An Internet of Things-based real-time monitoring and early warning system for the lubrication status of large machines

Abstract

The present invention discloses a real-time monitoring and early warning system for the lubrication state of large machines based on the Internet of Things, which relates to the technical field of real-time monitoring and early warning of states, and solves the following technical problems: firstly, it is difficult to use the sensor network of the Internet of Things to monitor the lubrication state of large machines in real time; secondly, it is difficult to intelligently identify and classify the lubrication state and then conduct early warning through time series analysis; thirdly, it is difficult to simulate the flow path of the lubricant and analyze the influence coefficient on the lubricant flow according to the fault identification model combined with the digital twin technology for early warning; finally, it is difficult to visually display the early warning information using the digital twin model. The present invention monitors the lubrication state in real time through sensors, transmits data to the cloud through the Internet of Things, analyzes the lubrication state with a deep learning model after being processed by a server, conducts simulation analysis and early warning in combination with the digital twin technology, and finally visually displays the early warning information through the digital twin model.

Description

Technical Field

[0001] The present invention belongs to the technical field of real-time monitoring and early warning of states, and specifically relates to a real-time monitoring and early warning system for the lubrication state of large machines based on the Internet of Things. Background Art

[0002] In industrial production, the lubrication state of large mechanical equipment, such as heavy construction machinery, high-speed electric spindles, etc., is directly related to its operating efficiency, service life, and safety. The Internet of Things technology realizes the real-time perception, dynamic monitoring, and intelligent decision-making of the physical world through means such as sensor networks and data transmission. Through the Internet of Things technology, the lubrication state parameters of equipment can be monitored in real time; by integrating technologies such as deep learning algorithms and digital twins, it is possible to achieve comprehensive, accurate, and real-time monitoring and early warning of the lubrication state of large machine equipment.

[0003] The following problems exist in the prior art: First, it is difficult to use the sensor network of the Internet of Things to monitor the lubrication state of large machines in real time; second, it is difficult to intelligently identify and classify the lubrication state and then conduct early warning through time series analysis; then, it is difficult to simulate the flow path of lubricants and analyze the influence coefficient on the lubricant flow according to the fault identification model combined with digital twin technology for early warning; finally, it is difficult to visually display early warning information using the digital twin model. Summary of the Invention

[0004] To solve the problems existing in the above prior art, the first aspect of the present invention provides a real-time monitoring and early warning system for the lubrication state of large machines based on the Internet of Things, including the following modules:

[0005] Data acquisition module: Real-time monitor the lubrication state of large machines through sensors and collect monitoring data in real time;

[0006] Data transmission module: Use the Internet of Things communication protocol to transmit the monitoring data collected by the sensors to the cloud server through a wireless network or a wired network;

[0007] Data processing module: On the cloud server, store, clean, and standardize the monitoring data;

[0008] Data analysis and early warning module: Intelligently identify and classify the lubrication state using the deep belief network of the deep learning model; conduct time series analysis on the lubrication state according to the recognition and classification results of the current lubrication state, and issue an early warning by calculating the lubrication state value under the predicted lubrication state; combine digital twin technology to build a digital twin model of the lubrication state of large machines to simulate the lubrication state of large machines; establish a fault identification model based on the monitoring data; according to the fault identification results combined with the digital twin model, simulate the flow path of lubricants, and issue an early warning by calculating the influence coefficient of the fault point on the lubricant flow;

[0009] Visual warning display module: Visualize and display warning information through a digital twin model.

[0010] Furthermore: Real-time monitor the lubrication status of the large machine through sensors, and collect monitoring data in real time, including the following steps:

[0011] Install various sensors at various parts of the lubrication system of the large machine, including: bearings, gears, oil tanks, oil pumps, oil pipes, lubrication points; among them, install temperature sensors to monitor the temperature of the large machine equipment and lubricant liquid; install pressure sensors to monitor the pressure of the lubricant; install oil quality sensors to monitor the density, conductivity and dielectric constant of the lubricant; install vibration sensors to monitor the vibration information of the equipment; install noise sensors to monitor the noise signal of the equipment; install moisture sensors to monitor the moisture content in the lubricant; real-time monitor the viscosity of the lubricant and use a particle counter to monitor the number of particles in the lubricant; introduce a laser displacement sensor to real-time monitor the oil film thickness of the lubricant.

[0012] Furthermore: Use the deep belief network of the deep learning model to intelligently identify and classify the lubrication status, including the following steps:

[0013] According to the processed monitoring data, use the deep belief network of the deep learning model to automatically learn the feature representation of the monitoring data, and identify and classify the lubrication status; the categories of lubrication status include: hydrodynamic lubrication, hydrostatic lubrication, elastohydrodynamic lubrication, thin film lubrication, boundary lubrication, dry friction and mixed lubrication;

[0014] Adopt a deep belief network model to construct an identification and classification model; use a restricted Boltzmann machine as the basic building block of the deep belief network model; through a layer-by-layer greedy training algorithm, first train each layer of the restricted Boltzmann machine separately; when training each layer of the restricted Boltzmann machine, use the contrastive divergence algorithm to optimize the parameters of the deep belief network model; stack the trained restricted Boltzmann machines layer by layer to form a complete deep belief network model;

[0015] After the layer-by-layer greedy training is completed, use the backpropagation algorithm to globally optimize the entire deep belief network model; add a softmax classifier to the output layer of the deep belief network model to classify the lubrication status;

[0016] Collect historical monitoring data to generate a training data set, and label the training data set with the corresponding category labels of the lubrication status;

[0017] Input the labeled training data set into the deep belief network model for training, including the training of the softmax classifier;

[0018] The monitored data after real-time collection and processing is input into the trained deep belief network model, and the recognition and classification results of the current lubrication state of the large machine for the monitored data collected in real time are output, including: hydrodynamic lubrication, hydrostatic lubrication, elastohydrodynamic lubrication, thin-film lubrication, boundary lubrication, dry friction, and mixed lubrication.

