Gear reducer performance optimization method based on multi-source data analysis

Abstract

The present application provides a gear reducer performance optimization method based on multi-source data analysis, which relates to the field of performance optimization technology, including: real-time monitoring of the gear reducer to obtain multi-source monitoring data; adaptive sensor abnormality correction of the multi-source monitoring data to generate reducer monitoring results; transmission feature identification and integration to establish a transmission feature monitoring matrix; optimization of transmission performance deviation of the gear reducer in combination with the transmission feature monitoring matrix to obtain a first strategy for performance optimization of the reducer; life loss prediction of the gear reducer to obtain a life loss prediction result; life performance compensation optimization of the first strategy for performance optimization of the reducer to generate a second strategy for performance optimization of the reducer; and performance optimization of the gear reducer. This application can solve the technical problems in the prior art caused by the low accuracy and strong singleness of the monitoring and maintenance of the gear reducer, thereby achieving the technical effect of improving monitoring accuracy and realizing diversified monitoring.

Description

Technical Field

[0001] The present application relates to the technical field of performance optimization, and in particular to a gear reducer performance optimization method based on multi-source data analysis. Background Art

[0002] With the continuous development of mechanical equipment and industrial production, speed reducers, as one of the core transmission components, are widely used in various industrial equipment, such as automated production lines, power generation, petrochemicals, mining, and other fields. Speed ​​reducers are subject to heavy loads and complex working environments during operation. Therefore, their performance stability and reliability are crucial to the overall operation of the equipment. However, existing technologies still have many shortcomings in speed reducer monitoring, fault prediction, and performance optimization, and improvements are urgently needed.

[0003] At present, most traditional monitoring methods rely on a single sensor, which makes it difficult to fully and accurately reflect the various operating parameters of the reducer. For example, vibration sensors can effectively detect mechanical failures, but their ability to identify problems such as excessive temperature and poor lubrication is poor. In addition, although some advanced prediction methods have been proposed, such as machine learning-based prediction models and data-driven analysis methods, these technologies still face problems such as low data quality and poor model generalization in practical applications. Most prediction models often rely on historical data for training, but in actual production environments, factors such as data incompleteness and noise interference often affect the accuracy of the model, resulting in inaccurate prediction results or delayed fault warnings. Therefore, existing reducer life prediction technologies have the disadvantages of low accuracy and poor real-time performance, making it difficult to achieve accurate fault warnings and life predictions.

[0004] In summary, the prior art has a technical problem in that the performance of the gear reducer is degraded due to the low accuracy and strong singleness of the monitoring and maintenance of the gear reducer. Summary of the Invention

[0005] The purpose of this application is to provide a gear reducer performance optimization method based on multi-source data analysis to solve the technical problem in the prior art that the performance of the gear reducer is reduced due to the low accuracy and strong singleness of the monitoring and maintenance of the gear reducer.

[0006] In view of the above problems, the present application provides a gear reducer performance optimization method based on multi-source data analysis, including: real-time monitoring of the gear reducer according to a pre-deployed multi-source sensor array to obtain multi-source monitoring data; adaptively correcting the influence of sensor anomalies on the multi-source monitoring data according to a sensor anomaly correction model to generate a reducer monitoring result; identifying and integrating transmission characteristics according to the reducer monitoring result to establish a transmission characteristic monitoring matrix; reading the control scheme of the gear reducer, and optimizing the transmission performance deviation of the gear reducer in combination with the transmission characteristic monitoring matrix to obtain a first strategy for reducer performance optimization; based on the reducer monitoring result, predicting the life loss of the gear reducer according to the reducer life loss prediction channel to obtain a life loss prediction result; performing life performance compensation optimization on the first reducer performance optimization strategy based on the life loss prediction result to generate a second reducer performance optimization strategy; and optimizing the performance of the gear reducer according to the second reducer performance optimization strategy.

[0007] The technical solution provided in the present application has at least the following technical effects or advantages: multi-source monitoring data is obtained by real-time monitoring of the gear reducer according to a pre-deployed multi-source sensor array; adaptive sensor anomaly influence correction is performed on the multi-source monitoring data according to a sensor anomaly correction model to generate a reducer monitoring result; transmission characteristics are identified and integrated according to the reducer monitoring result to establish a transmission characteristic monitoring matrix; the control scheme of the gear reducer is read, and the transmission performance deviation of the gear reducer is optimized in combination with the transmission characteristic monitoring matrix to obtain a first strategy for reducer performance optimization; based on the reducer monitoring result, the life loss of the gear reducer is predicted according to the reducer life loss prediction channel to obtain a life loss prediction result; based on the life loss prediction result, the life performance compensation optimization of the first reducer performance optimization strategy is performed to generate a second reducer performance optimization strategy; the performance of the gear reducer is optimized according to the second reducer performance optimization strategy. That is to say, by achieving the technical goal of improving the monitoring accuracy of the gear reducer and multi-dimensional fault detection, the technical effect of improving monitoring accuracy, realizing diversified monitoring and intelligent maintenance is achieved.

