Aluminum alloy processing equipment and power prediction method thereof

Abstract

The invention relates to the technical field of mechanical power prediction, in particular to aluminum alloy processing equipment and a power prediction method thereof. The method comprises the following steps: acquiring motion data of the industrial robot and performing attitude reconstruction and power parameter estimation to obtain output power parameters of the robot; meanwhile, load gravitational torque data are obtained, joint rotating speed and torque analysis is conducted through the load gravitational torque data, and the actual acting power of the robot is calculated; and comparing the power calculation and the actual power in the ideal environment to obtain an actual difference power value. The root of the gap, including a connection structure and a load influence factor, is positioned through disassembly analysis; based on the key influence factors, a robot power prediction model is established to execute a task of predicting the power of the industrial robot; according to the invention, a more accurate and more reliable power prediction method for the aluminum alloy processing equipment is constructed.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical power prediction, and in particular to aluminum alloy processing equipment and a power prediction method thereof. Background Art

[0002] The development of power prediction methods for aluminum alloy processing equipment can be traced back to the early days of industrial automation. In the initial stage, engineers mainly relied on their own experience and intuitive judgment to roughly estimate the power consumption of the robot during operation. Although this experience-based prediction method is simple and practical, due to the lack of scientific basis, its prediction accuracy is often unsatisfactory and it is difficult to truly guide the energy management and optimization design of the robot. Subsequently, with the in-depth study of robot dynamics and motor control, engineers began to try to establish more rigorous mathematical models to try to predict the power consumption of robots theoretically. These prediction methods based on mathematical models have improved the prediction accuracy to a certain extent, but due to the assumptions and simplifications of the model itself, it is difficult to fully cover the influence of various complex factors under actual working conditions, and there is still a certain deviation between the prediction results and the actual situation. In order to further improve the accuracy of the prediction, researchers began to introduce advanced simulation technology and use robot simulation software to simulate and predict power consumption. This simulation-based method can better reflect various factors under actual working conditions. However, since the simulation model itself has certain deviations from reality and only focuses on the power performance under actual working conditions, it lacks comparative analysis with ideal environments, and focuses more on the analysis of overall power levels, while ignoring the key influencing factors of connection structure and load, the prediction results are difficult to be convincing. Summary of the invention

[0003] Based on this, it is necessary to provide an aluminum alloy processing equipment and a power prediction method thereof to solve at least one of the above technical problems.

[0004] To achieve the above object, a power prediction method for aluminum alloy processing equipment comprises the following steps:

[0005] Step S1: acquiring motion data of the industrial robot; reconstructing the machine posture of the industrial robot motion data to obtain three-dimensional robot posture data; estimating power parameters of the three-dimensional robot posture data based on the motion data of the industrial robot to obtain robot output power parameters;

[0006] Step S2: Obtain load gravity torque data; use the load gravity torque data to perform a load speed test on the robot output power parameter to generate actual joint speed data; perform joint torque analysis on the actual joint speed data and the load gravity torque data to obtain actual joint torque data; perform overall performance calculation on the actual joint speed data and the actual joint torque data to generate the actual power value of the robot;

[0007] Step S3: performing an ideal environment simulation on the output power parameters of the robot to obtain an ideal lossless environment; performing an ideal power calculation on the output power parameters of the robot based on the ideal lossless environment to generate an ideal action power value of the robot;

[0008] Step S4: compare the power difference between the actual power value of the robot and the ideal power value of the robot to obtain the actual power difference value; disassemble and trace the actual power difference value to generate a gap root cause location report;

[0009] Step S5: Based on the gap root location report, the connection impact factor is derived for the actual gap power value to obtain the connection structure power impact factor; based on the gap root location report, the load impact analysis is performed on the actual gap power value to generate the load proportion power impact factor;

[0010] Step S6: Perform power prediction drive modeling on the connection structure power influence factor and the load ratio power influence factor to obtain a robot power prediction model to perform the task of predicting the power of the industrial robot.

[0011] The present invention starts from the steps of obtaining robot motion data, three-dimensional posture reconstruction, and power parameter estimation, comprehensively considers various factors that affect power prediction, and can more accurately reflect the power consumption characteristics under actual working conditions. Not only does it predict the actual power, but it also simulates the ideal power in an ideal lossless environment. By comparing and analyzing the gap between the actual and ideal power, the source of power loss can be better located. On the basis of analyzing the gap between the actual and ideal power, the method further disassembles the specific impact of the connection structure and load factors on power, providing an important basis for problem diagnosis and optimization. A power prediction model based on the above analysis is established, which can effectively predict the power consumption of industrial robots under actual working conditions and provide support for energy management and optimization design. It takes into account comprehensiveness, comparison and depth, and provides a more complete technical framework for accurately predicting robot power consumption, diagnosing the root cause of the problem, and optimizing the design, with stronger practical value and application prospects.

[0012] In the present specification, an aluminum alloy processing equipment is provided, which is used to execute the power prediction method of the aluminum alloy processing equipment as described above, and the aluminum alloy processing equipment comprises:

[0013] The data acquisition module is used to obtain the motion data of the industrial robot; reconstruct the machine posture of the industrial robot motion data to obtain the three-dimensional robot posture data; estimate the power parameters of the three-dimensional robot posture data based on the industrial robot motion data to obtain the robot output power parameters;

[0014] The actual power calculation module is used to obtain the load gravity torque data; use the load gravity torque data to perform load speed test on the robot output power parameters to generate actual joint speed data; perform joint torque analysis on the actual joint speed data and load gravity torque data to obtain actual joint torque data; perform overall performance calculation on the actual joint speed data and actual joint torque data to generate the actual power value of the robot;

[0015] The ideal power calculation module is used to simulate the ideal environment of the robot's output power parameters to obtain an ideal loss-free environment; based on the ideal loss-free environment, the ideal power calculation of the robot's output power parameters is performed to generate the robot's ideal action power value;

[0016] The power gap comparison module is used to compare the power difference between the actual power value of the robot and the ideal power value of the robot to obtain the actual power gap value; disassemble and trace the actual power gap value to generate a gap root cause location report;

[0017] The power impact analysis module is used to derive the connection impact factor of the actual gap power value based on the gap root source location report to obtain the connection structure power impact factor; based on the gap root source location report, the load impact analysis is performed on the actual gap power value to generate the load proportion power impact factor;

[0018] The power prediction module is used to perform power prediction driven modeling on the power influence factor of the connection structure and the power influence factor of the load ratio, and obtain a robot power prediction model to perform the task of predicting the power of the industrial robot.

[0019] The invention is beneficial in that it acquires and reconstructs the three-dimensional robot posture data, obtains the robot output power parameters through power parameter estimation, and calculates the actual joint speed, torque and actual power value. These data provide an important basis for subsequent power prediction. The ideal power value of the robot is calculated in an ideal lossless environment, which lays the foundation for the subsequent comparative analysis with the actual power value. The actual power value and the ideal power value are compared to obtain the actual gap power value, and the root of the gap is located by disassembling and tracing, which provides an important basis for the subsequent analysis of influencing factors. Based on the gap root location report, the influence factors of the connection structure and load on the power are analyzed, laying the foundation for establishing a more accurate power prediction model. Incorporating the influencing factors into the power prediction modeling can build a power prediction model of industrial robots that is closer to reality, and improve the accuracy and reliability of the prediction. It makes full use of the robot motion data, load data and the comparative analysis of the ideal and actual power values, finds out the key factors affecting the power, and integrates them into the final prediction model. It is a relatively systematic and comprehensive method with better practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A schematic diagram of the steps of a power prediction method for aluminum alloy processing equipment;

[0021] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0022] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0023] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0024] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0025] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0026] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0027] To achieve this, please refer to Figures 1 to 3 , a power prediction method for aluminum alloy processing equipment, comprising the following steps:

[0028] Step S1: Acquire the motion data of the industrial robot; reconstruct the machine posture of the industrial robot motion data to obtain three-dimensional robot posture data; estimate the power parameters of the three-dimensional robot posture data based on the industrial robot motion data to obtain the robot output power parameters; for example, the aluminum alloy processing equipment can be an aluminum alloy processing robot or an industrial robot for aluminum alloy processing.