[0019] Furthermore: according to the recognition and classification results of the current lubrication state, perform time series analysis on the lubrication state, and issue a warning by calculating the lubrication state value under the predicted lubrication state, including the following steps:

[0020] Based on the trained deep belief network model, identify the current lubrication state; by continuously monitoring the lubrication system of the large machine, identify whether there is a transition from one lubrication state to another;

[0021] When there is a transition, collect the monitored data of the lubrication state of the large machine at different time points, arrange the collected monitored data in chronological order to form a time series data set;

[0022] Construct a long short-term memory network model, input the time series data set into the long short-term memory network model for training; by inputting the monitored data of the current lubrication state identified by the deep belief network model into the trained long short-term memory network model, predict the lubrication state at the next time step;

[0023] Adopt the sliding window method to gradually predict the lubrication states at multiple future time steps. Starting from the current window of real-time monitoring, use the long short-term memory network model to predict the lubrication state at the next time step; by sliding the window forward by one time step, adding the latest real-time monitoring data, and removing the earliest monitored data, repeat the prediction process; by continuously sliding the window and using the long short-term memory network model for prediction, gradually obtain the lubrication states at multiple future time steps;

[0024] For the lubrication states at multiple future time steps obtained by prediction, obtain the monitored data under the predicted lubrication state; by calculating the formula for the lubrication state value under the predicted lubrication state: ,

[0025] where, represents the lubrication state value, represents the weight of the i-th sensor, represents the monitored value of the i-th sensor at the j-th time step, represents the minimum value of the historical monitored values of the i-th sensor, represents the maximum value of the historical monitored values of the i-th sensor; t is the length of the time window;

[0026] Set an end - of - life threshold for the lubrication state. When the lubrication state value is lower than the threshold, the end of life is reached, and the warning mechanism is triggered for warning.

[0027] Furthermore: Combining digital twin technology, construct a digital twin model of the large machine's lubrication state to simulate the lubrication state of the large machine, including the following steps:

[0028] According to the actual geometry and dimensions of the large machine's lubrication system, use 3D modeling software to construct a geometric model, which describes the various components of the lubrication system, including: the shapes, positions, and connection relationships of bearings, gears, oil tanks, oil pumps, oil pipes, and lubrication points;

[0029] Based on the constructed geometric model, by solving the hydrodynamic equations, simulate the flow state of the lubricant in the large machine's lubrication system, including: oil pressure distribution and flow velocity distribution; by solving the heat conduction equation, obtain the temperature distribution of the lubricant in the large machine's lubrication system and the heat transfer process; by solving the mechanical equilibrium equation, obtain the distribution of oil pressure, oil film thickness, and friction force parameters in the lubrication system;

[0030] By integrating the above geometric model and the results of hydrodynamic, heat conduction, and mechanical equilibrium analyses, construct a digital twin model of the large machine's lubrication state;

[0031] Real - time collect the monitoring data of the large machine's lubrication system through IoT sensors, associate the digital twin model with the monitoring data collected by IoT sensors, and synchronize the actual operating state of the large machine's lubrication system in the digital twin model in real - time through the data interface;

[0032] Deploy the trained deep belief network model into the digital twin model, output the current real - time lubrication state through the deep belief network model, and perform simulation display through the digital twin model.

[0033] Furthermore: Establish a fault identification model based on the monitoring data, including the following steps:

[0034] Use the convolutional neural network model in deep learning combined with the recurrent neural network model to establish a fault identification model;

[0035] Collect and process the historical monitoring data set of the large machine's lubrication state, and label the fault types including: leakage and blockage; input the labeled historical monitoring data set into the fault identification model for training;

[0036] Real - time monitor the collected and processed monitoring data through IoT, input it into the trained fault identification model, output whether a fault is identified in the current lubrication state of the large machine, and output the fault type.

[0037] Further: According to the fault identification result combined with the digital twin model, simulate the flow path of the lubricant, and issue a warning by calculating the influence coefficient of the fault point on the lubricant flow, including the following steps:

[0038] Deploy the trained fault identification model into the digital twin model. According to the fault identification model combined with the digital twin model, identify whether there are fault points in the lubricant flow path, including: leakage points and blockage points; according to the identified results, count the total number K of all fault points in the flow path;

[0039] According to the digital twin model, simulate the flow path of the lubricant in the large machine equipment, which is divided into a normal flow path and an abnormal propagation path; the abnormal propagation path is the flow path of the large machine lubrication state under fault output by the fault identification model;

[0040] During the simulation process, by using the computational fluid dynamics method, collect the flow conditions of the lubricant on the normal flow path and the abnormal propagation path, including: flow velocity, temperature and pressure distribution;

[0041] By calculating the influence coefficient formula of the fault point on the lubricant flow: ,

[0042] Among them, is the influence coefficient, is the pressure under the k-th fault point, is the flow velocity under the k-th fault point, is the temperature under the k-th fault point, is the pressure of the normal flow path, is the flow velocity of the normal flow path, is the temperature of the normal flow path, , and are the rates of change of the pressure, flow velocity and temperature with time under the k-th fault point respectively. K represents the total number of all fault points on the flow path; when the influence coefficient exceeds the threshold of 0.2, trigger the warning mechanism to issue a warning.

[0043] Further: Visualize and display the warning information through the digital twin model, including the following steps: According to the calculated results of the lubrication state value under the predicted lubrication state and the influence coefficient of the fault point on the lubricant flow, when the warning mechanism is triggered, integrate the generated warning information into the digital twin model, and the warning information includes: warning type, warning location and warning time.