[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, which can be implemented in accordance with the contents of the description, and 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 specifically listed below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in this application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and a person of ordinary skill in the art can obtain other drawings based on the provided drawings without creative work.

[0010] Figure 1 A schematic flow chart of the gear reducer performance optimization method based on multi-source data analysis in this application;

[0011] Figure 2 This is a flow chart of obtaining the first strategy for optimizing reducer performance in the gear reducer performance optimization method based on multi-source data analysis in this application. DETAILED DESCRIPTION

[0012] This application provides a gear reducer performance optimization method based on multi-source data analysis, resolving the technical problem in the prior art of gear reducer performance degradation caused by low accuracy and single-source monitoring and maintenance. This method achieves the technical goal of improving gear reducer monitoring accuracy and multi-dimensional fault detection, achieving the technical effects of improved monitoring accuracy, diversified monitoring, and intelligent maintenance.

[0013] Below, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited to the example embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. It should also be noted that, for the convenience of description, only the parts related to this application, rather than all of them, are shown in the accompanying drawings.

[0014] Please see the attached Figure 1 , this application provides a gear reducer performance optimization method based on multi-source data analysis, which specifically includes the following steps:

[0015] Step 1: Monitor the gear reducer in real time based on the pre-deployed multi-source sensor array to obtain multi-source monitoring data.

[0016] Specifically, a pre-placed multi-source sensor array enables real-time monitoring of the gear reducer's operating status. Pre-placement refers to the pre-installation of sensors at key locations or locations within the gear reducer. These sensors are capable of collecting multiple types of signals. A multi-source sensor array is a collection of different types of sensors, such as vibration sensors, temperature sensors, and acoustic emission sensors. These sensors operate independently but work together to provide more comprehensive data. The multi-source sensor array enables real-time monitoring of the gear reducer's operating status. The multi-source sensor array operates in real time, allowing data to be quickly acquired and analyzed, enabling the timely identification of potential problems. In practice, for example, five vibration sensors, two temperature sensors, and one acoustic emission sensor are deployed, each collecting different signals. Specifically, the vibration sensor records the frequency and amplitude of the reducer's vibrations during operation, the temperature sensor measures temperature changes in the bearings, and the acoustic emission sensor captures high-frequency noise signals generated by internal gear meshing. The data provided by these sensors collectively constitutes multi-source monitoring data, reflecting the health of the gear reducer from different perspectives.

[0017] Step 2: Adaptively correct the impact of sensor anomalies on the multi-source monitoring data according to the sensor anomaly correction model to generate a reducer monitoring result.

[0018] Specifically, the system uses a sensor anomaly correction model to adaptively correct the effects of sensor anomalies on multi-source monitoring data. This system adjusts and corrects monitoring data collected from different sensors to eliminate errors and deviations that may occur when sensors are in abnormal conditions, generating reducer monitoring results. The system automatically adjusts its correction method based on different data inputs to adapt to different monitoring environments and sensor conditions, ultimately achieving more accurate reducer monitoring results that more realistically reflect the reducer's operating status. This helps maintenance personnel promptly identify and address potential faults, thereby improving equipment reliability and safety.

[0019] Step 3: Identify and integrate transmission characteristics based on the reducer monitoring results to establish a transmission characteristic monitoring matrix.

[0020] Specifically, by analyzing monitoring data obtained from the reducer, key characteristics related to the reducer's transmission are identified, including data indicators such as vibration frequency, temperature changes, and speed. These characteristics reflect the transmission status of the reducer during operation. The identified characteristic information is then integrated to form a transmission characteristic monitoring matrix. This data set arranges different transmission characteristics into a data set containing transmission characteristic information at each moment of the reducer's operation. This allows operation and maintenance personnel to more easily compare, analyze, and judge the reducer's overall operating status, promptly identify potential transmission problems, and improve equipment maintenance efficiency.

[0021] Step 4: Read the control scheme of the gear reducer, optimize the transmission performance deviation of the gear reducer in combination with the transmission characteristic monitoring matrix, and obtain a first strategy for optimizing the performance of the reducer.

