A crane double winch synchronous control method

Abstract

The present application relates to the technical field of crane control, and discloses a crane double-hoist synchronous control method, comprising the following steps: collecting working condition data of the crane double-hoist in real time, calculating the synchronous speed error and displacement error of the crane double-hoist, inputting the working condition data of the crane double-hoist into a pre-trained decision tree working condition classification model, outputting a working condition type, calling a corresponding synchronous control model based on an LSTM neural network model trained for multiple working conditions according to the working condition type; based on the monitored synchronous speed error and synchronous displacement error, if the synchronous speed error and the synchronous displacement error are less than the expected control precision, further fine-tuning the control parameters, and re-examining the accuracy of the decision tree working condition classification model and the LSTM neural network synchronous control model, and optimizing and improving the model.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of crane control, in particular to a crane double-hoist synchronous control method. BACKGROUND

[0002] In the prior art, the working conditions of the crane double-hoist are analyzed in detail, and a unified control mode is adopted. However, there are deficiencies in analyzing the particularity of the double-hoist synchronous control under different working conditions. The synchronous requirements of the double-hoist are different, and the classification control is not performed. The synchronous control effect is poor, which affects the stability and operation efficiency of the crane. The synchronous speed error and the synchronous displacement error cannot be monitored and adjusted in real time and accurately. The error exceeds the expected control accuracy. The pre-trained decision tree working condition classification model and the LSTM neural network model trained for multiple working conditions are not used. The model is not optimized and improved according to real-time data. The control model is difficult to adapt to complex and variable working conditions. With the change of working conditions and the operation of equipment, the control effect gradually deteriorates. The long-term stable synchronous control cannot be guaranteed. The control is performed according to the preset parameters. The actual synchronous error cannot be adjusted in time, which leads to poor control effect. The problems existing in the model cannot be found and corrected in time, which affects the reliability and stability of the system.

[0003] Therefore, the present application provides a crane double-hoist synchronous control method. SUMMARY

[0004] The present application aims to provide a crane double-hoist synchronous control method to solve at least one of the above-mentioned problems in the prior art.

[0005] A crane double-hoist synchronous control method, comprising the following steps:

[0006] By collecting the working condition data of the crane double-hoist in real time, the speed synchronization error and the displacement error of the crane double-hoist are calculated. The working condition data of the crane double-hoist is input into the pre-trained decision tree working condition classification model, and the working condition type is output.

[0007] According to the working condition category, the corresponding synchronous control model is called based on the LSTM model trained for multiple working conditions.

[0008] Based on the monitored synchronous speed error and the synchronous displacement error, the error curve is drawn, and the control effect value is calculated to analyze and judge the size of the synchronous speed error and the expected control accuracy. The size of the synchronous displacement error and the expected control accuracy is analyzed and judged.

[0009] The control parameter gradient is calculated and analyzed, and the control parameter is fine-tuned. The accuracy of the decision tree working condition classification model and the LSTM model is re-considered, and the model is optimized and improved.

[0010] The beneficial effects of the present application are:

[0011] 1. By collecting the working condition data of the double hoist of the crane in real time, the running state of the system is obtained, and the real-time accurate control of the double hoist system is realized; the working condition data and the real-time error are spliced into a feature vector to improve the decision accuracy of the model; the working condition data and the real-time error are spliced into a feature vector to improve the decision accuracy of the model, the corresponding LSTM model handle is obtained by querying the model library through the hash table, and the appropriate model is quickly located and called; based on the time sequence input matrix of the real-time synchronization error, the dynamic change trend of the error is analyzed, the predicted error change rate and the system dynamic gain G are output, and the parameters in the fuzzy control algorithm are dynamically corrected according to the error change trend, so that the system can automatically adjust the control parameters according to the error change trend; the proportional coefficient is adjusted according to the predicted error change rate, and the integral coefficient is adjusted in combination with the system dynamic gain G, so as to avoid over-adjustment or overshoot of the system;

