Optimal control system for planetary gear reduction device based on dynamic load prediction

Abstract

The present invention discloses an optimization control system for a planetary gear reduction device based on dynamic load prediction, which belongs to the field of intelligent control technology, wherein the method includes: a real-time operation information acquisition module, which is used for dynamic monitoring to obtain real-time operation information; an initial prediction simulation load acquisition module, which is used for traversing and matching real-time power feature information in a reduction simulation database; a non-power feature visual graph acquisition module, which is used for reading a predetermined visualization strategy; a target non-power index acquisition module, which is used for analyzing the non-power visual graph to obtain the target non-power index of the target point; a target prediction value acquisition module, which is used for adjusting the target initial prediction value; and a control adjustment module, which is used for issuing an optimization control instruction to control and adjust the planetary gear reduction device. The present application solves the technical problems in the prior art of inaccurate load prediction of planetary gear reduction devices and untimely control optimization, which lead to low equipment efficiency, high failure rate and unstable operation.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to an optimization control system for a planetary gear reduction device based on dynamic load prediction. Background Art

[0002] As a common mechanical transmission device, planetary gear reduction devices are widely used in various types of industrial machinery. They have the advantages of large transmission ratio, small size and high efficiency. However, with the complexity of the operating conditions of industrial equipment, the load fluctuations and unstable working conditions faced by reduction devices have become increasingly prominent. Under high load conditions, the efficiency and life of the reduction device will be affected, and excessive load will cause mechanical failure.

[0003] At present, the optimization control technology based on load prediction has achieved initial results in the application of planetary gear reduction devices, but there are still many challenges. Existing technologies mostly rely on preset models for load prediction. This static prediction method has great limitations under complex dynamic conditions and is difficult to respond to actual working condition changes in a timely manner. Therefore, how to combine real-time power and non-power characteristic information and improve the accuracy of load prediction through dynamic monitoring and matching analysis of simulation databases has become a key problem that needs to be solved in existing technologies. Summary of the invention

[0004] The present application provides a planetary gear reduction device optimization control system based on dynamic load prediction, aiming to solve the technical problems in the prior art of inaccurate load prediction of planetary gear reduction devices and untimely control optimization, resulting in low equipment efficiency, high failure rate and unstable operation.

[0005] In view of the above problems, the present application provides an optimization control system for a planetary gear reduction device based on dynamic load prediction, the system comprising a real-time operation information acquisition module, used for dynamically monitoring and obtaining real-time operation information, the real-time operation information comprising real-time power characteristic information and real-time non-power characteristic information of the planetary gear reduction device; an initial prediction simulation load acquisition module, used for traversing and matching the real-time power characteristic information in a reduction simulation database to obtain an initial prediction simulation load, and screening to obtain a target initial prediction value in the initial prediction simulation load; a non-power characteristic visual graph acquisition module, used for reading a predetermined visualization strategy, and visualizing the real-time non-power characteristic information according to the predetermined visualization strategy to obtain a non-power characteristic visual graph; a target non-power index acquisition module, used for analyzing the non-power visual graph to obtain a target non-power index of a target point, wherein the target point refers to a point corresponding to the target initial prediction value; a target prediction value acquisition module, used for adjusting the target initial prediction value using the target non-power index as a weight coefficient to obtain a target prediction value; a control adjustment module, used for issuing an optimization control instruction when the target prediction value reaches a prediction value threshold, and controlling and adjusting the planetary gear reduction device based on the optimization control instruction.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] Due to the adoption of a planetary gear reduction device optimization control system based on dynamic load prediction and real-time monitoring, the technical solution of obtaining real-time power and non-power characteristic information, matching the simulation database for load prediction and combining visual analysis for adjustment, solves the technical problems in the prior art of inaccurate monitoring of the reduction device operating status, lagging load prediction and untimely control adjustment, resulting in low equipment operating efficiency and high failure rate, and achieves the technical effect of improving equipment operating efficiency, optimizing control accuracy and extending equipment service life.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic structural diagram of an optimization control system for a planetary gear reduction device based on dynamic load prediction is provided for an embodiment of the present application.

[0010] Figure 2 A schematic diagram of the execution flow of a target non-dynamic index acquisition module in an optimization control system of a planetary gear reduction device based on dynamic load prediction is provided for an embodiment of the present application.