[0022] Specifically, the control scheme of the gear reducer is read. The control scheme refers to the operating rules and parameter settings established to ensure the efficient and stable operation of the gear reducer during operation. This includes the start and stop of the reducer, load adjustment during operation, temperature control, and possible fault protection measures. For example, the reducer will automatically slow down or shut down in the event of high temperature or overload to avoid damage.

[0023] By analyzing the reducer's transmission characteristic monitoring matrix, the various performance parameters of the reducer during operation are evaluated and deviations from normal operating conditions are identified. For example, if the monitoring data shows that the reducer's temperature is consistently high or the vibration is excessive, it indicates that its transmission performance may have problems. Based on this deviation information, transmission performance deviation optimization is performed. By adjusting the control strategy, improving operating conditions, or optimizing the maintenance plan, the normal transmission performance of the reducer is restored. Ultimately, the first strategy for reducer performance optimization is generated. This is a specific solution for addressing reducer performance deviations, thereby improving the operating efficiency and reliability of the reducer through effective optimization methods.

[0024] Step 5: Based on the reducer monitoring result, the life loss prediction of the gear reducer is performed according to the reducer life loss prediction channel to obtain a life loss prediction result.

[0025] Specifically, by analyzing the reducer monitoring results, the reducer life loss prediction channel is used to predict the life loss of the gear reducer, predict the possible life loss of the reducer in the future, and obtain the life loss prediction result. The reducer monitoring results include various data of the reducer during operation, such as temperature, vibration, speed, etc., which are used to identify the operating status of the reducer and perform corresponding analysis. The life loss prediction channel is a prediction model or calculation method that combines the historical data of the reducer and the operating environment, and is used to predict the life loss of the reducer. The life loss prediction result will show the service life that the reducer may lose within a certain period of time in the future. For example, if the vibration frequency of the reducer is high, it may cause premature wear. The prediction model will estimate the degree of life loss based on this information and provide a specific prediction result.

[0026] Step 6: Based on the life loss prediction result, the first strategy for optimizing the performance of the reducer is optimized for life performance compensation to generate a second strategy for optimizing the performance of the reducer.

[0027] Specifically, based on the life loss prediction results, the first strategy for optimizing reducer performance is optimized for life performance compensation. By adjusting the reducer's operating parameters, improving maintenance cycles, and optimizing load control, the impact of life loss on performance is reduced. This generates a second strategy for optimizing reducer performance to compensate for the potential impact of life loss. The second strategy is more effective than the first strategy in addressing the impact of life loss, thereby extending the overall service life of the equipment.

[0028] Step seven: Optimize the performance of the gear reducer according to the second strategy for optimizing the performance of the reducer.

[0029] Specifically, the second strategy for optimizing the reducer's performance that has been generated is applied to the actual operation of the reducer. By adjusting the operating conditions of the reducer, such as reducing the load, optimizing the speed, improving lubrication, etc., its performance is improved, the working efficiency of the reducer is improved, the failure rate is reduced, and the service life of the equipment is extended.

[0030] The gear reducer performance optimization method based on multi-source data analysis can achieve the technical goal of improving the monitoring accuracy of the gear reducer and multi-dimensional fault detection, and achieve the technical effect of improving monitoring accuracy and realizing diversified monitoring and intelligent maintenance.

[0031] For further information, please see the attached Figure 2 , this application also includes: executing the expected deviation identification of the transmission characteristics of the gear reducer according to the control scheme and the transmission characteristic monitoring matrix to obtain a transmission deviation identification result; performing a transmission performance impact evaluation according to the transmission deviation identification result to obtain a transmission performance impact coefficient; judging whether the transmission performance impact coefficient is greater than or equal to a transmission performance impact threshold; if the transmission performance impact coefficient is greater than or equal to the transmission performance impact threshold, activating a transmission performance deviation optimization map; inputting the transmission deviation identification result into the transmission performance deviation optimization map to generate the first strategy for optimizing the reducer performance.

[0032] Specifically, the gear reducer's transmission characteristic deviation identification is performed based on the control scheme and transmission characteristic monitoring matrix to determine whether the current transmission characteristic of the reducer deviates from the expected value under normal operation. It also detects whether the reducer has any unexpected abnormal conditions during operation, and obtains transmission deviation identification results. For example, the reducer's vibration exceeds the normal range or the speed is too low.

[0033] Based on the transmission deviation identification results, the transmission performance impact assessment is performed to obtain the transmission performance impact coefficient, which is used to quantify the severity of performance degradation or deviation and determine whether the reducer's operating condition has reached the critical value for warning or failure. For example, the greater the deviation of the parameter, the higher the impact coefficient, and vice versa.