[0012] 2. The proportional coefficient and the integral coefficient in the fuzzy control algorithm are adjusted based on gradient calculation; the control parameters are optimized, the error data stages before and after adjustment are divided and detailed statistical analysis is performed, the error curve is drawn, and the error change trend is displayed; the sample coverage rate is calculated, and the sample insufficient working condition category data is supplemented in time to ensure the adaptability of the model to various working conditions; the threshold and the decision tree branch condition are adjusted according to the classification accuracy to improve the accuracy of model classification, whether to prune is determined according to the gain rate in the decision tree growth process, the model overfitting is prevented, the model weight variance is calculated, the L2 regularization method is used to optimize the weight distribution, and overfitting is avoided. BRIEF DESCRIPTION OF DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0014] Figure 1 is a step flow chart of a crane double hoist synchronization control method provided by embodiment 1 of the present application;

[0015] Figure 2 is a step flow chart of a crane double hoist synchronization control method provided by embodiment 2 of the present application. DETAILED DESCRIPTION

[0016] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only 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 ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention. Example 1

[0017] like Figure 1 As shown, an embodiment of the present invention provides a method for synchronously controlling dual winches of a crane, which specifically includes the following steps:

[0018] Step 1: By collecting the working condition data of the crane's dual winches in real time, the speed synchronization error and displacement error of the crane's dual winches are calculated. The working condition data of the crane's dual winches is input into a pre-trained decision tree working condition classification model, and the working condition type is output. Based on the working condition type, the corresponding synchronization control model is called based on the LSTM neural network model trained for multiple working conditions.

[0019] It should be noted that the working condition data refers to the working status data collected in real time when the crane double winches are working, including but not limited to: winch speed , winch displacement , cargo weight , ambient wind speed and ambient temperature ;

[0020] In some specific embodiments, the working condition data of the double hoist of the crane is obtained, and the hoisting speed of the double hoist of the crane is marked as and , the winch displacement of the crane double winch is marked as and , based on the obtained winch speed and Calculate the difference and take the absolute value to get the speed synchronization error ;

[0021] Based on the obtained winch speed and Calculate the difference and take the absolute value to get the speed synchronization error ;

[0022] Received cargo weight , ambient wind speed and ambient temperature Input the pre-trained decision tree working condition classification model and output the working condition category;

[0023] Specifically, the weight of the cargo According to the weight threshold, it is divided into three types: light load, normal load and heavy load. According to the wind speed threshold, it is divided into three types: weak wind, normal wind and strong wind. According to the temperature threshold, it is divided into three types: low temperature, normal temperature and high temperature;

[0024] It should be noted that the weight threshold, wind speed threshold, and temperature threshold are classification thresholds preset by professionals in the present invention based on historical experience and can be adjusted according to actual conditions;

[0025] The decision tree classification model is used to output discrete working condition labels, and each working condition label corresponds to a unique working condition code;

[0026] For example, working condition -01 corresponds to normal load - normal wind speed, and working condition -10 corresponds to heavy load - normal temperature;

[0027] A key-value pair database is used, where the key is the working condition code and the value is the binary file path of the pre-trained LSTM model;

[0028] Synchronize the working condition data with the real-time error speed error and speed synchronization error Concatenate into feature vectors;

[0029] Call the decision tree classification interface to obtain the working condition code, query the model library through the hash table, and obtain the handle of the corresponding LSTM model;

[0030] Collected in real time 、 , W, v and T are normalized according to the mean and variance corresponding to the working condition category; the data of the latest k=50 time steps are retained to form a time series input matrix;

[0031] Based on the real-time synchronous speed error Synchronous displacement error The time series input matrix is ​​used to analyze the dynamic change trend of the error and output the predicted error change rate. and system dynamic gain G;

[0032] Based on the obtained prediction error change rate and system dynamic gain G, it is used to adjust the parameters in the fuzzy control algorithm. and Make dynamic corrections;

[0033] According to the forecast error change rate, analyze and judge the trend of the forecast error change rate. If the forecast error change rate increases, it indicates that there is a risk of further increase in the error, and the proportional coefficient should be increased. , speed up the system's response to the error and enhance the control effect; if the predicted error change rate decreases, it can be reduced , prevent the system from over-adjusting;

[0034] For the integral coefficient , adjust in combination with the system dynamic gain G;

[0035] Compare the system dynamic gain G with the preset gain threshold;

[0036] If the system dynamic gain G is greater than the preset gain threshold , it indicates that the system is sensitive to the input signal, and the , avoid the system from overshooting due to the rapid accumulation of the integral term;

[0037] If the system dynamic gain G is less than the preset gain threshold , it indicates that the system is sensitive to the input signal, and the , enhance the integral action to eliminate the steady-state error;

