear at a certain meshing angle is abnormally increased.

[0048] Furthermore, based on the collected temperature physical quantities, the temperature evolution trend inside the gearbox can be obtained. For example, temperature data from multiple temperature measurement points inside the gearbox can be recorded and analyzed over a long period. By plotting temperature-time curves, the overall temperature change trend can be observed, such as whether it continues to rise, whether abnormal fluctuations occur, and whether the temperature gradient changes. These trends can reflect changes in the friction state, lubrication condition, or heat dissipation performance inside the gearbox.

[0049] Next, the accumulated deviation value is correlated with vibration variation characteristics and temperature evolution trends. Based on the correlation results, it is determined whether the gearbox exhibits early, hidden, non-uniform wear. For example, a multi-parameter fusion model can be established, using the accumulated deviation value, vibration variation characteristics, and temperature evolution trend as inputs. This model can be constructed based on machine learning algorithms (such as support vector machines or neural networks) or expert system rules. By analyzing the interrelationships and co-variations among these parameters, it is determined whether early, hidden, non-uniform wear exists. For example, when the accumulated deviation value continuously increases, while the vibration amplitude of a specific gear meshing phase shows a slight but continuous increase, and the internal temperature of the gearbox shows a slow upward trend, these correlated information may collectively indicate the presence of early, hidden, non-uniform wear in the gearbox.

[0050] Finally, the gearbox's operating control parameters are adjusted based on the assessment results. If early, hidden, uneven wear is detected, the system can automatically or prompt the operator to adjust the robot's operating control parameters. For example, the robot's operating speed, acceleration, or maximum torque can be appropriately reduced to decrease the load and impact on the gearbox; the robot's motion trajectory can be optimized to avoid prolonged operation in specific high-stress areas; the cycle time of the task unit can also be adjusted to increase the gearbox's cooling or recovery time. These adjustments aim to slow down the wear process, extend the gearbox's service life, and provide early warnings for subsequent maintenance.

[0051] This application offers significant advantages and innovations compared to existing technologies. Traditional methods primarily rely on real-time monitoring and adjustment of direct operating information such as motor current and speed. While effective in handling known load changes, these methods fall short in addressing early, insidious, non-uniform wear caused by minute changes in operating conditions. Existing technologies struggle to effectively correlate small, accumulated adjustment deviations over long periods with the slow evolution of deeper physical quantities such as vibration and temperature, thus failing to promptly detect and address potential equipment hazards.

[0052] In some embodiments described above in this application, the presence of early, hidden, non-uniform wear in the gearbox is determined by correlating cumulative deviation values, vibration variation characteristics, and temperature evolution trends. However, in practical applications, the motion state of the robot body may introduce additional vibration components, which may couple with the wear characteristics of the gearbox, thereby affecting the accuracy and reliability of the determination.

[0053] In this regard, refer to Figure 2 , Figure 2 This is a flowchart of a method for determining whether a gearbox has early, hidden, non-uniform wear based on correlation results, provided by an embodiment of the present invention. S16 includes: S161, Collect the real-time motion status information of the robot; S162, Based on the vibration physical quantity, combined with the instantaneous meshing phase of the gear and the real-time motion state information of the robot, the vibration change characteristics corresponding to the specific gear meshing phase and the structural resonance characteristics corresponding to the robot's body motion phase are obtained through superposition and averaging processing. S163, based on the vibration change characteristics and the structural resonance characteristics, combined with the cumulative deviation value and the temperature evolution trend, determine whether the gearbox has early hidden non-uniform wear.

[0054] Specifically, the real-time motion state information refers to the robot's current pose, velocity, acceleration, joint angles, joint torques, and other dynamic parameters. This information reflects the specific motion mode and force conditions of the robot body when performing a task. Collecting this information provides an important reference benchmark for subsequent vibration analysis. In acquiring vibration change characteristics, in addition to combining the instantaneous meshing phase of the gears, the robot's real-time motion state information is further incorporated. Through superposition and averaging, not only can vibration change characteristics corresponding to a specific gear meshing phase be obtained, but also structural resonance characteristics corresponding to the robot's body motion phase can be acquired. The structural resonance characteristics refer to the inherent vibration response generated by the robot's body structure under specific motion states; these responses may be superimposed with the gearbox's vibration signal. By distinguishing and extracting these structural resonance characteristics, interference from the robot's body motion on gearbox wear assessment can be effectively avoided.

[0055] In practical applications, based on the acquired vibration change characteristics and structural resonance characteristics, combined with the cumulative deviation value and temperature evolution trend, a more comprehensive and accurate judgment can be made. This means that when judging whether there is early hidden non-uniform wear in the gearbox, not only are the wear signs of the gear itself considered (through vibration change characteristics and cumulative deviation value), but also the thermal state of the gearbox (through temperature evolution trend), and the interference caused by the robot's own motion is excluded (through structural resonance characteristics).

[0056] This application's solution effectively solves the problem of robot motion interfering with gearbox wear assessment in traditional methods by introducing real-time robot motion state information and further extracting structural resonance features corresponding to the robot's motion phase. Through this technical solution, the accuracy and reliability of early-stage, hidden, non-uniform wear assessment of the gearbox are significantly improved. By eliminating the structural resonance interference introduced by robot motion, the obtained vibration variation features more realistically reflect the wear state inside the gearbox. This allows the system to identify potential wear problems earlier and more accurately, avoiding misjudgments or omissions, thus providing a solid technical foundation for timely adjustment of operating control parameters and extending gearbox service life.