[0029] Step S2: Obtain load gravity torque data; use the load gravity torque data to perform a load speed test on the robot output power parameter to generate actual joint speed data; perform joint torque analysis on the actual joint speed data and the load gravity torque data to obtain actual joint torque data; perform overall performance calculation on the actual joint speed data and the actual joint torque data to generate the actual power value of the robot;

[0030] Step S3: performing an ideal environment simulation on the output power parameters of the robot to obtain an ideal lossless environment; performing an ideal power calculation on the output power parameters of the robot based on the ideal lossless environment to generate an ideal action power value of the robot;

[0031] Step S4: compare the power difference between the actual power value of the robot and the ideal power value of the robot to obtain the actual power difference value; disassemble and trace the actual power difference value to generate a gap root cause location report;

[0032] Step S5: Based on the gap root location report, the connection impact factor is derived for the actual gap power value to obtain the connection structure power impact factor; based on the gap root location report, the load impact analysis is performed on the actual gap power value to generate the load proportion power impact factor;

[0033] Step S6: Perform power prediction drive modeling on the connection structure power influence factor and the load ratio power influence factor to obtain a robot power prediction model to perform the task of predicting the power of the industrial robot.

[0034] Starting from the steps of obtaining robot motion data, three-dimensional posture reconstruction, and power parameter estimation, the present invention comprehensively considers various factors that affect power prediction, and can more accurately reflect the power consumption characteristics under actual working conditions. Not only does it predict the actual power, but it also simulates the ideal power in an ideal lossless environment. By comparing and analyzing the gap between the actual and ideal power, the source of power loss can be better located. On the basis of analyzing the gap between the actual and ideal power, the method further disassembles the specific impact of the connection structure and load factors on power, providing an important basis for problem diagnosis and optimization. A power prediction model based on the above analysis is established, which can effectively predict the power consumption of industrial robots under actual working conditions and provide support for energy management and optimization design. It takes into account comprehensiveness, comparison and depth, and provides a more complete technical framework for accurately predicting robot power consumption, diagnosing the root cause of the problem, and optimizing the design, with stronger practical value and application prospects.

[0035] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of the power prediction method for aluminum alloy processing equipment of the present invention. In this example, the power prediction method for aluminum alloy processing equipment includes the following steps:

[0036] Step S1: acquiring motion data of the industrial robot; reconstructing the machine posture of the industrial robot motion data to obtain three-dimensional robot posture data; estimating power parameters of the three-dimensional robot posture data based on the motion data of the industrial robot to obtain robot output power parameters;

[0037] In an embodiment of the present invention, during the actual working process of the robot, a high-precision sensor system (such as an encoder, an accelerometer) is used to collect the position, velocity, and acceleration motion data of each joint of the robot in real time. The sampling frequency of the collected data is not less than 1kHz to ensure that the detailed changes in the rapid movement of the robot are captured. Based on the collected joint motion data, the actual position and posture of each component of the robot in three-dimensional space are obtained by forward kinematics calculation through the rigid body dynamics model and geometric relationship. This process involves the precise measurement and modeling of the robot link length and mass parameters. The three-dimensional robot posture data is substituted into the robot dynamics model, and the least squares method or extended Kalman filter parameter identification technology is used to estimate the driving torque and power parameters of each joint of the robot. This process requires the robot's connecting rod parameters and mass distribution to be determined in advance.

[0038] Step S2: Obtain load gravity torque data; use the load gravity torque data to perform a load speed test on the robot output power parameter to generate actual joint speed data; perform joint torque analysis on the actual joint speed data and the load gravity torque data to obtain actual joint torque data; perform overall performance calculation on the actual joint speed data and the actual joint torque data to generate the actual power value of the robot;

[0039] In an embodiment of the present invention, when the robot is working, a force / torque sensor is used to measure and record the external load gravity torque borne by each joint of the robot. The range and accuracy of the sensor should be able to cover the entire load range required by the process, and the sampling frequency should not be less than 1kHz to ensure that the detailed characteristics of the load change are captured. The robot power parameters (such as joint drive torque) and the load gravity torque data are input into the robot dynamics model, and the actual speed response curve of each joint under a given load is obtained by simulation calculation. This process requires the transmission coefficient and load additional inertia parameters of the robot to be calibrated in advance. Combined with the load gravity torque data and the joint speed data, the inverse dynamics formula is used to calculate the actual output torque of each joint of the robot under actual working conditions. This step also needs to consider the coupling effect between joints. Multiply the joint speed and torque data to obtain the actual output power of each joint of the robot. By summing up the power of each joint, the overall power consumption of the whole machine under actual working conditions can be obtained.

[0040] Step S3: performing an ideal environment simulation on the output power parameters of the robot to obtain an ideal lossless environment; performing an ideal power calculation on the output power parameters of the robot based on the ideal lossless environment to generate an ideal action power value of the robot;

[0041] In an embodiment of the present invention, an ideal lossless robot dynamics model is constructed using robot dynamics parameters (such as joint drive torque and inertia). The model assumes that there is no friction or damping loss between the components of the robot, the transmission system efficiency is 100%, and the influence of external loads is ignored. By establishing such an ideal model, the influence of various loss factors on the robot power under actual working conditions can be eliminated. The robot dynamics parameters are input into the above-mentioned ideal dynamics model, and the speed and torque curves of each joint under ideal conditions are obtained by forward dynamics calculation. Then the joint speed and torque are multiplied to obtain the output power of each joint of the robot under ideal conditions. By summing up the power of each joint, the overall power consumption of the whole machine under ideal conditions can be obtained.

[0042] Step S4: compare the power difference between the actual power value of the robot and the ideal power value of the robot to obtain the actual power difference value; disassemble and trace the actual power difference value to generate a gap root cause location report;

[0043] In an embodiment of the present invention, the actual power value of the robot is compared with the ideal power value of the robot. The specific operation is to compare the actual power and the ideal power of each joint, and calculate the absolute difference and relative difference between the two. This step can intuitively reflect the degree of influence of various loss factors on the overall power performance of the robot under actual working conditions. With respect to the power difference, the root cause of its generation is further analyzed. First, combined with the joint speed and torque data, the contribution of transmission efficiency, mechanical friction, and electromagnetic loss factors to power loss is quantitatively evaluated. Secondly, combined with the mechanism parameters and control parameters, the influence of structural stiffness, control accuracy, and load change factors on power loss is analyzed. Finally, the above analysis results are combined to form a complete gap root cause positioning report.