[0034] Determine whether the transmission performance impact coefficient is greater than or equal to the transmission performance impact threshold. The calculated transmission performance impact coefficient is then evaluated to determine whether it exceeds the preset threshold. If the transmission performance impact coefficient is greater than or equal to the transmission performance impact threshold, it indicates that the transmission component may have a serious performance issue. The transmission performance deviation map is activated, and further optimization or adjustment measures are taken. If the transmission performance deviates from the optimization map, this may include adjusting operating parameters, replacing components, or performing other maintenance operations to restore the reducer to optimal operating condition.

[0035] The transmission deviation identification results are fed into the transmission performance deviation optimization map. This information is then combined with the optimization map to generate specific performance optimization recommendations for the reducer. The transmission performance deviation optimization map includes optimization measures for different deviation scenarios. The transmission performance deviation optimization map can help maintenance personnel find the best adjustment plan to restore the reducer's normal performance. For example, if wear on a transmission component causes increased load, the optimization map may recommend adjusting load distribution or replacing the component to improve performance.

[0036] Furthermore, the present application also includes: performing transmission characteristic normal parameter retrieval on the gear reducer according to the control scheme to obtain a transmission characteristic parameter retrieval set; performing central trend analysis on the transmission characteristic parameter retrieval set to establish a transmission characteristic expectation matrix; performing deviation identification on the transmission characteristic monitoring matrix according to the transmission characteristic expectation matrix to generate the transmission deviation identification result.

[0037] Specifically, based on the gear reducer's control scheme, a transmission characteristic parameter search is performed on the gear reducer. Typical parameters expected under normal operating conditions are identified, resulting in a transmission characteristic parameter search set. Transmission characteristics include various important operating data such as speed, load, temperature, and vibration, which are used to evaluate reducer performance. This normal transmission characteristic parameter search yields a transmission characteristic parameter search set containing the standard values ​​for each transmission characteristic of the reducer when operating in a healthy state.

[0038] Perform central tendency analysis based on the transmission characteristic parameter retrieval set to establish a transmission characteristic expectation matrix. Central tendency analysis determines the typical range and expected value of the parameters under normal conditions by calculating statistics such as the mean and median of the transmission characteristic parameters. Establishing the transmission characteristic expectation matrix helps to obtain the typical behavior of each transmission characteristic when the reducer is operating under various conditions.

[0039] Deviations from the transmission characteristic monitoring matrix are identified based on the transmission characteristic expectation matrix. By comparing the standard values ​​in the transmission characteristic expectation matrix with the actual monitoring data, the deviated transmission characteristics are identified, and the transmission deviation identification result is generated, indicating that the operating parameters of the reducer are abnormal, which is used to help operation and maintenance personnel identify potential faults or abnormal behaviors.

[0040] Furthermore, the present application also includes: performing life-related cleaning according to the reducer monitoring results to obtain life-related monitoring data; the reducer life loss prediction channel includes P reducer life loss prediction models, where P is a positive integer greater than 1; inputting the life-related monitoring data into the P reducer life loss prediction models to obtain P life loss prediction coefficients; performing a proportion calculation based on the P life loss prediction accuracies corresponding to the P reducer life loss prediction models to obtain P life loss prediction incentive coefficients; performing a weighted calculation on the P life loss prediction coefficients based on the P life loss prediction incentive coefficients to generate the life loss prediction result.

[0041] Specifically, by cleaning the monitoring data of the reducer and removing irrelevant or noise-interfered data, effective monitoring data related to the reducer life is obtained, ensuring that only data that accurately reflects the health status and life changes of the reducer is retained, such as temperature, vibration, load and other factors, thereby helping to more accurately predict the remaining service life of the reducer and providing a basis for subsequent life prediction and optimization.

[0042] The life loss prediction channel uses P life loss prediction models to predict the life loss of the reducer. Each model is based on a different algorithm and data processing method. P is a positive integer greater than 1, indicating that multiple prediction models are involved in the life loss prediction. The combined results of multiple predictions improve the accuracy and reliability of the prediction.

[0043] The life-related monitoring data is fed into P reducer life-loss prediction models. Data processing and calculation are performed to obtain P corresponding life-loss prediction coefficients for each of the P reducer life-loss prediction models. These coefficients are used to quantify the life loss that a reducer may experience under specific conditions. For example, one model might predict life loss based on temperature changes, while another might focus on vibration data.

[0044] Based on the P life loss prediction accuracies corresponding to the P reducer life loss prediction models, a percentage calculation is performed to obtain P life loss prediction incentive coefficients. The prediction accuracy of each model may vary, and the P life loss prediction incentive coefficients obtained from the percentage calculation can be assigned different weights based on the reliability of each model, making the final prediction result more accurate. For example, if a model has higher prediction accuracy, its incentive coefficient may be larger, indicating that the model has greater influence in the overall prediction.

