Bearing capacity optimization control system and method for horizontal tool magazine

Abstract

The invention relates to the technical field of numerical control machine tools, in particular to a bearing capacity optimization control system and method for a horizontal tool magazine. The modeling module constructs a dynamic mechanical field model, receives instant sensing data and outputs mechanical behavior data; the instantaneous load prediction module generates an instantaneous dynamic load prediction value according to the cutter quality difference and the mechanical behavior data; the fatigue damage evaluation module calculates total fatigue damage according to the stress variable amplitude sequence and outputs a bearing capacity attenuation coefficient; and the multi-mode control module is used for generating a transmission chain control instruction according to the instantaneous dynamic load predicted value and the bearing capacity attenuation coefficient, and optimizing the bearing capacity of the horizontal tool magazine. According to the invention, through cooperative work of a plurality of modules, real-time monitoring and optimization of the bearing capacity of the tool magazine are realized, and the operation efficiency and reliability of the system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of numerical control machine tools, and particularly to an optimized control system and method for the load-bearing capacity of a horizontal tool magazine. Background Art

[0002] In numerical control machine tools, as a main tool management system, the horizontal tool magazine realizes the automatic storage, management and replacement of tools through a horizontal layout method, significantly improving the machining efficiency and flexibility of the machine tool. The existing technologies mainly focus on the optimization of the overall load-bearing capacity of the tool magazine system, and ensure the stable operation of the system under various working conditions through means such as structural design, load balancing, sensor monitoring and dynamic simulation.

[0003] However, in the actual tool change process, the weights of the tools often vary greatly, resulting in a sudden load during the tool change instant. This sudden change in weight will cause a sharp stress fluctuation on key components such as the chain in the transmission system, which will in turn cause accelerated fatigue damage. In severe cases, it may affect the system stability and service life. Most of the traditional technical means focus on the overall load distribution and static structure optimization of the tool magazine system, and pay insufficient attention to the transient stress changes caused by the sudden change in tool weight during the tool change process, and cannot effectively relieve the impact of this sudden load on the load-bearing capacity of the chain.

[0004] Therefore, aiming at the problem of the sharp change in chain stress caused by the sudden change in tool weight during the tool change process, there is an urgent need for an optimized control method and system that can monitor, predict and dynamically adjust the system state in real time to ensure the stable operation of key components during the tool change process of the tool magazine and extend the overall service life of the system. Summary of the Invention

[0005] In view of at least one of the above technical problems, the present invention provides an optimized control system and method for the load-bearing capacity of a horizontal tool magazine.

[0006] According to a first aspect of the present invention, there is provided an optimized control system for the load-bearing capacity of a horizontal tool magazine, comprising:

[0007] A tool mass measurement module that measures the tool mass in real time to obtain instant sensing data;

[0008] A modeling module that constructs a dynamic mechanical field model of the chain drive system, receives the instant sensing data, and outputs mechanical behavior data;

[0009] An instantaneous load prediction module that generates an instantaneous dynamic load prediction value according to the mass difference between the current tool and the target tool and the mechanical behavior data;

[0010] The fatigue damage assessment module extracts the maximum stress values within each tool change cycle based on the historical mechanical behavior data to form a stress amplitude sequence, calculates the total fatigue damage according to the stress amplitude sequence, and outputs the bearing capacity attenuation coefficient.

[0011] The multi-mode control module generates a transmission chain control command based on the predicted value of the instantaneous dynamic load and the bearing capacity attenuation coefficient to optimize the bearing capacity of the horizontal tool magazine.

[0012] In some embodiments of the present invention, it further includes:

[0013] The calibration module constructs a digital twin model that runs synchronously with the physical entity of the horizontal tool magazine, compares the dynamic mechanical field model with the actual operation data of the horizontal tool magazine in real time, calculates the modeling error, and triggers global parameter calibration when the modeling error exceeds the set threshold.

[0014] In some embodiments of the present invention, the tool quality measurement module includes the following units:

[0015] The sensor unit includes a strain sensor fixedly installed in the tool clamping mechanism, and the strain sensor measures the quality of the tool by detecting the deformation generated by the tool clamping mechanism during the clamping process.

[0016] The signal acquisition and processing unit acquires the output signal of the sensor unit, converts it into a digital signal, and filters and calibrates the signal to obtain the instant sensing data.

[0017] In some embodiments of the present invention, the dynamic mechanical field model includes the following units:

[0018] The data input and preprocessing unit receives the instant sensing data and performs preprocessing.

[0019] The dynamic modeling unit constructs a dynamic mechanical field model of the chain drive system based on an adaptive fuzzy neural network.

[0020] The stress output unit outputs the mechanical behavior data of the chain drive system at different times or spatial positions based on the dynamic mechanical field model.