[0044] Step S5: Based on the gap root location report, the connection impact factor is derived for the actual gap power value to obtain the connection structure power impact factor; based on the gap root location report, the load impact analysis is performed on the actual gap power value to generate the load proportion power impact factor;

[0045] In an embodiment of the present invention, the contribution of the mechanical connection structure (such as spherical bearings, gear pairs, and couplings) to the power loss is analyzed in combination with the gap root location report. First, for each main connection component, its friction power loss is calculated through a theoretical model or measured data to obtain the power influence factor of each connection component. Then, these power influence factors are weighted and summed to obtain the comprehensive influence factor of the entire connection structure on the total power loss. This factor reflects the degree of influence of the loss of the connection structure itself on the actual power performance of the robot. According to the gap root location report, the impact of external load changes on the power performance of the robot is analyzed. Specifically, for different workloads, the corresponding joint torque and power consumption are calculated, and compared with the power under ideal no-load conditions to obtain the power influence factors corresponding to different load ratios. This factor can reflect the degree of influence of external load changes on the actual power performance of the robot.

[0046] Step S6: Perform power prediction drive modeling on the connection structure power influence factor and the load ratio power influence factor to obtain a robot power prediction model to perform the task of predicting the power of the industrial robot.

[0047] In an embodiment of the present invention, a robot power prediction model is constructed based on the connection structure power influence factor and the load ratio power influence factor. Specifically, these two factors are first used as the input variables of the model, and the ideal power of the robot is used as the output target of the model. Then, a prediction model is established by using multiple linear regression or artificial intelligence algorithm (such as neural network, support vector machine) to fit the mapping relationship between input and output. A large amount of measured data is required for parameter optimization during model training to ensure the accuracy and robustness of model prediction. After the preliminary prediction model is established, the model needs to be verified using independent test data. The prediction accuracy of the model is evaluated by calculating the error index (such as root mean square error, relative error) between the predicted value and the actual value. If the prediction accuracy cannot meet the requirements, it is necessary to further optimize the model structure and adjust the model parameters until the prediction performance reaches the expected target. After completing the model training and verification, the prediction model can be used to perform the industrial robot power prediction task. In practical applications, only the connection structure parameters and load information of the robot are input into the model to quickly obtain the predicted power value of the robot under the working condition. This function is of great significance for robot energy management and power optimization application scenarios.

[0048] Preferably, step S1 comprises the following steps:

[0049] Step S11: acquiring the working data of the industrial robot, and performing continuous motion extraction on the working data of the industrial robot to obtain the motion data of the industrial robot;

[0050] Step S12: tracking feature points of the industrial robot motion data to generate motion feature point data;

[0051] Step S13: reconstructing the robot posture based on the motion feature point data to obtain three-dimensional robot posture data;

[0052] Step S14: simulating the motion trajectory of the three-dimensional robot posture data based on the industrial robot motion data, and outputting the robot motion trajectory data;

[0053] Step S15: Estimating the power parameters of the three-dimensional robot posture data according to the robot motion trajectory data to obtain the robot output power parameters.

[0054] The invention is useful in that it obtains the working data of the industrial robot, and generates the motion data and feature point data of the industrial robot by continuous motion extraction and motion feature point tracking. These data lay the foundation for the subsequent machine posture reconstruction. The robot is reconstructed in three dimensions using the motion feature point data to obtain the three-dimensional robot posture data. This provides the necessary input for the subsequent motion trajectory imitation and power parameter estimation. Based on the three-dimensional robot posture data, the robot motion trajectory data is generated by simulating the trajectory of the industrial robot motion data. These data help to estimate the power parameters of the robot more accurately. Using the robot motion trajectory data, the power parameters of the three-dimensional robot posture data are estimated to obtain the output power parameters of the robot. These power parameters provide the key input basis for the subsequent power prediction modeling. It is beneficial to start from the actual working data of the industrial robot, and finally obtain the power parameters of the robot through feature extraction, posture reconstruction and trajectory imitation. These data lay a solid foundation for the subsequent power prediction modeling and help to improve the accuracy and reliability of the prediction model.

[0055] In an embodiment of the present invention, various data of the robot in the actual working process, including joint angle, joint speed, and joint torque, are obtained from the robot control system or other sensing devices. The obtained raw data is subjected to signal processing and segmented analysis to extract a continuous robot motion sequence, i.e., industrial robot motion data. This step ensures that the subsequent analysis is based on the actual robot running trajectory. Computer vision or other technologies are used to automatically track the key feature points (such as joint center and end effector) in the obtained robot motion data. Through the time series changes of the feature points, feature point data describing the robot motion can be generated to provide basic data for subsequent three-dimensional posture reconstruction. The three-dimensional motion posture of the robot at each moment is reconstructed based on the feature point data using three-dimensional reconstruction technology in computer vision, such as structured light scanning and multi-view stitching. The robot motion data is associated with the three-dimensional posture data, and three-dimensional robot motion trajectory data that is highly consistent with the actual motion trajectory is generated through data-driven methods, such as motion capture and trajectory planning. The output torque and power dynamics parameters of each joint of the robot are estimated by combining the robot dynamics theory with the three-dimensional motion trajectory data.

[0056] Preferably, step S2 comprises the following steps:

[0057] Step S21: acquiring load cargo data; performing gravity load analysis on the load cargo data to generate load gravity moment data;

[0058] Step S22: using the load gravity torque data to perform a dynamic load test on the robot output power parameters to obtain a torque-speed characteristic curve;

[0059] Step S23: Calculating the speed of the robot output power parameter based on the torque-speed characteristic curve to generate actual joint speed data;

[0060] Step S24: performing momentum change rate identification on the actual joint speed data to obtain the momentum change rate;

[0061] Step S25: performing momentum-torque analysis on the load gravity moment data and momentum change rate to obtain actual joint torque data;

[0062] Step S26: Perform overall performance calculation on the actual joint speed data and the actual joint torque data to generate the actual power value of the robot.

[0063] The invention is useful in that it obtains the load cargo data, performs gravity load analysis on it, and generates the load gravity moment data. These data provide necessary input for the subsequent dynamic load test. Using the load gravity moment data, the robot output power parameter is subjected to dynamic load test, and the torque-speed characteristic curve is obtained. These data reflect the power output characteristics of the robot under actual load conditions. Based on the torque-speed characteristic curve, the robot output power parameter is subjected to speed calculation, and the actual joint speed data is generated. These data provide a basis for the subsequent momentum change rate identification. The momentum change rate is identified for the actual joint speed data, and momentum-torque analysis is performed in combination with the load gravity moment data, and the actual joint torque data is finally obtained. These data lay the foundation for the overall performance calculation. Using the actual joint speed data and the actual joint torque data, the actual power value of the robot is calculated. This result is the key output of the entire power prediction method, and provides the necessary reference basis for the subsequent power prediction modeling. Combined with the actual load situation, the actual power value of the robot is finally obtained through dynamic load testing, momentum change rate identification and momentum-torque analysis. These data are critical to improving the accuracy and reliability of power prediction models and help better guide industrial production practices.

[0064] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0065] Step S21: acquiring load cargo data; performing gravity load analysis on the load cargo data to generate load gravity moment data;

[0066] In an embodiment of the present invention, sensing equipment (such as force sensors and visual systems) is used to perform real-time measurement and data collection on the cargo carried by the robot when it is working. Specifically, the mass, geometric dimensions, and center of gravity position parameter information of the cargo are obtained to form the load cargo data. These data describe the external load carried by the robot during actual work. Based on the acquired load cargo data, combined with the robot kinematics and dynamics model, the gravitational torque generated by the load is calculated. First, a coordinate transformation relationship between the robot and the load is established to determine the position and posture of the load relative to each joint of the robot. Then, the Newton-Euler equation is used to calculate the gravitational torque generated by the load at each joint based on the mass and position information of the load.

