A method and system for the operation and maintenance of lubrication hydraulic equipment based on an online hydraulic model.

Abstract

This invention discloses a method and system for the operation and maintenance of lubrication hydraulic equipment based on an online hydraulic model, belonging to the field of intelligent operation and maintenance technology. The method includes the following steps: obtaining the structural diagram and operating principle description of the lubrication hydraulic equipment; constructing an online hydraulic model, acquiring current-time observation information, and comparing it with the predicted data of the online hydraulic model to generate observation features and dependency features; making pre-decision based on the observation features and dependency features to determine potential operation and maintenance schemes; generating a hierarchical communication network based on the minimum spanning tree algorithm; generating several joint decision schemes based on the potential operation and maintenance schemes, optimizing the generated joint decision schemes using the hierarchical communication network, evaluating all joint decision schemes using a state-action value function, and generating the final operation and maintenance scheme based on the evaluation results. This ensures that the operation and maintenance scheme can be efficiently transmitted and implemented during execution, ensuring the scientific nature and reliability of the decision-making.

Description

Technical Field

[0001] This invention relates to the field of intelligent operation and maintenance technology, and more specifically, to a method and system for the operation and maintenance of lubrication hydraulic equipment based on an online hydraulic model. Background Technology

[0002] Hydraulic lubrication equipment is a type of device used to control the pressure, flow rate, and direction of fluids. It transmits power through hydraulic oil and utilizes lubricating oil to reduce wear on mechanical parts, ensuring efficient system operation. These devices are widely used in industrial manufacturing, construction machinery, transportation, and other fields.

[0003] Operation and maintenance (O&M) of lubrication hydraulic equipment refers to the process of operating, maintaining, and managing the lubrication hydraulic system and its various components to ensure the normal operation of the equipment, extend its service life, and improve work efficiency. For example, regular inspections are the foundation for ensuring the normal operation of the lubrication hydraulic system; taking certain maintenance measures before equipment failure occurs prevents malfunctions; and when equipment failures do occur, rapid diagnosis and repair are crucial to minimize downtime.

[0004] For example, Chinese Patent 202210679014.5 discloses a method and system for fault diagnosis and operation and maintenance in the manufacturing process of intelligent hydraulic forming equipment. This method constructs an expert system and knowledge graph for the intelligent equipment to achieve rapid and accurate matching of operation and maintenance solutions. However, the above method still has the following shortcomings: it does not construct an online hydraulic model, resulting in an insufficient understanding of the equipment, and it does not generate multiple operation and maintenance solutions for comparison and optimization. This may lead to the final selected operation and maintenance solution being neither comprehensive nor optimal.

[0005] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0006] To address the problems in related technologies, this invention proposes a method and system for the operation and maintenance of lubrication hydraulic equipment based on an online hydraulic model, in order to overcome the aforementioned technical problems existing in the existing related technologies.

[0007] Therefore, the specific technical solution adopted by the present invention is as follows:

[0008] According to one aspect of the present invention, a method for the operation and maintenance of lubrication hydraulic equipment based on an online hydraulic model is provided, the method comprising the following steps:

[0009] S1. Obtain structural data and operating principle data of the lubrication hydraulic equipment, identify key components and their interrelationships, and obtain structural diagrams and operating principle descriptions.

[0010] S2. Based on the structural diagram and operating principle description, construct an online hydraulic model, obtain the observation information at the current moment, compare it with the prediction data of the online hydraulic model, and generate observation features and dependency features.

[0011] S3. Make pre-decision based on observation characteristics and dependency characteristics to determine potential operation and maintenance solutions.

[0012] S4. Based on the minimum spanning tree algorithm, generate a hierarchical communication network to achieve selective directed communication.

[0013] S5. Generate several joint decision schemes based on potential operation and maintenance schemes, optimize the generated joint decision schemes using a hierarchical communication network, evaluate all joint decision schemes using a state action value function, and generate the final operation and maintenance scheme based on the evaluation results.