[0021] In some embodiments of the present invention, the instantaneous load prediction module includes the following units:

[0022] The mass difference calculation unit receives the current tool quality and the target tool quality and calculates the mass difference between the two.

[0023] The stress data acquisition unit receives the mechanical behavior data and filters the data related to the tool change process.

[0024] A hybrid prediction calculation unit calculates and outputs a predicted value of the instantaneous dynamic load based on the quality difference and the mechanical behavior data by using a collaborative prediction algorithm.

[0025] In some embodiments of the present invention, the collaborative prediction algorithm includes:

[0026] Receiving historical mechanical behavior data for the tool change process;

[0027] Decomposing the historical mechanical behavior data by using the seasonal trend decomposition method to obtain a trend component and a residual component;

[0028] Modeling and predicting the trend component by using an autoregressive integrated moving average model to obtain a trend prediction value;

[0029] Modeling and predicting the residual component by using a long short-term memory neural network to obtain a residual prediction value;

[0030] Calculating a fusion weight output by the autoregressive integrated moving average model and the long short-term memory neural network model according to the quality difference, and performing weighted fusion correspondingly to obtain the predicted value of the instantaneous dynamic load.

[0031] In some embodiments of the present invention, the fatigue damage assessment module further includes:

[0032] Extracting the maximum stress value within each tool change cycle based on the historical mechanical behavior data to form a maximum stress value sequence;

[0033] Scanning the stress time series data within this cycle and finding the lowest point to extract, forming a minimum stress value sequence; Converting the extracted maximum stress value sequence and minimum stress value sequence into a stress amplitude variation sequence Z, and the formula is:

[0034] ;

[0035] where, is the maximum stress value of the i-th tool change cycle, is the minimum stress value of the i-th tool change cycle, and n is the total number of tool change cycles;

[0036] Calculating the total fatigue damage according to the stress amplitude variation sequence Z, and the formula is as follows:

[0037] ,

[0038] where, is the total fatigue damage, is the number of cycles corresponding to the i-th stress amplitude cycle, is the fatigue life of the i-th stress amplitude cycle, and n is the total number of tool change cycles;

[0039] According to the calculated total fatigue damage, the bearing capacity attenuation coefficient is output, and the formula is:

[0040] ,

[0041] where, is the bearing capacity attenuation coefficient, and its value range is .

[0042] In some embodiments of the present invention, the generating of the transmission chain control instruction includes:

[0043] Receiving the actual speed and actual acceleration of the chain drive system, as well as the desired speed and desired acceleration of the system;

[0044] Calculating the speed error and acceleration error of the chain drive system;

[0045] Performing a fuzzification process on the speed error and the acceleration error, and applying the fuzzy PID control rule to calculate the speed control instruction of the chain drive system;

[0046] Calculating the weights of each control instruction through a weight allocation algorithm;

[0047] Calculating the final cooperative control instruction according to the weight allocation result.

[0048] In some embodiments of the present invention, the triggering of the global parameter calibration when the modeling error exceeds the set threshold further includes:

[0049] Calculating the sensitivity of each parameter in the model to the error according to the modeling error distribution;

[0050] Constructing an optimization objective function according to the sensitivity, and using a multi-objective optimization algorithm to generate parameter adjustment instructions and update the control parameters;

[0051] Injecting the updated dynamic mechanical field model into the historical load data for forward simulation, and calculating the updated model prediction error;

[0052] If the error reduction rate is less than the set tolerance threshold, stop the optimization process; if the error reduction rate is still large, the control parameters need to be adjusted continuously.

[0053] According to the second aspect of the present invention, there is also provided a bearing capacity optimization control method for a horizontal tool magazine, including the following steps:

[0054] Measuring the tool mass in real time to obtain instant sensing data;

[0055] Constructing a dynamic mechanical field model of the chain drive system, receiving the instant sensing data, and outputting mechanical behavior data;

[0056] Generate an instantaneous dynamic load prediction value based on the mass difference between the current tool and the target tool and the mechanical behavior data.

[0057] Based on the historical mechanical behavior data, extract the maximum stress values within each tool change cycle to form a stress amplitude sequence, calculate the total fatigue damage according to the stress amplitude sequence, and output the bearing capacity attenuation coefficient.

[0058] Generate a transmission chain control command according to the instantaneous dynamic load prediction value and the bearing capacity attenuation coefficient, and optimize the bearing capacity of the horizontal tool magazine.