[0067] Step S22: using the load gravity torque data to perform a dynamic load test on the robot output power parameters to obtain a torque-speed characteristic curve;

[0068] In an embodiment of the present invention, the robot output power parameters, such as joint drive torque, power and load gravity torque data are combined to design a series of dynamic load conditions to simulate the situation where the robot is subjected to different external loads in actual operation. The robot is controlled to operate at different joint angular velocities through a console, and the actual output torque of each joint is measured. The joint drive torque and angular velocity data obtained from the test are sorted and analyzed, and the torque-speed characteristic curve of each joint is drawn to describe the operating characteristics of the robot under the action of dynamic loads, reflecting the performance limit of the joint driver. The curve can be used for subsequent energy consumption analysis and optimization, providing a key reference basis. It can also be used for the design and debugging of robot control systems.

[0069] Step S23: Calculating the speed of the robot output power parameter based on the torque-speed characteristic curve to generate actual joint speed data;

[0070] In an embodiment of the present invention, the torque-speed characteristic curve of the joint driver is used in combination with the load gravity torque data to determine the actual speed of each joint of the robot under the actual working load conditions through calculation. Specifically, for a given joint load torque, the corresponding characteristic curve is searched to obtain the actual speed of the joint. This process requires the use of numerical calculation methods, such as interpolation or curve fitting techniques, to ensure the accuracy of the speed calculation and consider the influence of dynamic loads, rather than simply using the rated speed parameters under no-load conditions. The actual speed data of each joint is sorted and output to form a complete actual joint speed data set, which reflects the actual operating status of each joint actuator during the actual working process of the robot and is the key input for subsequent energy consumption analysis and optimization.

[0071] Step S24: performing momentum change rate identification on the actual joint speed data to obtain the momentum change rate;

[0072] In the embodiment of the present invention, the angular momentum of each joint is calculated using the actual joint speed data. Angular momentum is an important physical quantity of the rotational motion of an object, which can reflect the momentum characteristics of the joint motion. The momentum change rate of each joint can be obtained by performing time derivation on these angular momentum data. The momentum change rate represents the speed of change of momentum during the joint motion process, and is a key indicator for evaluating the joint motion characteristics. The digital differential mathematical operation method is used to convert the discrete speed data into a continuous momentum change rate signal. At the same time, it is also necessary to design an appropriate filtering algorithm to eliminate noise interference in the data and ensure the accuracy of the momentum change rate calculation. The calculated momentum change rate data of each joint are sorted and output to form a complete momentum change rate data set, which describes the dynamic characteristics of the momentum change of the robot joint in the actual working process, and is used to analyze the stability and energy transfer efficiency of the joint motion, and provides an important basis for subsequent energy consumption analysis and optimization.

[0073] Step S25: performing momentum-torque analysis on the load gravity moment data and momentum change rate to obtain actual joint torque data;

[0074] In an embodiment of the present invention, the load gravity moment data and momentum change rate data are used, and the inverse solution method of the motion equation is adopted to calculate the actual torque of each joint. Specifically, a joint motion model based on the Newton-Euler equation is established, and the load gravity moment and momentum change rate are used as known quantities to reversely solve the actual torque of each joint. The rigid body dynamics theory is applied, and numerical calculation methods such as iteration method and matrix inversion technology are adopted to ensure the accuracy and stability of torque calculation. It should be noted that this torque calculation must take into account the coupling effect between joints. The actual torque data of each joint are sorted and output to form a complete actual joint torque data set, which reflects the actual torque load borne by each joint actuator during the actual working process of the robot, and is the key input for subsequent energy consumption analysis and optimization.

[0075] Step S26: Perform overall performance calculation on the actual joint speed data and the actual joint torque data to generate the actual power value of the robot.

[0076] In an embodiment of the present invention, the actual joint speed data is used to perform an independent kinematic characteristic analysis on the speed of each joint. Specifically, it is necessary to use a forward kinematic model to calculate the independent angular velocity of each joint according to the joint speed data. Consider the coupling relationship between the joints, and use a matrix operation numerical calculation method. Use the actual joint torque data to perform an independent dynamic characteristic analysis on the torque of each joint, and use the inverse dynamics model to calculate the independent torque of each joint according to the joint torque data. This requires considering the coupling relationship between the joints and using a matrix operation numerical calculation method. The obtained independent joint torque data can more accurately reflect the actual load of each joint and provide reliable input for subsequent power calculations. The independent joint angular velocity data and the independent joint torque data are combined and calculated according to the power calculation formula P=τ.ω. The total power consumption of the entire robot system is integrated and calculated using the three-dimensional robot posture data combined with the actual power value of the independent joints. Considering the coupling relationship between the joints and the three-dimensional motion characteristics of the robot as a whole, a matrix operation numerical method is used for comprehensive analysis.

[0077] Preferably, step S26 comprises the following steps:

[0078] Step S261: performing kinematic feature analysis on the actual joint speed data to obtain independent joint angular velocity data;

[0079] Step S262: Calculate the dynamic characteristics of the actual joint torque data to generate independent joint torque data;

[0080] Step S263: performing combined calculation on the independent joint angular velocity data and the independent joint torque data to obtain the independent joint actual power value;

[0081] Step S264: Integrate the actual power values ​​of the independent joints based on the three-dimensional robot posture data to generate the actual effective power value of the robot.

[0082] The invention is beneficial in that the kinematic characteristics of the actual joint speed data are analyzed to obtain the independent joint angular velocity data. This provides the necessary input for the subsequent dynamic characteristics calculation. The dynamic characteristics of the actual joint torque data are calculated to generate the independent joint torque data. These data and the angular velocity data obtained previously constitute the actual power calculation basis of the independent joint. The independent joint angular velocity data and the independent joint torque data are combined and calculated to obtain the actual power value of the independent joint. This lays the foundation for the final calculation of the actual power value of the robot. Using the three-dimensional robot posture data obtained previously, the actual power value of the independent joint is integrated to generate the actual power value of the robot. This result is the final output of the entire power prediction method. The actual joint data is analyzed and calculated more carefully from the two dimensions of kinematics and dynamics to obtain the actual power value of the independent joint. This not only enhances the accuracy of the calculation, but also provides more abundant input data for the subsequent power prediction modeling. The actual power values ​​of the independent joints are integrated using the three-dimensional robot posture data to generate the actual power value of the robot as a whole. This result provides a reliable reference for power prediction, which helps to improve the accuracy and applicability of the prediction model. It has made an important contribution to the improvement of power prediction methods for aluminum alloy processing equipment, which is conducive to better guiding actual production applications.

[0083] In an embodiment of the present invention, the actual joint speed data is used to perform an independent kinematic characteristic analysis on the speed of each joint. Specifically, it is necessary to use a forward kinematic model to calculate the independent angular velocity of each joint according to the joint speed data. Consider the coupling relationship between the joints, and use a matrix operation numerical calculation method. Use the actual joint torque data to perform an independent dynamic characteristic analysis on the torque of each joint, and use the inverse dynamics model to calculate the independent torque of each joint according to the joint torque data. This requires considering the coupling relationship between the joints and using a matrix operation numerical calculation method. The obtained independent joint torque data can more accurately reflect the actual load of each joint and provide reliable input for subsequent power calculations. The independent joint angular velocity data and the independent joint torque data are combined and calculated according to the power calculation formula P=τ.ω. The total power consumption of the entire robot system is integrated and calculated using the three-dimensional robot posture data combined with the actual power value of the independent joints. Considering the coupling relationship between the joints and the three-dimensional motion characteristics of the robot as a whole, a matrix operation numerical method is used for comprehensive analysis.

