Control method and system based on variable-frequency bucket wheel machine cable reel

Abstract

The invention discloses a control method and system based on a variable-frequency bucket wheel machine cable reel. The control method comprises the steps that a feature data recognition model and a data change curve prediction model are constructed; determining real-time feature data based on the feature data recognition model, if the real-time feature data is abnormal, calculating a first abnormal coefficient, if the real-time feature data is not abnormal, calculating a second abnormal coefficient of a prediction data change curve of the real-time feature data, and analyzing the first abnormal coefficient or the second abnormal coefficient based on a preset regulation and control analysis model, determining a first regulation and control strategy or a second regulation and control strategy, obtaining data change characteristics of the real-time characteristic data and real-time operation data, and generating a comprehensive evaluation value according to the data change characteristics of the real-time characteristic data and the real-time operation data; according to the comprehensive evaluation value, whether the regulation and control parameters are optimized or not is judged, if yes, a regulation and control instruction is generated according to the optimized regulation and control parameters, and high-precision control over the cable reel is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of cable reels, and in particular to a control method and system for a variable frequency bucket wheel excavator cable reel. Background Art

[0002] The cable reel of a variable frequency bucket wheel excavator is composed of a reel, a variable frequency motor, a controller, etc., wherein the reel is the main component for winding the cable, the variable frequency motor provides power, and the controller controls the speed and start and stop parameters of the variable frequency motor. With the improvement of the precision and quality requirements of mechanical equipment, the requirements for the tension control of the cable reel are also getting higher and higher. Therefore, there is an urgent need for a control method and system based on the cable reel of a variable frequency bucket wheel excavator to achieve high-precision control of the cable reel tension. Summary of the invention

[0003] In order to solve the above technical problems, the present application provides a control method and system based on a variable frequency bucket wheel excavator cable reel, by constructing a feature data recognition model and a data change curve prediction model, the real-time feature data is determined according to the feature data recognition model and it is judged whether it is abnormal. If so, a first control strategy is generated according to the first abnormal coefficient of the real-time feature data. If not, a predicted data change curve is generated according to the data change curve prediction model and it is judged whether it is abnormal. If so, a second control strategy is generated according to the second abnormal coefficient of the predicted data change curve to ensure the accuracy and timeliness of the control strategy, and calculate the comprehensive evaluation value to timely feedback and optimize the control parameters, so as to achieve high-precision control of the cable reel and improve the control effect.

[0004] In some embodiments of the present application, a control method based on a variable frequency bucket wheel machine cable reel is provided, comprising: Obtain and analyze multiple historical operation logs to determine the corresponding historical working conditions, historical demand data, and historical feature data, and build a feature data recognition model and a data change curve prediction model; Determine the real-time feature data based on the feature data recognition model, determine whether the real-time feature data is abnormal, and if so, calculate a first abnormality coefficient, analyze the first abnormality coefficient based on a preset control analysis model, and determine a first control strategy; If not, determine the predicted data change curve of the real-time feature data based on the data change curve prediction model, judge whether the predicted data change curve has an abnormality, and if so, calculate the second abnormality coefficient, analyze the second abnormality coefficient based on the preset control analysis model, and determine the second control strategy, wherein the first control strategy and the second control strategy include multiple control parameters; Generate a control instruction according to the control parameters and obtain data change characteristics of real-time characteristic data and real-time operation data, and generate a comprehensive evaluation value according to the data change characteristics of the real-time characteristic data and the real-time operation data; Whether to optimize the control parameters is determined based on the comprehensive evaluation value. If so, a control instruction is generated based on the optimized control parameters.

[0005] In some embodiments of the present application, the method further comprises: According to the equipment location information, structural information and preset panoramic map of the variable frequency bucket wheel machine cable reel, a static simulation scene is constructed; Obtain multiple historical job logs, and extract historical demand data, historical status data, and historical operation data from each historical job log, update the status of the static simulation scenario according to the historical status data in each historical job log, and obtain the historical working condition scenario of each historical job log; Construct an initial operation simulation model for each historical operation log according to the historical working condition scenario, input the historical demand data and historical operation data into the corresponding initial operation simulation model for dynamic operation simulation, and establish a dynamic relationship diagram between the historical tension value of the current historical operation log and the historical operation data; Analyze the dynamic relationship diagram, select dynamic operation periods in which the historical fluctuation value of the historical tension value is greater than the preset fluctuation value threshold, and set them as the focus period; Obtain the historical change value of each historical operation data in each focus period; Determine whether there is a linear or nonlinear relationship between the historical change value of the same historical operation data in all focus periods and the corresponding historical fluctuation value. If so, set the corresponding historical operation data as the historical feature data in the corresponding historical job log, and calculate the feature coefficient of the historical feature data.

[0006] In some embodiments of the present application, constructing a feature data recognition model and a data change curve prediction model includes: The historical working conditions and historical demand data of the same historical operation log are used as training input data, and the historical feature data of the corresponding historical operation log is used as training output data to perform neural network training to obtain a feature data recognition model; According to the historical feature data of multiple dynamic operation nodes, a data change curve of all historical feature data of the corresponding historical operation log is established. Based on the historical working conditions, historical demand data and historical feature data as training input data and the corresponding data change curve as training output data, a neural network model is trained to obtain a data change curve prediction model.