[0014] Based on the minimum spanning tree algorithm, generating a hierarchical communication network and realizing selective directed communication includes the following steps:

[0015] S41. Treat each key component in the lubrication hydraulic equipment as a node in the network, and establish edges between nodes according to the dependencies and functional connections between key components to obtain the initial graph;

[0016] S42. Set the weights of each edge to obtain a weighted device hierarchy network diagram;

[0017] S43. Determine the number of cluster partitions and use a clustering algorithm to cluster the nodes to obtain the clustering results;

[0018] S44. Using the nodes in the clustering results as nodes in the communication network, the minimum spanning tree algorithm is used to find the minimum spanning tree in the clustering network, and a hierarchical communication network based on the minimum spanning tree is obtained.

[0019] S45. Define directed communication rules and configure the network to enable selective directed communication.

[0020] Furthermore, an online hydraulic model is constructed, and the observation information at the current moment is obtained. This information is then compared with the prediction data of the online hydraulic model to generate observation features and dependency features, including the following steps:

[0021] S21. Based on the structural diagram and operating principle description, obtain the internal fluid flow of the lubricating hydraulic equipment and construct a fluid dynamics model;

[0022] S22. Construct a thermodynamic model based on the thermodynamic characteristics of the lubricating hydraulic equipment;

[0023] S23. Integrate the fluid dynamics model and the thermodynamics model to form an online hydraulic model, and input the initial conditions and parameters at the current moment into the online hydraulic model to generate prediction data of the current operating state.

[0024] S24. Obtain the observation information at the current moment, calculate the difference between the observation information and the predicted data, and identify outliers that exceed the expected range;

[0025] S25. Calculate the residuals between the observed data and the predicted data, and extract the observation features and dependency features based on the difference analysis and residual calculation.

[0026] Furthermore, based on observational and dependency characteristics, pre-decision making to determine potential operation and maintenance solutions includes the following steps:

[0027] S31. Extract key features from observation features and dependency features;

[0028] S32. Based on the extracted key features, use machine learning algorithms to classify fault modes;

[0029] S33. Analyze the potential causes of the identified fault modes;

[0030] S34. Based on the failure mode and potential causes, match suitable potential operation and maintenance solutions from the existing operation and maintenance solution library.

[0031] Furthermore, based on the extracted key features, the failure mode classification using machine learning algorithms includes the following steps:

[0032] S321. Collect observational features and dependency features, as well as corresponding key features, from historical normal operation and various failure modes.

[0033] S322. Mark the key features in S321 to indicate the operating status of the lubrication hydraulic equipment at that time;

[0034] S323. Divide the labeled dataset into a training set and a test set;

[0035] S324. Train the support vector machine model using the training and test sets, and evaluate and adjust the support vector machine model.

[0036] S325. Given new observation features and dependent features, use a well-trained and evaluated support vector machine model to predict the current state of the lubrication hydraulic equipment.

[0037] Furthermore, determining the number of cluster partitions and using a clustering algorithm to cluster the nodes to obtain the clustering results includes the following steps:

[0038] S431. Use the elbow rule to determine the number of cluster partitions;

[0039] S432. Randomly select k nodes as initial cluster centers;

[0040] S433. Use Euclidean distance to calculate the distance from each node to each cluster center;

[0041] S434. Based on the calculated distance, assign each node to the nearest cluster;

[0042] S435. Calculate the new centers for each cluster;

[0043] S436. Repeat steps S433-S435 until the preset number of iterations is reached or the cluster centers no longer change significantly.

[0044] Furthermore, using the nodes in the clustering results as nodes in the communication network, the minimum spanning tree algorithm is used to find the minimum spanning tree in the clustering network, resulting in a hierarchical communication network based on the minimum spanning tree, including the following steps:

[0045] S441. Treat all nodes in each cluster partition as independent nodes in the network;

[0046] S442. Construct an undirected graph, where the edges of the undirected graph represent the connections between independent nodes, and assign weights to each edge to obtain a weighted undirected graph.

[0047] S443. Based on a weighted undirected graph and executing the minimum spanning tree algorithm, the tree is gradually constructed by selecting the edge with the minimum weight until all nodes are included.