[0084] Preferably, step S3 comprises the following steps:

[0085] Step S31: performing power parameter simulation on the robot output power parameters to obtain parameter simulation results;

[0086] Step S32: performing loss compensation environment simulation on the parameter simulation results to obtain an ideal lossless environment;

[0087] Step S33: constructing an ideal dynamics model for the robot output power parameters based on an ideal lossless environment to generate an ideal dynamics model;

[0088] Step S34: Utilize the ideal dynamics model to perform ideal power calculation on the robot output power parameters, and output the ideal action power value of the robot.

[0089] The benefit of the present invention is that the power parameter simulation of the robot output power parameter is performed to obtain the parameter simulation result. This provides basic data for the subsequent construction of the ideal environment, and the loss compensation environment is simulated for the parameter simulation result to obtain the ideal lossless environment. This step realizes the transition from the actual environment to the ideal environment. Based on the ideal lossless environment, the ideal dynamic model of the robot output power parameter is constructed to generate the ideal dynamic model. This provides the necessary theoretical basis for the ideal power calculation. Using the ideal dynamic model constructed above, the robot output power parameter is calculated to calculate the ideal power, and the ideal power value of the robot is output. This result is another key output of the entire power prediction method. Through the power parameter simulation and the ideal environment construction, the transition from the actual environment to the ideal environment is realized, which lays the foundation for the subsequent construction of the ideal dynamic model. Based on the ideal environment, the ideal dynamic model of the robot is constructed, which provides theoretical support for the ideal power calculation, helps to more accurately evaluate the power output capacity of the robot under ideal conditions, and outputs the ideal power value of the robot, which provides an important basis for the comparative analysis of the actual power and the ideal power, and helps to more comprehensively predict and evaluate the power performance of the robot. By gradually transitioning the actual environment to the ideal environment, an ideal dynamics model was constructed, and ultimately the ideal power value of the robot was obtained, making an important contribution to the improvement of the power prediction method for aluminum alloy processing equipment.

[0090] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0091] Step S31: performing power parameter simulation on the robot output power parameters to obtain parameter simulation results;

[0092] In an embodiment of the present invention, the desired output power parameters are used as input, and the power output characteristics of the robot are simulated and calculated by establishing a robot dynamics simulation model. Specifically, it is necessary to establish a multi-rigid body dynamics model of the robot, and according to the Newton-Euler equation or the Lagrange equation, the desired position, velocity, acceleration, and torque parameters are brought into the model to calculate the corresponding dynamic response. Consider the coupling effect between the joints of the robot, as well as the dynamic characteristics of the transmission system and actuator components. Commercial robot simulation software can be used, or a dynamics simulation program based on a physical engine can be developed by yourself. After this step of dynamic simulation, the actual response characteristics of the robot under the action of the desired output power parameters can be obtained.

[0093] Step S32: performing loss compensation environment simulation on the parameter simulation results to obtain an ideal lossless environment;

[0094] In an embodiment of the present invention, the dynamic simulation results are used as input, and a series of loss compensation algorithms are used to correct the simulated data in a virtual environment to obtain an ideal lossless dynamic response, and corresponding loss models are established for the transmission system, actuator, and sensor links of each joint of the robot. These loss models include loss factors at multiple levels of mechanical, electrical, and control, such as friction, gear clearance, motor efficiency, and noise. According to these loss models, an inverse compensation algorithm is used to correct and compensate the dynamic simulation results, eliminate the influence of various loss factors, and obtain a dynamic response in an ideal lossless environment. Various loss factors are modeled in detail, and numerical optimization methods are used for compensation calculations. Artificial intelligence algorithms, such as neural networks, can be used to improve the accuracy and robustness of compensation.

[0095] Step S33: constructing an ideal dynamics model for the robot output power parameters based on an ideal lossless environment to generate an ideal dynamics model;

[0096] In an embodiment of the present invention, the ideal dynamic model of the robot is constructed using the dynamic response data in an ideal lossless environment. Specifically, it is necessary to establish the ideal dynamic equation of the robot multi-rigid body system, and model and express the position, velocity, acceleration, and torque parameters of each joint according to the Newton-Euler equation or the Lagrange equation. In this process, it is necessary to determine the mass and moment of inertia parameters of each connecting rod of the robot, define the kinematic constraints of each joint, and establish a complete set of ideal dynamic equations. In order to simplify the model, some minor nonlinear factors, such as gear clearance and elastic deformation, can be reasonably ignored, and only the key dynamic characteristics are retained. By performing parameter identification through the compensated ideal response data, the various parameters of the ideal dynamic model can be accurately estimated.

[0097] Step S34: Utilize the ideal dynamics model to perform ideal power calculation on the robot output power parameters, and output the ideal action power value of the robot.

[0098] In an embodiment of the present invention, based on the ideal dynamics model, the ideal power value output by each joint of the robot is calculated. Specifically, it is necessary to substitute the angle, angular velocity, angular acceleration and joint torque parameters of each joint into the power calculation formula according to the ideal dynamics equation to obtain the instantaneous ideal power of each joint. The power calculation formula is: P = τ.ω, where P is power, τ is joint torque, and ω is joint angular velocity. For each joint, its ideal power is calculated separately, and the ideal power values ​​of all joints are superimposed to obtain the total ideal action power of the entire robot system. This total ideal power value reflects the ideal power output level required for the robot to perform a certain action in an ideal lossless environment. In order to ensure the accuracy of the calculation, according to the complete ideal dynamics model, the coupling effect of each joint and the influence of various nonlinear factors are considered.

[0099] Preferably, step S4 comprises the following steps:

[0100] Step S41: performing semantic synchronization processing on the actual power value of the robot and the ideal power value of the robot to obtain a synchronized actual power value and a synchronized ideal power value;

[0101] Step S42: Compare the actual synchronization power value with the ideal synchronization power value to obtain the actual difference power value;

[0102] Step S43: performing gap correlation mining on the actual gap power value based on the robot output power parameter to generate a gap motivation vector;

[0103] Step S44: locate the root cause of the gap on the gap driver vector and obtain a gap root cause location report.

[0104] The benefit of the present invention lies in that the actual power value of the robot and the ideal power value of the robot are semantically synchronized to obtain the synchronized actual power value and the synchronized ideal power value. This step ensures the correspondence and comparability of the data of the two. The power difference between the synchronized actual power value and the synchronized ideal power value is compared to obtain the actual gap power value. This provides a quantitative basis for the subsequent location of the root cause of the gap. The gap correlation of the actual gap power value is mined based on the output power parameters of the robot to generate a gap driver vector. This step identifies the potential causes of the gap between the actual power and the ideal power. The gap root cause is located on the gap driver vector to obtain a gap root cause location report. This result provides an important basis for solving problems and optimizing robot performance. Through semantic synchronization processing, the comparability of actual power values ​​and ideal power values ​​is ensured, laying the foundation for subsequent gap analysis. The difference comparison of actual and ideal power values ​​quantitatively reflects the gap between reality and ideal, providing a quantitative basis for problem diagnosis. The output power parameters are used to explore the potential causes of the gap, generate gap dynamic vectors, and lay the foundation for gap root location. Finally, a gap root location report is obtained, which is of great reference value for targeted problem solving and optimization of robot performance. Through in-depth analysis of actual and ideal power values, the root cause of the gap is identified, making an important contribution to the improvement and application optimization of power prediction methods for aluminum alloy processing equipment.