[0044] Compared with the prior art, the beneficial effects of the present invention are:

[0045] The present invention monitors the lubrication status of large machines in real time through multiple sensors, and can comprehensively and accurately obtain various monitoring data of the lubrication system of large machine equipment; by using the Internet of Things communication protocol to transmit the monitoring data to the cloud server, the stability, efficiency and security of data transmission are ensured, supporting the concurrent transmission of data of large-scale equipment, and being compatible with wireless networks and wired networks, reducing the risk of data transmission interruption or loss.

[0046] After processing the monitoring data on the cloud server, the present invention intelligently identifies and classifies the lubrication status through a deep belief network model, and combines time series analysis and prediction techniques to obtain the lubrication status value under the predicted lubrication status and be able to give an early warning accordingly.

[0047] The present invention constructs a digital twin model of the lubrication status of large machines by using digital twin technology to visually present the equipment lubrication system; by combining the fault identification model and the digital twin model, the lubricant flow path is simulated; by analyzing the influence coefficient of the fault point on the lubricant flow on the flow path, an early warning is given. Finally, the early warning information is visually displayed through the digital twin model, realizing the comprehensive monitoring and early warning of the lubrication status of large machines. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0049] Figure 1 It is a system module diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] The following will clearly and completely describe the technical solutions of the present invention in combination with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0051] Please refer to Figure 1 , the first aspect embodiment of the present invention provides an Internet of Things-based real-time monitoring and early warning system for the lubrication status of large machines, including the following modules:

[0052] Data acquisition module: monitors the lubrication status of large machines in real time through sensors and collects monitoring data in real time;

[0053] Data transmission module: Using the Internet of Things communication protocol, it transmits the monitoring data collected by sensors to the cloud server through a wireless network or a wired network;

[0054] Data processing module: On the cloud server, it stores, cleans, and standardizes the monitoring data;

[0055] Data analysis and early warning module: Using the deep belief network of the deep learning model to intelligently identify and classify the lubrication state; According to the identification and classification results of the current lubrication state, it conducts time series analysis on the lubrication state, and issues an early warning by calculating the lubrication state value under the predicted lubrication state; Combining digital twin technology, it constructs a digital twin model of the lubrication state of the large machine to simulate the lubrication state of the large machine; Based on the monitoring data, it establishes a fault identification model; According to the fault identification results and combined with the digital twin model, it simulates the flow path of the lubricant, and issues an early warning by calculating the influence coefficient of the fault point on the lubricant flow;

[0056] Visualization early warning display module: It visualizes and displays the early warning information through the digital twin model.

[0057] Specifically, according to the characteristics of the lubrication system of the large machine, select appropriate sensor types, including but not limited to: temperature sensors, pressure sensors, flow sensors, etc. By installing sensors at key lubrication parts of the large machine, ensure that the lubrication state can be comprehensively and accurately monitored. The sensors collect lubrication state data in real time and use lightweight Internet of Things communication protocols such as MQTT and CoAP to ensure low power consumption, real-time performance, and reliability of data transmission. According to the on-site environment, select appropriate network transmission methods, including: in areas covered by wired networks, give priority to using wired networks; in areas covered by wireless networks, use wireless networks such as 4G / 5G and Wi-Fi. During data transmission, use encryption technologies such as TLS / SSL to ensure data security and prevent data leakage or tampering. Establish a database on the cloud server to store the collected monitoring data. The database design should meet the requirements of efficient storage, retrieval, and analysis of data. Perform preprocessing operations on the received monitoring data, such as denoising, duplicate removal, and filling missing values, to improve data quality and standardize the data to ensure the consistency and comparability of data collected by different sensors. Use a deep belief network model to intelligently identify and classify the lubrication state. By training the deep belief network model, enable it to accurately identify different lubrication states. Conduct trend analysis on the time series data of the lubrication state to predict the future lubrication state. According to the prediction results, analyze the lubrication state values in the predicted lubrication state for early warning. Combine the actual structure of the large machine and the characteristics of the lubrication system to construct a digital twin model of the lubrication state of the large machine, which should be able to simulate the lubrication system. Establish a fault identification model based on the monitoring data to identify potential fault points. Combine the digital twin model to simulate the flow path of the lubricant and calculate the influence coefficient of the fault point on the lubricant flow. When the influence coefficient exceeds the preset threshold, trigger an early warning. Finally, visually display the early warning information through the digital twin model.

[0058] In this embodiment, the lubrication state of the large machine is monitored in real time by sensors, and monitoring data is collected in real time, including the following steps:

[0059] Install various sensors at various parts of the lubrication system of the large machine, including: bearings, gears, oil tanks, oil pumps, oil pipes, lubrication points; among them, install temperature sensors to monitor the temperature of the large machine equipment and the lubricant liquid; install pressure sensors to monitor the pressure of the lubricant; install oil quality sensors to monitor the density, conductivity, and dielectric constant of the lubricant; install vibration sensors to monitor the vibration information of the equipment; install noise sensors to monitor the noise signal of the equipment; install moisture sensors to monitor the moisture content in the lubricant; monitor the viscosity of the lubricant in real time and use a particle counter to monitor the number of particles in the lubricant; introduce a laser displacement sensor to monitor the oil film thickness of the lubricant in real time.

[0060] Specifically, based on the structure of the large equipment and the characteristics of the lubrication system, determine the installation positions and quantities of the sensors. Use appropriate tools and methods to fix the sensors at the predetermined positions to ensure the stability and accuracy of the sensors. Connect the signal lines and power supply lines of the sensors to ensure the normal operation of the sensors. After the installation of the sensors is completed, conduct debugging and calibration work to ensure the measurement accuracy and reliability of the sensors. Adjust parameters such as the sensitivity and zero point of the sensors by comparing with the actual measured values or standard values.

