An AI-based multi-motor winch optimization control method
Through the AI-based multi-motor winch optimization control method, dynamically adjusting load distribution, the traditional control method is solved in the lack of load unbalanced and unexpected situations, and the efficient synchronization and stable operation of the motor system is achieved, and the production efficiency is improved. Security.
Patent Information
- Application Number
- CN202411332018.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-24
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-09-24
AI Technical Summary
Traditional multi-motor control methods lack immediate and predictive adjustment capabilities in the face of load imbalance and emergencies, resulting in poor system synchronization and coordination, affecting production efficiency and safety.
Using AI-based multi-motor winch optimization control method, collect real-time load data, analyze load proportion and distribution differences, predict load change trends, generate predicted load distribution models, and dynamically adjust the motor speed and torque to achieve load balance and optimize control.
It significantly improves the working synchronization between motors, avoids energy waste and equipment losses, ensures the stable operation of the system and efficient energy use in a changing environment, and improves the stability and efficiency of the production process.
Smart Images

Figure CN118833749B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of AI optimized motor control technology, and in particular to an AI-based multi-motor winch optimization control method. Background Art
[0002] The field of AI optimized motor control technology combines artificial intelligence technology with traditional motor control technology to achieve more efficient and precise control performance. In this technology field, the application of AI aims to optimize motor control strategies by learning and adapting to patterns discovered during the process, thereby improving the overall performance of the system. This includes the use of AI technologies such as machine learning algorithms, neural networks, and genetic algorithms to analyze and process complex data in the control process to achieve more efficient and precise control effects. This technology field is suitable for many application scenarios, such as automated manufacturing, robotics, transportation systems, etc., especially in those occasions where multiple motors need to work together.
[0003] Among them, the AI-based multi-motor winch optimization control method refers to the use of artificial intelligence technology to optimize the control strategy of the multi-motor winch system. The purpose of this method is to automatically adjust the control parameters through intelligent algorithms to improve the synchronization, stability and efficiency of the multi-motor system. In a multi-motor winch system, precise coordination between different motors is essential, and any slight synchronization deviation leads to a decrease in system efficiency or mechanical failure. Therefore, this optimization control method aims to ensure a high degree of synchronization between motors and optimize the overall system performance through the learning and adaptability of AI, so as to achieve the effects of energy saving and consumption reduction, improving production efficiency and ensuring operational safety.
[0004] Traditional multi-motor control methods cannot effectively respond to the regulation needs of load imbalance and emergencies, lack instant and predictive adjustment capabilities, resulting in poor system synchronization and coordination, and easily causing low energy efficiency and equipment damage. For example, without the support of a predictive load distribution model, it is difficult for multiple motors on a production line to work synchronously in an optimal state, resulting in reduced production efficiency, difficulty in quality control, and even production stagnation. In addition, the traditional methods are not adaptable enough when facing changes in working conditions, which aggravates the ineffective use of energy and operational risks, and significantly affects the continuity and safety of production. These shortcomings reflect the limitations of system operation efficiency and safety assurance without intelligent support. Summary of the invention
[0005] The purpose of the present invention is to solve the shortcomings existing in the prior art and to propose an AI-based multi-motor winch optimization control method.
[0006] In order to achieve the above object, the present invention adopts the following technical solution: an AI-based multi-motor winch optimization control method, comprising the following steps:
[0007] S1: Collect real-time load data of the multi-motor system, including the speed, torque, current and voltage parameters of each motor, synchronize timestamps, and generate real-time data snapshots;
[0008] S2: Based on the real-time data snapshot, analyze the load proportion of each motor, compare the difference in load distribution among multiple motors, identify the unbalanced area of load distribution, and generate a load inequality index;
[0009] S3: Analyze historical load data and current imbalance conditions according to the load inequality index, predict the load change trend in the short term in the future, derive an optimized load distribution plan, and generate a predictive load distribution model;
[0010] S4: generating a separate load adjustment instruction for each motor according to the predicted load distribution model and adjusting the speed and torque to balance the overall load and generate an adjustment strategy;
[0011] S5: Implement the adjustment strategy, issue adjustment instructions to the control system of each motor, affect the running state of the motor, monitor the response of each motor, and generate optimized motor control parameters;
[0012] S6: Using the optimized motor control parameters, the motor winch system performance is monitored again, and system operation indicators including energy efficiency, stability and response speed are collected and analyzed to generate system performance analysis results.
[0013] The real-time data snapshot specifically includes speed mark, torque record, current data, and voltage value. The load inequality index includes load share difference, load comparison between motors, and imbalance area positioning. The predictive load distribution model specifically refers to load change trend analysis and optimization distribution plan derivation. The adjustment strategy includes speed adjustment instructions, torque adjustment plan, and load balancing measures. The optimized motor control parameters specifically include adjusted speed settings, updated torque configuration, and motor response status. The system performance analysis results include energy efficiency evaluation, stability analysis, and response speed measurement.