[0057] In some embodiments described above, this application proposes a method to determine whether a gearbox exhibits early, hidden, non-uniform wear by correlating the cumulative deviation of operating parameters of the robot drive components, the vibration characteristics of the gearbox, and the temperature evolution trend. However, in practical applications, the operating state of the gearbox, its remaining lifespan, and current production demands are dynamically changing. If the judgment logic and response strategy remain fixed, it may lead to misjudgments or an inability to respond promptly and effectively to wear conditions under different operating conditions, thereby affecting the accuracy of the judgment and the adaptability of the control strategy.

[0058] In this regard, this application further proposes that the steps of correlating the cumulative deviation value with vibration change characteristics and temperature evolution trend, and determining whether the gearbox has early hidden non-uniform wear based on the correlation results, also include: Obtain information on the robot's current operating status, the expected lifespan parameters of the gearbox, and production requirements; Based on the operating condition information, the judgment thresholds used to determine the cumulative deviation value, vibration change characteristics, and temperature evolution trend are dynamically adjusted. By combining the expected life parameters of the gearbox and the accumulated operating time, the weights of vibration change characteristics and temperature evolution trend in the correlation judgment process are adjusted. Adjust the early warning level and response strategy according to production needs; Based on the adjusted judgment threshold, weight, and warning level, the cumulative deviation value, vibration change characteristics, and temperature evolution trend are correlated and judged to determine whether there is early hidden non-uniform wear in the gearbox.

[0059] Specifically, obtaining the robot's current operating status information refers to acquiring real-time data such as the robot's current load status, operating speed, acceleration, and ambient temperature. This information reflects the actual workload and environmental impact currently borne by the gearbox. The gearbox's expected lifespan parameter refers to the theoretical service life determined during the design or testing phase of that gearbox model, while production requirements refer to the current production tasks' demands on the robot's operating efficiency, downtime, and product quality.

[0060] The dynamic adjustment of the judgment threshold based on operating condition information can be understood as follows: when the robot is under harsh conditions such as high load and high speed, the wear process of the gearbox may accelerate. In this case, the wear judgment threshold should be appropriately lowered to detect potential problems earlier. Conversely, under light load or standby conditions, the threshold can be appropriately increased to avoid oversensitivity. The purpose is to make the wear judgment more consistent with actual operating conditions and improve the accuracy and sensitivity of the judgment.

[0061] In practical applications, the weights of vibration variation characteristics and temperature evolution trends are adjusted by combining the gearbox's expected life parameters and accumulated operating time. For example, when a gearbox is nearing the end of its expected life, even if the vibration or temperature changes are not significant, they should be given higher weight, because even minor anomalies may indicate severe wear at this stage. Conversely, in the early stages of gearbox operation, parameters such as cumulative deviation may be given more emphasis. The aim is to incorporate the gearbox's life cycle factors into the assessment, making wear judgment more proactive.

[0062] Furthermore, early warning levels and response strategies can be adjusted based on production needs. For example, when production tasks have extremely high downtime requirements, even if minor wear is detected, a lower early warning level may be selected, and non-downtime intervention measures such as flexible task planning may be prioritized. Conversely, when production tasks allow for downtime maintenance, a higher early warning level can be set, triggering a more proactive maintenance response. The aim is to balance production efficiency and equipment health, enabling more flexible decision-making.

[0063] The proposed solution introduces robot operating condition information, gearbox expected life parameters, and production requirements. Based on this information, it dynamically adjusts the judgment threshold, feature weights, warning levels, and response strategies. This makes the judgment process for early-stage hidden non-uniform wear of the gearbox no longer static, but adaptively adjustable according to the actual operating environment, equipment health status, and business objectives.

[0064] In some embodiments described above in this application, the presence of early, hidden non-uniform wear in the gearbox is identified by collecting operating parameters of the robot's drive components, vibration physical quantities, and temperature physical quantities of the gearbox, and then combining these with task units, cumulative deviation values, vibration change characteristics, and temperature evolution trends for correlation judgment. However, relying solely on passively collected vibration and temperature physical quantities may make it difficult to accurately and comprehensively capture hidden non-uniform wear inside the gearbox in the early stages. In particular, for microscopic damage such as surface wear and subsurface fatigue, their characteristic signals may be masked by other operating noises, leading to misjudgments or omissions, thereby affecting the timeliness and accuracy of diagnosis.

[0065] In response, this application further proposes a method for controlling parameters during gearbox operation, which includes, in the step of obtaining vibration change characteristics corresponding to a specific gear meshing phase, the following: In each task unit of the robot, during the non-critical motion phases, a small load disturbance is actively applied to the gearbox, and the small load disturbance has different frequencies and amplitudes. The vibration response and ultrasonic emission response of the gearbox to the minute load disturbance are collected simultaneously. The collected vibration response is synchronously averaged, and the amplitude and phase changes of the frequency components of the vibration response are analyzed to extract surface wear characteristics. Transient event counting and energy distribution analysis were performed on the collected ultrasonic acoustic emission response to extract subsurface fatigue characteristics; Based on the surface wear characteristics and the subsurface fatigue characteristics, the damage trend prediction of the gearbox is corrected; Using the corrected damage trend prediction as an auxiliary basis for judgment, and combining the cumulative deviation value, the vibration change characteristics, and the temperature evolution trend, it is determined whether the gearbox has early hidden non-uniform wear.