[0061] Temperature sensors: Install temperature sensors at key parts such as the bearings, gears, and oil tanks of the large equipment to comprehensively monitor the temperature status of the equipment. Install pressure sensors at positions such as the outlet of the oil pump and the oil pipe to monitor the pressure changes of the lubricant in real time. Install oil quality sensors in the oil tank or oil pipe to obtain the quality information of the lubricant in real time. Install vibration sensors on key transmission parts such as bearings and gears to monitor the vibration of the equipment. Install noise sensors on the outer shell of the equipment or nearby to obtain the noise information of the equipment in real time. Monitor the water content in the lubricant to prevent the decline of lubrication performance and equipment corrosion caused by excessive water content. Install water sensors in the oil tank or oil pipe to monitor the water content in the lubricant in real time. Monitor the viscosity of the lubricant by using a dedicated viscometer or the viscosity monitoring function integrated in the oil quality sensor. Install particle counters in the oil tank or oil pipe to monitor the number of particles in the lubricant in real time. Install laser displacement sensors near key lubrication points including bearings, gears, etc. to measure the thickness of the oil film.

[0062] In this embodiment, the deep belief network of the deep learning model is used to intelligently identify and classify the lubrication state, including the following steps:

[0063] According to the processed monitoring data, use the deep belief network of the deep learning model to automatically learn the feature representation of the monitoring data, identify and classify the lubrication state; the categories of the lubrication state include: hydrodynamic lubrication, hydrostatic lubrication, elastohydrodynamic lubrication, thin film lubrication, boundary lubrication, dry friction, and mixed lubrication;

[0064] Adopt a deep belief network model to construct an identification and classification model; use a restricted Boltzmann machine as the basic building block of the deep belief network model; through a layer-by-layer greedy training algorithm, first train each layer of the restricted Boltzmann machine separately; when training each layer of the restricted Boltzmann machine, use the contrastive divergence algorithm to optimize the parameters of the deep belief network model; stack the trained restricted Boltzmann machines layer by layer to form a complete deep belief network model;

[0065] After the layer-by-layer greedy training is completed, use the backpropagation algorithm to globally optimize the entire deep belief network model; add a softmax classifier to the output layer of the deep belief network model to classify the lubrication state;

[0066] Collect historical monitoring data to generate a training data set, and label the training data set to label the corresponding lubrication state class labels;

[0067] Input the labeled training data set into a deep belief network model for training, including the training of a softmax classifier;

[0068] Input the monitored data after real-time acquisition and processing into the trained deep belief network model, and output the recognition and classification results of the current large machine lubrication state at the time point of the large machine monitoring data collected in real time, including: hydrodynamic lubrication, hydrostatic lubrication, elastohydrodynamic lubrication, thin film lubrication, boundary lubrication, dry friction, and mixed lubrication.

[0069] Specifically, a deep belief network model is adopted as the recognition and classification model, and a restricted Boltzmann machine is used as the basic building block of the deep belief network model. Through the layer-by-layer greedy training algorithm, each layer of the restricted Boltzmann machine is trained separately. When training each layer of the restricted Boltzmann machine, the contrastive divergence algorithm is used to optimize the parameters of the deep belief network model. The trained restricted Boltzmann machines are stacked layer by layer to form a complete deep belief network model. After the layer-by-layer greedy training is completed, the backpropagation algorithm is used to globally optimize the entire deep belief network model to further improve the performance of the model. A softmax classifier is added to the output layer of the deep belief network model to classify the lubrication state. The softmax classifier can convert the output of the model into a probability distribution, thereby achieving accurate classification of different lubrication states. Historical monitoring data is collected, and this data should include monitoring data under different lubrication states. The collected training data set is labeled with the corresponding category labels of the lubrication state, including: hydrodynamic lubrication, hydrostatic lubrication, elastohydrodynamic lubrication, thin-film lubrication, boundary lubrication, dry friction, and mixed lubrication, etc. The collected data is preprocessed, including operations such as data cleaning and normalization, to ensure the quality and consistency of the data. The labeled training data set is input into the deep belief network model for training, including the training of the softmax classifier. During the training process, attention should be paid to suppressing overfitting phenomena. Techniques such as data augmentation, Dropout mechanism, L1 / L2 regularization, etc. can be used to improve the generalization ability of the model; methods such as grid search, random search, or Bayesian optimization are used to find the optimal combination of hyperparameters; by adjusting hyperparameters such as the parameters and learning rate of the model, the performance and accuracy of the model are optimized. After the training is completed, the validation data set should be used to evaluate and validate the model to ensure the accuracy and reliability of the model. At the same time, methods such as cross-validation can be used to further evaluate the performance of the model. Through real-time acquisition and processed monitoring data, this data should include the same features and information as the training data set. The processed real-time monitoring data is input into the trained deep belief network model, and the model will identify and classify the lubrication state based on the learned feature representation, and output the recognition and classification results of the lubrication state, including: hydrodynamic lubrication, hydrostatic lubrication, elastohydrodynamic lubrication, thin-film lubrication, boundary lubrication, dry friction, and mixed lubrication, etc. categories.