[0059] The beneficial effects of the present invention are as follows: Through the collaborative work of multiple modules, the present invention realizes the real-time monitoring and optimization of the bearing capacity of the tool magazine. An adaptive fuzzy neural network is used to construct a dynamic mechanical field model to accurately predict the instantaneous dynamic load, and the collaborative prediction algorithm is combined to improve the load prediction accuracy. Through the fatigue damage assessment based on the stress amplitude sequence, potential fatigue problems can be identified in a timely manner, reducing the occurrence of failures. Also, through the real-time calibration of the digital twin model, the modeling error is ensured to be within an acceptable range, improving the accuracy of the model. The multi-mode control module optimizes the transmission chain control command according to the instantaneous load prediction value and the bearing capacity attenuation coefficient, flexibly adjusts the system operation strategy, thereby maximizing the bearing capacity of the tool magazine, extending the service life of the equipment, and improving the operation efficiency and reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0061] Figure 1 It is a schematic structural diagram of the bearing capacity optimization control system of the horizontal tool magazine in the embodiment of the present invention.

[0062] Figure 2 It is a schematic structural diagram of the tool quality measurement module in the bearing capacity optimization control system of the horizontal tool magazine in the embodiment of the present invention.

[0063] Figure 3 It is a schematic structural diagram of the dynamic mechanical field model in the bearing capacity optimization control system of the horizontal tool magazine in the embodiment of the present invention.

[0064] Figure 4 It is a schematic structural diagram of the instantaneous load prediction module in the bearing capacity optimization control system of the horizontal tool magazine in the embodiment of the present invention.

[0065] Figure 5Schematic flow chart of the collaborative prediction algorithm in the bearing capacity optimization control system of the horizontal tool magazine in the embodiments of the present invention;

[0066] Figure 6 Schematic flow chart of the fatigue damage assessment module in the bearing capacity optimization control system of the horizontal tool magazine in the embodiments of the present invention;

[0067] Figure 7 Schematic flow chart of generating the transmission chain control instruction in the bearing capacity optimization control system of the horizontal tool magazine in the embodiments of the present invention;

[0068] Figure 8 Schematic flow chart of the global parameter calibration in the bearing capacity optimization control system of the horizontal tool magazine in the embodiments of the present invention;

[0069] Figure 9 Schematic flow chart of the bearing capacity optimization control method of the horizontal tool magazine in the embodiments of the present invention. Detailed implementation manners

[0070] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments.

[0071] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly on the other element or there may also be a middle element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be a middle element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are only for the purpose of illustration and do not represent the only implementation manners.

[0072] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used in the description of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0073] Embodiment 1

[0074] As Figure 1 shown, the bearing capacity optimization control system of the horizontal tool magazine includes:

[0075] The tool quality measurement module is integrated into the tool clamping mechanism and measures the tool quality in real time through strain sensors to obtain immediate sensing data. The strain sensors provide accurate data on the current quality of the tool, which serves as the basis for subsequent system analysis and prediction. It can monitor the quality status of the tool in real time, provide accurate input data for load prediction and fatigue damage assessment. Through this module, the system can detect the wear condition of the tool in a timely manner, ensure that the tool is used in the best working state, reduce the occurrence of equipment failures, and thus extend the service life of the tool magazine.

[0076] The modeling module constructs a dynamic mechanical field model of the chain drive system based on an adaptive fuzzy neural network, receives the immediate sensing data, and outputs mechanical behavior data. It receives the immediate sensing data transmitted from the tool quality measurement module and outputs mechanical behavior data according to these data. The fuzzy neural network can handle complex system behaviors and uncertainties, provide real-time and accurate predictions of mechanical behaviors. The system can adapt to different working conditions and environmental changes, improving the robustness and prediction accuracy of the entire system.

[0077] The instantaneous load prediction module generates an instantaneous dynamic load prediction value through a collaborative prediction algorithm based on the quality difference between the current tool and the target tool and the mechanical behavior data within a set time window before the tool change command is triggered. This module can accurately predict the instantaneous load of the tool magazine during the tool change process, ensure that the system can adjust the control strategy in a timely manner to avoid overload situations and prevent damage to the tool magazine. Through this prediction step, the system can anticipate changes in the load in advance to adjust the operation of the tool magazine, thereby effectively improving the safety and stability of the operation.

[0078] The fatigue damage assessment module extracts the maximum stress values within each tool change cycle from the historical mechanical behavior data to form a stress amplitude sequence, calculates the total fatigue damage based on the stress amplitude sequence, and outputs the bearing capacity attenuation coefficient. During the long-term operation of the tool and the tool magazine, due to instantaneous loads and long working hours, the tool magazine will suffer varying degrees of fatigue damage. The main function of this module is to assess the fatigue damage in different working cycles, help predict potential structural damage. Through the assessment step, the system can identify fatigue problems in advance, avoid premature equipment failure, and provide a scientific basis for equipment maintenance and repair, extending the service life of the equipment.

[0079] The multi-mode control module generates a transmission chain control command according to the instantaneous dynamic load prediction value and the bearing capacity attenuation coefficient through a weight distribution algorithm to optimize the bearing capacity of the horizontal tool magazine. Through the multi-mode control module, the system can flexibly adjust the operation strategy under different load conditions, improve the stability and reliability of the system, help reduce unnecessary energy consumption and mechanical wear, thereby improving the overall operation efficiency of the system, and ensuring that the equipment is always in the best load state to avoid damage caused by excessive load.