[0105] In an embodiment of the present invention, the actual power and ideal power data are time-synchronized and aligned to ensure that the time series between the data are consistent. Through interpolation or resampling methods, the two sets of data are mapped to the same time series, so that there are corresponding actual power and ideal power values ​​at each time point, ensuring that the subsequent power difference comparison and gap analysis are based on the same time reference, avoiding analysis deviations caused by time dislocation. The synchronized actual power value and the synchronized ideal power value are subtracted one by one to obtain the power difference at each time point. These power difference values ​​reflect the gap between the power output of the robot under the actual operating state and the ideal state. The power difference values ​​at all time points are summarized to obtain a function curve describing the actual gap, that is, the actual gap power value. This actual gap power value provides a key measurement basis for the subsequent gap root cause analysis. Using the actual kinematic parameters of each joint of the robot, such as position, velocity, acceleration, and actual power value, the actual gap power value is analyzed for correlation with the kinematic parameters of each joint and the actual power value, and the dynamic parameters with significant correlation with the actual gap power value are excavated to form a gap dynamic vector, which reflects the key dynamic factors that cause the gap between the actual power and the ideal power. Based on the gap dynamic vector, combined with the overall structure and working environment information of the robot, the root cause of the power gap is analyzed to find out the key fault points or failure causes that cause the power gap, such as wear of mechanical transmission components, sensor measurement errors, and incompatibility of control algorithms. The root causes of the gap between the actual power and the ideal power are systematically summarized to form a detailed gap root cause location report.

[0106] Preferably, step S5 comprises the following steps:

[0107] Step S51: extracting connection structure features from the three-dimensional robot posture data to generate connection structure feature data;

[0108] Step S52: performing contact surface friction power consumption analysis on the connection structure characteristic data to obtain connection point power consumption data;

[0109] Step S53: Calculate the power loss ratio of the connection point power consumption data and the robot output power parameter to obtain the connection structure power influence factor;

[0110] Step S54: performing load bearing capacity deduction on the connection structure characteristic data according to the gap root location report to obtain the maximum allowable dynamic load data;

[0111] Step S55: Perform load impact analysis on the actual difference power value based on the maximum allowable load data to obtain a load proportional power impact factor.

[0112] The benefit of the present invention lies in that the connection structure features are extracted from the three-dimensional robot posture data to generate connection structure feature data. This lays the foundation for subsequent power analysis. The connection structure feature data is subjected to contact surface friction power consumption analysis to obtain the connection point power consumption data. This reflects the actual impact of the connection structure on power. The power loss ratio of the connection point power consumption data and the robot output power parameters is calculated to obtain the connection structure power impact factor. This provides a basis for quantitatively evaluating the impact of the connection structure on power. The load bearing capacity of the connection structure feature data is deduced based on the gap root cause location report to obtain the maximum allowable dynamic load data. This provides a reference for subsequent load impact analysis. Based on the maximum allowable load data, a load impact analysis is performed on the actual gap power value to obtain the load proportional power impact factor. This helps to quantify the impact of the load on the power gap. Through the extraction and analysis of connection structure features, the actual impact of the connection structure on power is deeply explored, which lays the foundation for the analysis of the causes of power loss, and quantitatively evaluates the power influencing factor of the connection structure, providing a reference for optimizing the design of the connection structure. Combined with the gap root cause location report, the load-bearing capacity of the connection structure is evaluated, which provides a basis for analyzing the impact of the load on the power gap, and the load proportional power impact factor is quantified, which helps to more comprehensively diagnose and solve the causes of the power gap. Through in-depth analysis of the impact of the connection structure on power, it provides important support for more accurately predicting and optimizing the power performance of industrial robots.

[0113] In the embodiment of the present invention, the load range is divided into a plurality of discrete scale levels by using the maximum allowable dynamic load data, each scale level represents the allowable working range of the connection structure under the load level, such as 0-20%, 20-40%, and these scale levels and corresponding load values ​​are compiled into an allowable load scale table to provide a reference basis for subsequent association mapping. This scale table quantifies the strength limit of the connection structure and provides a standardized reference system for load bearing capacity assessment. Combined with the actual gap power value, it is associated with the allowable load scale table to determine the load level corresponding to the actual gap power value, establish the corresponding relationship between the load level and the power gap, and form a set of data describing the "load-power gap" association, which provides an input basis for subsequent sensitivity analysis. This association data provides a quantitative reference standard for optimization design and can intuitively evaluate the impact of load on power loss. Based on the load-power gap correlation data, the power gap sensitivity coefficients corresponding to different load levels are calculated using mathematical analysis methods. These sensitivity coefficients quantitatively describe the degree of influence of load changes on power loss and reflect the nonlinear relationship between the two. These sensitivity coefficients are organized into a matrix form to obtain a coefficient matrix that describes the load-power gap sensitivity in the entire working range. This coefficient matrix provides an important basis for the subsequent determination of influencing factors and can assist in optimizing design decisions. Combined with the gap power sensitivity coefficient matrix, the power influence factor values ​​corresponding to each load level are determined. According to the actual working conditions, the appropriate influence factor value is selected as the final load proportional power influence factor. This influence factor quantitatively describes the degree of influence of the load level on power loss and provides an important basis for fault diagnosis and performance optimization.

[0114] Preferably, step S55 comprises the following steps:

[0115] Step S551: performing allowable load scale cutting on the maximum allowable load data to generate an allowable load scale table;

[0116] Step S552: performing correlation mapping on the actual gap power value based on the allowable load scale table to obtain load-power gap correlation data;

[0117] Step S553: ​​performing gap power sensitivity coefficient analysis on the load-power gap correlation data to obtain a gap power sensitivity coefficient matrix;

[0118] Step S554: finalize the power impact factor of the gap power sensitivity coefficient matrix to generate the load proportion power impact factor.

[0119] The benefit of the present invention lies in that the maximum allowable load data is cut into allowable load scales to generate an allowable load scale table. This provides a data basis for subsequent correlation mapping. Based on the allowable load scale table, the actual gap power value is correlated and mapped to obtain load-power gap correlation data. This establishes the association between load and power gap, and the gap power sensitivity coefficient analysis is performed on the load-power gap correlation data to obtain a gap power sensitivity coefficient matrix. This can reflect the sensitivity of the power gap under different load conditions, and the power impact factor of the gap power sensitivity coefficient matrix is ​​finalized to generate a load ratio power impact factor. This is a key indicator for quantitatively evaluating the impact of load on power gap. By allowing load scale cutting, a set of standard load data reference system is established, which provides a basis for subsequent correlation mapping. Based on the load-power gap correlation data, the sensitivity of the power gap under different load conditions is analyzed, which helps to diagnose the cause of the power gap more accurately. Finally, the load proportional power influence factor is generated, which is a key quantitative indicator and can provide valuable reference for optimizing the power performance of the robot. These steps conduct in-depth correlation analysis between load factors and power gaps, provide important support for comprehensive diagnosis and resolution of power gap problems, further enrich and improve the methodology of industrial robot power prediction and analysis, help to more accurately evaluate the impact of load on power performance, and provide an important basis for robot power optimization.