[0070] In this embodiment, according to the recognition and classification results of the lubrication state, time series analysis is performed on the lubrication state, and early warning is carried out by calculating the lubrication state value under the predicted lubrication state, including the following steps:

[0071] Based on the trained deep belief network model, the current lubrication state is identified; by continuously monitoring the lubrication system of the large machine, it is identified whether there is a transition from one lubrication state to another;

[0072] When there is a transition, collect the monitoring data of the large machine lubrication state at different time points, arrange the collected monitoring data in chronological order to form a time series data set;

[0073] Construct a long short-term memory network model, input the time series data set into the long short-term memory network model for training; by inputting the monitoring data of the current lubrication state identified by the deep belief network model into the trained long short-term memory network model, predict the lubrication state at the next time step;

[0074] Adopt a sliding window method to gradually predict the lubrication states at multiple future time steps. Starting from the current window of real-time monitoring, use the long short-term memory network model to predict the lubrication state at the next time step; by sliding the window forward by one time step, adding the latest real-time monitoring data, and removing the earliest monitoring data, repeat the prediction process; by continuously sliding the window and using the long short-term memory network model for prediction, gradually obtain the lubrication states at multiple future time steps;

[0075] For the lubrication states at multiple future time steps obtained by prediction, obtain the monitoring data under the predicted lubrication state; by calculating the lubrication state value formula under the predicted lubrication state: ,

[0076] where, represents the lubrication state value, represents the weight of the i-th sensor, represents the monitoring value of the i-th sensor at the j-th time step, represents the minimum value of the historical monitoring values of the i-th sensor, represents the maximum value of the historical monitoring values of the i-th sensor; t is the length of the time window;

[0077] Set a life end threshold for the lubrication state. When the lubrication state value is lower than the threshold, the end of life is reached, and a warning mechanism is triggered for warning.

[0078] Specifically, collect the historical monitoring data of the large machine lubrication system, including various monitoring data in sensors under different lubrication states, information such as timestamps, etc. Use the collected historical data to train a deep belief network model. Utilize the trained deep belief network model to perform real-time identification on the monitoring data of the current large machine lubrication system to determine the current lubrication state. By continuously monitoring the lubrication system of the large machine, identify whether there is a transition from one lubrication state to another. When a state transition is detected, start collecting the monitoring data of the large machine lubrication state at different time points. Arrange the collected monitoring data in chronological order to form a time series data set. This data set will be used to train a long short-term memory network model. Construct a long short-term memory network model that can process time series data and capture the long-term dependencies in the data. Input the time series data set into the long short-term memory network model for training so that the model can learn the law of the lubrication state changing over time. Input the monitoring data of the current lubrication state identified by the deep belief network model into the trained long short-term memory network model to predict the lubrication state at the next time step. Adopt a sliding window method to gradually predict the lubrication states at multiple future time steps. Starting from the current window of real-time monitoring, use the long short-term memory network model to predict the lubrication state at the next time step. Then, slide the window forward by one time step, add the latest real-time monitoring data, remove the earliest monitoring data, and repeat the prediction process. According to the predicted lubrication states at multiple future time steps, calculate the lubrication state values of the monitoring data under the predicted lubrication states. By setting an end-of-life threshold for the lubrication state, first collect a large amount of monitoring data of the large machine lubrication state at different time points, including data during normal operation and when lubrication problems occur; then analyze the collected historical data to determine the change trends and characteristics of the lubrication state during normal operation and failure of the equipment; then, according to the analysis results, select a suitable threshold setting method, including statistical methods such as mean ± standard deviation, methods based on equipment operation experience such as referring to the maintenance manual provided by the equipment manufacturer, and machine learning-based methods such as training a classification model using historical data to distinguish normal and abnormal states; then apply the set threshold to the actual monitoring data, observe the triggering situation of the early warning mechanism, and verify whether the threshold can accurately reflect the end of life of the lubrication state. If the early warning is too frequent or untimely, the threshold needs to be adjusted. In this embodiment, the end-of-life threshold for the lubrication state is set to 0.3, and the specific threshold needs to be comprehensively adjusted and determined according to the above methods and actual situations. When the calculated lubrication state value is lower than this threshold, it is considered that the large machine equipment has reached the end of life, and the early warning mechanism needs to be triggered for early warning.

[0079] In this embodiment, in combination with digital twin technology, a digital twin model of the large machine lubrication state is constructed to simulate the lubrication state of the large machine, including the following steps:

[0080] According to the actual geometry and dimensions of the large machine lubrication system, a geometric model is constructed using 3D modeling software, which describes the various components of the lubrication system, including: the shapes, positions, and interconnected relationships of bearings, gears, oil tanks, oil pumps, oil pipes, and lubrication points;

[0081] Based on the constructed geometric model, by solving the hydrodynamic equations, the flow state of the lubricant in the large machine lubrication system is simulated, including: oil pressure distribution and flow velocity distribution; by solving the heat conduction equation, the temperature distribution of the lubricant in the large machine lubrication system and the heat transfer process are obtained; by solving the mechanical equilibrium equation, the distributions of oil pressure, oil film thickness, and friction force parameters in the lubrication system are obtained;

[0082] By integrating the above geometric model and the analysis results of hydrodynamics, heat conduction, and mechanical equilibrium, a digital twin model of the lubrication state of the large machine is constructed;

[0083] The monitoring data of the large machine lubrication system are collected in real time through IoT sensors, and the digital twin model is associated with the monitoring data collected by the IoT sensors, and the actual operating state of the large machine lubrication system in the digital twin model is synchronized in real time through the data interface;

[0084] The trained deep belief network model is deployed into the digital twin model, and the current real-time lubrication state is output through the deep belief network model and simulated and displayed through the digital twin model.