[0045] Performing a weighted calculation on P life loss prediction coefficients according to the P life loss prediction incentive coefficients to generate a life loss prediction result can make the final life loss prediction result more consistent with the actual situation, thereby helping to make more accurate maintenance and replacement decisions.

[0046] Furthermore, the present application also includes: judging whether the life loss prediction result is greater than or equal to the life loss threshold; if the life loss prediction result is greater than or equal to the life loss threshold, taking the life loss prediction result as the life performance optimization target, performing a life performance optimization decision on the gear reducer according to the life-related monitoring data, and obtaining a first life performance optimization strategy; performing a life impact analysis on the gear reducer according to the first reducer performance optimization strategy, and obtaining a first strategy life impact analysis result; correcting the first life performance optimization strategy according to the first strategy life impact analysis result, and generating a second life performance optimization strategy; performing compensation optimization on the first reducer performance optimization strategy according to the second life performance optimization strategy, and obtaining the second reducer performance optimization strategy.

[0047] Specifically, the predicted life loss of the reducer is compared with a pre-set life loss threshold to determine whether the predicted life loss result is greater than or equal to the threshold. The life loss prediction result represents the estimated loss of operating life of the reducer at a specific moment or under certain conditions. The life loss threshold is set based on the equipment's usage requirements and safety requirements.

[0048] If the life loss prediction result is greater than or equal to the life loss threshold, the life loss prediction result is judged to have reached the threshold, indicating that the reducer may need maintenance or optimization to prevent premature failure. By using the life loss prediction result as the life performance optimization target, life performance optimization decisions are made for the gear reducer based on the life-related monitoring data. For example, the optimization decision may include adjusting operating conditions, changing maintenance cycles, or optimizing the operating environment to reduce or delay life loss. Ultimately, the first life performance optimization strategy is generated, that is, a set of implementation plans for optimizing the reducer life, and finally the first life performance optimization strategy is obtained.

[0049] Based on the first strategy for optimizing reducer performance, a lifespan impact analysis is performed on the gear reducer. This evaluates the actual impact of the optimization strategy on the reducer's lifespan and provides the first strategy's lifespan impact analysis results. This analysis then determines whether the strategy's implementation effectively extends the reducer's lifespan or changes its operating status. For example, assume that after adjusting the reducer's workload, the lifespan analysis results indicate an extension of the equipment's lifespan. This is the first strategy's lifespan impact analysis result.

[0050] Based on the results of the first strategy's life impact analysis, the first strategy for optimizing life performance is corrected to generate a second strategy for optimizing life performance, further improving the reducer's life performance. If the analysis indicates that the adjustments were not as effective as expected, the optimization strategy is corrected, adjusting the corresponding parameters or conditions to generate a new optimization strategy, the second strategy for optimizing life performance, to ensure the equipment operates under optimal conditions and maximize its service life.

[0051] Furthermore, the present application also includes: identifying aging characteristics based on the reducer monitoring results to obtain reducer aging identification results; evaluating the degree of aging based on the reducer aging identification results to obtain a reducer aging coefficient; if the reducer aging coefficient is greater than or equal to the reducer aging threshold, making a performance optimization decision based on the reducer aging identification results to obtain a first aging characteristic performance optimization scheme; analyzing the aging impact of the gear reducer based on the second reducer performance optimization strategy to obtain a second strategy aging impact analysis result; adjusting the first aging characteristic performance optimization scheme based on the aging impact analysis result of the second strategy to obtain a second aging characteristic performance optimization scheme; optimizing the second reducer performance optimization strategy based on the second aging characteristic performance optimization scheme to generate a third reducer performance optimization strategy; and optimizing the gear reducer performance based on the third reducer performance optimization strategy.

[0052] Specifically, by processing and comparing sensor data, aging characteristics are identified based on the reducer monitoring results, namely wear, corrosion or other phenomena that may cause performance degradation, and the reducer aging identification results are obtained, such as temperature increase, increased vibration, etc.

[0053] Based on the results of the reducer aging identification, the aging degree is assessed and the reducer aging coefficient is obtained. This coefficient is used to quantify the degree of performance loss caused by aging during the entire use of the reducer. The aging of the reducer is determined by analyzing various operating data of the reducer, such as vibration amplitude and temperature changes.

[0054] If the reducer aging coefficient is greater than or equal to the reducer aging threshold, it means that the reducer is aged to a high degree and performance optimization is required. Then, a performance optimization decision is made based on the reducer aging identification result. By adjusting the operating mode, control strategy or performing maintenance, the first aging characteristic performance optimization solution is obtained to extend the service life of the reducer and improve the operating efficiency.