[0038] After the control parameter adjustment is completed, the adjusted control parameter is applied to the double-hoist synchronous control system;

[0039] The technical scheme of the embodiment of the present application is: by real-time acquisition of the working condition data of the double-hoist of the crane, the running state of the system is obtained, and real-time accurate control of the double-hoist system is realized; the speed synchronization error and the displacement error are calculated, the deviation in the running process of the double-hoist system is found, and a quantitative basis is provided for control strategy adjustment; the working condition data and the real-time error are spliced into a feature vector, the decision accuracy of the model is improved; the working condition data and the real-time error are spliced into a feature vector, the decision accuracy of the model is improved, the corresponding LSTM model handle is obtained by querying the model library through the hash table, and the appropriate model is quickly located and called; based on the time sequence input matrix of the real-time synchronization error, the dynamic change trend of the error is analyzed, the predicted error change rate and the system dynamic gain G are output, and the parameters in the fuzzy control algorithm are dynamically corrected accordingly, so that the system can automatically adjust the control parameters according to the change trend of the error; the proportional coefficient is adjusted according to the predicted error change rate, and the integral coefficient is adjusted in combination with the system dynamic gain G, to avoid over-adjustment or overshooting of the system.

[0040] Embodiment 2

[0041] As shown in Figure 1 , the embodiment of the present application provides a crane double-hoist synchronous control method

[0042] Step 2: based on the monitored synchronization speed error and synchronization displacement error, if the synchronization speed error and synchronization displacement error are less than the expected control accuracy, further fine-tune the control parameters, and re-consider the accuracy of the decision tree working condition classification model and the LSTM neural network synchronous control model, and optimize and improve the model;

[0043] It should be noted that the control parameter includes a proportional coefficient and an integral coefficient;

[0044] In some specific embodiments, the real-time synchronous speed error and the synchronous displacement error are continuously monitored, and the error data before and after adjustment are compared to evaluate the control effect;

[0045] After completing the adaptive particle swarm algorithm optimization model weight or triggering parameter retraining and applying new control parameters, the monitored error data is divided into two stages before and after adjustment in chronological order;

[0046] Based on the obtained error data before and after adjustment, the mean, variance, maximum and minimum of the error data before and after adjustment are calculated respectively;

[0047] Taking time as the horizontal axis and synchronous speed error and synchronous displacement error as the vertical axis, error curves before and after adjustment are drawn; by intuitive curve comparison, the change trend of synchronous speed error and synchronous displacement error in different stages is observed;

[0048] Based on the obtained synchronous speed error and the synchronous displacement error , the control effect value is obtained by formula , wherein and are preset proportional coefficients, is the monitoring time interval;

[0049] Based on the obtained control effect value, the gradient of the proportional coefficient is calculated by formula , and the gradient of the integral coefficient is calculated by formula ;

[0050] Based on the obtained gradient of the proportional coefficient and the gradient of the integral coefficient , the adjustment direction of the proportional coefficient and the integral coefficient is analyzed and judged;

[0051] Specifically, if the gradient of the proportional coefficient is greater than 0, the proportional coefficient is reduced, and if the gradient of the proportional coefficient is less than 0, the proportional coefficient is increased;

[0052] If the gradient of the proportional coefficient is greater than 0, the proportional coefficient is reduced, and if the gradient of the proportional coefficient is less than 0, the proportional coefficient ;

[0053] Based on the gradient descent method, the parameters are adjusted specifically, and the adjusted proportional coefficient is obtained by formula The adjusted integral coefficient is obtained by formula , wherein is the proportional coefficient before adjustment, is the integral coefficient before adjustment, is the preset learning rate;

[0054] The adjusted control parameters are applied to the crane double-hoist synchronization control system, and the performance index is recalculated , if decreases, it indicates that the adjustment direction is correct, and fine tuning is continued; if increases, the adjustment direction is changed and the learning rate is adjusted;

[0055] Based on cross-validation, the threshold boundaries of the cargo weight W, the environmental wind speed v and the environmental temperature T are verified, and the classification accuracy is calculated by formula , wherein TP is the true positive, TN is the true negative, FP is the false positive, and FN is the false negative;