[0085] Specifically, collect the actual geometry, dimensions of the large machine lubrication system, and detailed parameters of each component such as bearings, gears, oil tanks, oil pumps, oil pipes, lubrication points, etc. Use 3D modeling software such as SolidWorks, AutoCAD, etc. to construct a geometric model of the large machine lubrication system based on the collected data. Describe the shape, position, and interconnection relationship of each component in detail in the geometric model to ensure that the model is consistent with the actual situation. Based on the constructed geometric model, by solving the hydrodynamic equations, simulate the flow state of the lubricant in the large machine lubrication system, analyze the oil pressure distribution and flow velocity distribution to understand the flow of the lubricant in the system. By solving the heat conduction equation, obtain the temperature distribution of the lubricant in the large machine lubrication system and the heat transfer process, which helps to evaluate the thermal performance of the lubrication system and ensure that the lubricant works within an appropriate temperature range. By solving the mechanical equilibrium equation, obtain the distribution of key parameters such as oil pressure, oil film thickness, and friction force parameters in the lubrication system. These parameters are of great significance for evaluating the performance of the lubrication system and predicting potential failures. Integrate the geometric model and the results of hydrodynamic, heat conduction, and mechanical equilibrium analyses to construct a digital twin model of the large machine lubrication state, which should be able to comprehensively reflect the actual situation of the large machine lubrication system, including the geometric shape, positional relationship, and physical properties of each component. Design a data interface to achieve real-time data synchronization between the digital twin model and the IoT sensors, which are used to collect the monitoring data of the large machine lubrication system in real time. Collect the monitoring data of the large machine lubrication system in real time through the IoT sensors and synchronize the collected data to the digital twin model in real time through the data interface to reflect the actual operating state of the large machine lubrication system. The trained deep belief network model is used to predict the real-time lubrication state of the large machine lubrication system. Deploy the trained deep belief network model to the digital twin model, and output the current real-time lubrication state through the deep belief network model; use the digital twin model for simulation display to visually present the lubrication state of the large machine lubrication system.

[0086] In this embodiment, a fault identification model is established based on the monitoring data, including the following steps:

[0087] Use a convolutional neural network model combined with a recurrent neural network model in deep learning to establish a fault identification model;

[0088] Collect and process the historical monitoring data set of the large machine lubrication state, and label the fault types including: leakage and blockage; input the labeled historical monitoring data set into the fault identification model for training;

[0089] The monitored data collected and processed through the IoT in real time is input into the trained fault identification model, and it outputs whether a fault is identified in the current large machine lubrication state and outputs the fault type.

[0090] Specifically, collect historical monitoring data on the lubrication status of large machines. These data should cover different operating conditions and fault situations. Ensure the integrity, accuracy, and consistency of the data for subsequent analysis and processing. Clean the collected historical monitoring data to remove noise, outliers, and missing values; perform normalization or standardization on the data to improve the convergence speed and performance of the model. Label the monitoring data according to fault types including but not limited to: leakage and blockage to provide supervision information for model training. Use a convolutional neural network model combined with a recurrent neural network model in deep learning to establish a fault identification model. The convolutional neural network model is good at extracting local features from data with strong spatial correlation, while the recurrent neural network model is suitable for processing time series data to capture time-dependent relationships. Design a network structure containing multiple convolutional layers and pooling layers in the convolutional neural network model to extract spatial features from the lubrication status monitoring data. The recurrent neural network model is connected to the recurrent neural network after the convolutional neural network model to process time series data and capture the patterns of fault development over time. Set corresponding output nodes according to fault types and use the softmax function for multi-classification. Input the labeled historical monitoring data set into the fault identification model for training. Use the cross-entropy loss function as the optimization objective and use optimization algorithms such as gradient descent to update the model parameters. Optimize the training process and performance of the model by adjusting hyperparameters such as the learning rate and batch size. Use Internet of Things technology to monitor the lubrication status of large machines in real time and collect real-time monitoring data. Preprocess the real-time data to maintain the same data format and feature space as the training data. Input the preprocessed real-time monitoring data into the trained fault identification model. The model outputs whether a fault is identified under the current lubrication status of the large machine and gives the specific fault types including: leakage or blockage.

[0091] In this embodiment, according to the fault identification result combined with the digital twin model, simulate the flow path of the lubricant and give an early warning by calculating the influence coefficient of the fault point on the lubricant flow, including the following steps:

[0092] Deploy the trained fault identification model into the digital twin model. According to the fault identification model combined with the digital twin model, identify whether there are fault points in the lubricant flow path, including: leakage points and blockage points; according to the identification result, count the total number K of all fault points in the flow path;

[0093] According to the digital twin model, simulate the flow path of the lubricant in the large machine equipment, which is divided into a normal flow path and an abnormal propagation path; the abnormal propagation path is the flow path of the lubrication status of the large machine output by the fault identification model under the fault.

[0094] During the simulation, by using computational fluid dynamics methods, the flow conditions of the lubricant on the normal flow path and the abnormal propagation path are collected, including: flow velocity, temperature, and pressure distribution;

[0095] By calculating the influence coefficient formula of the fault point on the lubricant flow: ,

[0096] where, is the influence coefficient, is the pressure under the k-th fault point, is the flow velocity under the k-th fault point, is the temperature under the k-th fault point, is the pressure of the normal flow path, is the flow velocity of the normal flow path, is the temperature of the normal flow path, , and are the rates of change of pressure, flow velocity, and temperature with time under the k-th fault point, respectively. K represents the total number of all fault points on the flow path; when the influence coefficient exceeds the threshold of 0.2, the early warning mechanism is triggered for early warning.