[0048] S444. Based on the structure of the minimum spanning tree, determine the hierarchical relationship between nodes and identify the root node in the minimum spanning tree;

[0049] S445. Determine the paths from the root node to all other nodes to form a hierarchical structure.

[0050] Furthermore, several joint decision-making schemes are generated based on potential operation and maintenance solutions. These joint decision-making schemes are then optimized using a hierarchical communication network. A state-action value function is used to evaluate all joint decision-making schemes. Based on the evaluation results, the final operation and maintenance solution is generated, including the following steps:

[0051] S51. Combine different potential operation and maintenance solutions to generate several joint decision-making solutions;

[0052] S52. Optimize the generated joint decision-making scheme using a hierarchical relationship network;

[0053] S53. Use the state-action value function to evaluate each joint decision-making scheme and calculate the state-action value;

[0054] S54. Sort all joint decision schemes according to the state action values ​​and select the optimal scheme.

[0055] Furthermore, each joint decision-making scheme is evaluated using a state-action value function. The calculation of the state-action value includes the following steps:

[0056] S531. Define the status and operation and maintenance actions of the operation and maintenance plan;

[0057] S532. Initialize the status action value table for each potential operation and maintenance solution, and design the reward function;

[0058] S533, Execute the operation and maintenance actions in each joint decision-making scheme and obtain new status information;

[0059] S534. Based on the new state information, the acquisition of rewards, and the prediction of future returns, update the state action values ​​of each joint decision-making scheme.

[0060] Furthermore, the formula for the reward function is:

[0061] ;

[0062] The formula for updating the state action value is:

[0063] ;

[0064] In the formula, R imd ( t () indicates at a point in time t Instant rewards;

[0065] P t Indicates a point in time t Actual measured values ​​of performance indicators P ref Indicates a reference value for a performance indicator. C R represents the cost of implementing the operation and maintenance plan, and R represents the cost at a given point in time. t Risk indicators R long Indicates long-term returns;

[0066] w 1 represents the weighting coefficient of the performance indicator. w 2 represents the weighting coefficient for operation and maintenance costs. w 3 indicates the weighting coefficient of the risk indicator;

[0067] Q old ( S , A This indicates that maintenance actions are being taken. A Previous status S Q value under, Q new ( S, A () represents the updated Q value;

[0068] γ Indicates the discount factor. α Indicates the learning rate;

[0069] Q ( S next , A next ) indicates the new state S next The following maintenance actions A next The expected Q value;

[0070] max Anext Q ( S next , A next ) indicates the new state S next The maximum Q value among all maintenance actions.

[0071] According to another aspect of the present invention, a lubrication hydraulic equipment operation and maintenance system based on an online hydraulic model is also provided. The lubrication hydraulic equipment operation and maintenance system based on an online hydraulic model includes an equipment data acquisition module, a feature acquisition module, a potential solution generation module, a network generation module, and a final solution generation module; wherein the equipment data acquisition module, the feature acquisition module, the potential solution generation module, the network generation module, and the final solution generation module are sequentially connected.

[0072] The equipment data acquisition module is used to acquire structural data and operating principle data of lubrication hydraulic equipment, identify key components and their interrelationships, and obtain structural diagrams and operating principle descriptions.

[0073] The feature acquisition module is used to construct an online hydraulic model based on the structure diagram and operating principle description, acquire the observation information at the current moment, compare it with the prediction data of the online hydraulic model, and generate observation features and dependency features.

[0074] The potential solution generation module is used to make pre-decision based on observation features and dependency features to determine potential operation and maintenance solutions.

[0075] The network generation module is used to generate hierarchical communication networks based on the minimum spanning tree algorithm, enabling selective directed communication.

[0076] The final solution generation module is used to generate several joint decision-making schemes based on potential operation and maintenance schemes, optimize the generated joint decision-making schemes using a hierarchical communication network, evaluate all joint decision-making schemes using a state action value function, and generate the final operation and maintenance scheme based on the evaluation results.