[0011] Explanation of the reference numerals: real-time operation information acquisition module 11, initial predicted simulated load acquisition module 12, non-power characteristic visual graph acquisition module 13, target non-power index acquisition module 14, target predicted value acquisition module 15, control adjustment module 16. DETAILED DESCRIPTION

[0012] The overall idea of ​​the technical solution provided by this application is as follows:

[0013] The embodiment of the present application provides an optimization control system for a planetary gear reduction device based on dynamic load prediction. By dynamically monitoring the real-time operating information of the planetary gear reduction device, including power and non-power characteristics, load prediction analysis is performed in combination with a simulation database, and the predicted values ​​are visualized and adjusted. Optimized control instructions are performed based on the adjusted predicted values ​​to achieve intelligent optimization control of the equipment to improve its operating efficiency and stability.

[0014] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0015] Examples, such as Figure 1 As shown, the embodiment of the present application provides an optimization control system for a planetary gear reduction device based on dynamic load prediction, the system comprising:

[0016] The real-time operation information acquisition module 11 is used for dynamically monitoring and obtaining real-time operation information, wherein the real-time operation information includes real-time power characteristic information and real-time non-power characteristic information of the planetary gear reduction device.

[0017] Specifically, real-time dynamic characteristic information refers to physical quantities that directly affect the operating efficiency of the equipment, such as torque, speed, load, etc., which can reflect the dynamic state of the equipment. For example, the current speed and load of the planetary gear reducer. Real-time non-dynamic characteristic information: refers to characteristic information related to the operation of the equipment but does not directly affect the power, such as temperature, vibration, sound, wear, etc. These characteristics affect the health status and service life of the equipment.

[0018] First, install sensors to monitor the operating status of the planetary gear reduction device in real time. These sensors can include speed sensors, torque sensors, temperature sensors, vibration sensors, etc. Through these sensors, real-time dynamic characteristic information and non-dynamic characteristic information related to the operation of the equipment can be obtained respectively.

[0019] Dynamic characteristic information can be monitored by torque sensors and speed sensors. At the same time, non-dynamic characteristic information is obtained through temperature sensors, vibration sensors, etc. These real-time operation information will be transmitted to a data acquisition system and stored in the equipment's control system or cloud database for subsequent processing and analysis.

[0020] Through this module, the system can achieve accurate load prediction and fault warning, allowing the equipment to adjust before load changes or potential faults occur. This can not only improve the operating stability and efficiency of the planetary gear reducer, but also extend the life of the equipment and reduce the impact of sudden faults on production. At the same time, by combining power and non-power information, the operating status of the equipment can be more comprehensively understood, making the control system have stronger prediction and response capabilities.

[0021] The initial predicted simulated load acquisition module 12 is used to traverse and match the real-time power characteristic information in the deceleration simulation database to obtain the initial predicted simulated load, and screen to obtain the target initial predicted value in the initial predicted simulated load.

[0022] Specifically, the deceleration simulation database is a database that stores a large amount of historical simulation data, including data such as the load conditions and dynamic response of the planetary gear reduction device under different working conditions. The database can be obtained by large-scale calculation of the simulation model of the equipment, or it can be constructed through historical operation data. The initial predicted simulation load is a set of one or more sets of load data that match the current dynamic characteristic information obtained by traversing and matching the simulation database. The load data predicts the load conditions that the equipment will bear under the current working conditions. The target initial prediction value is the load prediction value that is most representative or closest to the current operating state in the initial prediction simulation load, which is selected through further screening as the final reference data.

[0023] During actual operation, the planetary gear reduction device collects real-time dynamic characteristic information such as torque, speed and power through the installed sensors. These data are continuously transmitted to the control system or data processing platform. At this time, the system will input these real-time dynamic characteristic information into a pre-built reduction simulation database.

[0024] The database contains a large amount of historical simulation data, covering the performance of the reduction device under different loads and different environments. In order to predict the equipment load under the current working conditions, the system will traverse and match these simulation data in the database to find the historical simulation cases that are most similar to the current equipment status (such as speed, torque, etc.). After finding these simulation data, the system generates an initial predicted simulation load, which is a set of simulation results under different load conditions. In this set of initial prediction data, the system will obtain a target initial prediction value based on preset filtering rules, such as selecting the load value closest to the current operating condition, or selecting the load condition that represents the greatest risk.

[0025] This module quickly and accurately predicts the load that the equipment will bear under the current working conditions by referring to historical simulation data. By screening the target initial prediction value, the accuracy of load prediction can be improved and the operation safety of the equipment can be ensured. When the equipment load is too high, the system can issue an alarm or make adjustments in advance to avoid mechanical damage caused by overload. It can effectively extend the life of the equipment, reduce production interruptions caused by sudden failures, and improve the operating efficiency of the entire system.

[0026] The non-power feature visible graph obtaining module 13 is used to read a predetermined visualization strategy, and perform visualization processing on the real-time non-power feature information according to the predetermined visualization strategy to obtain a non-power feature visible graph.