[0014] As a further solution of the present invention, based on the real-time data snapshot, the load proportion of each motor is analyzed, the difference in load distribution among multiple motors is compared, and the unbalanced area of load distribution is identified. The steps of generating the load inequality index are specifically as follows:
[0015] S201: Based on the real-time data snapshot, calculate the actual operating load of each motor, compare the actual operating load with the rated power of the motor, set the operating efficiency standard of each motor according to the comparison result, evaluate the operating efficiency of each motor, and generate motor efficiency data;
[0016] S202: using the motor efficiency data, calculating the motor load ratio, drawing a graph of the relationship between efficiency and load ratio, identifying abnormal efficiency areas in the graph, marking motor data points with efficiencies lower than expected, analyzing the load characteristics of the motor data points, and screening inefficient motor load data;
[0017] S203: calculating the average value of the load rates of multiple motors through the load data of the inefficient motors, determining the standard deviation of the load rates, identifying motors whose deviations exceed a preset proportion of the average value as areas with uneven load distribution, making a load distribution map, and marking the areas with uneven loads as focus points in the load distribution map;
[0018] S204: Using the load distribution diagram, analyze the focus points, identify the load peak area, measure the load values of multiple areas, compare the load differences between the areas, calculate the load deviation degree, sort the areas according to the load deviation degree, select the area with the highest load and analyze the cause of the load deviation, and generate a load inequality index.
[0019] As a further solution of the present invention, for the load inequality index, historical load data and current imbalance are analyzed, the load change trend in the short term in the future is predicted, and an optimized load distribution plan is derived. The steps of generating a predictive load distribution model are specifically as follows:
[0020] S301: Based on the load inequality indicator, collect historical load data, compare current data with historical data, identify load fluctuation patterns, analyze the time point when the load peak occurs, measure load changes in differentiated time periods, and generate load fluctuation pattern data;
[0021] S302: using the load fluctuation pattern data, analyzing the fluctuation trend, identifying the load changes occurring in the short term and predicting the direction and degree, analyzing the load characteristics of the differentiated time periods according to the fluctuation pattern, measuring the historical load level in the same period, comparing the current load state with the historical data, and obtaining the short-term load change prediction data;
[0022] S303: Analyze the load distribution possibility between differentiated motors through the short-term load change prediction data, calculate the average load and peak load of each motor, compare the load bearing capacity of differentiated motors, identify the uneven load distribution, and generate preliminary load optimization solution data;
[0023] S304: Using the preliminary load optimization scheme data, refining the optimization strategy, evaluating the impact of multiple optimization strategies on load distribution, adjusting key parameters to match the predicted load changes, simulating the motor operating state under differentiated parameter settings, and generating a predicted load distribution model.
[0024] As a further solution of the present invention, according to the predicted load distribution model, a separate load adjustment instruction is generated for each motor and the speed and torque are adjusted to balance the overall load. The steps of generating the adjustment strategy are specifically as follows:
[0025] S401: Generate a separate load adjustment instruction for each motor and adjust the speed and torque according to the predicted load distribution model to balance the overall load, collect the current operating status information of multiple motors and analyze the load capacity, calculate the speed and torque value that needs to be adjusted for each motor, and generate a preliminary load adjustment instruction;
[0026] S402: Based on the preliminary load adjustment instruction, analyze the load difference between the multiple motors, use the load difference analysis technology to adjust the speed and torque parameter values, refine the multi-motor operation status data, use the load difference analysis module to identify the load imbalance between the multiple motors, and generate an optimized load adjustment instruction according to the performance characteristics and current operation status of each motor;
[0027] S403: According to the optimized load adjustment instruction, the speed and torque of multiple motors are adjusted, and the adjustment response and operating status of each motor are monitored using a real-time data feedback mechanism. The adjusted motor operating efficiency and load distribution balance are evaluated, and an adjustment strategy is generated based on the monitoring results.
[0028] As a further solution of the present invention, the load difference analysis technology adopts the formula
[0029]
[0030] in, is the improved standard deviation, is the number of motors, For the The load of the motor, is the average value of the load, For the The weight factor of each motor.
[0031] As a further solution of the present invention, the steps of implementing the adjustment strategy, issuing adjustment instructions to the control system of each motor, affecting the running state of the motor, monitoring the response of each motor, and generating optimized motor control parameters are specifically as follows:
[0032] S501: Implement the adjustment strategy, issue adjustment instructions to the control system of each motor, use real-time data acquisition to monitor the motor operation status, record the speed, temperature and current response data of the motor, and obtain preliminary motor response data;
[0033] S502: Analyze the change trend of the motor speed, temperature and current based on the preliminary motor response data, identify the data points that deviate from the expected operation mode, adjust the motor control parameters to match the expected performance standards, and obtain the adjusted motor control parameters;
[0034] S503: Based on the adjusted motor control parameters, send instructions to the motor control system again, monitor and record the changes in the motor's operating status, use PID control technology to track the motor's operating parameters in real time, analyze the deviation between the parameters and the expected targets, adjust the control strategy according to the deviation results, optimize the motor's operating performance, and obtain optimized motor control parameters.