[0066] Specifically, in each task unit of the robot, during non-critical motion phases, minute load disturbances are actively applied to the gearbox. These minute load disturbances have different frequencies and amplitudes. The "non-critical motion phases" refer to stages in the robot's task execution where the load or motion accuracy requirements are relatively low, such as the unloaded or lightly loaded phase when the robot moves from one workstation to another, or the start or end phases of a task unit. Applying disturbances during these phases avoids significantly impacting the robot's normal operating flow. By actively applying minute load disturbances, subtle damage characteristics that are difficult to detect within the gearbox under normal operating conditions can be stimulated. These disturbances are typically preset excitation signals with known frequencies and amplitudes, such as sweep signals or pulse signals, with the aim of probing the gearbox's dynamic response under controlled conditions.

[0067] Furthermore, the vibration response and ultrasonic emission response of the gearbox to the minor load disturbance are simultaneously acquired. The "vibration response" primarily reflects the overall macroscopic motion and structural resonance characteristics of the gearbox, and is crucial for identifying gear meshing impacts, bearing failures, and other issues. The "ultrasonic emission response" is more sensitive to microscopic damage within materials, such as crack initiation and propagation, surface friction, and wear. Its frequency is typically much higher than the audible range, providing earlier damage signals. Simultaneous acquisition ensures the temporal consistency of the two response data, facilitating subsequent correlation analysis.

[0068] Based on this, the acquired vibration response is synchronously averaged, and the amplitude and phase changes of the frequency components of the vibration response are analyzed to extract surface wear characteristics. Synchronous averaging effectively filters out random noise and enhances periodic signals related to gear meshing, thus revealing more clearly the vibration characteristics caused by gear surface wear, such as the amplitude and phase changes of the meshing frequency and its harmonics. These changes directly reflect the alterations in the gear surface geometry and contact state.

[0069] Simultaneously, transient event counting and energy distribution analysis were performed on the acquired ultrasonic acoustic emission responses to extract subsurface fatigue characteristics. Transient event counting refers to counting the number of acoustic emission signal pulses within a certain time period; a high count rate is usually associated with microcrack propagation, friction, or impact events within the material. Energy distribution analysis can reveal the intensity and duration of acoustic emission events, thereby distinguishing different types of damage sources. These analytical methods can effectively identify the initiation and propagation of subsurface fatigue cracks in gears, providing crucial evidence for early, hidden damage.

[0070] Therefore, based on the surface wear characteristics and the subsurface fatigue characteristics, the damage trend prediction of the gearbox is corrected. By combining macroscopic vibration characteristics and microscopic acoustic emission characteristics, the current health condition and future damage development trend of the gearbox can be assessed more comprehensively and accurately, improving the reliability of the prediction.

[0071] Finally, the revised damage trend prediction is used as an auxiliary judgment criterion, combined with the cumulative deviation value, vibration change characteristics, and temperature evolution trend, to determine whether the gearbox exhibits early, hidden, non-uniform wear. This means that, based on the original judgment based on operating parameter deviations, passive vibration, and temperature trends, more refined and earlier damage characteristics are introduced, forming a multi-source information fusion judgment mechanism, thereby significantly improving the detection capability of early, hidden, non-uniform wear.

[0072] The solution proposed in this application overcomes the limitations of traditional passive monitoring in detecting early, concealed, non-uniform wear by actively applying a small load disturbance to the gearbox during non-critical motion phases and simultaneously acquiring vibration and ultrasonic emission responses.

[0073] Through the above technical solution, this application can significantly improve the detection capability of early-stage, concealed, non-uniform wear in gearboxes. Specifically, by actively applying minute load disturbances, weak damage signals that are difficult to detect with traditional passive monitoring can be effectively excited and captured, thereby achieving accurate identification of early, concealed damage such as surface wear and subsurface fatigue in gearboxes. This simultaneous acquisition and in-depth analysis of multi-source responses (vibration and ultrasonic acoustic emission) makes the extraction of damage features more comprehensive and refined, enabling the differentiation of different types of damage mechanisms. Therefore, the corrected damage trend prediction has higher accuracy and reliability, providing a more solid foundation for subsequent correlation judgments. Compared to methods relying solely on passive monitoring, the solution in this application can detect potential faults earlier, preventing further damage deterioration, thereby effectively reducing the risk of sudden gearbox failures, extending its service life, and providing more timely and accurate decision-making basis for preventative maintenance and flexible task planning.

[0074] In some of the embodiments described above in this application, although the damage trend prediction of the gearbox can be corrected by analyzing surface wear characteristics and subsurface fatigue characteristics, thereby more accurately assessing the health status of the gearbox, if the focus remains solely on predicting the damage trend and the corrected prediction results are not effectively translated into optimization of the robot's operating strategy, it may be impossible to maximize the service life of the gearbox or proactively take measures to avoid potential failure risks in the early stages of damage.

[0075] In this regard, this application further proposes that, after correcting the damage trend prediction of the gearbox based on surface wear characteristics and subsurface fatigue characteristics, it also includes: adaptively adjusting the flexible task planning strategy based on the corrected damage trend prediction.

[0076] Specifically, the "adaptive and flexible task planning strategy" refers to the system's ability to dynamically and intelligently adjust the robot's task execution method based on real-time updated and corrected gearbox damage trend predictions. The "flexible task planning strategy" can be understood as a strategy that flexibly adjusts the robot's operational parameters, such as motion trajectory, operating speed, acceleration, torque limits, task sequence, and cycle time, without affecting core production goals or causing only acceptable impacts. This adjustment aims to mitigate further damage or provide the gearbox with recovery time by changing the load pattern, working time, or environmental conditions applied by the robot to the gearbox.