[0007] In some embodiments of the present application, determining whether the real-time feature data is abnormal, and if so, calculating a first abnormality coefficient includes: Obtain real-time demand data and real-time status data, and determine the real-time working condition scenario based on the real-time status data; Input the real-time working condition scenario and real-time demand data into the feature data recognition model to obtain real-time feature data; Compare the real-time feature data with the corresponding standard feature data interval to obtain the real-time feature data difference; If the real-time feature data difference is greater than the preset difference threshold, it is determined that the real-time feature data is abnormal, and the real-time feature difference between the real-time feature data difference and the preset difference threshold is calculated and set as the real-time feature data is abnormal feature data; A first abnormality coefficient of the real-time dynamic operation node is generated according to the real-time feature difference of the abnormal feature data of the real-time dynamic operation node and the feature coefficient of the abnormal feature data.

[0008] In some embodiments of the present application, analyzing the first abnormal coefficient based on a preset control analysis model to determine the first control strategy includes: According to the real-time demand data and the real-time working condition scenario, a corresponding preset control analysis model is matched in a preset control analysis model database, wherein the preset control analysis model database includes a plurality of historical demand instructions and preset control analysis models corresponding to the corresponding historical working condition scenarios, and the preset control analysis model includes a plurality of preset abnormality coefficients, and each preset abnormality coefficient is associated with a corresponding preset tension deviation amount and a preset control parameter of the preset tension deviation amount; When there is an abnormality in the real-time feature data, a similarity analysis is performed on the first abnormality coefficient of the real-time dynamic operation node and the preset abnormality coefficient in the corresponding preset control analysis model, and the preset control parameter of the preset abnormality coefficient with a similarity greater than a preset similarity threshold and the largest similarity is set as the control parameter of the real-time dynamic operation node; Predict the first predicted characteristic data of the next preset dynamic operation node according to the real-time characteristic data, control parameters and the influence relationship between the corresponding historical control parameters and historical characteristic data of the real-time dynamic operation node, and calculate the difference between the first predicted characteristic data of the next preset dynamic operation node and the first predicted characteristic data of the corresponding standard characteristic data interval; If the first predicted feature data difference is greater than a preset difference threshold, calculating a first predicted feature difference between the first predicted feature data difference and the preset difference threshold and setting the corresponding first predicted feature data as abnormal feature data; Generate a first abnormality coefficient corresponding to the next preset dynamic operation node according to the first predicted feature difference of the abnormal feature data of the next preset dynamic operation node and the feature coefficient of the corresponding abnormal feature data, and perform similarity analysis with the preset abnormality coefficient in the corresponding preset control analysis model, and determine the control parameters of the next preset dynamic operation node according to the analysis results; Preset a plurality of preset dynamic operation nodes, and sequentially generate control parameters of the preset dynamic operation nodes according to the time sequence of the preset dynamic operation nodes until the first predicted characteristic data differences of the preset dynamic operation nodes are all less than the preset difference threshold; The first control strategy is constructed by sequentially generating control parameters according to the time sequence of preset dynamic operation nodes.

[0009] In some embodiments of the present application, determining whether the predicted data change curve is abnormal, and if so, calculating a second abnormality coefficient includes: If the difference of the real-time feature data is less than the preset difference threshold, it is determined that there is no abnormality in the real-time feature data; Input the real-time characteristic data, the real-time working condition scenario and the real-time demand data into the data change curve prediction model to obtain a predicted data change curve of the real-time characteristic data, wherein the predicted data change curve includes a plurality of preset dynamic operation nodes, and each preset dynamic operation node has a plurality of second predicted characteristic data; The second prediction characteristic data of each preset dynamic operation node in the prediction data change curve is compared with the corresponding standard characteristic data interval to obtain multiple second prediction characteristic data differences. If all the second prediction characteristic data differences are less than the preset difference threshold, the first control strategy is not generated; If the second predicted feature data difference is greater than the preset difference threshold, calculating the second predicted feature difference between the second predicted feature data difference and the preset difference threshold and setting the corresponding second predicted feature data as abnormal feature data; Generate a second abnormality coefficient corresponding to the preset dynamic operation node according to the second predicted characteristic difference value of the abnormal characteristic data of the same preset dynamic operation node and the characteristic coefficient corresponding to the abnormal characteristic data; The second abnormal coefficient of each preset dynamic operation node in the predicted data change curve is generated in sequence.

[0010] In some embodiments of the present application, analyzing the second abnormal coefficient based on a preset control analysis model to determine the second control strategy includes: When there is no abnormality in the real-time feature data, the second abnormality coefficient at each preset dynamic operation node in the predicted data change curve is respectively analyzed for similarity with the preset abnormality coefficient in the preset control analysis model, and the preset control parameter of the preset abnormality coefficient with the greatest similarity greater than the preset similarity threshold is set as the control parameter corresponding to the preset dynamic operation node; According to the time sequence of the preset dynamic operation nodes in the predicted data change curve, the control parameters corresponding to the preset dynamic operation nodes are used to construct a second control strategy.