[0097] Specifically, the trained fault identification model is seamlessly integrated into the digital twin model, which ensures that the fault identification module can obtain data from the digital twin model in real time and output fault point information including: leakage points and blockage points. Using the integrated fault identification model and combining with the real-time data of the digital twin model, the fault points in the lubricant flow path are accurately identified. The fault point information will be used for subsequent lubricant flow simulation and influence coefficient calculation. In the digital twin model, the complete flow path of the lubricant in the large machine equipment is simulated, including: the normal flow path and the abnormal propagation path caused by faults. According to the output of the fault identification model, the lubricant flow path is divided into the normal flow path and the abnormal propagation path. The abnormal propagation path specifically refers to the flow path of the lubricant in the fault state. During the simulation, computational fluid dynamics methods are used to accurately collect the temperature, flow velocity, and pressure distribution data of the lubricant on the normal flow path and the abnormal propagation path. The collected temperature, flow velocity, and pressure distribution data are substituted into the influence coefficient formula to understand the specific influence of all fault points on the lubricant flow. In this embodiment, the threshold of the influence coefficient is set to 0.2, and the threshold of the influence coefficient is dynamically adjusted specifically by combining the actual situation, expert experience, simulation experiments, or historical data. When the calculated influence coefficient exceeds this threshold, the early warning mechanism is triggered for early warning.

[0098] In this embodiment, visual display of warning information is carried out through a digital twin model, including the following steps: According to the calculated results of the lubrication state value under the predicted lubrication state and the influence coefficient of the fault point on the lubricant flow, when the warning mechanism is triggered, the generated warning information is integrated into the digital twin model, and the warning information includes: warning type, warning location, and warning time.

[0099] Specifically, when the calculated predicted lubrication state value is lower than the threshold, the end of life is reached and the warning mechanism is triggered; when the fault identification model combines with the digital twin model to detect a fault point and calculates that the influence coefficient of the fault point on the lubricant flow exceeds the preset threshold, the warning mechanism is triggered. According to the above warning mechanism, warning information is generated, including warning types such as: the lubrication state has reached the end of life, leakage point warning, blockage warning, etc., the warning location is the specific location of the fault point, and the warning time is the exact time point when the warning is triggered. The generated warning information is seamlessly integrated into the digital twin model, ensuring that the warning information can be synchronized with the real-time data of the digital twin model and accurately displayed.

[0100] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A real-time monitoring and early warning system for lubrication status of large machines based on the Internet of Things, characterized in that: Includes the following modules: Data acquisition module: monitors the lubrication status of large machines in real time through sensors and collects monitoring data in real time; Data transmission module: using the IoT communication protocol, the monitoring data collected by the sensor is transmitted to the cloud server through a wireless network or a wired network; Data processing module: Store, clean and standardize monitoring data on cloud servers; Data analysis and early warning module: Use the deep belief network of the deep learning model to intelligently identify and classify the lubrication status; perform time series analysis on the lubrication status based on the identification and classification results of the current lubrication status, and issue an early warning by calculating and predicting the lubrication status value under the lubrication status; Combined with digital twin technology, a digital twin model of the lubrication status of the large machine is constructed to simulate the lubrication status of the large machine; a fault identification model is established based on the monitoring data; based on the fault identification results combined with the digital twin model, the flow path of the lubricant is simulated, and an early warning is issued by calculating the influence coefficient of the fault point on the lubricant flow; Visual warning display module: Visually display warning information through the digital twin model; Among them, according to the fault identification results combined with the digital twin model, the flow path of the lubricant is simulated, and the influence coefficient of the fault point on the lubricant flow is calculated to issue an early warning, including the following steps: Deploy the trained fault identification model to the digital twin model, and identify the fault points in the lubricant flow path, including leakage points and blockage points, based on the fault identification model combined with the digital twin model. According to the identification results, count the total number K of all fault points in the flow path. The flow path of the lubricant in the large machine equipment is simulated according to the digital twin model and divided into a normal flow path and an abnormal propagation path; the abnormal propagation path is the flow path of the large machine lubrication state output by the fault identification model under the fault; During the simulation, computational fluid dynamics methods are used to collect the flow conditions of the lubricant on the normal flow path and the abnormal propagation path, including: flow velocity, temperature and pressure distribution; The influence coefficient formula of the fault point on the lubricant flow is calculated by: , in, is the influence coefficient, is the pressure under the kth fault point, is the flow rate at the kth fault point, is the temperature at the kth fault point, is the pressure of the normal flow path, is the flow velocity of the normal flow path, is the temperature of the normal flow path, , and are the rates of change of pressure, flow rate and temperature at the kth fault point over time, respectively, and K represents the total number of all fault points on the flow path; when the influence coefficient exceeds the threshold of 0.2, the early warning mechanism is triggered to issue an early warning.

2. According to the Internet of Things-based real-time monitoring and early warning system for lubrication status of large machines according to claim 1, it is characterized in that: The lubrication status of the large machine is monitored in real time by sensors, and monitoring data is collected in real time, including the following steps: Various sensors are installed in various parts of the lubrication system of the large machine, including bearings, gears, oil tanks, oil pumps, oil pipes, and lubrication points. Among them, temperature sensors are installed to monitor the temperature of the large machine equipment and lubricant liquid; pressure sensors are installed to monitor the pressure of the lubricant; oil quality sensors are installed to monitor the density, conductivity and dielectric constant of the lubricant; vibration sensors are installed to monitor the vibration information of the equipment; noise sensors are installed to monitor the noise signal of the equipment; moisture sensors are installed to monitor the moisture content in the lubricant; the viscosity of the lubricant is monitored in real time and a particle counter is used to monitor the number of particles in the lubricant; laser displacement sensors are introduced to monitor the oil film thickness of the lubricant in real time.