[0027] Specifically, a predetermined visualization strategy refers to a pre-established scheme or rule for presenting complex data in the form of graphs, charts, or other visual forms. Different data features and information categories require different visualization methods so that users can intuitively and clearly understand the changing trends or states of the data. Visualization strategies can include different display methods such as bar charts, line charts, heat maps, and dashboards. The non-dynamic feature visualization is the result of visualization processing and is an intuitive presentation of non-dynamic feature information. Through graphical presentation, users can quickly identify abnormal points or trends in the equipment status.

[0028] In the control system, real-time non-power characteristic information is continuously collected through sensors, including vibration, temperature, noise, lubrication status, etc. Non-power characteristic information often exists in the form of a large amount of raw data, which is difficult to analyze and judge directly, so it needs to be processed through a predetermined visualization strategy.

[0029] First, the system will select an appropriate visualization strategy based on the equipment operating conditions or user-set requirements, usually based on different types of non-dynamic data characteristics. For example, assume that the monitored non-dynamic information includes vibration and temperature data. The vibration sensor collects the vibration intensity of the equipment at different time points. The system can display the vibration intensity distribution of each position of the equipment through a heat map, and intuitively show which parts have abnormally enhanced vibration. Similarly, the temperature change of the equipment detected by the temperature sensor can use a line graph to show the temperature change trend over time. If the temperature suddenly rises at a certain moment, the line graph can clearly show the abnormal fluctuation.

[0030] Through visualization tools, a non-power feature visual graph is generated. The visual graph can not only reflect the non-power feature of the equipment in the current operating state, but also display the operating trend of the equipment by overlaying historical data. By viewing these visual graphs, users can quickly identify whether there are potential risks or abnormal points in the equipment.

[0031] The target non-power index obtaining module 14 is used to analyze the non-power visibility graph to obtain the target non-power index of the target point, wherein the target point refers to the point corresponding to the target initial prediction value.

[0032] Specifically, a target point refers to a specific physical location or component in the equipment, which is associated with the load status or predicted value of the equipment. For example, a specific component such as a gear, bearing or seal of a planetary gear reducer can be used as a target point. The target non-power index is a comprehensive indicator obtained by analyzing the non-power characteristic information of the equipment at a certain target point. The index reflects the health status or operating status of the point and is often used to evaluate the operating risk or maintenance needs of the equipment.

[0033] Analyze the status of the target point from the non-power visual graph. The non-power visual graph is a chart or heat map generated based on non-power characteristics such as vibration and temperature, showing the distribution of non-power characteristics of different parts of the equipment. Focus on the visual graph data corresponding to the target point to obtain the target non-power index.

[0034] For example, suppose the target point is a gear bearing in a reduction gear. Through the temperature sensor and vibration sensor, the system will generate a vibration heat map and temperature line graph of the bearing position. If the temperature graph shows that the bearing temperature has a significant upward trend in a short period of time, and the vibration intensity of the bearing part in the vibration heat map also increases significantly, it indicates that there is a problem of wear or insufficient lubrication at the point.

[0035] Data fusion and analysis of multiple non-power information are performed to calculate a comprehensive target non-power index. This index can be obtained by weighted calculation, for example, different weights are assigned according to the different importance of data such as temperature, vibration, and noise. Ultimately, the target non-power index can be used as a key indicator to reflect the health status of the point.

[0036] By calculating the target non-power index, the system can detect abnormal points in the operation of the equipment in a timely manner to avoid greater damage. For example, when the temperature and vibration index of a gear exceeds the set threshold, the system can issue a warning in advance and arrange for maintenance.

[0037] The target prediction value obtaining module 15 is used to adjust the target initial prediction value by using the target non-power index as a weight coefficient to obtain a target prediction value.

[0038] Specifically, the weight coefficient here is an adjustment coefficient determined by the target non-power index, which is used to correct the target initial prediction value. The weight coefficient can increase or decrease the prediction value according to the non-power index of the target point to reflect the impact of the non-power characteristics of the current working condition on the load or state. The target prediction value is the adjusted prediction result, which is a more accurate load or performance prediction value after considering the real-time non-power characteristics of the equipment. This value will be used for subsequent control decisions.

[0039] The target non-power index reflects the health status of the equipment at a specific point. If the target non-power index is high (such as excessive vibration or high temperature), it means that the equipment is not in good operating condition and the weight coefficient needs to be increased to increase the sensitivity to load prediction. On the contrary, if the target non-power index is low, it means that the equipment is in a relatively stable operating condition and the weight coefficient can be relatively reduced. The target prediction value after weight adjustment will be closer to the actual operating status of the equipment. Since the non-power status of the equipment (such as abnormal vibration or temperature change) has been taken into account at this time, the prediction value is more accurate than the initial prediction value based only on the power characteristic information, and can better reflect the actual load and operating conditions of the equipment.