[0077] This application's solution uses the corrected damage trend prediction results as input to drive adaptive adjustments to the flexible task planning strategy, thereby transforming passive monitoring and prediction of gearbox health status into proactive intervention and optimization. When an accelerated damage trend in the gearbox is predicted, the system can intelligently select and implement the most suitable flexible task planning strategy based on the type, location, and severity of the damage. For example, if it is predicted that a specific gear pair will experience accelerated wear under high-speed, heavy-load conditions, the system can adjust the robot's movement speed or trajectory to reduce the instantaneous load on that gear pair or shorten the duration of high stress. This proactive adjustment mechanism allows the system to dynamically optimize the gearbox's operating conditions based on its actual health status, thereby effectively mitigating damage accumulation and extending component life.

[0078] In some embodiments described above, this application proposes a scheme for adaptively adjusting a flexible task planning strategy based on a modified damage trend prediction. However, in its implementation, focusing solely on the damage mitigation effect on the gearbox may adversely affect other critical components of the robot, such as increasing their load or accelerating their wear, thereby impacting the overall reliability and lifespan of the robot. Therefore, this application further proposes that when adaptively adjusting the flexible task planning strategy, the operating status and damage sensitivity of other critical robot components be comprehensively considered to optimize the entire robot system.

[0079] In this regard, this application further proposes the following steps for adaptively adjusting the flexible task planning strategy based on the modified damage trend prediction: Obtain the current operating status information of other key components of the robot; Based on the current operating status information of the other key components, assess the current health status and damage sensitivity of the other key components; The effects of different flexible task planning strategies on damage mitigation of the gearbox and the potential increase in load on other key components were simulated. Based on the simulation results, a flexible task planning strategy that has the least impact on the load of the other key components is selected, and the robot's operating parameters are adjusted according to the selected flexible task planning strategy.

[0080] Specifically, obtaining the current operating status information of other key robot components means, in addition to the gearbox, collecting real-time data from other key components in the robot system, such as the current, voltage, speed, vibration, temperature, and stress parameters of components like drive motors, bearings, linkage mechanisms, and end effectors. This data is used to comprehensively understand the operating status of each part of the robot. Assessing the current health status and damage sensitivity of these other key components based on their current operating status information can be understood as using a pre-set health assessment model or expert system to evaluate the current wear level, fatigue accumulation, and remaining life of these key components based on the collected operating status information. Simultaneously, it is also necessary to analyze the damage sensitivity of these components under different loads or operating conditions, that is, their responsiveness to changes in specific operating parameters.

[0081] Furthermore, simulating the damage mitigation effect of different flexible task planning strategies on the gearbox and the impact on the potential increase in load on other key components refers to pre-simulating multiple possible flexible task planning strategies in a virtual simulation environment. For each strategy, it is necessary to calculate its degree of improvement on the gearbox damage trend and the amount of additional load, stress, or wear increase that it may cause to other key components of the robot. For example, simulations can be performed using methods such as finite element analysis and multibody dynamics simulation. Based on the simulation results, the flexible task planning strategy with the least impact on the load of other key components is selected, and the robot's operating parameters are adjusted according to the selected flexible task planning strategy. Specifically, considering both the gearbox damage mitigation effect and the impact on the load of other components, priority is given to task planning strategies that effectively mitigate gearbox damage while minimizing the potential negative impact on other key components. Once selected, the corresponding operating parameters (such as speed, acceleration, trajectory, cycle time, etc.) are sent to the robot controller to guide the robot's actual operation.

[0082] This application's solution extends the local optimization of the gearbox to the global optimization of the entire robot system by acquiring the operational status information of other key robot components and evaluating their health status and damage sensitivity. By simulating the effects of different task planning strategies on gearbox damage mitigation and the potential increase in load on other key components, the system can comprehensively weigh the health status of each component during decision-making, avoiding situations where over-protecting one component damages others. Finally, by selecting a flexible task planning strategy that minimizes the load on other key components and adjusting the robot's operating parameters accordingly, the overall balance and reliability of the robot's operation are ensured.

[0083] In some embodiments described above in this application, a flexible task planning strategy is adaptively adjusted based on the modified damage trend prediction, and the robot's operating parameters are adjusted according to the selected flexible task planning strategy. However, in its implementation, how to specifically adjust the robot's operating parameters to effectively achieve damage mitigation and load optimization still requires more detailed guidance. Therefore, this application further proposes that the aforementioned adjustment of the robot's operating parameters based on the selected flexible task planning strategy may include the following methods: Based on the selected flexible task planning strategy, adjust the operating speed, acceleration, or torque limits of specific motion axes of the robot to reduce the instantaneous load or impact on the gearbox. And / or, based on the selected flexible task planning strategy, optimize the robot's motion trajectory or posture to avoid the gearbox being subjected to high stress or resonant frequency for a long time under a specific motion phase; And / or, depending on the selected flexible task planning strategy, adjust the cycle time or task sequence of the robot's task units to increase the gearbox's cooling or recovery time.

[0084] Specifically, adjusting the operating speed, acceleration, or torque limits of specific motion axes of a robot refers to fine-tuning the motion parameters of each joint or axis of the robot according to the instructions of a flexible task planning strategy. For example, when it is predicted that the gearbox may be subjected to excessive load during a certain motion phase, the operating speed or acceleration during that phase can be appropriately reduced, or a stricter torque limit can be set, thereby effectively reducing the instantaneous load or impact on the gearbox and avoiding accelerated wear or fatigue due to overload.