[0080] The calibration module constructs a digital twin model that runs synchronously with the physical entity of the horizontal tool magazine, compares the dynamic mechanical field model with the actual operation data of the horizontal tool magazine in real time, calculates the modeling error, and triggers global parameter calibration when the modeling error exceeds the set threshold. The calibration module ensures the consistency between the system model and the physical entity, and corrects the errors in the model in real time, thereby ensuring the accuracy of the system prediction results.

[0081] Through the collaborative work of multiple modules, the present invention realizes the real-time monitoring and optimization of the load-bearing capacity of the tool magazine. It uses an adaptive fuzzy neural network to construct a dynamic mechanical field model to accurately predict the instantaneous dynamic load, combines a collaborative prediction algorithm to improve the load prediction accuracy, and through the fatigue damage assessment based on the stress amplitude sequence, can timely identify potential fatigue problems, reduce the occurrence of faults. It also ensures that the modeling error is within an acceptable range through the real-time calibration of the digital twin model, improving the accuracy of the model. The multi-mode control module optimizes the transmission chain control instructions according to the instantaneous load prediction value and the load-bearing capacity attenuation coefficient, flexibly adjusts the system operation strategy, thereby maximizing the load-bearing capacity of the tool magazine, extending the service life of the equipment, and improving the operation efficiency and reliability of the system.

[0082] In some embodiments of the present invention, as Figure 2 shown, the tool quality measurement module further includes the following units:

[0083] The sensor unit includes a strain sensor fixedly installed in the tool clamping mechanism. The strain sensor measures the mass of the tool by detecting the deformation generated by the tool clamping mechanism during the clamping process. The calculation formula for the tool mass is:

[0084] ,

[0085] where m is the mass of the tool, in kg, is the strain value, measured by the strain sensor, A is the cross-sectional area of the area where the strain sensor acts, in , E is the elastic modulus of the material of the tool clamping mechanism, in N / , g is the acceleration due to gravity, in m / ;

[0086] The signal acquisition and processing unit is used to collect the output signal of the strain sensor unit, convert it into a digital signal, and filter and calibrate the signal to obtain the instant sensing data of the tool quality; the filtering and calibration functions ensure the quality and accuracy of the signal. By filtering, noise can be removed and interference can be reduced to ensure more accurate data, while calibration ensures that the system can still maintain high precision under different environments and conditions. It should be noted here that there are many ways of filtering, which can be a low-pass filter, a high-pass filter, or median filtering or average filtering and other forms. It should also be noted that there are also many ways of calibration, which can be proportional calibration, multi-point calibration, or calibration based on a reference standard or other forms that can be calibrated in this technical solution.

[0087] The storage unit is used to store the instant sensing data. There are many forms of the storage unit, which can be a read-only memory, cloud storage, or an external memory card or hard disk storage and other forms.

[0088] As Figure 3 shown, in some embodiments of the present invention, the dynamic mechanical field model further includes the following units:

[0089] The data input and preprocessing unit is used to receive the instant sensing data and perform normalization or standardization processing on the instant sensing data; the normalization and standardization processing can eliminate the influence of the dimension difference and outliers of the data, enable the input data to adapt to the calculation requirements of the model, avoid inaccurate prediction or modeling caused by the scale problem of the data, and can more effectively utilize the data from the tool quality measurement module and other sensors, reducing the interference caused by data inconsistency or outliers.

[0090] The dynamic modeling unit is used to construct a dynamic mechanical field model of the chain drive system based on an adaptive fuzzy neural network; through the adaptive fuzzy neural network, the system can process complex and non-linear mechanical behavior data and self-adjust according to the input data. The dynamic modeling unit constructs an accurate dynamic mechanical field model by combining real-time data, including tool quality and mechanical behavior data, which enables the system to dynamically adapt to load changes and accurately predict the mechanical behavior under different working conditions.

[0091] The stress output unit is used to output the mechanical behavior data of the chain drive system at different times or spatial positions based on the dynamic mechanical field model.

[0092] As Figure 4 shown, in some embodiments of the present invention, the instantaneous load prediction module further includes the following units:

[0093] The mass difference calculation unit is used to receive the current tool quality and the target tool quality, and calculate the mass difference between the two. The formula is:

[0094] ,

[0095] Among them, is the poor quality of the tool, in kg, is the target tool quality, in kg, is the current tool quality, in kg; calculating the quality difference is crucial for predicting the instantaneous load change, because the quality of the tool directly affects the load change generated during its replacement in the tool magazine. After the quality difference calculation unit is combined with other modules, it can help the instantaneous load prediction module predict the load change more accurately, thereby avoiding the impact of excessive instantaneous loads on the stability and load-bearing capacity of the tool magazine system.