[0040] The control adjustment module 16 is used to issue an optimization control instruction when the target prediction value reaches the prediction value threshold, and to control and adjust the planetary gear reduction device based on the optimization control instruction.

[0041] Specifically, the prediction value threshold refers to a safety threshold set by the system for the equipment. When the target prediction value exceeds or reaches this threshold, it means that the equipment is in a dangerous state or needs to be adjusted. The optimization control instruction is an adjustment instruction automatically generated by the system. When the target prediction value exceeds the threshold, the system issues this instruction to adjust the operating state of the equipment. Such instructions usually include operations such as reducing the speed, reducing the load, and changing the lubrication state to avoid equipment damage or reduced efficiency.

[0042] The system continuously monitors the target prediction value of the planetary gear reducer. For example, the target prediction value is 120 Nm and the preset prediction value threshold is 100 Nm. The system automatically compares these two values. When the target prediction value exceeds the prediction value threshold (e.g. 120Nm > 100 Nm), the system generates and issues an optimization control instruction to notify the equipment control system to make adjustments.

[0043] After receiving the control command, the equipment control system will make necessary adjustments to the planetary gear reduction device according to the command. For example, if the load of the equipment is too large, the control system will automatically reduce the speed of the motor to reduce the load. In addition, the system will also increase the flow of lubricating oil to reduce the heat generated by friction.

[0044] By setting the prediction value threshold, the system can promptly identify abnormal conditions of the equipment and take measures to significantly reduce the risk of equipment failure due to overload, overheating, etc., thereby improving the safety of the overall equipment.

[0045] Furthermore, the initial predicted simulated load obtaining module 12 is further configured to perform the following steps:

[0046] Read a predetermined power index, and obtain a real-time power index parameter based on the predetermined power index in the real-time power characteristic information; extract the first simulation information in the deceleration simulation database, the first simulation information including a first simulation power index parameter and a first simulation distributed load; when a first power similarity obtained by comparing the real-time power index parameter with the first simulation power index parameter meets a similarity threshold, use the first simulation distributed load as the initial predicted simulation load.

[0047] Specifically, the predetermined power index refers to a parameter standard related to the power characteristics of the planetary gear reduction device set in advance. The real-time power index parameter is a parameter corresponding to the predetermined power index extracted from the real-time power characteristic information. These parameters are used to compare and analyze the power index in the simulation data to determine whether the power condition of the current device is consistent with the simulation model. The speed simulation database is a database that stores a large amount of simulation information, including the power characteristics and load data of the device under different working conditions. These simulation data provide a reference for real-time monitoring. The first simulation information is a group of simulation data similar to the current device operating state extracted from the speed reduction simulation database, including simulated power index parameters and distributed load data. The power similarity indicates the degree of match between the real-time power index parameters of the current device and the power index parameters in the simulation database. The similarity is used to determine whether the current operating state of the device is similar to a certain simulation model. The simulated distributed load refers to the load distribution of the device at different times or different points during the simulation process. It provides an overall load prediction model to help the system perform more accurate real-time load assessment.

[0048] First, the system reads a set of pre-set power indicators, such as torque, power, and speed, etc. Next, the system extracts real-time power indicator parameters that match these pre-set power indicators from the real-time power characteristic information of the equipment.

[0049] The system extracts the first simulation information that is closest to the current working condition from the deceleration simulation database. This simulation information includes the first simulation power index parameters (such as simulated torque and speed) of the equipment under similar operating conditions and the corresponding first simulation distributed load. These simulation data are simulated under various working conditions by simulation software (such as MATLAB Simulink, ANSYS, etc.), providing the load distribution and power characteristics of the equipment under different states.

[0050] By comparing the real-time power index parameters with the first simulated power index parameters, the system calculates a power similarity. For example, by calculating the difference between the two sets of torque and speed, a similarity value is obtained. If the similarity reaches a predetermined similarity threshold (such as a similarity greater than 90%), the system considers that the current device state is very close to the simulation model. At this time, the system uses the first simulated distributed load in the simulation model as the initial predicted simulated load of the current device for subsequent prediction and control adjustments.

[0051] Furthermore, the initial predicted simulated load obtaining module 12 is further configured to perform the following steps:

[0052] The initial predicted simulated load includes a plurality of loads with point identifiers; a target load with the largest load among the plurality of loads with point identifiers is selected, and the target load corresponds to the target point; and the target load is used as the target initial predicted value.