[0085] Optimizing the robot's motion trajectory or posture during task execution can be understood as improving the stress conditions of the gearbox by changing the robot's motion path or the posture of the end effector without affecting task completion. For example, if a certain motion trajectory causes the gearbox to be under high stress or close to the resonant frequency for a long time at a specific meshing phase, the trajectory can be adjusted to avoid these unfavorable conditions, thereby reducing fatigue damage to the gearbox.

[0086] In practical applications, adjusting the cycle time or task sequence of a robot's task units specifically refers to flexibly adjusting the completion time of individual task units or the execution order of tasks throughout the entire production process. For example, when the gearbox temperature is too high or wear is accelerating, the cycle time of the task unit can be appropriately extended, or short idle or low-load operation phases can be inserted into the task sequence to provide the gearbox with additional cooling or recovery time, thereby effectively extending its service life.

[0087] The solution presented in this application optimizes the gearbox's operating state from multiple dimensions through fine-tuning of the robot's operating parameters. By adjusting the operating speed, acceleration, or torque limits, the instantaneous load and impact on the gearbox can be directly controlled, preventing it from bearing excessive stress in a short period and effectively suppressing the occurrence and development of early, hidden, uneven wear. It is precisely due to the precise control of these dynamic parameters that the gearbox can remain within a safer load range during operation.

[0088] Furthermore, by optimizing the robot's motion trajectory or posture, the risk of the gearbox being subjected to high stress or resonant frequencies for extended periods under specific motion phases can be avoided both spatially and temporally. This strategic adjustment can fundamentally reduce external excitations that lead to gearbox fatigue and wear, thereby extending its healthy operating cycle. It is precisely due to the intelligent planning of the motion path that the gearbox can avoid being in adverse stress conditions for extended periods.

[0089] Meanwhile, by adjusting the cycle time or task sequence of the task units, the gearbox is provided with the necessary cooling and recovery time. This flexible adjustment in time helps dissipate the heat accumulated inside the gearbox and allows the material to recover between stress cycles, thereby effectively alleviating thermal stress concentration and fatigue accumulation, and further delaying the wear process. It is precisely because of the flexible control of the task rhythm that the gearbox can get sufficient rest between high-intensity working intervals.

[0090] In some embodiments described above, this application proposes simulating the damage mitigation effects of different flexible task planning strategies on the gearbox and their impact on potential load increases in other critical components to select the optimal strategy. However, in practical applications, simulations based purely on preset models may not accurately reflect the actual operating state and damage evolution process of the robot and its components under complex and variable working conditions. This may lead to deviations between simulation results and actual conditions, thus affecting the optimization effect of the flexible task planning strategy. If the above problems are not addressed, the selected strategy may fail to achieve the expected damage mitigation effect and may even cause unnecessary load increases in other critical components.

[0091] In response, this application further proposes the following steps for simulating the damage mitigation effects of different flexible task planning strategies on the gearbox and the impact on the potential increase in load on other critical components: Acquire real-time operational data of the robot, gearbox, and other key components; By comparing real-time operating data with corresponding parameters in the simulation environment, the deviation between the simulation environment and the actual operating environment can be identified and quantified. Based on the identified and quantified deviations, the component damage evolution parameters in the simulation environment are adjusted. These component damage evolution parameters include the material fatigue coefficient, wear rate constant, and thermal conductivity coefficient. In the adjusted simulation environment, different flexible task planning strategies were implemented, and the damage mitigation effect of the gearbox and the impact on the potential increase in load on other critical components were calculated.

[0092] Specifically, acquiring real-time operational data of the robot, gearbox, and other key components refers to continuously monitoring and collecting various performance indicators and status information of the robot body, gearbox, and other key components such as motors and bearings during actual operation through a sensor network. This data may include, but is not limited to, instantaneous load, rotational speed, vibration signals, temperature, current, and voltage of each component, with the aim of providing realistic and dynamic input for subsequent simulation environment calibration.

[0093] This process involves comparing real-time operational data with corresponding parameters in the simulation environment to identify and quantify the deviations between the simulation and actual operating environments. This can be understood as comparing the actual collected real-time operational data with the corresponding parameters in a pre-established simulation model. For example, the measured gearbox temperature can be compared with the temperature predicted by the simulation model, or the actual vibration spectrum can be compared with the spectrum predicted by the model. Through statistical methods or machine learning algorithms, the differences between the simulation model and actual operation can be identified and quantified into specific deviation values. The purpose is to evaluate the accuracy of the current simulation model and provide a basis for subsequent parameter adjustments.

[0094] In practical applications, based on the identified and quantified deviations, the component damage evolution parameters in the simulation environment are adjusted. These parameters include the material fatigue coefficient, wear rate constant, and thermal conductivity coefficient. Specifically, if real-time data indicates that the gearbox wear rate is higher than the simulation prediction, the wear rate constant in the simulation model can be increased accordingly. If there is a significant difference between the actual temperature and the simulation prediction, the thermal conductivity coefficient can be adjusted. Adjusting the material fatigue coefficient reflects the true accumulation of fatigue damage under actual working conditions. The purpose of dynamically adjusting these parameters is to enable the simulation environment to more accurately reflect the actual damage evolution of the component, improving the realism of the simulation and the accuracy of the prediction.

[0095] Furthermore, in the adjusted simulation environment, different flexible task planning strategies were executed, and the damage mitigation effect on the gearbox and the impact on the potential increase in load on other critical components were calculated. This means that in the simulation model after real-time data calibration and parameter adjustment, a variety of different flexible task planning strategies can be run, such as adjusting the robot's movement speed, trajectory, or task cycle. By simulating these strategies, the degree to which each strategy mitigates the cumulative damage to the gearbox and the potential additional load or stress increase on other critical components (such as motors and bearings) can be quantitatively evaluated. The purpose is to provide a quantitative basis for selecting the optimal flexible task planning strategy.