[0056] Exemplarily, the data close to the weight threshold is divided into two groups, if the cargo weight W is less than the weight threshold, it is marked as the data small group, if the cargo weight W is greater than the weight threshold, it is marked as the data large group, the two groups of data are input into the decision tree model, and the classification accuracy is calculated to analyze and judge the rationality of the threshold; if the classification accuracy is less than the standard accuracy, the weight threshold is adjusted and the decision tree branch condition is adjusted;

[0057] After the optimization based on the threshold, during the growth process of the decision tree, the gain rate GR of the node is calculated by formula , wherein is the information gain of attribute A to sample set T, is the split information metric of attribute A to sample set T;

[0058] Based on the obtained gain rate GR of the node, if the gain rate GR is less than the preset standard gain rate, the splitting of the corresponding node is stopped and pruning is performed;

[0059] After the optimization based on the decision tree, the variance of the LSTM model weight is calculated by formula , wherein is the number of weights, is the weight, is the mean of the weights;

[0060] If the variance of the LSTM model weights If the variance is greater than the preset threshold, it indicates that the weight distribution is uneven, and the model is marked as uneven weight model, and regularization method is used for adjustment;

[0061] Based on the obtained weight imbalance model, the formula Get the optimized loss function, where is the regularization coefficient, by adjusting the regularization coefficient The value of optimizes the weight distribution;

[0062] The technical solution of the embodiment of the present invention is: by constructing a performance indicator including synchronous speed error and synchronous displacement error, and adjusting the proportional coefficient and integral coefficient in the fuzzy control algorithm based on gradient calculation; optimizing control parameters to effectively reduce errors; continuously monitoring real-time synchronous speed error and synchronous displacement error, dividing the error data stages before and after adjustment and performing detailed statistical analysis, drawing error curves, and displaying error change trends; reviewing and optimizing the decision tree working condition classification model from the two aspects of sample coverage and boundary conditions, and by calculating the sample coverage rate, timely supplementing the working condition category data with insufficient samples to ensure the adaptability of the model to various working conditions; using cross-validation to verify at the threshold boundary, adjusting the threshold and decision tree branch conditions according to the classification accuracy to improve the accuracy of model classification, and in the process of decision tree growth, deciding whether to prune according to the gain rate to prevent model overfitting, calculating the model weight variance, and using the L2 regularization method to optimize the weight distribution to avoid overfitting.

[0063] The above is a detailed description of an embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention; all equivalent changes and improvements made within the scope of application of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A crane double winch synchronization control method, characterized in that: The following steps are involved: By collecting the working condition data of the crane's dual winches in real time, the speed synchronization error and displacement error of the crane's dual winches are calculated, and the working condition data of the crane's dual winches are input into the pre-trained decision tree working condition classification model to output the working condition type; The specific method for obtaining the decision tree working condition classification model is: Obtain the working condition data of the crane double winch, and mark the winch speed of the crane double winch as and , the winch displacement of the crane double winch is marked as and , based on the obtained winch speed and Calculate the difference and take the absolute value to get the speed synchronization error ; Based on the obtained winch speed and Calculate the difference and take the absolute value to get the speed synchronization error ; Received cargo weight , ambient wind speed and ambient temperature Input the pre-trained decision tree working condition classification model and output the working condition category; According to the working condition category, the corresponding synchronous control model is called based on the LSTM model trained for various working conditions; The specific method for obtaining the LSTM model is: The decision tree classification model is used to output discrete working condition labels, and each working condition label corresponds to a unique working condition code; A key-value pair database is used, where the key is the working condition code and the value is the binary file path of the pre-trained LSTM model; Synchronize the working condition data with the real-time error speed error and speed synchronization error Splice into feature vector; Call the decision tree classification interface to obtain the working condition code, query the model library through the hash table, and obtain the handle of the corresponding LSTM model; Based on the synchronous speed error and synchronous displacement error obtained through monitoring, the error curve is drawn, and the control effect value is calculated to analyze and judge the size of the synchronous speed error and the expected control accuracy. The size of the synchronous displacement error and the expected control accuracy is analyzed and judged. The control parameter gradient is calculated and analyzed and the control parameters are fine-tuned. The accuracy of the decision tree working condition classification model and the LSTM model are re-examined to optimize and improve the model.