[0053] Specifically, in the above steps, the system obtains the initial predicted simulated load from the simulation database. In the simulated load information, each load will be accompanied by a corresponding point mark, indicating at which specific position or time point the load is generated. For example, different tooth surfaces of a gear bear different loads at different operating stages, and the point mark is used to distinguish the load distribution at these different positions.

[0054] The system will traverse these loads with point identification and find the one with the largest load value, which is the target load. This load often reflects the part of the equipment that is subject to the highest stress during operation.

[0055] For example, one tooth surface of a gear bears a higher load than other tooth surfaces due to local stress concentration; or during a certain period of time of the equipment, the system is subjected to the highest instantaneous load due to changes in working conditions.

[0056] After finding the maximum load, the system determines the corresponding target point, that is, the specific location or time point where the load is the largest. At this point, the maximum load value is considered to be the target initial prediction value under the current state of the equipment and is used in subsequent optimization control steps.

[0057] By screening the simulated loads at multiple points, the system can identify the maximum load that the equipment will bear during operation. This ensures that the predicted load has a high degree of accuracy and contributes to the safe operation of the equipment. After the system screens out the maximum load, it can perform real-time optimization control based on this load value to avoid overloading the equipment at certain points with large loads and extend the service life of the equipment.

[0058] Furthermore, the non-dynamic feature visible graph obtaining module 13 is also used to perform the following steps:

[0059] Acquire first characteristic information of a first non-power indicator in the real-time non-power characteristic information, wherein the first non-power indicator refers to any one of the predetermined non-power indicators, and the first characteristic information includes a plurality of parameter values ​​with point identifications; visualize the plurality of parameter values ​​with point identifications according to the predetermined visualization strategy to obtain a first visible graph; and fuse the first visible graph to obtain the non-power characteristic visible graph.

[0060] Specifically, the first non-power indicator is one of the predetermined non-power related parameters, including vibration, temperature, sound, electrical signals, etc. This indicator is used to evaluate the operating status of the equipment. For example, temperature, as one of the non-power indicators, can be used to monitor the temperature changes of the equipment. The first characteristic information is specific data related to the first non-power indicator, including multiple parameter values ​​with point identifiers. The point identifier refers to the equipment location, component or time point corresponding to the non-power characteristic data, which is used to distinguish the measurement data at different locations or times. The first visual graph is a visualization graph generated based on the first non-power indicator, showing multiple characteristic information with point identifiers.

[0061] First, a first non-power indicator is selected from the real-time non-power feature information of the equipment. For example, if temperature is selected as the non-power indicator, the system will obtain temperature-related parameter data from temperature sensors at different locations. These data will be accompanied by point identifiers, such as the temperature data of each gear contact point of the equipment. Using the pre-set visualization strategy, the collected multiple parameter values ​​with point identifiers are converted into graphical data. After completing the visualization processing, the system generates the first visual graph, which intuitively shows the status of a non-power indicator of the equipment.

[0062] The first visual graphs of multiple non-power indicators are further fused together to generate a more comprehensive non-power feature visual graph. For example, the visual graphs of three different non-power indicators, temperature, vibration and sound, are fused to finally generate an integrated graph, which can simultaneously observe the changes in these parameters. In this way, the system can monitor the status of the equipment more comprehensively and detect potential problems in a timely manner.

[0063] By visualizing the non-power characteristics of the equipment, the intuitiveness and operability of the equipment operation status are improved. The equipment status no longer relies solely on complex data analysis, but is directly displayed in a graphical way, allowing maintenance personnel to judge the equipment status more quickly and accurately. This not only improves the efficiency of equipment management, but also helps to find problems in time and make optimization adjustments, thereby extending the service life of the equipment, reducing maintenance costs, and improving the overall reliability and stability of the system.

[0064] Furthermore, the predetermined non-dynamic indicators include a vibration dimension indicator set, a sound dimension indicator set, a temperature dimension indicator set, an electrical dimension indicator set, a lubrication dimension indicator set and a wear dimension indicator set.

[0065] Specifically, vibration characteristics include: Vibration amplitude: load changes will cause changes in vibration amplitude, which can be monitored by vibration sensors. Frequency components: The frequency components of the vibration signal can reveal the operating status of the gear and the fluctuation of the load. Kurtosis, form factor, peak factor, etc. These statistical features can be used to describe the characteristics of the vibration signal, thereby reflecting the load status.

[0066] Acoustic characteristics include: Sound intensity: When the load increases, the sound intensity generated by the gear meshing will increase. Sound frequency: Under different load conditions, the sound frequency components emitted by the gearbox are different.

[0067] Temperature characteristics include, Oil temperature: An increase in lubricating oil temperature is a sign of increased friction due to increased load. Gear surface temperature: Changes in gear surface temperature can reflect changes in load and thermal conditions.