[0011] In some embodiments of the present application, generating a comprehensive evaluation value according to the data change characteristics of the real-time characteristic data and the real-time operation data includes: Acquire the real-time characteristic data of the current preset dynamic operation node, compare the real-time characteristic data of the current preset dynamic operation node with the real-time characteristic data of the last preset dynamic operation node, and obtain the data change characteristics of the real-time characteristic data at the current preset dynamic operation node, wherein the data change characteristics include the data change trend and the data change amount; Generate a first evaluation value of the current preset dynamic operation node according to the data change characteristics of the real-time characteristic data at the current preset dynamic operation node and the corresponding characteristic coefficient; The real-time demand data is divided into a plurality of real-time demand sub-data, and corresponding standard operation data is set according to each real-time demand sub-data and mapped to a corresponding preset dynamic operation node; Obtain the real-time operation data of the current preset dynamic operation node, and compare it with the corresponding standard operation data to obtain the difference of the real-time operation data; Generate a second evaluation value according to the real-time operation data difference at the current preset dynamic operation node and the corresponding characteristic coefficient; Generate a comprehensive evaluation value of the current preset dynamic operation node according to the first evaluation value and the second evaluation value; The calculation formula of the comprehensive evaluation value is: ; Among them, H is the comprehensive evaluation value, g1 is the first evaluation conversion coefficient, p1 is the weight coefficient of the first evaluation value, is the selection coefficient of the ith real-time feature data. When the data change trend of the ith real-time feature data is a positive change trend, , when the data change trend of the i-th real-time feature data is a negative change trend, , bi is the data change value of the i-th real-time feature data, t1i is the feature coefficient of the i-th real-time feature data, g2 is the second evaluation conversion coefficient, p2 is the weight coefficient of the second evaluation value, is the difference of the sth real-time operation data, and t2s is the weight coefficient of the sth real-time operation data.

[0012] In some embodiments of the present application, whether to optimize the control parameters is determined according to the comprehensive evaluation value, and if so, generating a control instruction according to the optimized control parameters includes: Presetting a comprehensive evaluation value threshold; If the comprehensive evaluation value is less than the preset comprehensive evaluation value threshold, the comprehensive evaluation value difference between the comprehensive evaluation value and the preset comprehensive evaluation value threshold is calculated, and the control parameters of the next preset dynamic operation node of the current preset dynamic operation node are optimized according to the comprehensive evaluation value difference, and the control instructions are generated according to the optimized control parameters; If the comprehensive evaluation value is greater than the preset comprehensive evaluation value threshold, the control parameters will not be optimized.

[0013] In some embodiments of the present application, a control system based on a variable frequency bucket wheel machine cable reel is also included: An acquisition module is used to acquire and analyze multiple historical operation logs, determine the corresponding historical working conditions, historical demand data, and historical feature data, and build a feature data recognition model and a data change curve prediction model; A first judgment module is used to determine the real-time feature data based on the feature data recognition model, determine whether the real-time feature data is abnormal, and if so, calculate a first abnormality coefficient, analyze the first abnormality coefficient based on a preset control analysis model, and determine a first control strategy; A second judgment module is used to determine the predicted data change curve of the real-time feature data based on the data change curve prediction model, and determine whether the predicted data change curve is abnormal. If yes, calculate the second abnormal coefficient, analyze the second abnormal coefficient based on the preset control analysis model, and determine the second control strategy, wherein the first control strategy and the second control strategy include multiple control parameters; A generation module is used to generate control instructions according to the control parameters and obtain data change characteristics of real-time feature data and real-time operation data, and generate a comprehensive evaluation value according to the data change characteristics of the real-time feature data and the real-time operation data; The optimization module is used to determine whether to optimize the control parameters according to the comprehensive evaluation value, and if so, generate control instructions according to the optimized control parameters.

[0014] A control method and system based on a variable frequency bucket wheel machine cable reel in an embodiment of the present application has the following beneficial effects compared with the prior art: By constructing a feature data recognition model and a data change curve prediction model, the real-time feature data is determined according to the feature data recognition model and it is judged whether it is abnormal. If so, a first control strategy is generated according to the first abnormality coefficient of the real-time feature data. If not, a predicted data change curve is generated according to the data change curve prediction model and it is judged whether it is abnormal. If so, a second control strategy is generated according to the second abnormality coefficient of the predicted data change curve to ensure the accuracy and timeliness of the control strategy, calculate the comprehensive evaluation value, timely feedback and optimize the control parameters, so as to achieve high-precision control of the cable reel and improve the control effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of a control method based on a variable frequency bucket wheel machine cable reel in an embodiment of the present application; Figure 2 It is a schematic diagram of a control system based on a variable frequency bucket wheel excavator cable reel in an embodiment of the present application. DETAILED DESCRIPTION

[0016] The specific implementation methods of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.

[0017] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0018] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0019] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0020] like Figure 1 As shown, a control method for a cable reel based on a variable frequency bucket wheel machine according to an embodiment of the present application includes: Step S101: obtaining and analyzing multiple historical operation logs, determining corresponding historical working conditions, historical demand data, and historical feature data, and constructing a feature data recognition model and a data change curve prediction model; Step S102: determining the real-time feature data based on the feature data recognition model, judging whether the real-time feature data is abnormal, and if so, calculating a first abnormality coefficient, analyzing the first abnormality coefficient based on a preset control analysis model, and determining a first control strategy; Step S103: If not, determine the predicted data change curve of the real-time feature data based on the data change curve prediction model, and judge whether the predicted data change curve has an abnormality. If so, calculate the second abnormality coefficient, analyze the second abnormality coefficient based on the preset control analysis model, and determine the second control strategy, wherein the first control strategy and the second control strategy include multiple control parameters; Step S104: Generate a control instruction according to the control parameters and obtain data change characteristics of the real-time feature data and the real-time operation data, and generate a comprehensive evaluation value according to the data change characteristics of the real-time feature data and the real-time operation data; Step S105: judging whether to optimize the control parameters according to the comprehensive evaluation value, and if so, generating control instructions according to the optimized control parameters.