[0096] The stress data acquisition unit is used to receive the mechanical behavior data output by the modeling module and screen the data related to the tool replacement process; the stress data acquisition unit extracts useful signals related to the current operating conditions from a large amount of mechanical behavior data to ensure that the data on which subsequent predictions and evaluations are based is relevant and of high quality, which can reduce the problem of inaccurate load prediction caused by errors or irrelevant data.

[0097] The hybrid prediction calculation unit is used to calculate the predicted value of the instantaneous dynamic load based on the quality difference and the mechanical behavior data by using the collaborative prediction algorithm and output the predicted value of the instantaneous dynamic load. By integrating data from different sources (such as quality difference, stress data, etc.), the collaborative prediction algorithm can effectively improve the prediction accuracy of the dynamic load. This accurate load prediction can not only prevent the system from being overloaded, but also effectively improve the working efficiency and long-term reliability of the tool magazine.

[0098] In some embodiments of the present invention, as Figure 5 shown, the collaborative prediction algorithm includes:

[0099] A10: Receive the historical real-time detection data for the tool change process, and the formula is:

[0100] ,

[0101] Among them, D represents the historical real-time detection data, represents the detection data at the i-th moment, and n represents the total duration of the data;

[0102] A20: Use the seasonal trend decomposition method to decompose the historical real-time detection data to obtain the trend component and the residual component, and the decomposition formula is:

[0103] ,

[0104] Among them, Denoted as the detection data at a certain moment \(t\) in the historical real-time detection data, \(T(t)\) is the trend component, and \(R(t)\) is the residual component;

[0105] A30: For the decomposed trend component, an autoregressive integrated moving average model is used for modeling and prediction to obtain the trend prediction value. The formula of the autoregressive integrated moving average model is:

[0106] (t - 1) (t - 2) (t - p) (t - 1)+ (t - 2) (t - q),

[0107] where, is the predicted value of the trend component, (t - 1), (t - 2), …, (t - n) represents the historical trend component, (t - 1), (t - 2), …, (t - q) are the residual terms, are the coefficients of the autoregressive model, are the coefficients of the moving average model, \(p\) is the order of the autoregressive model, and \(q\) is the order of the moving average model;

[0108] A40: For the residual component, a long short-term memory neural network is used for modeling and prediction to obtain the residual prediction value. The prediction formula of the long short-term memory neural network is:

[0109] ,

[0110] where, is the historical residual data, is the predicted future residual value, is the time window length of the historical data;

[0111] A50: According to the tool quality difference value, calculate the fusion weights output by the autoregressive integrated moving average model and the long short-term memory neural network model. The weight calculation formula is:

[0112] ,

[0113] ,

[0114] where, and are the fusion weights of the autoregressive integrated moving average model and the long short-term memory neural network model, is the tool quality difference value, is the sum of the quality differences of all tools;

[0115] A60: According to the fusion weight and , the predicted values of the autoregressive integrated moving average model and the long short-term memory neural network model are weighted and fused to obtain the final predicted value of the instantaneous dynamic load. The formula is:

[0116] ,

[0117] where, is the final predicted value of the instantaneous dynamic load, representing the predicted load at the time step moment, is the trend component predicted by the autoregressive integrated moving average model, is the residual component predicted by the long short-term memory neural network model.

[0118] The seasonal trend decomposition method can extract the long-term trend and short-term fluctuations in the data, effectively removing noise interference; the ARIMA model can accurately capture the linear changes in historical trends, and the LSTM model can handle non-linear and complex dynamic patterns, further improving the prediction ability. By weighted fusing the prediction results of these two models, the system can dynamically adjust the contribution degree of the prediction model under different tool states, so as to obtain a more accurate predicted value of the instantaneous load.

[0119] In some embodiments of the present invention, as Figure 6 shown, the fatigue damage assessment module includes:

[0120] B10: Based on the historical mechanical behavior data, extract the maximum stress value in each tool change cycle to form a maximum stress value sequence. The formula is:

[0121] ,

[0122] where, is the maximum stress value sequence, is the maximum stress value of the i-th tool change cycle, and n is the total number of tool change cycles;

[0123] B20: Scan the stress time series data in this cycle and find the lowest point to extract, forming a minimum stress value sequence. The formula is:

[0124] ,

[0125] where, is the minimum stress value sequence, is the maximum stress value of the i-th tool change cycle, and n is the total number of tool change cycles;

[0126] B30: Convert the extracted maximum stress value sequence and minimum stress value sequence into a stress amplitude variation sequence Z, with the formula:

[0127] ;

[0128] B40: Calculate the total fatigue damage based on the stress amplitude variation sequence Z, with the formula as follows:

[0129] ,

[0130] where, is the total fatigue damage, is the number of cycles corresponding to the i-th stress amplitude cycle, is the fatigue life of the i-th stress amplitude cycle, and n is the total number of tool change cycles;

[0131] B50: Output the bearing capacity attenuation coefficient based on the calculated total fatigue damage, with the formula:

[0132] ,

[0133] where, is the bearing capacity attenuation coefficient, and its value range is .