[0068] Electrical characteristics, if applicable, include: Motor current: Changes in the motor current reflect changes in load, since an increase in load generally results in an increase in motor current. Power consumption: The power consumption of the planetary gear reduction unit is directly related to the load.

[0069] Lubrication characteristics include, Oil pressure: Changes in the pressure of the lubricating oil can reflect changes in load, because an increase in load will cause the oil film thickness to decrease and the oil pressure to increase. Oil quality: Changes in oil quality, such as the degree of contamination, will affect the lubrication effect and thus the load-bearing capacity.

[0070] Wear characteristics include: Gear wear: The degree of gear wear will increase with increasing load.

[0071] By integrating multiple non-power dimension indicators, the monitoring accuracy of the planetary gear reducer operating status is improved. The fusion of multiple dimensions of data provides a more comprehensive and in-depth analysis of the equipment status. Not only can potential faults be discovered earlier, but it can also help achieve optimized control of the equipment, effectively extend the equipment life, reduce downtime and maintenance costs, and improve the overall operating efficiency of the system.

[0072] Further, such as Figure 2 As shown, the target non-power index obtaining module 14 is further used to perform the following steps:

[0073] The first-level neighbor nodes of the target point are obtained in the first visible graph, and the first-level neighbor nodes include multiple first-level nodes; the first node among the multiple first-level nodes is randomly extracted, and the neighbor nodes of the first node are obtained in the first visible graph, which are recorded as second-level neighbor nodes, and the second-level neighbor nodes include multiple second-level nodes; the multiple second-level nodes are sampled to obtain a first sampling set, and the first sampling set includes multiple characteristic parameters of the first non-power indicator; the multiple characteristic parameters are aggregated and analyzed to obtain a first aggregation result, and the first characteristic parameters corresponding to the first node are averaged to obtain a first target parameter; the first target parameter is aggregated and analyzed to obtain a second aggregation result, and the second aggregation result is used as the first target non-power indicator data of the first node; the first target non-power indicator data is normalized and weighted, and combined with the target non-power characteristic parameters of the target point, to obtain the target non-power index.

[0074] Specifically, the first-level neighbor node refers to the direct neighbor node of the target point, indicating the physical location or data point that has a certain relationship with the target point when the device is running. The second-level neighbor node refers to the indirect neighbor node extending from the first-level neighbor node, indicating that it has a more distant relationship with the target point, but still has a certain influence relationship.

[0075] First, by analyzing the non-dynamic characteristic visual graph of the equipment, the corresponding target point is found. Usually, it is the key part of the equipment during operation (such as the heating point or the load concentration point). Then, the system will analyze the first-level neighbor nodes directly related to the target point. These neighbor nodes have physical or data associations with the target point. For example, if a part has a high temperature, the adjacent parts will also have a trend of rising temperature.

[0076] The system randomly selects one of the multiple first-level neighbor nodes as the first node, and then further tracks the second-level neighbor nodes of the node, which are located in a more distant area. Taking temperature as an example, if the temperature of a gear is abnormally high, the system will first check the temperature of its directly adjacent components (first-level neighbor nodes), and then check the adjacent components of the adjacent components (i.e., second-level neighbor nodes) to determine whether the temperature increase has a tendency to spread.

[0077] A sampling set is extracted from the secondary neighbor nodes, and multiple non-dynamic characteristic parameters (such as temperature, vibration, sound, etc.) of these nodes are collected. Next, the system aggregates and analyzes these parameters to obtain a comprehensive result (the first aggregate result), and calculates the mean value with the characteristic parameters of the first node to obtain a more accurate first target parameter.

[0078] The first target parameter is aggregated again to generate a second aggregate result. This result will be used as the target non-power indicator data of the first node to further help determine the state of the target point. Through this hierarchical analysis, the non-power state of the entire equipment system can be evaluated more accurately.

[0079] In order to ensure the rationality between different data, the non-power index data of each node is normalized, and weighted calculation is performed according to the weight of its overall impact on the equipment to obtain a more reasonable target non-power index.

[0080] Through hierarchical, random sampling and aggregate analysis methods, potential problems in equipment operation can be effectively captured, especially interrelated problems in complex systems. Multi-level analysis methods can be more comprehensive and identify the hidden fault propagation between different parts of the equipment. At the same time, normalized weighted calculation ensures the accuracy of the analysis results, improves the reliability of fault warning and the stability of equipment operation.

[0081] Further, it is determined whether the first characteristic parameter is within a predetermined threshold, and if not, the first node is removed from the first-level neighbor nodes.