[0035] As a further solution of the present invention, the PID control technology adopts the formula
[0036]
[0037] Calculate the optimized output of the motor control, where, is the proportional gain, is the integral gain, is the differential gain, is the deviation between the target parameter and the actual parameter, , , They are respectively the dynamic adjustment values of proportional, integral and differential gains is the newly introduced frequency response coefficient, It is the real-time change value of the motor operating frequency.
[0038] Compared with the prior art, the advantages and positive effects of the present invention are:
[0039] In the present invention, the load distribution model is predicted to actively identify and predict the load imbalance trend in a changing working environment. This active prediction and adjustment mechanism significantly optimizes the working synchronization between motors and effectively avoids energy waste and equipment loss caused by slow response. The introduction of dynamic adjustment strategy ensures the system's immediate response and adjustment to various working conditions, maintains the system's stable operation and efficient energy use, and thus significantly improves the stability and efficiency of the production process, achieving the optimization of resource use. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Figure 1 It is a schematic diagram of the main steps of the present invention;
[0041] Figure 2 It is a schematic diagram of the refinement of S1 of the present invention;
[0042] Figure 3 It is a schematic diagram of the refinement of S2 of the present invention;
[0043] Figure 4 It is a schematic diagram of the refinement of S3 of the present invention;
[0044] Figure 5 It is a schematic diagram of the refinement of S4 of the present invention;
[0045] Figure 6 It is a detailed schematic diagram of S5 of the present invention;
[0046] Figure 7 It is a detailed schematic diagram of S6 of the present invention. DETAILED DESCRIPTION
[0047] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0048] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.
[0049] Embodiment 1
[0050] See also Figure 1 The present invention provides a technical solution: an AI-based multi-motor winch optimization control method, comprising the following steps:
[0051] S1: Collect real-time load data of the multi-motor system, including the speed, torque, current and voltage parameters of each motor, synchronize timestamps, and generate real-time data snapshots;
[0052] S2: Based on real-time data snapshots, analyze the load share of each motor, compare the differences in load distribution among multiple motors, identify the unbalanced areas of load distribution, and generate load inequality indicators;
[0053] S3: Analyze the historical load data and current imbalance situation according to the load inequality index, predict the load change trend in the short term in the future, derive the optimized load distribution plan, and generate the predictive load distribution model;
[0054] S4: Generate a separate load adjustment command for each motor and adjust the speed and torque according to the predicted load distribution model to balance the overall load and generate an adjustment strategy;
[0055] S5: Implementing the adjustment strategy, issuing adjustment instructions to the control system of each motor, affecting the operating state of the motor, monitoring the response of each motor, and generating optimized motor control parameters;
[0056] S6: Using the optimized motor control parameters, the motor winch system performance is monitored again, and the system operation indicators including energy efficiency, stability and response speed are collected and analyzed to generate system performance analysis results.
[0057] The real-time data snapshots specifically include speed marks, torque records, current data, and voltage values. The load inequality indicators include load share differences, load comparisons between motors, and imbalanced area positioning. The predictive load distribution model specifically refers to load change trend analysis and optimization distribution plan derivation. The adjustment strategy includes speed adjustment instructions, torque adjustment plans, and load balancing measures. The optimized motor control parameters specifically include adjusted speed settings, updated torque configurations, and motor response status. The system performance analysis results include energy efficiency evaluation, stability analysis, and response speed measurement.
[0058] See also Figure 2 , collect the real-time load data of the multi-motor system, including the speed, torque, current and voltage parameters of each motor, synchronize the timestamp, and generate the real-time data snapshot steps as follows:
[0059] S101: Based on the real-time load data of the multi-motor system, sensors are deployed to collect the speed, torque, current and voltage parameters of each motor, and a synchronization mechanism is used to mark the timestamp of the collected data to obtain a motor parameter data set;
[0060] Based on the real-time load data of the multi-motor system, sensors are deployed to collect the speed, torque, current and voltage parameters of each motor. The Network Time Protocol (NTP) client is used to timestamp the collected data to ensure the accuracy and consistency of the time. In this process, the NTP client communicates with the NTP server and uses the Stratum hierarchy to determine the accuracy of the time source. Through the request-response mechanism, the client sends a time request to the NTP server, and the server returns the current time based on its UTC time source. The client receives the server time and calculates the round-trip delay. Finally, the local time is adjusted according to this delay to ensure that the collected data has an accurate timestamp and generate a motor parameter data set.