[0021] In this embodiment, the historical working condition scenario refers to the operating status of the cable reel when receiving different historical operation instructions, including but not limited to the reel status, cable collection or release status of the cable reel, etc. The historical demand data refers to multiple demand data corresponding to the historical operation instructions, including but not limited to operation limit time, motion path planning data, target torque value, etc. The operation path planning data refers to the vehicle moving path, speed change curve, etc. The historical feature data refers to the operation data that affects the tension value under different historical operation instructions and corresponding historical operating conditions.

[0022] In this embodiment, real-time operation data refers to data that can evaluate operation efficiency, operation completion and operation operation. A comprehensive evaluation value is obtained by comprehensive evaluation of data change characteristics and real-time operation data. According to the comprehensive evaluation value, timely feedback is provided and corresponding control parameters are optimized to improve control accuracy and control efficiency.

[0023] In this embodiment, a feature data recognition model is constructed based on historical operating scenarios and historical feature data, laying the foundation for the subsequent identification of real-time feature data of real-time operating scenarios, improving the tension control accuracy of the cable reel, accurately judging whether the cable reel is in a constant tension state, and performing timely regulation to ensure the stable operation of the cable reel.

[0024] In some embodiments of the present application, the method further comprises: According to the equipment location information, structural information and preset panoramic map of the variable frequency bucket wheel machine cable reel, a static simulation scene is constructed; Obtain multiple historical job logs, and extract historical demand data, historical status data, and historical operation data from each historical job log, update the status of the static simulation scenario according to the historical status data in each historical job log, and obtain the historical working condition scenario of each historical job log; Construct an initial operation simulation model for each historical operation log according to the historical working condition scenario, input the historical demand data and historical operation data into the corresponding initial operation simulation model for dynamic operation simulation, and establish a dynamic relationship diagram between the historical tension value of the current historical operation log and the historical operation data; Analyze the dynamic relationship diagram, select dynamic operation periods in which the historical fluctuation value of the historical tension value is greater than the preset fluctuation value threshold, and set them as the focus period; Obtain the historical change value of each historical operation data in each focus period; Determine whether there is a linear or nonlinear relationship between the historical change value of the same historical operation data in all focus periods and the corresponding historical fluctuation value. If so, set the corresponding historical operation data as the historical feature data in the corresponding historical job log, and calculate the feature coefficient of the historical feature data.

[0025] In this embodiment, the static simulation scenario refers to the scenario simulation of the equipment connection structure, layout method and surrounding operating conditions involved in the variable frequency bucket wheel excavator cable reel, and the status update refers to combining the historical status data in the historical operation log with the static simulation scenario to obtain the actual operating status of the cable reel corresponding to the historical operation log before the operation.

[0026] In this embodiment, when there is a linear or nonlinear relationship between the historical change value and the corresponding historical fluctuation value, it means that the change of the corresponding historical operation data has a greater impact on the fluctuation of the tension, so the corresponding historical operation data is set as the historical characteristic data, which lays the foundation for the subsequent judgment of whether the tension is constant, and improves the high-precision regulation strategy, control accuracy and control efficiency.

[0027] In this embodiment, the characteristic coefficient refers to the degree of influence of historical characteristic data on tension changes. When the historical fluctuation value of the historical tension value fluctuates greatly with the increase of the historical change value of the historical characteristic data, the corresponding characteristic coefficient is larger, and vice versa.

[0028] In some embodiments of the present application, constructing a feature data recognition model and a data change curve prediction model includes: The historical working conditions and historical demand data of the same historical operation log are used as training input data, and the historical feature data of the corresponding historical operation log is used as training output data to perform neural network training to obtain a feature data recognition model; According to the historical feature data of multiple dynamic operation nodes, a data change curve of all historical feature data of the corresponding historical operation log is established. Based on the historical working conditions, historical demand data and historical feature data as training input data and the corresponding data change curve as training output data, a neural network model is trained to obtain a data change curve prediction model.

[0029] In this embodiment, by determining the historical characteristic data under different historical operating conditions and historical demand data, and constructing a data change curve prediction model, the foundation is laid for the subsequent determination of real-time characteristic data and the predicted data change curve of real-time characteristic data in future time periods, thereby determining the first control strategy and improving the tension control accuracy of the cable reel.

[0030] In some embodiments of the present application, determining whether the real-time feature data is abnormal, and if so, calculating a first abnormality coefficient includes: Obtain real-time demand data and real-time status data, and determine the real-time working condition scenario based on the real-time status data; Input the real-time working condition scenario and real-time demand data into the feature data recognition model to obtain real-time feature data; Compare the real-time feature data with the corresponding standard feature data interval to obtain the real-time feature data difference; If the real-time feature data difference is greater than the preset difference threshold, it is determined that the real-time feature data is abnormal, and the real-time feature difference between the real-time feature data difference and the preset difference threshold is calculated and set as the real-time feature data is abnormal feature data; A first abnormality coefficient of the real-time dynamic operation node is generated according to the real-time feature difference of the abnormal feature data of the real-time dynamic operation node and the feature coefficient of the abnormal feature data.

[0031] In this embodiment, by analyzing the real-time feature data, a corresponding analysis mode is selected according to whether the real-time feature data is abnormal. When the real-time feature data is abnormal, the first predicted feature data of the next preset dynamic operation node is predicted based on the abnormal feature data in the real-time feature data and the first abnormal coefficient, thereby obtaining the first control strategy, improving the accuracy and timeliness of the control parameters in the first control strategy, and ensuring that the cable reel is in a constant tension state.