[0055] After executing the first aging characteristic performance optimization solution, the impact of the new performance optimization strategy on the reducer aging is further analyzed based on the current state of the reducer to evaluate whether the strategy can effectively slow the aging process, and obtain the results of the second strategy aging impact analysis. For example, the workload of the reducer may be adjusted, or the cooling system may be used to reduce the temperature to slow the aging phenomenon.

[0056] Based on the results of the second strategy's aging impact analysis, the first aging characteristic performance optimization plan is adjusted. Further adjustments are made to the reducer's aging characteristic treatment methods, optimizing the identification and response measures for aging characteristics. This can continuously improve the reducer's operating status and prevent further deterioration. For example, if the analysis results indicate that the cooling system is ineffective, the cooling method may be adjusted or auxiliary heat dissipation equipment may be added to reduce the temperature and slow the aging process.

[0057] Based on the second performance optimization plan based on aging characteristics, the second strategy for speed reducer performance optimization is optimized, including adjusting operating parameters and optimizing maintenance plans, which then generates the third strategy for speed reducer performance optimization. This third strategy better addresses speed reducer aging, thereby extending the equipment's service life and improving its operating efficiency. For example, if aging causes increased wear on certain speed reducer components, the optimization strategy may include premature replacement of the components.

[0058] According to the third strategy of speed reducer performance optimization, the gear speed reducer is optimized, that is, actual performance optimization operations are performed on the speed reducer, including adjusting the workload of the speed reducer, optimizing the settings of the cooling system or replacing components with larger wear and tear, etc., so that the performance of the speed reducer can be further improved, and it can maintain high efficiency and reliability during the aging process, thereby reducing downtime and improving the overall equipment value.

[0059] Furthermore, the present application also includes: extracting the kth parameter of reducer monitoring based on the multi-source monitoring data, where k is a positive integer; collecting the sensor device status information corresponding to the kth parameter of reducer monitoring to obtain the kth device status data; performing anomaly detection on the kth device status data to obtain the kth device status anomaly data; based on the kth device status anomaly data and the kth parameter of reducer monitoring, obtaining the kth parameter correction result according to the sensor anomaly correction model; and adding the kth parameter correction result to the reducer monitoring result.

[0060] Specifically, multi-source monitoring data refers to various types of data collected from reducers via multiple sensors, such as vibration, temperature, or acoustic signals. Based on this multi-source monitoring data, we extract the kth parameter of the reducer monitoring. This means selecting specific information from the overall data: the kth parameter of all monitored parameters, where k is a positive integer, such as 1 or 2. By extracting this parameter, we can focus on analyzing its changing trends.

[0061] Next, the sensor device status information corresponding to the kth parameter monitored by the reducer is collected to obtain the kth device status data. Collection refers to obtaining the corresponding data; the sensor device status information represents the operating status of the sensor associated with the kth parameter, such as the sensor's voltage, sensitivity, or operating temperature, and is used to determine whether the sensor itself is functioning properly.

[0062] Then, anomaly detection is performed on the k-th device status data to obtain the k-th device status abnormal data, where anomaly detection refers to analyzing whether the data deviates from the normal range, such as monitoring the sensitivity changes of the temperature sensor. If the sensitivity decreases, it may be considered that the device is abnormal, and then the data marked as abnormal in the detection is obtained to provide a basis for subsequent processing.

[0063] Next, the sensor anomaly correction model is used to correct for deviations caused by sensor anomalies, such as interpolation to correct intermittent sensor signals or parameter adjustments to correct for reduced sensitivity. Based on the k-th device status anomaly data and the k-th reducer monitoring parameter, the sensor anomaly correction model generates the k-th parameter correction result. This correction results in data that is closer to the actual situation.

[0064] Finally, following the method for obtaining the kth parameter correction result, the multi-source monitoring data is accessed sequentially to obtain the parameter correction results corresponding to each reducer monitoring parameter. Each parameter correction result is combined into the reducer monitoring result. The reducer monitoring result is a comprehensive representation of all monitored parameters, such as the health status of vibration, temperature, and noise signals. This ensures that the final monitoring data is more reliable, providing a more scientific basis for decision-making in equipment management.

[0065] Furthermore, the present application also includes: performing a sensor abnormality impact evaluation based on the kth device status abnormality data to obtain the kth sensor abnormality impact coefficient; judging whether the kth sensor abnormality impact coefficient is greater than or equal to the sensor abnormality impact threshold; if the kth sensor abnormality impact coefficient is greater than or equal to the sensor abnormality impact threshold, activating the sensor abnormality correction model; inputting the kth device status abnormality data and the kth parameter monitored by the reducer into the sensor abnormality correction model to obtain the kth parameter correction result.