[0061] S102: using the motor parameter data set, checking the timestamp of each data, identifying and removing the data with wrong or missing timestamps, performing time calibration on the remaining data in the motor parameter data set, and generating a time synchronization data set;
[0062] Using the motor parameter dataset, the timestamp of each data is checked. By writing a Python script and using the dropna function and query function of the pandas library, the data with incorrect or missing timestamps are identified and removed. The dropna function is used to remove missing timestamp data, and the query function is used to filter out and remove data with incorrect timestamp formats. For time calibration, the to_datetime function of pandas is used to convert the timestamps of the remaining data into a unified date and time format. This step ensures the format consistency and accuracy of all data timestamps and generates a time-synchronized dataset.
[0063] S103: Based on the time synchronization data set, the data structure is sorted and formatted, the format of multi-motor data records is unified, a unified time tag is given to the data at each time point, the motor data is arranged in chronological order, and a real-time data snapshot is generated.
[0064] Based on the time-synchronized data set, the data structure is organized and formatted. By writing a Python script, the DataFrame object of the pandas library is used to unify the format of multi-motor data records, and a unified time label is given to the data at each time point. The sort_values function is used to arrange the motor data in chronological order. In this process, the DataFrame object is used as a data container to store the motor parameter data. Each column represents a parameter (speed, torque, current, voltage), and each row represents a data record at a time point. The sort_values function sorts all rows according to the time label column to ensure that the data is arranged in chronological order and generate a real-time data snapshot.
[0065] See also Figure 3 , based on real-time data snapshots, analyze the load share of each motor, compare the differences in load distribution among multiple motors, identify the unbalanced areas of load distribution, and generate the load inequality index in the following steps:
[0066] S201: Calculate the actual operating load of each motor based on the real-time data snapshot, compare the actual operating load with the rated power of the motor, set the operating efficiency standard of each motor according to the comparison result, evaluate the operating efficiency of each motor, and generate motor efficiency data;
[0067] Based on the real-time data snapshot, the actual operating load of each motor is calculated. The numpy library is used to perform numerical calculations by writing Python scripts to compare the actual operating load with the rated power of the motor. In this process, the mean function of the numpy library is used to calculate the average load of each motor, and the subtract function is used to calculate the difference between the actual operating load and the rated power. The operating efficiency standard of each motor is set according to the comparison results, and the operating efficiency of each motor is evaluated. The evaluation process uses conditional statements to classify the load and power comparison results of each motor into different efficiency levels to generate motor efficiency data.
[0068] S202: using the motor efficiency data, calculating the motor load ratio, drawing a graph of the relationship between efficiency and load ratio, identifying abnormal efficiency areas in the graph, marking motor data points with efficiencies lower than expected, analyzing the load characteristics of the motor data points, and screening inefficient motor load data;
[0069] The motor efficiency data is used to calculate the motor load ratio, and the efficiency and load ratio relationship chart is drawn through the Matplotlib library. In this process, the scatter function is used to draw a scatter plot to represent the efficiency and load ratio of each motor. The xlabel and ylabel functions set the X-axis and Y-axis labels of the chart respectively. The identify function is used to identify the efficiency abnormal areas in the chart. The motor data points with efficiency lower than expected are marked through logical judgment, the load characteristics of the motor data points are analyzed, and the inefficient motor load data are screened.
[0070] S203: calculating the average value of the load rates of multiple motors through the load data of the inefficient motors, determining the standard deviation of the load rates, identifying the motors whose deviations exceed a preset proportion of the average value as areas with uneven load distribution, making a load distribution map, and marking the areas with uneven loads as focus points in the load distribution map;
[0071] The average value of the load rates of multiple motors is calculated using the load data of inefficient motors. The standard deviation of the load rate is determined using the mean function of numpy. The std function of numpy is used to identify motors whose deviations exceed a preset proportion of the average value as areas with uneven load distribution. The bar function of Matplotlib is used to create a load distribution graph, and the areas with uneven loads are marked as focus points in the load distribution graph. In this process, the bar function is used to draw a bar graph to represent the load rate of each motor, and the focus points are marked in the graph using the annotate function.
[0072] S204: Use the load distribution diagram to analyze the focus points, identify the load peak area, measure the load values of multiple areas, compare the load differences between the areas, calculate the load deviation degree, sort the areas according to the load deviation degree, select the area with the highest load and analyze the cause of the load deviation, and generate a load inequality index.
[0073] Use the load distribution diagram to analyze the focus points, identify the load peak areas, measure the load values of multiple areas, compare the load differences between areas through the max function of the numpy library, calculate the degree of load deviation, use the subtract function and mean function to calculate the difference and average of the load between areas, sort the areas according to the degree of load deviation, use the argsort function of numpy to sort, select the area with the highest load and analyze the cause of load deviation, and generate a load inequality index.