[0096] This application's solution addresses the problem of prediction bias caused by discrepancies between the model and reality in traditional simulations by introducing real-time operational data for dynamic calibration and parameter adjustment of the simulation environment. By acquiring real-time operational data from the robot, gearbox, and other key components, the dynamic changes and actual responses of the components under real-world conditions can be captured. Comparing this real-time data with corresponding parameters in the simulation environment and identifying and quantifying deviations allows the simulation model to promptly detect and correct these deviations. Based on this, and according to the quantified deviations, parameters related to component damage evolution, such as material fatigue coefficients, wear rate constants, and thermal conductivity coefficients, are adjusted in the simulation environment, ensuring that the simulation model more accurately reflects the actual damage accumulation and thermal behavior of the components. Finally, executing different flexible task planning strategies in the calibrated and adjusted simulation environment yields more realistic damage mitigation effects and load impact assessments, thus avoiding the risk of making decisions based on inaccurate simulation results.

[0097] The above technical solutions significantly improve the accuracy and reliability of the simulated environment, making the evaluation of flexible task planning strategies more realistic and effective. This not only allows for more accurate prediction of gearbox damage mitigation effects but also more precise assessment of potential load impacts on other critical components, thus avoiding suboptimal or erroneous decisions due to inaccurate simulations. Consequently, the selected flexible task planning strategy can more effectively extend the gearbox's service life while ensuring the stability and reliability of the overall robot system, achieving more refined and intelligent control of operational parameters.

[0098] In some embodiments described above, this application proposes to evaluate the damage mitigation effect of the gearbox and its impact on the potential increase in load on other critical components by simulating different flexible task planning strategies, and to select the optimal strategy accordingly. However, in actual operating environments, external environmental factors, such as ambient temperature, humidity, dust, and corrosive gases, may significantly affect the damage evolution process of the gearbox and other critical components. If this external environmental information is not incorporated into the simulation and evaluation process, the actual effect of the selected strategy may deviate from expectations, or even fail to achieve the optimal damage mitigation and load balancing effect, thereby affecting the long-term reliability and lifespan of the robot.

[0099] In this regard, this application further proposes that the steps for simulating the damage mitigation effect of different flexible task planning strategies on the gearbox and the impact on the potential increase in load on other key components also include: Obtain external environmental information about the robot's operating environment; Based on external environmental information, adjust the environmental impact parameters in the damage trend prediction of the gearbox, and adjust the environmental impact parameters in the damage sensitivity of other key components. Based on the adjusted environmental impact parameters, the evaluation results of the flexible mission planning strategy are revised, including the evaluation of the damage mitigation effect on the gearbox and the evaluation of the potential increase in load on other key components. Based on the revised evaluation results, the flexible task planning strategy that best mitigates gearbox damage and has the least impact on the load of other critical components is selected.

[0100] Specifically, acquiring external environmental information about the robot's operating environment refers to obtaining environmental parameters of the robot's working area in real time or periodically through sensors, environmental monitoring systems, or external data interfaces. These parameters include ambient temperature, humidity, concentration of particulate matter in the air, and composition of corrosive gases. This information is crucial for accurately assessing the actual wear, fatigue, and aging processes of components.

[0101] Adjusting the environmental impact parameters in the damage trend prediction of the gearbox, and adjusting the environmental impact parameters in the damage sensitivity of other key components, based on external environmental information, can be understood as using collected external environmental data as input to dynamically correct the correlation coefficients or weights in the damage prediction model. For example, in high-temperature and high-humidity environments, the performance of the gearbox lubricating oil may decrease, increasing the wear rate; in dusty environments, seal wear intensifies, and the risk of foreign objects entering the gearbox increases. Therefore, it is necessary to adjust parameters such as material fatigue coefficient, wear rate constant, and thermal conductivity coefficient according to these environmental factors to better reflect actual working conditions. Simultaneously, the sensitivity of other key components (such as motors, bearings, and connecting rods) to environmental factors should also be considered and adjusted.

[0102] In practical applications, revising the evaluation results of flexible task planning strategies based on adjusted environmental impact parameters means that the calculation results of the damage mitigation effect on the gearbox and the impact on the potential increase in load on other key components when different flexible task planning strategies are executed in a simulated environment will be based on the damage evolution model after the environmental parameters have been adjusted. For example, after considering the accelerated wear effect of high temperature environment on gearbox, the damage mitigation effect of a task planning strategy that was originally considered effective may be re-evaluated as less effective than expected, or more significant adjustments to operating parameters may be required to achieve the same mitigation effect.

[0103] Therefore, based on the revised evaluation results, the flexible task planning strategy that has the best effect on mitigating gearbox damage and has the least impact on the load of other key components is selected. The purpose is to ensure that the selected strategy is not only effective under ideal conditions, but also achieves the best overall performance in the actual variable environment. That is, while effectively extending the gearbox life, the negative impact on the overall performance and life of the robot is minimized.

[0104] This application addresses the limitation of existing technologies that may lead to inaccurate strategy evaluation due to the neglect of environmental factors by incorporating external environmental information of the robot's operating environment into the evaluation process of flexible task planning strategies. Through this technical solution, the application significantly improves the adaptability and robustness of flexible task planning strategies. By acquiring and utilizing external environmental information in real time, the damage trend prediction of the gearbox and the damage sensitivity of key components are dynamically adjusted, making the evaluation results of different task planning strategies more accurate and reliable. This avoids the problem of strategy failure or poor performance due to model discrepancies with reality under harsh or changing environmental conditions. Ultimately, this application can select the flexible task planning strategy that provides the best damage mitigation effect to the gearbox in the actual operating environment and has the least impact on the load of other key components, thereby effectively extending the service life of the gearbox and the entire robot, improving equipment reliability and operating efficiency, and reducing maintenance costs.