[0120] In the embodiment of the present invention, the load range is divided into a plurality of discrete scale levels by using the maximum allowable dynamic load data, each scale level represents the allowable working range of the connection structure under the load level, such as 0-20%, 20-40%, and these scale levels and corresponding load values ​​are compiled into an allowable load scale table to provide a reference basis for subsequent association mapping. This scale table quantifies the strength limit of the connection structure and provides a standardized reference system for load bearing capacity assessment. Combined with the actual gap power value, it is associated with the allowable load scale table to determine the load level corresponding to the actual gap power value, establish the corresponding relationship between the load level and the power gap, and form a set of data describing the "load-power gap" association, which provides an input basis for subsequent sensitivity analysis. This association data provides a quantitative reference standard for optimization design and can intuitively evaluate the impact of load on power loss. Based on the load-power gap correlation data, the power gap sensitivity coefficients corresponding to different load levels are calculated using mathematical analysis methods. These sensitivity coefficients quantitatively describe the degree of influence of load changes on power loss and reflect the nonlinear relationship between the two. These sensitivity coefficients are organized into a matrix form to obtain a coefficient matrix that describes the load-power gap sensitivity in the entire working range. This coefficient matrix provides an important basis for the subsequent determination of influencing factors and can assist in optimizing design decisions. Combined with the gap power sensitivity coefficient matrix, the power influence factor values ​​corresponding to each load level are determined. According to the actual working conditions, the appropriate influence factor value is selected as the final load proportional power influence factor. This influence factor quantitatively describes the degree of influence of the load level on power loss and provides an important basis for fault diagnosis and performance optimization.

[0121] Preferably, step S6 comprises the following steps:

[0122] Step S61: Perform power correlation network mapping on the connection structure power influence factor and the load ratio power influence factor to generate a multi-dimensional power correlation network;

[0123] Step S62: Perform power prediction drive modeling on the multi-dimensional power correlation network to obtain a robot power prediction model;

[0124] Step S63: Use the robot power prediction model to perform power prediction on the load gravity torque data and the industrial robot motion data to generate predicted industrial robot power to perform the task of predicting the industrial robot power.

[0125] The benefit of the present invention lies in that a power correlation network mapping is performed on the power influencing factors of the connection structure and the power influencing factors of the load ratio, and a multi-dimensional power correlation network is generated. This combines the robot structure and load factors to form a comprehensive power correlation model. The multi-dimensional power correlation network is subjected to power prediction drive modeling to obtain a robot power prediction model. This is a power prediction model based on multiple factors and can more accurately estimate the robot power. Using the robot power prediction model, power prediction is performed on the load gravity torque data and the industrial robot motion data to generate a predicted industrial robot power. This provides a basis for robot power optimization and can better perform the task of predicting the power of industrial robots. A multidimensional power correlation network was established by integrating the connection structure and load factors, which more comprehensively reflects the key factors affecting the robot power. A power prediction model was established based on this correlation network, which provides an effective means to accurately predict the robot power. The prediction model was used to analyze the operating data to generate the predicted industrial robot power, which provides an important basis for optimizing the robot power performance. These steps embody a systematic and multidimensional robot power prediction method, which can better reflect the complexity of robot power changes compared with the prediction of a single factor, further improve the analytical framework of industrial robot power prediction, and construct a prediction model based on multi-factor correlation, which provides important support for improving the robot power performance optimization.

[0126] In the embodiment of the present invention, the load ratio power influence factor is used in combination with the connection structure power influence factor to establish a mapping relationship between the two, and the artificial neural network machine learning method is used to input these influence factor data into the model training to learn the complex power association law. Through network training, a multidimensional power association network model that can describe the comprehensive influence of multiple factors on power is obtained. The network model can convert the load characteristics and motion trajectory multidimensional information into the power prediction output of the industrial robot, providing an important basis for subsequent steps. Based on the multidimensional power association network, the power prediction driven modeling method is further applied to extract a model that can be used for industrial robot power prediction, receive input information from load and motion, and finally output the power prediction value of the industrial robot after deep learning algorithm training. The complex association law implied in the multidimensional factors is learned by using historical data to improve the prediction accuracy. The robot power prediction model is applied to the actual industrial production environment, receiving the load characteristic data and motion trajectory data from the sensor as input. After the model's predictive calculation, the power prediction value of the industrial robot under the current working conditions is output, which provides a basis for subsequent power monitoring and optimization, helps the factory understand the power consumption of the robot in advance, provides an important reference for energy management and fault diagnosis, and establishes a comprehensive power analysis and prediction system for industrial robots, providing strong support for improving energy utilization efficiency and reliability.

[0127] In the present specification, an aluminum alloy processing equipment is provided, which is used to execute the power prediction method of the aluminum alloy processing equipment as described above, and the aluminum alloy processing equipment comprises:

[0128] The data acquisition module is used to obtain the motion data of the industrial robot; reconstruct the machine posture of the industrial robot motion data to obtain the three-dimensional robot posture data; estimate the power parameters of the three-dimensional robot posture data based on the industrial robot motion data to obtain the robot output power parameters;

[0129] The actual power calculation module is used to obtain the load gravity torque data; use the load gravity torque data to perform load speed test on the robot output power parameters to generate actual joint speed data; perform joint torque analysis on the actual joint speed data and load gravity torque data to obtain actual joint torque data; perform overall performance calculation on the actual joint speed data and actual joint torque data to generate the actual power value of the robot;

[0130] The ideal power calculation module is used to simulate the ideal environment of the robot's output power parameters to obtain an ideal loss-free environment; based on the ideal loss-free environment, the ideal power calculation of the robot's output power parameters is performed to generate the robot's ideal action power value;

[0131] The power gap comparison module is used to compare the power difference between the actual power value of the robot and the ideal power value of the robot to obtain the actual power gap value; disassemble and trace the actual power gap value to generate a gap root cause location report;

[0132] The power impact analysis module is used to derive the connection impact factor of the actual gap power value based on the gap root source location report to obtain the connection structure power impact factor; based on the gap root source location report, the load impact analysis is performed on the actual gap power value to generate the load proportion power impact factor;

[0133] The power prediction module is used to perform power prediction driven modeling on the power influence factor of the connection structure and the power influence factor of the load ratio, and obtain a robot power prediction model to perform the task of predicting the power of the industrial robot.

[0134] The invention is beneficial in that it acquires and reconstructs the three-dimensional robot posture data, obtains the robot output power parameters through power parameter estimation, and calculates the actual joint speed, torque and actual power value. These data provide an important basis for subsequent power prediction. The ideal power value of the robot is calculated in an ideal lossless environment, which lays the foundation for the subsequent comparative analysis with the actual power value. The actual power value and the ideal power value are compared to obtain the actual gap power value, and the root of the gap is located by disassembling and tracing, which provides an important basis for the subsequent analysis of influencing factors. Based on the gap root location report, the influence factors of the connection structure and load on the power are analyzed, laying the foundation for establishing a more accurate power prediction model. Incorporating the influencing factors into the power prediction modeling can build a power prediction model of industrial robots that is closer to reality, and improve the accuracy and reliability of the prediction. It makes full use of the robot motion data, load data and the comparative analysis of the ideal and actual power values, finds out the key factors affecting the power, and integrates them into the final prediction model. It is a relatively systematic and comprehensive method with better practical value.