[0074] See also Figure 4 ,For the load inequality index, analyze the historical load data and the current imbalance situation, predict the load change trend in the short term in the future, derive the optimized load distribution plan, and generate the predictive load distribution model in the following steps:
[0075] S301: Based on the load inequality index, collect historical load data, compare current data with historical data, identify load fluctuation patterns, analyze the time point when the load peak occurs, measure load changes in differentiated time periods, and generate load fluctuation pattern data;
[0076] Based on the load inequality index, historical load data is collected, and the historical load data of the motor operation is extracted from the database through SQL query. The current data is compared with the historical data. The data is compared using Python scripts and the pandas library. In this process, the pandas DataFrame object is used to store current and historical data, and the merge function is used to merge the DataFrame of current data and historical data. The load fluctuation pattern is identified through conditional filtering and the groupby method, and the time point when the load peak occurs is analyzed. The diff function is used to measure the load changes in differentiated time periods to generate load fluctuation pattern data.
[0077] S302: using load fluctuation pattern data, analyzing fluctuation trends, identifying load changes that occur in the short term and predicting the direction and degree, analyzing load characteristics of differentiated time periods according to the fluctuation pattern, measuring historical load levels over the same period, comparing current load status with historical data, and obtaining short-term load change prediction data;
[0078] The load fluctuation pattern data is used to analyze the fluctuation trend. The signal module in the SciPy library is used to apply filters to smooth the data, identify the load changes that occur in the short term, and predict the direction and degree. In this process, the savgol_filter function of the signal module is used to smooth the data, thereby identifying the fluctuation trend. The load characteristics of differentiated time periods are analyzed according to the fluctuation pattern, and the historical load level in the same period is measured. The current load status is compared with the historical data through the compare function of DataFrame to obtain the short-term load change prediction data.
[0079] S303: Analyze the load distribution possibility between differentiated motors through short-term load change prediction data, calculate the average load and peak load of each motor, compare the load bearing capacity of differentiated motors, identify the uneven load distribution, and generate preliminary load optimization plan data;
[0080] Through the short-term load change prediction data, the load distribution possibility between differentiated motors is analyzed, the average load and peak load of each motor are calculated, and the mean and max functions of the numpy library are used for calculation. The load bearing capacity of differentiated motors is compared, and the uneven load distribution is identified by building logical judgment to generate preliminary load optimization plan data.
[0081] S304: Using preliminary load optimization solution data, refine the optimization strategy, evaluate the impact of multiple optimization strategies on load distribution, adjust key parameters to match predicted load changes, simulate the motor operating status under differentiated parameter settings, and generate a predictive load distribution model.
[0082] Use preliminary load optimization solution data to refine the optimization strategy, evaluate the impact of multiple optimization strategies on load distribution, adjust key parameters to match the predicted load changes, and use simulation software to simulate the motor operating status under differentiated parameter settings. In this process, use Simulink or similar tools to set the motor model parameters, including load, speed, and power parameters, and generate a predictive load distribution model by running simulation experiments.
[0083] See also Figure 5 , according to the predicted load distribution model, a separate load adjustment command is generated for each motor and the speed and torque are adjusted to balance the overall load. The specific steps for generating the adjustment strategy are:
[0084] S401: Generate a separate load adjustment instruction for each motor and adjust the speed and torque according to the predicted load distribution model to balance the overall load, collect the current operating status information of multiple motors and analyze the load capacity, calculate the speed and torque value that needs to be adjusted for each motor, and generate a preliminary load adjustment instruction;
[0085] According to the predicted load distribution model, a separate load adjustment instruction is generated for each motor and the speed and torque are adjusted to balance the overall load. The current operating status information of multiple motors is collected through a real-time monitoring system, and the load capacity is analyzed using MATLAB software to calculate the speed and torque values that need to be adjusted for each motor. In this process, MATLAB's Control System Toolbox is used to design and analyze the speed and torque control system, the simulink module is used to simulate the motor operating status, and the pidtune function is used to automatically adjust the PID controller parameters to achieve the predetermined speed and torque, and generate preliminary load adjustment instructions.
[0086] S402: Based on the preliminary load adjustment instruction, analyze the load difference between the multiple motors, use the load difference analysis technology to adjust the speed and torque parameter values, refine the multi-motor operation status data, use the load difference analysis module to identify the load imbalance between the multiple motors, and generate an optimized load adjustment instruction according to the performance characteristics and current operation status of each motor;
[0087] Based on the preliminary load adjustment instructions, the load differences between multiple motors are analyzed, and the load difference analysis technology is used to adjust the speed and torque parameter values. This module uses Python scripts and numpy libraries to refine the multi-motor operating status data, and calculates the standard deviation of the multi-motor load through numpy's std function to identify load imbalance. According to the performance characteristics and current operating status of each motor, conditional judgment and loop statements are used to generate optimized load adjustment instructions.
[0088] The load difference analysis technique uses the formula
[0089]
[0090] in, is the improved standard deviation, is the number of motors, For the The load of the motor, is the average value of the load, For the The weight factor of each motor.