[0134] The fatigue damage assessment module accurately evaluates the fatigue damage of the tool magazine during different tool change cycles through the analysis of historical mechanical behavior data. The module extracts the maximum stress value and minimum stress value within each tool change cycle to form a maximum stress value sequence and a minimum stress value sequence. These data can reflect the stress changes endured by the tool magazine during operation. Finally, by calculating the stress amplitude variation sequence, the variation law of the stress amplitude can be further analyzed, providing a data basis for the assessment of fatigue damage.

[0135] Based on the stress amplitude variation sequence, the module calculates the fatigue damage corresponding to each stress amplitude cycle and finally obtains the total fatigue damage. This index can help accurately evaluate the degree of fatigue damage that the tool magazine may encounter during long-term use. The bearing capacity attenuation coefficient calculated from the total fatigue damage can quantify the decline of the tool magazine's bearing capacity over time, providing a reference for the service life of the tool magazine in practical applications, predicting when the equipment may experience overload or failure, and then carrying out reasonable maintenance and optimization.

[0136] In some embodiments of the present invention, the formula of the weight distribution algorithm is as follows:

[0137] ,

[0138] where, represents the weight of the i-th control instruction, is the predicted value of the instantaneous dynamic load, is the i-th bearing capacity attenuation coefficient, and m is the total number of control commands in the transmission chain.

[0139] The weight allocation algorithm combines the predicted value of the instantaneous dynamic load and the bearing capacity attenuation coefficient to accurately allocate weights to each control command, thereby optimizing the load management of the transmission chain, ensuring that the system can flexibly respond under different load conditions, avoiding the risks of overloading or underloading, improving the stability and safety of the system, reducing the fatigue damage of the equipment, and extending the service life. In addition, the algorithm also optimizes the load distribution, improving the operating efficiency and resource utilization rate of the system.

[0140] In some embodiments of the present invention, as Figure 7 shown, generating the transmission chain control commands according to the weight allocation algorithm includes:

[0141] C10: Receiving the actual speed V(t) and actual acceleration A(t) of the chain drive system, as well as the desired speed and desired acceleration ;

[0142] C20: Calculating the speed error and acceleration error , where

[0143] ,

[0144] ;

[0145] C30: Fuzzifying the speed error and acceleration error to obtain the fuzzy variables and ;

[0146] C40: According to the fuzzified errors and , applying the fuzzy PID control rule to calculate the speed control command of the chain drive system, and the formula is:

[0147] ,

[0148] ,

[0149] where are the proportional, integral, and differential gain coefficients of the fuzzy PID control respectively, and are the control commands for speed and acceleration respectively;

[0150] C50: Calculating the weights of each control command through the weight allocation algorithm, and the formula is:

[0151] ,

[0152] ,

[0153] wherein, is the weight of the speed control command, is the weight of the acceleration control command;

[0154] C60: Calculate the final cooperative control command according to the weight distribution result, and the formula is:

[0155] ,

[0156] wherein, is the final control command of the chain drive system.

[0157] By combining fuzzy PID control and weight distribution algorithm, precise control of the chain drive system is achieved, significantly improving the response performance and stability of the system. By fuzzifying the speed and acceleration errors, the system can effectively cope with uncertainties and fluctuations, thus achieving smooth and highly adaptable control. Fuzzy PID control further optimizes the regulation of speed and acceleration, precisely adjusting the control command to ensure the efficient operation of the system. At the same time, the weight distribution algorithm dynamically adjusts the priority of the control command according to the speed and acceleration errors, ensuring the balance and stability of the system under various working conditions. The cooperative control combining the speed and acceleration control commands optimizes the overall performance of the system, reduces overshoot, oscillation and hysteresis phenomena, and improves the working efficiency, precision and robustness of the chain drive system.

[0158] In some embodiments of the present invention, as Figure 8 shown, when the modeling error exceeds the set threshold, global parameter calibration is triggered, and it further includes:

[0159] D10: Calculate the sensitivity of each parameter in the model to the error according to the modeling error distribution , and the formula is:

[0160] ,

[0161] wherein, is the model prediction error, is the i-th parameter in the model;

[0162] D20: Construct an optimization objective function according to the sensitivity, and use a multi-objective optimization algorithm to generate parameter adjustment instructions. The optimization objective function is:

[0163] ,

[0164] wherein, is the model prediction error, is the adjustment amount of the i-th parameter, is the weight coefficient between optimization objectives, T is the time step, and n is the number of parameters;

[0165] D30: Update the control parameters according to the optimization results. The update formula is:

[0166] ,

[0167] where, is the value of the i-th parameter at time t + 1, is the value of the i-th parameter at time t, is the parameter increment calculated according to the optimization algorithm;

[0168] D40: Inject the updated dynamic mechanical field model into the historical load data for forward simulation, and calculate the updated model prediction error. The error reduction rate of the forward simulation result is:

[0169] ,

[0170] where, is the error reduction rate, is the error of the actual load data, is the predicted error of the updated model;

[0171] D50: If the error reduction rate is less than the set tolerance threshold, stop the optimization process; if the error reduction rate is still large, the control parameters need to be adjusted continuously.