[0032] In some embodiments of the present application, analyzing the first abnormal coefficient based on a preset control analysis model to determine the first control strategy includes: According to the real-time demand data and the real-time working condition scenario, a corresponding preset control analysis model is matched in a preset control analysis model database, wherein the preset control analysis model database includes a plurality of historical demand instructions and preset control analysis models corresponding to the corresponding historical working condition scenarios, and the preset control analysis model includes a plurality of preset abnormality coefficients, and each preset abnormality coefficient is associated with a corresponding preset tension deviation amount and a preset control parameter of the preset tension deviation amount; When there is an abnormality in the real-time feature data, a similarity analysis is performed on the first abnormality coefficient of the real-time dynamic operation node and the preset abnormality coefficient in the corresponding preset control analysis model, and the preset control parameter of the preset abnormality coefficient with a similarity greater than a preset similarity threshold and the largest similarity is set as the control parameter of the real-time dynamic operation node; Predict the first predicted characteristic data of the next preset dynamic operation node according to the real-time characteristic data, control parameters and the influence relationship between the corresponding historical control parameters and historical characteristic data of the real-time dynamic operation node, and calculate the difference between the first predicted characteristic data of the next preset dynamic operation node and the first predicted characteristic data of the corresponding standard characteristic data interval; If the first predicted feature data difference is greater than a preset difference threshold, calculating a first predicted feature difference between the first predicted feature data difference and the preset difference threshold and setting the corresponding first predicted feature data as abnormal feature data; Generate a first abnormality coefficient corresponding to the next preset dynamic operation node according to the first predicted feature difference of the abnormal feature data of the next preset dynamic operation node and the feature coefficient of the corresponding abnormal feature data, and perform similarity analysis with the preset abnormality coefficient in the corresponding preset control analysis model, and determine the control parameters of the next preset dynamic operation node according to the analysis results; Preset a plurality of preset dynamic operation nodes, and sequentially generate control parameters of the preset dynamic operation nodes according to the time sequence of the preset dynamic operation nodes until the first predicted characteristic data differences of the preset dynamic operation nodes are all less than the preset difference threshold; The first control strategy is constructed by sequentially generating control parameters according to the time sequence of preset dynamic operation nodes.

[0033] In this embodiment, the similarity analysis refers to comparing the coefficient difference between the first abnormal coefficient and the preset abnormal coefficient. When the coefficient difference is smaller, the similarity is greater, and vice versa.

[0034] In this embodiment, the first predicted characteristic data and the corresponding first abnormal coefficient of the next preset dynamic operation node are predicted by the abnormal characteristic data and the first abnormal coefficient in the real-time characteristic data, so as to construct a first control strategy to improve the control accuracy and control efficiency of the cable reel.

[0035] In some embodiments of the present application, determining whether the predicted data change curve is abnormal, and if so, calculating a second abnormality coefficient includes: If the difference of the real-time feature data is less than the preset difference threshold, it is determined that there is no abnormality in the real-time feature data; Input the real-time characteristic data, the real-time working condition scenario and the real-time demand data into the data change curve prediction model to obtain a predicted data change curve of the real-time characteristic data, wherein the predicted data change curve includes a plurality of preset dynamic operation nodes, and each preset dynamic operation node has a plurality of second predicted characteristic data; The second prediction characteristic data of each preset dynamic operation node in the prediction data change curve is compared with the corresponding standard characteristic data interval to obtain multiple second prediction characteristic data differences. If all the second prediction characteristic data differences are less than the preset difference threshold, the first control strategy is not generated; If the second predicted feature data difference is greater than the preset difference threshold, calculating the second predicted feature difference between the second predicted feature data difference and the preset difference threshold and setting the corresponding second predicted feature data as abnormal feature data; Generate a second abnormality coefficient corresponding to the preset dynamic operation node according to the second predicted characteristic difference value of the abnormal characteristic data of the same preset dynamic operation node and the characteristic coefficient corresponding to the abnormal characteristic data; The second abnormal coefficient of each preset dynamic operation node in the predicted data change curve is generated in sequence.

[0036] In this embodiment, when there is no abnormality in the real-time characteristic data, a predicted data change curve is generated and it is determined whether there is an abnormality. The data change trend of the cable reel in the future period is predicted in advance and the control parameters are set in advance to improve the operating stability and efficiency of the cable reel. Under the premise of ensuring that the cable reel is in a constant tension state, the control efficiency and control effect are improved.

[0037] In some embodiments of the present application, analyzing the second abnormal coefficient based on a preset control analysis model to determine the second control strategy includes: When there is no abnormality in the real-time feature data, the second abnormality coefficient at each preset dynamic operation node in the predicted data change curve is respectively analyzed for similarity with the preset abnormality coefficient in the preset control analysis model, and the preset control parameter of the preset abnormality coefficient with the greatest similarity greater than the preset similarity threshold is set as the control parameter corresponding to the preset dynamic operation node; According to the time sequence of the preset dynamic operation nodes in the predicted data change curve, the control parameters corresponding to the preset dynamic operation nodes are used to construct a second control strategy.