[0091] First, a weight coefficient is assigned to each motor by observing its operating efficiency and failure rate. , where motors with high motor efficiency and low failure rate will get higher weights, and vice versa, they will get lower weights. Then, the actual load of each motor is calculated. , and find the average value of all motor loads , then use the difference between the load of each motor and the average load, multiply it by the weight coefficient of the corresponding motor, calculate the sum of the squares of the weighted differences, and finally, divide it by the total number of motors , and take the square root to get the improved standard deviation ,To ensure the determination of the weight coefficient, it is ,through analyzing the historical operation data of the motor, including efficiency and ,failure rate, and using data analysis methods such as linear ,regression analysis, to determine the specific value of the weight coefficient of each ,motor.
[0092] S403: According to the optimized load adjustment instructions, the speed and torque of multiple motors are adjusted, and the adjustment response and operating status of each motor are monitored using a real-time data feedback mechanism. The operating efficiency and load distribution balance of the adjusted motors are evaluated, and an adjustment strategy is generated based on the monitoring results.
[0093] According to the optimized load adjustment instructions, the speed and torque of multiple motors are adjusted, and a real-time data feedback mechanism is used to monitor the adjustment response and operating status of each motor. This mechanism is implemented through PLC (Programmable Logic Controller) and SCADA (Supervisory Control and Data Acquisition System). The PLC is programmed to automatically execute load adjustment instructions. The SCADA system is used to monitor the motor status in real time and collect operating data. The collected data is analyzed through MATLAB's Data Acquisition Toolbox to evaluate the adjusted motor operating efficiency and load distribution balance. According to the monitoring results, an adjustment strategy is generated.
[0094] See also Figure 6 , implement the adjustment strategy, send adjustment instructions to the control system of each motor, affect the operating state of the motor, monitor the response of each motor, and generate the optimized motor control parameters in the following steps:
[0095] S501: Implementing the adjustment strategy, issuing adjustment instructions to the control system of each motor, using real-time data acquisition to monitor the motor operation status, recording the speed, temperature and current response data of the motor, and obtaining preliminary motor response data;
[0096] Implement adjustment strategies, send adjustment instructions to the control system of each motor, use real-time data acquisition system to monitor the motor operation status, record the motor speed, temperature and current response data, collect data in real time through SCADA system, and use Modbus protocol to transmit data to the central monitoring system. In this process, Modbus TCP is used for network communication to ensure real-time transmission and accuracy of data, so as to obtain preliminary motor response data.
[0097] S502: Analyze the change trend of motor speed, temperature and current based on the preliminary motor response data, identify data points that deviate from the expected operation mode, adjust the motor control parameters to match the expected performance standards, and obtain the adjusted motor control parameters;
[0098] Based on the preliminary motor response data, the changing trends of motor speed, temperature and current are analyzed. The collected data are analyzed using Python scripts and the pandas library. In this process, the rolling and mean functions of pandas are used to perform sliding average processing on the data to smooth the trend line. The identify_anomalies function is used to identify data points that deviate from the expected operating mode. The degree of deviation is determined by setting a threshold, and the motor control parameters are adjusted to match the expected performance standards. The tune function of the PID control algorithm is used to automatically adjust the P, I, and D parameters to obtain the adjusted motor control parameters.
[0099] S503: Based on the adjusted motor control parameters, send instructions to the motor control system again, monitor and record the changes in the motor's operating status, use PID control technology to track the motor's operating parameters in real time, analyze the deviation between the parameters and the expected targets, adjust the control strategy according to the deviation results, optimize the motor's operating performance, and obtain the optimized motor control parameters.
[0100] Based on the adjusted motor control parameters, instructions are issued to the motor control system again to monitor and record the changes in the motor's operating status. The real-time data feedback mechanism is used to receive the operating status data through the PID controller, track the motor operating parameters in real time, and analyze the deviation between the parameters and the expected targets. The deviation analysis is performed through the Simulink module of MATLAB, and the control strategy is adjusted according to the deviation results to optimize the motor's operating performance and obtain the optimized motor control parameters.
[0101] PID control technology uses the formula
[0102]
[0103] Calculate the optimized output of the motor control, where, is the proportional gain, is the integral gain, is the differential gain, is the deviation between the target parameter and the actual parameter, , , They are respectively the dynamic adjustment values of proportional, integral and differential gains is the newly introduced frequency response coefficient, It is the real-time change value of the motor operating frequency.
[0104] First, calculate the current deviation , analyze the difference between the target and the actual operating status, and then dynamically adjust , , Gain value, by introducing , , , according to the real-time operating conditions and historical performance data of the motor, fine-tune the PID control parameters to adapt to different operating environments, and then, through Multiply by the real-time change in the motor operating frequency , introduce the direct impact of motor frequency changes on PID output, ensure that the control system can respond more sensitively to small changes in motor state, and finally, combine all calculation results to obtain the optimized PID output to optimize the motor's operating performance. The specific value of can be adjusted through experiments or simulations based on the response characteristics of the motor and the working environment to ensure the best control effect.