[0172] In practical applications, with the long-term operation of the system or the change of the external environment, the original model may not accurately reflect the current state, resulting in the continuous accumulation of prediction errors. This kind of error will affect the performance of the control system. Especially in the case of load changes, equipment wear, etc., the traditional static parameter adjustment method cannot effectively adapt to these dynamic changes, resulting in a decrease in control accuracy and even system failures. By introducing the global parameter calibration and intelligent optimization mechanism, these problems can be effectively solved, enabling the system to automatically adjust when the error exceeds the set threshold, ensuring prediction accuracy and system stability.

[0173] Through intelligent error analysis and dynamic parameter adjustment, the prediction accuracy and stability of the system are improved. When the modeling error exceeds the set threshold, the system calculates the sensitivity of each parameter in the model to the error, identifies the key influencing parameters, and balances error minimization and parameter adjustment amount through a multi-objective optimization algorithm, thereby achieving precise parameter adjustment. The optimization process takes into account the reduction of error and the amount of parameter adjustment to ensure stable and efficient optimization effects. Through forward simulation and real-time feedback, the system can continuously evaluate the optimization results and determine whether further adjustment of control parameters is needed, avoiding ineffective optimization operations.

[0174] According to the second aspect of the present invention, there is also provided an optimization control method for the load-bearing capacity of a horizontal tool magazine, as Figure 9 shown, including the following steps:

[0175] S10: Measure the tool quality in real time to obtain instant sensing data;

[0176] S20: Construct a dynamic mechanical field model of the chain drive system, receive the instant sensing data, and output mechanical behavior data;

[0177] S30: Generate an instantaneous dynamic load prediction value according to the mass difference between the current tool and the target tool and the mechanical behavior data;

[0178] S40: Based on the historical mechanical behavior data, extract the maximum stress values within each tool change cycle to form a stress amplitude sequence, calculate the total fatigue damage according to the stress amplitude sequence, and output a load-bearing capacity attenuation coefficient;

[0179] S50: Generate a transmission chain control instruction according to the instantaneous dynamic load prediction value and the load-bearing capacity attenuation coefficient to optimize the load-bearing capacity of the horizontal tool magazine.

[0180] Through the above method, during the real-time working process of the horizontal tool magazine, within a set time window, the change in the load-bearing capacity generated in the future can be predicted based on the tool quality and the quality of the target tool. An optimization scheme is generated through the calculation of the load-bearing capacity to address the problems caused by sudden changes in the load-bearing capacity.

[0181] Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A load-bearing capacity optimization control system for a horizontal tool magazine, characterized in that: include: Tool quality measurement module, which measures tool quality in real time and obtains instant sensor data; A modeling module, constructing a dynamic mechanical field model of the chain transmission system, receiving the instant sensor data, and outputting mechanical behavior data; An instantaneous load prediction module generates an instantaneous dynamic load prediction value according to the mass difference between the current tool and the target tool and the mechanical behavior data; A fatigue damage assessment module, based on the historical mechanical behavior data, extracts the maximum stress value in each tool change cycle to form a stress amplitude sequence, calculates the total fatigue damage according to the stress amplitude sequence, and outputs the load-bearing capacity attenuation coefficient; The multi-mode control module generates a transmission chain control instruction according to the instantaneous dynamic load prediction value and the load-bearing capacity attenuation coefficient to optimize the load-bearing capacity of the horizontal tool magazine.

2. The load-bearing capacity optimization control system of the horizontal tool magazine according to claim 1 is characterized in that: Also includes: The calibration module constructs a digital twin model that runs synchronously with the physical entity of the flat tool magazine, compares the dynamic mechanical field model with the actual operating data of the flat tool magazine in real time, calculates the modeling error, and triggers the global parameter calibration when the modeling error exceeds the set threshold.

3. The load-bearing capacity optimization control system of the horizontal tool magazine according to claim 1 is characterized in that: The tool quality measurement module includes the following units: The sensor unit comprises a strain sensor fixedly mounted in the tool clamping mechanism, wherein the strain sensor measures the quality of the tool by detecting the deformation of the tool clamping mechanism during the clamping process; The signal acquisition and processing unit acquires the output signal of the sensor unit, converts it into a digital signal, and filters and calibrates the signal to obtain the real-time sensing data.