[0105] In this regard, refer to Figure 3 , Figure 3 This is a schematic diagram of a gearbox operation process parameter control system provided in an embodiment of the present invention. The system includes: The acquisition module is used to acquire the operating parameters of the robot drive components and the physical quantities of the gearbox, including vibration and temperature physical quantities. The task unit definition module is used to define the complete process of the robot handling one workpiece at a time as a task unit. The deviation accumulation module is used to obtain the deviation between the operating parameters of the robot drive component and the preset standard parameters according to the task unit, and accumulate the deviation to obtain the deviation accumulation value; The vibration feature acquisition module is used to acquire the vibration change features corresponding to a specific gear meshing phase by superimposing and averaging the collected vibration physical quantities of the gearbox and the instantaneous meshing phase of the gear. The temperature trend acquisition module is used to acquire the temperature evolution trend inside the gearbox based on the collected temperature physical quantities of the gearbox. The correlation judgment module is used to correlate the cumulative deviation value with the vibration change characteristics and the temperature evolution trend, determine whether the gearbox has early hidden non-uniform wear based on the correlation result, and output instructions for adjusting the gearbox operation control parameters based on the judgment result.

[0106] This application's system, through modular design, achieves multi-dimensional and in-depth monitoring and intelligent control of the robot gearbox's operating status. The system acquires operating parameters of the robot's drive components and physical quantities of the gearbox via a data acquisition module. A task unit definition module standardizes the workflow, a deviation accumulation module captures long-term minor anomalies, and vibration characteristic acquisition and temperature trend acquisition modules respectively analyze the gearbox's dynamic and thermodynamic characteristics. Finally, a correlation and judgment module fuses and intelligently analyzes this heterogeneous data, enabling accurate identification of early, hidden, non-uniform wear in the gearbox and timely output of adjustment commands. This effectively avoids sudden failures and improves the stability and reliability of the production line.

[0107] The gearbox operation process parameter control system of this application has significant advantages and innovations compared to existing technologies. Traditional systems mainly rely on real-time monitoring of single or a few direct operating parameters, making it difficult to effectively capture the cumulative damage caused by minute changes in operating conditions. For example, existing systems may only monitor whether the motor current or temperature exceeds a fixed threshold, but cannot effectively correlate the minute adjustment deviations accumulated over a long period of time with the slow evolution trend of deeper physical quantities.

[0108] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for controlling parameters during gearbox operation, characterized in that, include: The operating parameters of the robot's drive components and the physical quantities of the gearbox are collected, including vibration and temperature physical quantities. The complete process of a robot handling one workpiece at a time is defined as a task unit. According to the task unit, the deviation between the operating parameters of the robot drive component and the preset standard parameters is obtained, and the deviation is accumulated to obtain the cumulative deviation value; Based on the vibration physical quantities and combined with the instantaneous meshing phase of the gears, vibration change characteristics corresponding to a specific gear meshing phase are obtained through superposition and averaging. In the non-critical motion phases set in each task unit of the robot, a small load disturbance is actively applied to the gearbox, and the small load disturbance has different frequencies and amplitudes. The vibration response and ultrasonic acoustic emission response of the gearbox to the small load disturbance are collected synchronously. The collected vibration response is synchronously averaged, and the amplitude and phase changes of the frequency components of the vibration response are analyzed to extract surface wear characteristics. Transient event counting and energy distribution analysis are performed on the collected ultrasonic acoustic emission response to extract subsurface fatigue characteristics; the damage trend prediction of the gearbox is corrected based on the surface wear characteristics and the subsurface fatigue characteristics; the corrected damage trend prediction is used as an auxiliary judgment basis, combined with the cumulative deviation value, the vibration change characteristics and the temperature evolution trend, to determine whether the gearbox has early hidden non-uniform wear. Based on the aforementioned temperature physical quantity, the temperature evolution trend inside the gearbox is obtained; The cumulative deviation value is correlated with the vibration change characteristics and the temperature evolution trend. Based on the correlation results, it is determined whether the gearbox has early hidden non-uniform wear. Adjust the operating control parameters of the gearbox based on the judgment results.

2. The method for controlling gearbox operating parameters according to claim 1, characterized in that, The step of correlating the cumulative deviation value with the vibration change characteristics and the temperature evolution trend, and determining whether the gearbox has early, hidden, non-uniform wear based on the correlation result, includes: Collect the robot's real-time motion status information; Based on the vibration physical quantities, combined with the instantaneous meshing phase of the gear and the real-time motion state information of the robot, the vibration change characteristics corresponding to the specific gear meshing phase and the structural resonance characteristics corresponding to the robot's body motion phase are obtained through superposition and averaging. Based on the vibration change characteristics and the structural resonance characteristics, combined with the cumulative deviation value and the temperature evolution trend, it is determined whether the gearbox has early hidden non-uniform wear.