[0105] See also Figure 7 ,Using the optimized motor control parameters, the motor winch system performance is monitored again, and the system operation indicators including energy efficiency, stability and response speed are collected and analyzed. The specific steps for generating system performance analysis results are as follows:
[0106] S601: Using the optimized motor control parameters, the motor winch system is started, and a speed sensor, a power consumption sensor, a vibration sensor, and a temperature sensor are deployed to capture speed, power consumption, vibration, and time data, which are synchronously recorded in the system log to obtain system startup performance data;
[0107] Using the optimized motor control parameters, the motor winch system is started, and the speed sensor, power consumption sensor, vibration sensor, and temperature sensor are deployed to capture the speed, power consumption, vibration, and temperature data, which are then synchronously recorded in the system log through the data acquisition system. During this process, the RS-485 communication protocol is used to ensure accurate transmission of sensor data. The data acquisition system is programmed with LabVIEW software to achieve real-time data collection and synchronous recording, ensuring that accurate and real-time system startup performance data is collected.
[0108] S602: Based on the system startup performance data, synchronously capture the motor energy consumption and vibration frequency information, use the timer to accurately measure the time interval from startup to response completion, screen the key indicators of motor stability, and obtain system operation stability analysis data;
[0109] Based on the system startup performance data, the motor energy consumption and vibration frequency information are captured synchronously, and a high-precision timer is used to accurately measure the time interval from startup to response completion. In this process, Python scripts are combined with the numpy library for data processing and analysis. The mean function of numpy is used to calculate the average energy consumption, and the fft function is used to analyze the vibration frequency, thereby screening out the key indicators of motor stability and obtaining the system operation stability analysis data.
[0110] S603: Based on the system operation stability analysis data, energy efficiency, stability and response speed are selected as key evaluation indicators, and differentiated weights are set for each indicator. The weighted scores of multiple indicators are calculated using the weights and actual measurement values to obtain the system performance analysis results.
[0111] Based on the system operation stability analysis data, energy efficiency, stability and response speed are selected as key evaluation indicators, differentiated weights are set for each indicator, and the weighted average method is used to calculate the weighted scores of multiple indicators. In this process, the SUMPRODUCT and SUM functions of Excel are used to calculate the weighted scores. The weights are preset according to the previous system performance analysis and expert opinions. The actual measured values are obtained by the above steps. This method is used to calculate the weighted score of system performance and obtain the system performance analysis results.
[0112] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.
Claims
1. An AI-based multi-motor winch optimization control method, It is characterized in that The following steps are involved: Collect real-time load data of multi-motor systems, including the speed, torque, current and voltage parameters of each motor, synchronize timestamps, and generate real-time data snapshots; Based on the real-time data snapshot, analyze the load proportion of each motor, compare the difference in load distribution between multiple motors, identify the unbalanced area of load distribution, and generate a load inequality index; The steps of the load inequality indicator are specifically as follows: Based on the real-time data snapshot, the actual operating load of each motor is calculated, the actual operating load is compared with the rated power of the motor, the operating efficiency standard of each motor is set according to the comparison result, the operating efficiency of each motor is evaluated, and the motor efficiency data is generated; Using the motor efficiency data, calculate the motor load ratio, draw a graph of the relationship between efficiency and load ratio, identify the efficiency abnormality area in the graph, mark the motor data points with lower efficiency than expected, analyze the load characteristics of the motor data points, and screen the inefficient motor load data; By using the inefficient motor load data, an average value of the load rates of multiple motors is calculated, a standard deviation of the load rates is determined, motors whose deviations exceed a preset proportion of the average value are identified as areas with uneven load distribution, a load distribution map is prepared, and the areas with uneven loads are marked as focus points in the load distribution map; Using the load distribution diagram, analyzing the focus points, identifying the load peak area, measuring the load values of multiple areas, comparing the load differences between the areas, calculating the load deviation degree, sorting the areas according to the load deviation degree, selecting the area with the highest load and analyzing the cause of the load deviation, and generating a load inequality index; According to the load inequality index, historical load data and current imbalance are analyzed, the load change trend in the short term in the future is predicted, the optimized load distribution plan is derived, and a predictive load distribution model is generated; According to the predicted load distribution model, a separate load adjustment instruction is generated for each motor and the speed and torque are adjusted to balance the overall load and generate an adjustment strategy; Implementing the adjustment strategy, issuing adjustment instructions to the control system of each motor, affecting the operating state of the motor, monitoring the response of each motor, and generating optimized motor control parameters; The optimized motor control parameters are used to monitor the performance of the motor winch system again, collect and analyze system operation indicators including energy efficiency, stability and response speed, and generate system performance analysis results.