3. According to the Internet of Things-based real-time monitoring and early warning system for lubrication status of large machines according to claim 1, it is characterized in that: The deep belief network of the deep learning model is used to intelligently identify and classify the lubrication status, including the following steps: Based on the processed monitoring data, the deep belief network of the deep learning model is used to automatically learn the feature representation of the monitoring data, identify and classify the lubrication state; the categories of lubrication state include: fluid dynamic lubrication, liquid static lubrication, elastic fluid dynamic lubrication, thin film lubrication, boundary lubrication, dry friction and mixed lubrication; The recognition and classification model is constructed by using the deep belief network model; the restricted Boltzmann machine is used as the basic building block of the deep belief network model; each layer of the restricted Boltzmann machine is trained separately through the layer-by-layer greedy training algorithm; when training each layer of the restricted Boltzmann machine, the contrastive divergence algorithm is used to optimize the parameters of the deep belief network model; the trained restricted Boltzmann machines are stacked layer by layer to form a complete deep belief network model; After the layer-by-layer greedy training is completed, the back propagation algorithm is used to globally optimize the entire deep belief network model; a softmax classifier is added to the output layer of the deep belief network model to classify the lubrication state; Collect historical monitoring data to generate a training data set, and annotate the training data set to label the corresponding lubrication status category labels; Use the labeled training data set to input into the deep belief network model for training, including the training of the softmax classifier; The monitoring data collected and processed in real time is input into the trained deep belief network model, and the identification and classification results of the current large machine lubrication status at the time point of the real-time collected large machine monitoring data are output, including: fluid dynamic lubrication, liquid static lubrication, elastic fluid dynamic lubrication, thin film lubrication, boundary lubrication, dry friction and mixed lubrication.

4. According to the Internet of Things-based real-time monitoring and early warning system for lubrication status of large machines according to claim 3, it is characterized in that: According to the identification and classification results of the current lubrication state, the lubrication state is analyzed in time series, and the lubrication state value under the lubrication state is predicted by calculation to give an early warning, including the following steps: The current lubrication state is identified based on the trained deep belief network model; by continuously monitoring the lubrication system of the large machine, it is identified whether there is a transition from one lubrication state to another; When there is a transition, the monitoring data of the lubrication status of the large machine at different time points are collected, and the collected monitoring data are arranged in chronological order to form a time series data set; A long short-term memory network model is constructed, and the time series data set is input into the long short-term memory network model for training; the lubrication state of the next time step is predicted by inputting the monitoring data of the current lubrication state identified by the deep belief network model into the trained long short-term memory network model; The sliding window method is used to gradually predict the lubrication status of multiple time steps in the future. Starting from the current window of real-time monitoring, the long short-term memory network model is used to predict the lubrication status of the next time step; by sliding the window forward one time step, adding the latest real-time monitoring data, removing the earliest monitoring data, and repeating the prediction process; by continuously sliding the window and using the long short-term memory network model for prediction, the lubrication status of multiple time steps in the future is gradually obtained; For the predicted lubrication state of multiple time steps in the future, the monitoring data under the predicted lubrication state is obtained; the lubrication state value formula under the predicted lubrication state is calculated: , in, Indicates the lubrication status value, represents the weight of the i-th sensor, represents the monitoring value of the i-th sensor at the j-th time step, represents the minimum value of the historical monitoring value of the i-th sensor, represents the maximum value of the historical monitoring value of the i-th sensor; t is the length of the time window; A lubrication status life end threshold is set. When the lubrication status value is lower than the threshold, the life end is reached and the early warning mechanism is triggered to issue an early warning.

5. According to the Internet of Things-based real-time monitoring and early warning system for lubrication status of large machines according to claim 1, it is characterized in that: Combining digital twin technology, a digital twin model of the lubrication status of the large machine is constructed to simulate the lubrication status of the large machine, including the following steps: According to the actual geometric shape and size of the large machine lubrication system, a geometric model is constructed using 3D modeling software, which describes the various components of the lubrication system, including: the shape, position and interconnection relationship of the bearings, gears, oil tanks, oil pumps, oil pipes, and lubrication points; Based on the constructed geometric model, the flow state of lubricant in the large machine lubrication system is simulated by solving the fluid dynamics equation, including oil pressure distribution and flow velocity distribution; the temperature distribution of lubricant in the large machine lubrication system and the heat transfer process are obtained by solving the heat conduction equation; the distribution of oil pressure, oil film thickness and friction parameters in the lubrication system is obtained by solving the mechanical equilibrium equation; By integrating the above geometric models with the results of fluid dynamics, heat conduction and mechanical balance analysis, a digital twin model of the lubrication status of the large machine was constructed; The monitoring data of the large machine lubrication system is collected in real time through IoT sensors, and the digital twin model is associated with the monitoring data collected by the IoT sensors. The actual operating status of the large machine lubrication system in the digital twin model is synchronized in real time through the data interface; The trained deep belief network model is deployed into the digital twin model, the current real-time lubrication status is output through the deep belief network model, and simulated and displayed through the digital twin model.

6. The real-time monitoring and early warning system for lubrication status of large machines based on the Internet of Things according to claim 1 is characterized in that: Establishing a fault identification model based on monitoring data includes the following steps: The fault recognition model is established by combining the convolutional neural network model in deep learning with the recurrent neural network model; The historical monitoring data set of the lubrication status of the large machine is collected and processed, and the fault types are marked, including leakage and blockage; the annotated historical monitoring data set is input into the fault recognition model for training; The monitoring data collected and processed through real-time monitoring of the Internet of Things is input into the trained fault identification model, which outputs whether a fault is identified under the current lubrication state of the large machine and the type of fault.

7. The real-time monitoring and early warning system for lubrication status of large machines based on the Internet of Things according to claim 1 is characterized in that: The digital twin model is used to visualize the warning information, including the following steps: based on the calculated results of the lubrication state value and the influence coefficient of the fault point on the lubricant flow under the predicted lubrication state, when the warning mechanism is triggered, the generated warning information is integrated into the digital twin model, and the warning information includes: warning type, warning location and warning time.

Images