[0082] Specifically, it is determined whether the first characteristic parameter of each node exceeds a predetermined threshold to determine whether to retain the node as part of further analysis. If the characteristic parameter of a node exceeds a reasonable range, it represents an abnormal point of the device and needs to be retained for further analysis. If the parameter of the node does not exceed the threshold, it means that its status is within the normal range and can be removed from the first-level neighbor nodes.

[0083] Specifically, data will first be collected for the selected first node to obtain the first characteristic parameter of the node. For example, in a gear reduction device, the first node is the bearing of a certain gear, and the characteristic parameters extracted by the system are the temperature, vibration amplitude or wear degree of the bearing. After extracting the characteristic parameters, the system will compare these parameters with the set predetermined threshold. If the parameter exceeds the threshold (for example, the temperature is too high or the vibration amplitude is too large), it means that there is an abnormality in the node, and the system will continue to retain the node for further analysis. If the parameter is within the normal range, the system determines that it does not need to be focused on.

[0084] For those nodes that do not exceed the threshold, the system removes them from the first-level neighbor node list. This reduces the amount of data in the analysis process, saves computing resources, and makes the analysis process more targeted and efficient.

[0085] Further, performing aggregation analysis on the plurality of characteristic parameters to obtain a first aggregation result includes:

[0086] The maximum parameter and the minimum parameter among the multiple characteristic parameters are obtained, and the maximum parameter deviation is calculated; if the maximum parameter deviation exceeds the deviation threshold, a test instruction is issued, and the multiple characteristic parameters are monitored for rationality test based on the test instruction.

[0087] Specifically, the maximum parameter deviation refers to the difference between the maximum and minimum values ​​of multiple characteristic parameters, reflecting the amplitude of parameter fluctuation. The deviation threshold is a pre-set standard value. If the maximum parameter deviation exceeds this threshold, the system considers that there is an abnormal fluctuation and further detection is required. Inspection instruction When the system finds that the maximum parameter deviation exceeds the set threshold, it will issue an instruction to re-inspect the rationality of these parameters to check for abnormalities or errors.

[0088] First, extract different operating status parameters of the equipment from multiple monitoring points. Perform aggregate analysis on these different characteristic parameters. Common methods include calculating the mean, maximum, minimum, and standard deviation of these parameters. For example, if temperature data is obtained from 5 sensors, these data can be averaged to obtain the overall temperature performance of the system. In addition, it is necessary to calculate the maximum and minimum parameters to further evaluate the fluctuation of the parameters.

[0089] By extracting the maximum and minimum values ​​of all characteristic parameters and calculating their difference, this is the maximum parameter deviation. Next, the system compares the calculated maximum parameter deviation with the preset deviation threshold. If the maximum parameter deviation is greater than the set threshold (for example, the threshold value is 8°C and the deviation is 10°C), the system will consider that the equipment operation status has abnormal fluctuations.

[0090] If the deviation is detected to exceed the threshold, the system will issue a test command to require further inspection of these characteristic parameters. After the test command is triggered, the rationality of these characteristic parameters is checked to analyze whether there is sensor failure, environmental interference or equipment abnormality.

[0091] Through this analysis mechanism of maximum parameter deviation and deviation threshold, abnormal fluctuations in the operating status can be quickly identified, so that inspection instructions can be issued in time for troubleshooting. This effectively improves the safety and stability of equipment operation, ensures that a quick response can be made when an abnormality occurs, and avoids potential equipment damage and production stagnation.

[0092] In summary, the planetary gear reduction device optimization control system based on dynamic load prediction provided by the embodiment of the present application has the following technical effects:

[0093] 1. The optimization control method of the planetary gear reduction device based on real-time monitoring data can accurately predict and timely adjust the operating status of the equipment through the analysis of power and non-power characteristic information. This method can effectively improve the operating efficiency and reliability of the reduction device, reduce the failure rate, extend the life of the equipment, and realize dynamic control to avoid abnormal shutdown and equipment damage caused by load fluctuations.

[0094] 2. By matching real-time power indicators and simulation data, the accuracy of load prediction can be effectively improved. Comparing real-time power indicators with data in the simulation database can ensure the reliability of the initial predicted simulated load, thereby improving the optimization control accuracy of the system and avoiding unnecessary adjustments caused by misprediction.

[0095] 3. By acquiring the characteristic parameters of the first-level and second-level neighbor nodes for sampling and aggregation analysis, the non-dynamic characteristics of the target point can be more accurately reflected. Multi-level data aggregation analysis improves the intelligence of the system, enables more detailed status monitoring and adjustment optimization, avoids the limitations of a single data point, and improves the reliability of prediction.