[0066] Specifically, the kth sensor abnormality impact coefficient is obtained by analyzing the abnormal data generated by the kth device during operation and evaluating its potential impact on sensor performance. Abnormal data may result from a malfunction or abnormal condition during device operation, such as excessive temperature or vibration. The sensor abnormality impact coefficient, derived from this analysis, represents the degree of impact on the sensor during the abnormal condition. For example, if a device malfunction causes severe vibration, the vibration sensor may be significantly disturbed, resulting in a higher impact coefficient.

[0067] Next, the sensor anomaly impact coefficients are compared to determine whether the kth sensor anomaly impact coefficient is greater than or equal to the sensor anomaly impact threshold. The sensor anomaly impact threshold is a predefined standard used to determine whether the abnormal state of the equipment is serious enough to require further corrective measures.

[0068] If the kth sensor abnormality influence coefficient is greater than or equal to the sensor abnormality influence threshold, indicating that the impact of the abnormality is large enough, the sensor abnormality correction model is activated to adjust or correct the sensor readings and measurement results according to the abnormal state of the equipment and the influence coefficient, so as to obtain more accurate monitoring data.

[0069] After activating the sensor anomaly correction model, the kth device's status anomaly data and the kth reducer monitoring parameter associated with the reducer are input into the model. Based on this data, the sensor anomaly correction model adjusts the kth parameter's measured value to produce a kth parameter correction result. For example, if the reducer's temperature reading from the sensor is inaccurate, the correction result can be corrected to a value closer to the actual temperature.

[0070] In summary, the gear reducer performance optimization method based on multi-source data analysis provided in the present application has the following technical effects: the gear reducer is monitored in real time according to a pre-deployed multi-source sensor array to obtain multi-source monitoring data; the multi-source monitoring data is adaptively corrected for the influence of sensor anomalies according to a sensor anomaly correction model to generate a reducer monitoring result; transmission characteristics are identified and integrated according to the reducer monitoring result to establish a transmission characteristic monitoring matrix; the control scheme of the gear reducer is read, and the transmission performance deviation of the gear reducer is optimized in combination with the transmission characteristic monitoring matrix to obtain a first strategy for reducer performance optimization; based on the reducer monitoring result, the life loss of the gear reducer is predicted according to the reducer life loss prediction channel to obtain a life loss prediction result; based on the life loss prediction result, the first strategy for reducer performance optimization is optimized for life performance compensation to generate a second strategy for reducer performance optimization; the performance of the gear reducer is optimized according to the second strategy for reducer performance optimization. That is to say, by achieving the technical goal of improving the monitoring accuracy of the gear reducer and multi-dimensional fault detection, the technical effect of improving monitoring accuracy, realizing diversified monitoring and intelligent maintenance is achieved.

[0071] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0072] Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the spirit and 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 is intended to include these modifications and variations.

Claims

1. A gear reducer performance optimization method based on multi-source data analysis, characterized in that: include: The gear reducer is monitored in real time based on the pre-deployed multi-source sensor array to obtain multi-source monitoring data; Adaptively correct the impact of sensor anomalies on the multi-source monitoring data according to the sensor anomaly correction model to generate a reducer monitoring result; Identify and integrate transmission characteristics based on the reducer monitoring results to establish a transmission characteristic monitoring matrix; Reading the control scheme of the gear reducer, optimizing the transmission performance deviation of the gear reducer in combination with the transmission characteristic monitoring matrix, and obtaining a first strategy for optimizing the performance of the reducer; Based on the reducer monitoring result, predicting the life loss of the gear reducer according to the reducer life loss prediction channel to obtain a life loss prediction result; Performing life performance compensation optimization on the first reducer performance optimization strategy based on the life loss prediction result to generate a second reducer performance optimization strategy; Optimizing the performance of the gear reducer according to the second strategy for optimizing the performance of the reducer; Performing aging feature identification based on the reducer monitoring result to obtain a reducer aging identification result; Performing an aging degree assessment based on the reducer aging identification result to obtain a reducer aging coefficient; If the reducer aging coefficient is greater than or equal to the reducer aging threshold, a performance optimization decision is made based on the reducer aging identification result to obtain a first aging characteristic performance optimization solution; Performing an aging impact analysis on the gear reducer according to the second strategy for optimizing the performance of the reducer, and obtaining an aging impact analysis result of the second strategy; Adjust the first aging characteristic performance optimization scheme according to the aging impact analysis result of the second strategy to obtain a second aging characteristic performance optimization scheme; According to the second aging characteristic performance optimization scheme, the second reducer performance optimization strategy is optimized to generate a third reducer performance optimization strategy; The gear reducer is optimized for performance according to the third strategy for optimizing the reducer performance.