[0038] In some embodiments of the present application, generating a comprehensive evaluation value according to the data change characteristics of the real-time characteristic data and the real-time operation data includes: Acquire the real-time characteristic data of the current preset dynamic operation node, compare the real-time characteristic data of the current preset dynamic operation node with the real-time characteristic data of the last preset dynamic operation node, and obtain the data change characteristics of the real-time characteristic data at the current preset dynamic operation node, wherein the data change characteristics include the data change trend and the data change amount; Generate a first evaluation value of the current preset dynamic operation node according to the data change characteristics of the real-time characteristic data at the current preset dynamic operation node and the corresponding characteristic coefficient; The real-time demand data is divided into a plurality of real-time demand sub-data, and corresponding standard operation data is set according to each real-time demand sub-data and mapped to a corresponding preset dynamic operation node; Obtain the real-time operation data of the current preset dynamic operation node, and compare it with the corresponding standard operation data to obtain the difference of the real-time operation data; Generate a second evaluation value according to the real-time operation data difference at the current preset dynamic operation node and the corresponding characteristic coefficient; Generate a comprehensive evaluation value of the current preset dynamic operation node according to the first evaluation value and the second evaluation value; The calculation formula of the comprehensive evaluation value is: ; Among them, H is the comprehensive evaluation value, g1 is the first evaluation conversion coefficient, p1 is the weight coefficient of the first evaluation value, is the selection coefficient of the ith real-time feature data. When the data change trend of the ith real-time feature data is a positive change trend, , when the data change trend of the i-th real-time feature data is a negative change trend, , bi is the data change value of the i-th real-time feature data, t1i is the feature coefficient of the i-th real-time feature data, g2 is the second evaluation conversion coefficient, p2 is the weight coefficient of the second evaluation value, is the difference of the sth real-time operation data, and t2s is the weight coefficient of the sth real-time operation data.

[0039] In this embodiment, the first evaluation conversion coefficient and the second evaluation conversion coefficient refer to converting the data change characteristics of the real-time feature data and the real-time operation data difference into evaluation values ​​of the same dimension, and the first evaluation conversion coefficient converts the data change characteristics of all real-time feature data at the current preset dynamic operation node into a first evaluation value, and the second evaluation conversion coefficient converts all real-time action data differences at the current preset dynamic operation node into a second evaluation value.

[0040] In this embodiment, the control effect and operation efficiency of the control parameters of the current preset dynamic operation node are obtained by calculating the comprehensive evaluation value, and the control parameters of the next preset dynamic operation node are optimized according to the comprehensive evaluation value to improve the control accuracy and control effect of the cable reel.

[0041] In some embodiments of the present application, whether to optimize the control parameters is determined according to the comprehensive evaluation value, and if so, generating a control instruction according to the optimized control parameters includes: Presetting a comprehensive evaluation value threshold; If the comprehensive evaluation value is less than the preset comprehensive evaluation value threshold, the comprehensive evaluation value difference between the comprehensive evaluation value and the preset comprehensive evaluation value threshold is calculated, and the control parameters of the next preset dynamic operation node of the current preset dynamic operation node are optimized according to the comprehensive evaluation value difference, and the control instructions are generated according to the optimized control parameters; If the comprehensive evaluation value is greater than the preset comprehensive evaluation value threshold, the control parameters will not be optimized.

[0042] In some embodiments of the present application, Figure 2 As shown, a control system based on a variable frequency bucket wheel machine cable reel is also included: An acquisition module is used to acquire and analyze multiple historical operation logs, determine the corresponding historical working conditions, historical demand data, and historical feature data, and build a feature data recognition model and a data change curve prediction model; A first judgment module is used to determine the real-time feature data based on the feature data recognition model, determine whether the real-time feature data is abnormal, and if so, calculate a first abnormality coefficient, analyze the first abnormality coefficient based on a preset control analysis model, and determine a first control strategy; A second judgment module is used to determine the predicted data change curve of the real-time feature data based on the data change curve prediction model, and determine whether the predicted data change curve is abnormal. If yes, calculate the second abnormal coefficient, analyze the second abnormal coefficient based on the preset control analysis model, and determine the second control strategy, wherein the first control strategy and the second control strategy include multiple control parameters; A generation module is used to generate control instructions according to the control parameters and obtain data change characteristics of real-time feature data and real-time operation data, and generate a comprehensive evaluation value according to the data change characteristics of the real-time feature data and the real-time operation data; The optimization module is used to determine whether to optimize the control parameters according to the comprehensive evaluation value, and if so, generate control instructions according to the optimized control parameters.

[0043] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present application. These improvements and substitutions should also be regarded as the scope of protection of the present application.

Claims

1. A control method based on a variable frequency bucket wheel machine cable reel, characterized in that: include: Obtain and analyze multiple historical operation logs to determine the corresponding historical working conditions, historical demand data, and historical feature data, and build a feature data recognition model and a data change curve prediction model; Determine the real-time feature data based on the feature data recognition model, determine whether the real-time feature data is abnormal, and if so, calculate a first abnormality coefficient, analyze the first abnormality coefficient based on a preset control analysis model, and determine a first control strategy; If not, determine the predicted data change curve of the real-time feature data based on the data change curve prediction model, judge whether the predicted data change curve has an abnormality, and if so, calculate the second abnormality coefficient, analyze the second abnormality coefficient based on the preset control analysis model, and determine the second control strategy, wherein the first control strategy and the second control strategy include multiple control parameters; Generate a control instruction according to the control parameters and obtain data change characteristics of real-time characteristic data and real-time operation data, and generate a comprehensive evaluation value according to the data change characteristics of the real-time characteristic data and the real-time operation data; Whether to optimize the control parameters is determined based on the comprehensive evaluation value. If so, a control instruction is generated based on the optimized control parameters.