[0096] 4. Through the calculation and rationality test of the maximum parameter deviation, the system can identify anomalies in advance and start the inspection mechanism, improving the accuracy of equipment operation monitoring. This method can quickly respond to potential problems of the equipment, avoid misjudgments caused by data fluctuations or sensor failures, and effectively improve the safety and operation efficiency of the equipment.

[0097] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0098] Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept, and / or refer to multiple elements that can be selected individually or in full. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. The planetary gear reduction device optimization control system based on dynamic load prediction is characterized in that: include: A real-time operation information acquisition module, used for dynamically monitoring and obtaining real-time operation information, wherein the real-time operation information includes real-time power characteristic information and real-time non-power characteristic information of the planetary gear reduction device; An initial predicted simulated load acquisition module is used to traverse and match the real-time power characteristic information in the deceleration simulation database to obtain an initial predicted simulated load, and screen to obtain a target initial predicted value in the initial predicted simulated load; A non-power feature visible graph acquisition module is used to read a predetermined visualization strategy, and perform visualization processing on the real-time non-power feature information according to the predetermined visualization strategy to obtain a non-power feature visible graph; A target non-power index obtaining module, used for analyzing the non-power feature visual graph to obtain a target non-power index of a target point, wherein the target point refers to a point corresponding to the target initial prediction value; A target prediction value obtaining module, used to adjust the target initial prediction value by using the target non-power index as a weight coefficient to obtain a target prediction value; A control adjustment module, configured to issue an optimization control instruction when the target prediction value reaches a prediction value threshold, and to control and adjust the planetary gear reduction device based on the optimization control instruction; Wherein, the system further comprises: Reading a predetermined power index, and obtaining a real-time power index parameter by matching the predetermined power index in the real-time power characteristic information; Extracting first simulation information from the deceleration simulation database, wherein the first simulation information includes a first simulation power index parameter and a first simulation distributed load; When a first power similarity obtained by comparing the real-time power index parameter with the first simulation power index parameter meets a similarity threshold, taking the first simulation distributed load as the initial predicted simulation load; Acquire first characteristic information of a first non-power indicator in the real-time non-power characteristic information, wherein the first non-power indicator refers to any one of the predetermined non-power indicators, and the first characteristic information includes a plurality of parameter values ​​with point identifiers; Performing visualization processing on the plurality of parameter values ​​with point identifiers according to the predetermined visualization strategy to obtain a first visualization graph; fusing the first visible graph to obtain the non-dynamic feature visible graph; Among them, the predetermined non-dynamic indicators include a vibration dimension indicator set, a sound dimension indicator set, a temperature dimension indicator set, an electrical dimension indicator set, a lubrication dimension indicator set and a wear dimension indicator set.

2. The planetary gear reduction device optimization control system based on dynamic load prediction according to claim 1 is characterized in that: include: The initial predicted simulated load includes a plurality of loads with point identifications; Selecting a target load with the largest load among the multiple loads with point identifiers, the target load corresponding to the target point; The target load is used as the target initial prediction value.

3. The planetary gear reduction device optimization control system based on dynamic load prediction according to claim 1, characterized in that: include: Acquire a first-level neighbor node of the target point in the first visible graph, where the first-level neighbor node includes a plurality of first-level nodes; Randomly extract a first node from the plurality of first-level nodes, and obtain neighbor nodes of the first node in the first visible graph, which are recorded as second-level neighbor nodes, wherein the second-level neighbor nodes include a plurality of second-level nodes; Sampling the plurality of secondary nodes to obtain a first sampling set, wherein the first sampling set includes a plurality of characteristic parameters of the first non-power indicator; Performing aggregation analysis on the multiple characteristic parameters to obtain a first aggregation result, and performing mean calculation on the first characteristic parameter corresponding to the first node to obtain a first target parameter; Performing aggregation analysis on the first target parameter to obtain a second aggregation result, and using the second aggregation result as the first target non-power indicator data of the first node; The first target non-power indicator data is normalized and weighted, and combined with the target non-power characteristic parameters of the target point to obtain the target non-power index.

4. The planetary gear reduction device optimization control system based on dynamic load prediction according to claim 3 is characterized in that: It is determined whether the first characteristic parameter is within a predetermined threshold; if not, the first node is removed from the first-level neighbor nodes.

5. The planetary gear reduction device optimization control system based on dynamic load prediction according to claim 3, characterized in that: The first aggregation result is obtained by performing aggregation analysis on the plurality of characteristic parameters, which includes: Obtaining the maximum parameter and the minimum parameter among the plurality of characteristic parameters, and calculating the maximum parameter deviation; If the maximum parameter deviation exceeds the deviation threshold, a test instruction is issued, and a monitoring rationality test is performed on the multiple characteristic parameters based on the test instruction.

Images