2. A crane dual winch synchronous control method according to claim 1, characterized in that: The specific process of obtaining the control parameters is as follows: Collected in real time 、 , W, v and T are normalized according to the mean and variance corresponding to the working condition category; the data of the latest k=50 time steps are retained to form a time series input matrix; Based on the real-time synchronous speed error, Synchronous displacement error The time series input matrix is ​​used to analyze the dynamic change trend of the error and output the predicted error change rate. and system dynamic gain G; Based on the obtained prediction error change rate and system dynamic gain G, it is used to adjust the control parameters in the fuzzy control algorithm. and Make dynamic corrections.

3. A crane dual winch synchronous control method according to claim 1, characterized in that: The specific adjustment process of the control parameters is as follows: For the integral coefficient , adjusted in combination with the system dynamic gain G; Comparing the system dynamic gain G with a preset gain threshold; If the system dynamic gain G is greater than the preset gain threshold , indicating that the system is sensitive to the input signal, reducing , to avoid the system overshoot caused by the rapid accumulation of integral terms; If the system dynamic gain G is less than the preset gain threshold , indicating that the system is sensitive to the input signal and increases , enhance the integral effect and eliminate steady-state errors; After the control parameter adjustment is completed, the adjusted control parameters are applied to the dual winch synchronous control system.

4. A crane dual winch synchronous control method according to claim 1, characterized in that: The specific process of obtaining the error curve is as follows: Continuously monitor the real-time synchronous speed error and synchronous displacement error, compare the error data before and after adjustment, and evaluate the control effect; After completing the adaptive particle swarm algorithm to optimize the model weights or trigger parameter retraining and apply new control parameters, the monitored error data is divided into two stages in chronological order: before adjustment and after adjustment; Based on the obtained error data before and after adjustment, respectively calculating the mean, variance, maximum value and minimum value of the error data before and after adjustment; With time as the horizontal axis and synchronous speed error and synchronous displacement error as the vertical axes, the error curves before and after adjustment are plotted; through intuitive curve comparison, the changing trends of synchronous speed error and synchronous displacement error at different stages are observed.

5. A crane dual winch synchronous control method according to claim 1, characterized in that: The specific method for obtaining the control effect value is: Based on the obtained synchronous speed error and synchronous displacement error , through the formula Get the control effect value, where and is the preset scale factor, is the monitoring time interval.

6. A crane dual winch synchronous control method according to claim 1, characterized in that: The specific method for obtaining the control parameter gradient is: Based on the obtained control effect value, the formula Calculate the proportional coefficient The gradient of Calculate the integral coefficient gradient; Based on the obtained proportional coefficient The gradient and integral coefficient of Gradient, analyze and determine the proportional coefficient and the integral coefficient adjustment direction.

7. A crane dual winch synchronous control method according to claim 1, characterized in that: The specific optimization process of the control parameters is: The parameters are adjusted specifically based on the gradient descent method, and the formula Get the adjusted proportional coefficient through the formula Get the adjusted integral coefficient, where is the proportional coefficient before adjustment, is the integral coefficient before adjustment, is the preset learning rate; Apply the adjusted control parameters to the crane's dual winch synchronization control system, conduct real-time monitoring again, and recalculate performance indicators ,like If it decreases, it means the adjustment direction is correct and continue to make fine adjustments; if If it increases, adjust the direction and the learning rate; Based on cross-validation, the threshold boundaries of cargo weight W, ambient wind speed v and ambient temperature T are verified, and the formula Calculate the classification accuracy , where TP is a true positive example, TN is a true negative example, FP is a false positive example, and FN is a false negative example.

8. A crane dual winch synchronous control method according to claim 1, characterized in that: The specific optimization method of the model is: After the optimization based on the threshold, during the decision tree growth process, the formula Calculate the node gain rate GR, where is the information gain of attribute A on sample set T, is the split information measure of attribute A on sample set T; Based on the obtained node gain rate GR, if the gain rate GR is less than the preset standard gain rate, the corresponding node splitting is stopped and pruning is performed; After optimization based on the decision tree, the formula Calculate the variance of the LSTM model weights in, is the number of weights, For the weights, is the mean of the weights; If the variance of the LSTM model weights If the variance is greater than the preset threshold, it indicates that the weight distribution is uneven, and the model is marked as uneven weight model, and regularization method is used for adjustment; Based on the obtained weight imbalance model, the formula Get the optimized loss function, where is the regularization coefficient, by adjusting the regularization coefficient The value of optimizes the weight distribution.

Images