[0135] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0136] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A power prediction method for aluminum alloy processing equipment, characterized in that: The following steps are involved: Step S1: acquiring motion data of the industrial robot; reconstructing the machine posture of the industrial robot motion data to obtain three-dimensional robot posture data; estimating power parameters of the three-dimensional robot posture data based on the motion data of the industrial robot to obtain robot output power parameters; Step S2: Obtain load gravity torque data; use the load gravity torque data to perform a load speed test on the robot output power parameter to generate actual joint speed data; perform joint torque analysis on the actual joint speed data and the load gravity torque data to obtain actual joint torque data; Perform overall performance calculation on actual joint speed data and actual joint torque data to generate actual power value of the robot; Step S3: performing an ideal environment simulation on the output power parameters of the robot to obtain an ideal lossless environment; performing an ideal power calculation on the output power parameters of the robot based on the ideal lossless environment to generate an ideal action power value of the robot; Step S4: compare the power difference between the actual power value of the robot and the ideal power value of the robot to obtain the actual power difference value; disassemble and trace the actual power difference value to generate a gap root cause location report; Step S5: Based on the gap root location report, the connection impact factor is derived for the actual gap power value to obtain the connection structure power impact factor; based on the gap root location report, the load impact analysis is performed on the actual gap power value to generate the load proportion power impact factor; Step S6: Perform power prediction drive modeling on the connection structure power influence factor and the load ratio power influence factor to obtain a robot power prediction model to perform the task of predicting the power of the industrial robot.

2. The power prediction method of aluminum alloy processing equipment according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: acquiring the working data of the industrial robot, and performing continuous motion extraction on the working data of the industrial robot to obtain the motion data of the industrial robot; Step S12: tracking feature points of the industrial robot motion data to generate motion feature point data; Step S13: reconstructing the robot posture based on the motion feature point data to obtain three-dimensional robot posture data; Step S14: simulating the motion trajectory of the three-dimensional robot posture data based on the industrial robot motion data, and outputting the robot motion trajectory data; Step S15: Estimating the power parameters of the three-dimensional robot posture data according to the robot motion trajectory data to obtain the robot output power parameters.

3. The power prediction method of aluminum alloy processing equipment according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: acquiring load cargo data; performing gravity load analysis on the load cargo data to generate load gravity moment data; Step S22: using the load gravity torque data to perform a dynamic load test on the robot output power parameters to obtain a torque-speed characteristic curve; Step S23: Calculating the speed of the robot output power parameter based on the torque-speed characteristic curve to generate actual joint speed data; Step S24: performing momentum change rate identification on the actual joint speed data to obtain the momentum change rate; Step S25: performing momentum-torque analysis on the load gravity moment data and momentum change rate to obtain actual joint torque data; Step S26: Perform overall performance calculation on the actual joint speed data and the actual joint torque data to generate the actual power value of the robot.

4. The power prediction method for aluminum alloy processing equipment according to claim 3, characterized in that: Step S26 includes the following steps: Step S261: performing kinematic feature analysis on the actual joint speed data to obtain independent joint angular velocity data; Step S262: Calculate the dynamic characteristics of the actual joint torque data to generate independent joint torque data; Step S263: performing combined calculation on the independent joint angular velocity data and the independent joint torque data to obtain the independent joint actual power value; Step S264: Integrate the actual power values ​​of the independent joints based on the three-dimensional robot posture data to generate the actual effective power value of the robot.

5. The power prediction method of aluminum alloy processing equipment according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: performing power parameter simulation on the robot output power parameters to obtain parameter simulation results; Step S32: performing loss compensation environment simulation on the parameter simulation results to obtain an ideal lossless environment; Step S33: constructing an ideal dynamics model for the robot output power parameters based on an ideal lossless environment to generate an ideal dynamics model; Step S34: Utilize the ideal dynamics model to perform ideal power calculation on the robot output power parameters, and output the ideal action power value of the robot.

6. The power prediction method of aluminum alloy processing equipment according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: performing semantic synchronization processing on the actual power value of the robot and the ideal power value of the robot to obtain a synchronized actual power value and a synchronized ideal power value; Step S42: Compare the actual synchronization power value with the ideal synchronization power value to obtain the actual difference power value; Step S43: performing gap correlation mining on the actual gap power value based on the robot output power parameter to generate a gap motivation vector; Step S44: locate the root cause of the gap on the gap driver vector and obtain a gap root cause location report.

7. The power prediction method of aluminum alloy processing equipment according to claim 1, characterized in that: Step S5 includes the following steps: Step S51: extracting connection structure features from the three-dimensional robot posture data to generate connection structure feature data; Step S52: performing contact surface friction power consumption analysis on the connection structure characteristic data to obtain connection point power consumption data; Step S53: Calculate the power loss ratio of the connection point power consumption data and the robot output power parameter to obtain the connection structure power influence factor; Step S54: performing load bearing capacity deduction on the connection structure characteristic data according to the gap root location report to obtain the maximum allowable dynamic load data; Step S55: Perform load impact analysis on the actual difference power value based on the maximum allowable load data to obtain a load proportional power impact factor.

8. The power prediction method for aluminum alloy processing equipment according to claim 7, characterized in that: Step S55 The following steps are involved: Step S551: performing allowable load scale cutting on the maximum allowable load data to generate an allowable load scale table; Step S552: performing correlation mapping on the actual gap power value based on the allowable load scale table to obtain load-power gap correlation data; Step S553: ​​performing gap power sensitivity coefficient analysis on the load-power gap correlation data to obtain a gap power sensitivity coefficient matrix; Step S554: finalize the power impact factor of the gap power sensitivity coefficient matrix to generate the load proportion power impact factor.

9. The power prediction method of aluminum alloy processing equipment according to claim 1, characterized in that: Step S6 includes the following steps: Step S61: Perform power correlation network mapping on the connection structure power influence factor and the load ratio power influence factor to generate a multi-dimensional power correlation network; Step S62: Perform power prediction drive modeling on the multi-dimensional power correlation network to obtain a robot power prediction model; Step S63: Use the robot power prediction model to perform power prediction on the load gravity torque data and the industrial robot motion data to generate predicted industrial robot power to perform the task of predicting the industrial robot power.

10. An aluminum alloy processing equipment, characterized in that: Used to execute the power prediction method of aluminum alloy processing equipment according to claim 1, the aluminum alloy processing equipment comprising: The data acquisition module is used to obtain the motion data of the industrial robot; reconstruct the machine posture of the industrial robot motion data to obtain the three-dimensional robot posture data; estimate the power parameters of the three-dimensional robot posture data based on the industrial robot motion data to obtain the robot output power parameters; The actual power calculation module is used to obtain the load gravity torque data; use the load gravity torque data to perform load speed test on the robot output power parameters to generate actual joint speed data; perform joint torque analysis on the actual joint speed data and load gravity torque data to obtain actual joint torque data; perform overall performance calculation on the actual joint speed data and actual joint torque data to generate the actual power value of the robot; The ideal power calculation module is used to simulate the ideal environment of the robot output power parameters to obtain an ideal lossless environment; based on the ideal lossless environment, the ideal power calculation of the robot output power parameters is performed to generate the ideal action power value of the robot; The power gap comparison module is used to compare the power difference between the actual power value of the robot and the ideal power value of the robot to obtain the actual power gap value; disassemble and trace the actual power gap value to generate a gap root cause location report; The power impact analysis module is used to derive the connection impact factor of the actual gap power value based on the gap root source location report to obtain the connection structure power impact factor; based on the gap root source location report, the load impact analysis is performed on the actual gap power value to generate the load proportion power impact factor; The power prediction module is used to perform power prediction driven modeling on the power influence factor of the connection structure and the power influence factor of the load ratio, and obtain a robot power prediction model to perform the task of predicting the power of the industrial robot.

Images