2. The method according to claim 1, wherein The control scheme of the gear reducer is read, and the transmission performance deviation of the gear reducer is optimized in combination with the transmission characteristic monitoring matrix to obtain a first strategy for optimizing the performance of the reducer, including: performing transmission characteristic expected deviation identification of the gear reducer according to the control scheme and the transmission characteristic monitoring matrix to obtain a transmission deviation identification result; Performing transmission performance impact evaluation based on the transmission deviation identification result to obtain a transmission performance impact coefficient; Determining whether the transmission performance impact coefficient is greater than or equal to a transmission performance impact threshold; If the transmission performance impact coefficient is greater than or equal to the transmission performance impact threshold, activating the transmission performance deviation optimization map; The transmission deviation identification result is input into the transmission performance deviation optimization map to generate the first strategy for optimizing the performance of the reducer.

3. The method according to claim 2, wherein Performing transmission characteristic expected deviation identification of the gear reducer according to the control scheme and the transmission characteristic monitoring matrix to obtain a transmission deviation identification result includes: Performing a transmission characteristic normal parameter search on the gear reducer according to the control scheme to obtain a transmission characteristic parameter search set; Performing a central tendency analysis based on the transmission characteristic parameter retrieval set to establish a transmission characteristic expectation matrix; Deviation identification is performed on the transmission characteristic monitoring matrix according to the transmission characteristic expectation matrix to generate the transmission deviation identification result.

4. The method according to claim 1, wherein Based on the reducer monitoring result, the life loss prediction of the gear reducer is performed according to the reducer life loss prediction channel to obtain a life loss prediction result, including: Perform life-related cleaning according to the reducer monitoring results to obtain life-related monitoring data; The reducer life loss prediction channel includes P reducer life loss prediction models, where P is a positive integer greater than 1; Inputting the life-related monitoring data into the P reducer life loss prediction models to obtain P life loss prediction coefficients; Calculate the proportion of P life loss prediction accuracies corresponding to the P reducer life loss prediction models to obtain P life loss prediction incentive coefficients; The P life loss prediction coefficients are weightedly calculated according to the P life loss prediction incentive coefficients to generate the life loss prediction result.

5. The method according to claim 4, wherein Performing life performance compensation optimization on the first reducer performance optimization strategy based on the life loss prediction result to generate a second reducer performance optimization strategy, including: Determining whether the life loss prediction result is greater than or equal to a life loss threshold; If the life loss prediction result is greater than or equal to the life loss threshold, taking the life loss prediction result as the life performance optimization target, performing a life performance optimization decision on the gear reducer according to the life-related monitoring data, and obtaining a first life performance optimization strategy; Performing a life impact analysis on the gear reducer according to the first strategy for optimizing the performance of the reducer to obtain a life impact analysis result of the first strategy; Correcting the first life performance optimization strategy according to the life impact analysis result of the first strategy to generate a second life performance optimization strategy; According to the second life performance optimization strategy, the first reducer performance optimization strategy is compensated and optimized to obtain the second reducer performance optimization strategy.

6. The method according to claim 1, wherein Adaptively correcting the influence of sensor anomalies on the multi-source monitoring data according to the sensor anomaly correction model to generate a reducer monitoring result, including: Extracting the kth parameter of the reducer monitoring according to the multi-source monitoring data, where k is a positive integer; Collecting sensor device status information corresponding to the kth parameter monitored by the reducer to obtain kth device status data; Performing anomaly detection on the k-th device status data to obtain k-th device status abnormality data; Based on the kth device status abnormality data and the kth parameter monitored by the reducer, and according to the sensor abnormality correction model, a kth parameter correction result is obtained; The kth parameter correction result is added to the reducer monitoring result.

7. The method according to claim 6, wherein Based on the kth device state abnormality data and the kth parameter monitored by the reducer, and according to the sensor abnormality correction model, a kth parameter correction result is obtained, including: Performing a sensor abnormality impact evaluation based on the kth device status abnormality data to obtain a kth sensor abnormality impact coefficient; Determining whether the kth sensing abnormality influence coefficient is greater than or equal to a sensing abnormality influence threshold; If the kth sensor abnormality influence coefficient is greater than or equal to the sensor abnormality influence threshold, activating the sensor abnormality correction model; The kth device status abnormality data and the kth parameter monitored by the reducer are input into the sensor abnormality correction model to obtain the kth parameter correction result.

Images