2. The control method based on the variable frequency bucket wheel machine cable reel according to claim 1, characterized in that: Also includes: According to the equipment location information, structural information and preset panoramic map of the variable frequency bucket wheel machine cable reel, a static simulation scene is constructed; Obtain multiple historical job logs, and extract historical demand data, historical status data, and historical operation data from each historical job log, update the status of the static simulation scenario according to the historical status data in each historical job log, and obtain the historical working condition scenario of each historical job log; Construct an initial operation simulation model for each historical operation log according to the historical working condition scenario, input the historical demand data and historical operation data into the corresponding initial operation simulation model for dynamic operation simulation, and establish a dynamic relationship diagram between the historical tension value of the current historical operation log and the historical operation data; Analyze the dynamic relationship diagram, select dynamic operation periods in which the historical fluctuation value of the historical tension value is greater than the preset fluctuation value threshold, and set them as the focus period; Obtain the historical change value of each historical operation data in each focus period; Determine whether there is a linear or nonlinear relationship between the historical change value of the same historical operation data in all focus periods and the corresponding historical fluctuation value. If so, set the corresponding historical operation data as the historical feature data in the corresponding historical job log, and calculate the feature coefficient of the historical feature data.

3. The control method based on the variable frequency bucket wheel machine cable reel according to claim 2, characterized in that: Construct a feature data recognition model and a data change curve prediction model, including: The historical working conditions and historical demand data of the same historical operation log are used as training input data, and the historical feature data of the corresponding historical operation log is used as training output data to perform neural network training to obtain a feature data recognition model; According to the historical feature data of multiple dynamic operation nodes, a data change curve of all historical feature data of the corresponding historical operation log is established. Based on the historical working conditions, historical demand data and historical feature data as training input data and the corresponding data change curve as training output data, a neural network model is trained to obtain a data change curve prediction model.

4. The control method based on the variable frequency bucket wheel machine cable reel according to claim 3, characterized in that: Determine whether the real-time feature data is abnormal. If so, calculate the first abnormal coefficient, including: Obtain real-time demand data and real-time status data, and determine the real-time working condition scenario based on the real-time status data; Input the real-time working condition scenario and real-time demand data into the feature data recognition model to obtain real-time feature data; Compare the real-time feature data with the corresponding standard feature data interval to obtain the real-time feature data difference; If the real-time feature data difference is greater than the preset difference threshold, it is determined that the real-time feature data is abnormal, and the real-time feature difference between the real-time feature data difference and the preset difference threshold is calculated and set as the real-time feature data is abnormal feature data; A first abnormality coefficient of the real-time dynamic operation node is generated according to the real-time feature difference of the abnormal feature data of the real-time dynamic operation node and the feature coefficient of the abnormal feature data.

5. The control method based on the variable frequency bucket wheel machine cable reel according to claim 4, characterized in that: Analyzing the first abnormal coefficient based on a preset control analysis model to determine a first control strategy includes: According to the real-time demand data and the real-time working condition scenario, a corresponding preset control analysis model is matched in a preset control analysis model database, wherein the preset control analysis model database includes a plurality of historical demand instructions and preset control analysis models corresponding to the corresponding historical working condition scenarios, and the preset control analysis model includes a plurality of preset abnormality coefficients, and each preset abnormality coefficient is associated with a corresponding preset tension deviation amount and a preset control parameter of the preset tension deviation amount; When there is an abnormality in the real-time feature data, a similarity analysis is performed on the first abnormality coefficient of the real-time dynamic operation node and the preset abnormality coefficient in the corresponding preset control analysis model, and the preset control parameter of the preset abnormality coefficient with a similarity greater than a preset similarity threshold and the largest similarity is set as the control parameter of the real-time dynamic operation node; Predict the first predicted characteristic data of the next preset dynamic operation node according to the real-time characteristic data, control parameters and the influence relationship between the corresponding historical control parameters and historical characteristic data of the real-time dynamic operation node, and calculate the difference between the first predicted characteristic data of the next preset dynamic operation node and the first predicted characteristic data of the corresponding standard characteristic data interval; If the first predicted feature data difference is greater than a preset difference threshold, calculating a first predicted feature difference between the first predicted feature data difference and the preset difference threshold and setting the corresponding first predicted feature data as abnormal feature data; Generate a first abnormality coefficient corresponding to the next preset dynamic operation node according to the first predicted feature difference of the abnormal feature data of the next preset dynamic operation node and the feature coefficient of the corresponding abnormal feature data, and perform similarity analysis with the preset abnormality coefficient in the corresponding preset control analysis model, and determine the control parameters of the next preset dynamic operation node according to the analysis results; Preset a plurality of preset dynamic operation nodes, and sequentially generate control parameters of the preset dynamic operation nodes according to the time sequence of the preset dynamic operation nodes until the first predicted characteristic data differences of the preset dynamic operation nodes are all less than the preset difference threshold; The first control strategy is constructed by sequentially generating control parameters according to the time sequence of preset dynamic operation nodes.