2. According to the AI-based multi-motor winch optimization control method of claim 1, It is characterized in that The real-time data snapshot specifically includes speed mark, torque record, current data, and voltage value. The load inequality index includes load share difference, load comparison between motors, and imbalance area positioning. The predictive load distribution model specifically refers to load change trend analysis and optimization distribution plan derivation. The adjustment strategy includes speed adjustment instructions, torque adjustment plan, and load balancing measures. The optimized motor control parameters specifically include adjusted speed settings, updated torque configuration, and motor response status. The system performance analysis results include energy efficiency evaluation, stability analysis, and response speed measurement.
3. According to the AI-based multi-motor winch optimization control method of claim 1, It is characterized in that According to the load inequality index, the historical load data and the current imbalance are analyzed, the load change trend in the short term in the future is predicted, and the optimized load distribution plan is derived. The steps of generating the predicted load distribution model are as follows: Based on the load inequality index, historical load data is collected, current data is compared with historical data, load fluctuation patterns are identified, the time point when the load peak occurs is analyzed, load changes in differentiated time periods are measured, and load fluctuation pattern data is generated; The load fluctuation pattern data is used to analyze the fluctuation trend, identify the load changes that occur in the short term and predict the direction and degree, analyze the load characteristics of the differentiated time periods according to the fluctuation pattern, measure the historical load level in the same period, compare the current load state with the historical data, and obtain the short-term load change prediction data; By using the short-term load change prediction data, the load distribution possibility between the differentiated motors is analyzed, the average load and peak load of each motor are calculated, the load bearing capacity of the differentiated motors is compared, the uneven load distribution is identified, and preliminary load optimization plan data is generated; The preliminary load optimization scheme data is used to refine the optimization strategy, evaluate the impact of multiple optimization strategies on load distribution, adjust key parameters to match the predicted load changes, simulate the motor operating status under differentiated parameter settings, and generate a predictive load distribution model.
4. According to the AI-based multi-motor winch optimization control method of claim 1, It is characterized in that According to the predicted load distribution model, a separate load adjustment instruction is generated for each motor and the speed and torque are adjusted to balance the overall load. The steps of generating the adjustment strategy are specifically as follows: According to the predicted load distribution model, a separate load adjustment instruction is generated for each motor and the speed and torque are adjusted to balance the overall load, the current operating status information of multiple motors is collected and the load capacity is analyzed, the speed and torque values that need to be adjusted for each motor are calculated, and a preliminary load adjustment instruction is generated; Based on the preliminary load adjustment instruction, analyze the load difference between the multiple motors, use the load difference analysis technology to adjust the speed and torque parameter values, refine the multi-motor operation status data, use the load difference analysis module to identify the load imbalance between the multiple motors, and generate an optimized load adjustment instruction according to the performance characteristics and current operation status of each motor; According to the optimized load adjustment instructions, the speed and torque of multiple motors are adjusted, and a real-time data feedback mechanism is used to monitor the adjustment response and operating status of each motor, the adjusted motor operating efficiency and load distribution balance are evaluated, and an adjustment strategy is generated based on the monitoring results.
5. According to the AI-based multi-motor winch optimization control method of claim 4, It is characterized in that The load difference analysis technique uses the formula: ; in, is the improved standard deviation, is the number of motors, For the The load of the motor, is the average value of the load, For the The weight factor of each motor.
6. According to the AI-based multi-motor winch optimization control method of claim 1, It is characterized in that The steps of implementing the adjustment strategy, issuing adjustment instructions to the control system of each motor, affecting the running state of the motor, monitoring the response of each motor, and generating optimized motor control parameters are as follows: Implement the adjustment strategy, issue adjustment instructions to the control system of each motor, use real-time data acquisition to monitor the motor operation status, record the speed, temperature and current response data of the motor, and obtain preliminary motor response data; Based on the preliminary motor response data, analyzing the change trends of motor speed, temperature and current, identifying data points that deviate from the expected operating mode, adjusting motor control parameters to match the expected performance standards, and obtaining adjusted motor control parameters; Based on the adjusted motor control parameters, instructions are sent to the motor control system again to monitor and record changes in the motor's operating status. PID control technology is used to track the motor's operating parameters in real time, analyze the deviation between the parameters and the expected targets, adjust the control strategy according to the deviation results, optimize the motor's operating performance, and obtain optimized motor control parameters.
7. The AI-based multi-motor winch optimization control method according to claim 6, It is characterized in that The PID control technique uses the formula: ; Calculate the optimized output of the motor control, where, is the proportional gain, is the integral gain, is the differential gain, is the deviation between the target number and the actual parameter, , , They are respectively the dynamic adjustment values of proportional, integral and differential gains, is the newly introduced frequency response coefficient, It is the real-time change value of the motor operating frequency.
Citation Information
Patent Citations
Multi-motor control method and device based on particle swarm adaptation and electronic equipment
CN114362595A