4. The load-bearing capacity optimization control system of the horizontal tool magazine according to claim 3 is characterized in that: The dynamic mechanical field model includes the following units: A data input and preprocessing unit receives the instant sensor data and performs preprocessing; Dynamic modeling unit, which builds the dynamic mechanical field model of the chain transmission system based on adaptive fuzzy neural network; The stress output unit outputs the mechanical behavior data of the chain transmission system at different times or spatial positions based on the dynamic mechanical field model.

5. The load-bearing capacity optimization control system of the horizontal tool magazine according to claim 1, characterized in that: The instantaneous load prediction module includes the following units: A quality difference calculation unit receives the current tool quality and the target tool quality and calculates the quality difference between the two; A stress data acquisition unit receives the mechanical behavior data and screens data related to the tool replacement process; The hybrid prediction calculation unit calculates and outputs the instantaneous dynamic load prediction value by using a collaborative prediction algorithm based on the mass difference and the mechanical behavior data.

6. The load-bearing capacity optimization control system of the horizontal tool magazine according to claim 5, characterized in that: The collaborative prediction algorithm includes: receiving historical mechanical behavior data for tool changing process; Decomposing the historical mechanical behavior data using a seasonal trend decomposition method to obtain a trend component and a residual component; The trend component is modeled and predicted using an autoregressive integrated moving average model to obtain a trend prediction value; The residual component is modeled and predicted using a long short-term memory neural network to obtain a residual prediction value; According to the mass difference, the fusion weights of the outputs of the autoregressive integrated moving average model and the long short-term memory neural network model are calculated, and the instantaneous dynamic load prediction value is obtained by corresponding weighted fusion.

7. The load-bearing capacity optimization control system of the horizontal tool magazine according to claim 6, characterized in that: The Fatigue Damage Assessment module also includes: Based on the historical mechanical behavior data, the maximum stress value in each tool change cycle is extracted to form a maximum stress value sequence; Scan the stress time series data within the cycle and find the lowest point to extract, forming a minimum stress value sequence; The extracted maximum stress value sequence and minimum stress value sequence are converted into stress amplitude sequence Z, and the formula is: ； in, is the maximum stress value of the i-th tool change cycle, is the minimum stress value of the i-th tool change cycle, and n is the total number of tool change cycles; The total fatigue damage is calculated based on the stress amplitude sequence Z, and the formula is as follows: ， in, is the total fatigue damage, is the number of cycles corresponding to the i-th stress amplitude cycle, is the fatigue life of the i-th stress amplitude cycle, and n is the total number of tool change cycles; According to the calculated total fatigue damage, the load-bearing capacity attenuation coefficient is output, and the formula is: ， in, is the load capacity reduction coefficient, and its value range is .

8. The load-bearing capacity optimization control system of the horizontal tool magazine according to claim 1, characterized in that: The generating of the transmission chain control instruction comprises: receiving an actual speed and an actual acceleration of the chain drive system, and a desired speed and a desired acceleration of the system; Calculate the velocity error and acceleration error of the chain drive system; The speed error and the acceleration error are fuzzified, and a fuzzy PID control rule is applied to calculate a speed control instruction of the chain transmission system; The weight of each control instruction is calculated through a weight distribution algorithm, and the final collaborative control instruction is calculated based on the weight distribution result.

9. The load-bearing capacity optimization control system of the horizontal tool magazine according to claim 2, characterized in that: The global parameter calibration is triggered when the modeling error exceeds a set threshold, and further includes: According to the modeling error distribution, calculate the sensitivity of each parameter in the model to the error; Construct an optimization objective function based on sensitivity, use a multi-objective optimization algorithm to generate parameter adjustment instructions, and update control parameters; Inject the updated dynamic mechanical field model into the historical load data for forward simulation and calculate the updated model prediction error; If the error reduction rate is less than the set tolerance threshold, the optimization process is stopped; if the error reduction rate is still large, the control parameters need to be further adjusted.

10. The method for optimizing the load-bearing capacity of a horizontal tool magazine according to claim 1, characterized in that: The load capacity optimization control system using the horizontal tool magazine according to any one of claims 1 to 9 comprises the following steps: Real-time measurement of tool quality to obtain instant sensor data; Constructing a dynamic mechanical field model of the chain transmission system, receiving the instant sensor data, and outputting mechanical behavior data; generating an instantaneous dynamic load prediction value according to the mass difference between the current tool and the target tool and the mechanical behavior data; Based on the historical mechanical behavior data, the maximum stress value in each tool change cycle is extracted to form a stress amplitude sequence, the total fatigue damage is calculated according to the stress amplitude sequence, and the bearing capacity attenuation coefficient is output; According to the instantaneous dynamic load prediction value and the load-bearing capacity attenuation coefficient, a transmission chain control instruction is generated to optimize the load-bearing capacity of the horizontal tool magazine.

Images