6. The control method based on the variable frequency bucket wheel machine cable reel according to claim 5, characterized in that: Determine whether the predicted data change curve is abnormal. If so, calculate the second abnormal coefficient, including: If the difference of the real-time feature data is less than the preset difference threshold, it is determined that there is no abnormality in the real-time feature data; Input the real-time characteristic data, the real-time working condition scenario and the real-time demand data into the data change curve prediction model to obtain a predicted data change curve of the real-time characteristic data, wherein the predicted data change curve includes a plurality of preset dynamic operation nodes, and each preset dynamic operation node has a plurality of second predicted characteristic data; The second prediction characteristic data of each preset dynamic operation node in the prediction data change curve is compared with the corresponding standard characteristic data interval to obtain multiple second prediction characteristic data differences. If all the second prediction characteristic data differences are less than the preset difference threshold, the first control strategy is not generated; If the second predicted feature data difference is greater than the preset difference threshold, calculating the second predicted feature difference between the second predicted feature data difference and the preset difference threshold and setting the corresponding second predicted feature data as abnormal feature data; Generate a second abnormality coefficient corresponding to the preset dynamic operation node according to the second predicted characteristic difference value of the abnormal characteristic data of the same preset dynamic operation node and the characteristic coefficient corresponding to the abnormal characteristic data; The second abnormal coefficient of each preset dynamic operation node in the predicted data change curve is generated in sequence.

7. The control method based on the variable frequency bucket wheel machine cable reel according to claim 6, characterized in that: Analyzing the second abnormal coefficient based on the preset control analysis model to determine the second control strategy includes: When there is no abnormality in the real-time feature data, the second abnormality coefficient at each preset dynamic operation node in the predicted data change curve is respectively analyzed for similarity with the preset abnormality coefficient in the preset control analysis model, and the preset control parameter of the preset abnormality coefficient with the greatest similarity greater than the preset similarity threshold is set as the control parameter corresponding to the preset dynamic operation node; According to the time sequence of the preset dynamic operation nodes in the predicted data change curve, the control parameters corresponding to the preset dynamic operation nodes are used to construct a second control strategy.

8. The control method based on the variable frequency bucket wheel machine cable reel according to claim 7, characterized in that: Generate a comprehensive evaluation value based on the data change characteristics of real-time feature data and real-time operation data, including: Acquire the real-time characteristic data of the current preset dynamic operation node, compare the real-time characteristic data of the current preset dynamic operation node with the real-time characteristic data of the last preset dynamic operation node, and obtain the data change characteristics of the real-time characteristic data at the current preset dynamic operation node, wherein the data change characteristics include the data change trend and the data change amount; Generate a first evaluation value of the current preset dynamic operation node according to the data change characteristics of the real-time characteristic data at the current preset dynamic operation node and the corresponding characteristic coefficient; The real-time demand data is divided into a plurality of real-time demand sub-data, and corresponding standard operation data is set according to each real-time demand sub-data and mapped to a corresponding preset dynamic operation node; Obtain the real-time operation data of the current preset dynamic operation node, and compare it with the corresponding standard operation data to obtain the difference of the real-time operation data; Generate a second evaluation value according to the real-time operation data difference at the current preset dynamic operation node and the corresponding characteristic coefficient; Generate a comprehensive evaluation value of the current preset dynamic operation node according to the first evaluation value and the second evaluation value; The calculation formula of the comprehensive evaluation value is: ； Among them, H is the comprehensive evaluation value, g1 is the first evaluation conversion coefficient, p1 is the weight coefficient of the first evaluation value, is the selection coefficient of the ith real-time feature data. When the data change trend of the ith real-time feature data is a positive change trend, , when the data change trend of the i-th real-time feature data is a negative change trend, , bi is the data change value of the i-th real-time feature data, t1i is the feature coefficient of the i-th real-time feature data, g2 is the second evaluation conversion coefficient, p2 is the weight coefficient of the second evaluation value, is the difference of the sth real-time operation data, and t2s is the weight coefficient of the sth real-time operation data.

9. The control method based on the variable frequency bucket wheel machine cable reel according to claim 8, characterized in that: Whether to optimize the control parameters is determined based on the comprehensive evaluation value. If so, a control instruction is generated based on the optimized control parameters, including: Presetting a comprehensive evaluation value threshold; If the comprehensive evaluation value is less than the preset comprehensive evaluation value threshold, the comprehensive evaluation value difference between the comprehensive evaluation value and the preset comprehensive evaluation value threshold is calculated, and the control parameters of the next preset dynamic operation node of the current preset dynamic operation node are optimized according to the comprehensive evaluation value difference, and the control instructions are generated according to the optimized control parameters; If the comprehensive evaluation value is greater than the preset comprehensive evaluation value threshold, the control parameters will not be optimized.

10. A control system based on a variable frequency bucket wheel machine cable reel, characterized in that: include: An acquisition module is used to acquire and analyze multiple historical operation logs, determine the corresponding historical working conditions, historical demand data, and historical feature data, and build a feature data recognition model and a data change curve prediction model; A first judgment module is used to determine the real-time feature data based on the feature data recognition model, determine whether the real-time feature data is abnormal, and if so, calculate a first abnormality coefficient, analyze the first abnormality coefficient based on a preset control analysis model, and determine a first control strategy; A second judgment module is used to determine the predicted data change curve of the real-time feature data based on the data change curve prediction model, and determine whether the predicted data change curve is abnormal. If yes, calculate the second abnormal coefficient, analyze the second abnormal coefficient based on the preset control analysis model, and determine the second control strategy, wherein the first control strategy and the second control strategy include multiple control parameters; A generation module is used to generate control instructions according to the control parameters and obtain data change characteristics of real-time feature data and real-time operation data, and generate a comprehensive evaluation value according to the data change characteristics of the real-time feature data and the real-time operation data; The optimization module is used to determine whether to optimize the control parameters according to the comprehensive evaluation value, and if so, generate control instructions according to the optimized control parameters.

Images