3. The method for controlling parameters during gearbox operation according to claim 2, characterized in that, The step of correlating the cumulative deviation value with the vibration change characteristics and the temperature evolution trend, and determining whether the gearbox has early, hidden, non-uniform wear based on the correlation result, further includes: Obtain the current operating status information of the robot, the expected life parameters of the gearbox, and the production requirements; Based on the operating condition information, the judgment thresholds used to determine the cumulative deviation value, the vibration change characteristics, and the temperature evolution trend are dynamically adjusted. By combining the expected life parameters of the gearbox and the accumulated operating time, the weights of the vibration change characteristics and the temperature evolution trend are adjusted during the correlation judgment process. Adjust the early warning level and response strategy according to the production needs described; Based on the adjusted judgment threshold, the weight, and the warning level, the cumulative deviation value, the vibration change characteristics, and the temperature evolution trend are correlated and judged to determine whether the gearbox has early hidden non-uniform wear.

4. The method for controlling gearbox operating parameters according to claim 1, characterized in that, After correcting the damage trend prediction of the gearbox based on the surface wear characteristics and the subsurface fatigue characteristics, the method further includes: Based on the revised damage trend prediction, the flexible task planning strategy is adaptively adjusted.

5. The method for controlling gearbox operating parameters according to claim 4, characterized in that, The step of adaptively adjusting the flexible task planning strategy based on the corrected damage trend prediction includes: Obtain the current operating status information of other key components of the robot; Based on the current operating status information of the other key components, assess the current health status and damage sensitivity of the other key components; The effects of different flexible task planning strategies on damage mitigation of the gearbox and the potential increase in load on other key components were simulated. Based on the simulation results, a flexible task planning strategy that has the least impact on the load of the other key components is selected, and the robot's operating parameters are adjusted according to the selected flexible task planning strategy.

6. The method for controlling parameters during gearbox operation according to claim 5, characterized in that, Adjusting the robot's operating parameters according to the selected flexible task planning strategy includes: Based on the selected flexible task planning strategy, the operating speed, acceleration, or torque limit of a specific motion axis of the robot is adjusted to reduce the instantaneous load or impact on the gearbox. And / or, based on the selected flexible task planning strategy, optimize the motion trajectory or posture of the robot to perform the task, so as to avoid the gearbox being subjected to high stress or resonant frequency for a long time under a specific motion phase; And / or, depending on the selected flexible task planning strategy, adjust the cycle time or task sequence of the robot's task units to increase the cooling or recovery time of the gearbox.

7. The method for controlling parameters during gearbox operation according to claim 5, characterized in that, The simulation of the damage mitigation effect of different flexible task planning strategies on the gearbox and the impact on the potential increase in load on other key components includes: Acquire real-time operating data of the robot, the gearbox, and other key components; The real-time operating data is compared with the corresponding parameters in the simulation environment to identify and quantify the deviation between the simulation environment and the actual operating environment. Based on the identified and quantified deviation, the component damage evolution parameters in the simulation environment are adjusted, including the material fatigue coefficient, wear rate constant, and thermal conductivity coefficient. In the adjusted simulation environment, different flexible task planning strategies are executed, and the damage mitigation effect of the gearbox and the impact on the potential increase in load on other critical components are calculated.

8. The method for controlling parameters during gearbox operation according to claim 5, characterized in that, The steps of simulating the damage mitigation effect of different flexible task planning strategies on the gearbox and the impact on the potential increase in load on other key components also include: Obtain external environmental information about the robot's operating environment; Based on the external environment information, adjust the environmental impact parameters in the damage trend prediction of the gearbox, and adjust the environmental impact parameters in the damage sensitivity of other key components; Based on the adjusted environmental impact parameters, the evaluation results of the flexible task planning strategy are revised, including an evaluation of the damage mitigation effect on the gearbox and an evaluation of the potential increase in load on other critical components. Based on the revised evaluation results, a flexible task planning strategy that best mitigates damage to the gearbox and minimizes the impact on the load of other critical components is selected.

9. A gearbox operating process parameter control system, characterized in that, The system includes: The acquisition module is used to acquire the operating parameters of the robot drive components and the physical quantities of the gearbox, including vibration and temperature physical quantities. The task unit definition module is used to define the complete process of the robot handling one workpiece at a time as a task unit. The deviation accumulation module is used to obtain the deviation between the operating parameters of the robot drive component and the preset standard parameters according to the task unit, and accumulate the deviation to obtain the deviation accumulation value; The vibration feature acquisition module is used to acquire vibration change features corresponding to a specific gear meshing phase by superimposing and averaging the collected vibration physical quantities of the gearbox and combining them with the instantaneous meshing phase of the gears. In the non-critical motion phases set in each task unit of the robot, a small load disturbance with different frequencies and amplitudes is actively applied to the gearbox. The module simultaneously acquires the vibration response and ultrasonic acoustic emission response of the gearbox to the small load disturbance. The acquired vibration response is simultaneously averaged, and the amplitude and phase changes of the frequency components of the vibration response are analyzed to extract surface wear features. The acquired ultrasonic acoustic emission response undergoes transient event counting and energy distribution analysis to extract subsurface fatigue features. Based on the surface wear features and the subsurface fatigue features, the damage trend prediction of the gearbox is corrected. The corrected damage trend prediction is used as an auxiliary judgment criterion, combined with the cumulative deviation value, the vibration change features, and the temperature evolution trend, to determine whether the gearbox experiences early, hidden, non-uniform wear. The temperature trend acquisition module is used to acquire the temperature evolution trend inside the gearbox based on the collected temperature physical quantities of the gearbox. The correlation judgment module is used to correlate the cumulative deviation value with the vibration change characteristics and the temperature evolution trend, determine whether the gearbox has early hidden non-uniform wear based on the correlation result, and output instructions for adjusting the gearbox operation control parameters based on the judgment result.

Images