[0077] The beneficial effects of this invention are as follows:

[0078] (1) The present invention provides a lubrication hydraulic equipment operation and maintenance method and system based on an online hydraulic model. By acquiring the structural diagram and operating principle description of the lubrication hydraulic equipment, it provides basic data for subsequent steps and constructs an online hydraulic model, which can accurately simulate the operating status of the equipment and predict possible performance changes and failures. It uses observation features and dependency features to make pre-decision, ensuring that the pre-decision basis is sufficient and reducing blindness and subjectivity. Developing potential operation and maintenance plans in advance helps to better plan maintenance work.

[0079] (2) This invention generates a hierarchical communication network using the minimum spanning tree algorithm, ensuring the shortest and most efficient communication path, reducing information transmission delays and resource consumption, facilitating hierarchical processing and distribution of information, reducing communication conflicts and information redundancy, and ensuring efficient transmission and implementation of the joint decision-making scheme generated through the optimization of the hierarchical communication network, reducing execution errors and delays. Finally, a state action value function is used for quantitative evaluation to ensure the scientific nature and reliability of the decision, reducing biases caused by subjective judgment. The final generated operation and maintenance scheme has undergone multiple evaluations and optimizations, possessing high feasibility and execution efficiency, and contributing to achieving optimal maintenance results and equipment performance. Attached Figure Description

[0080] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0081] Figure 1 This is a flowchart of a lubrication hydraulic equipment operation and maintenance method based on an online hydraulic model according to an embodiment of the present invention;

[0082] Figure 2 This is a schematic diagram of a lubrication hydraulic equipment operation and maintenance system based on an online hydraulic model according to an embodiment of the present invention.

[0083] In the picture:

[0084] 1. Equipment data acquisition module; 2. Feature acquisition module; 3. Potential solution generation module; 4. Network generation module; 5. Final solution generation module. Detailed Implementation

[0085] To further illustrate the various embodiments, the present invention provides accompanying drawings, which are part of the disclosure of the present invention. These drawings are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these drawings, those skilled in the art should be able to understand other possible implementation methods and the advantages of the present invention. The components in the drawings are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0086] According to an embodiment of the present invention, a method and system for the operation and maintenance of lubrication hydraulic equipment based on an online hydraulic model is provided.

[0087] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 As shown, according to an embodiment of the present invention, a method for the operation and maintenance of lubrication hydraulic equipment based on an online hydraulic model is provided. This method includes the following steps:

[0088] S1. Obtain structural and operational principle data of the lubrication hydraulic equipment, identify key components and their interrelationships, and obtain structural diagrams and operational principle descriptions. Specifically:

[0089] Collect equipment documentation and equipment composition information for lubrication hydraulic equipment, and obtain data on equipment operating principles;

[0090] Identify key components and their interrelationships, such as key pumps, main valves, and key sensors; hydraulic oil flow paths, pressure change paths, and temperature change paths;

[0091] Demonstrate the physical layout of each component and the connection relationships of key parts. Explain the function and interaction of each part, and describe the workflow and operating procedures under different working conditions.

[0092] S2. Based on the structural diagram and operating principle description, construct an online hydraulic model, including a fluid dynamics model and a thermodynamic model, and obtain the current observation information (such as pressure, temperature, flow rate, etc.), and compare it with the prediction data of the online hydraulic model to generate observation features and dependency features.

[0093] In a further embodiment, constructing an online hydraulic model, obtaining the observation information at the current moment, and comparing it with the prediction data of the online hydraulic model to generate observation features and dependency features includes the following steps:

[0094] S21. Based on the structural diagram and operating principle description, obtain the internal fluid flow of the lubricating hydraulic equipment, and construct a fluid dynamics model, including the fluid flow equations and interactions of components such as pipes, valves, and pumps.

[0095] S22. Based on the thermodynamic characteristics of the lubricating hydraulic equipment, construct a thermodynamic model, including the thermodynamic equations for processes such as heat source, heat conduction, and heat dissipation.

[0096] S23. Integrate the fluid dynamics model and the thermodynamics model to form an online hydraulic model, and input the initial conditions and parameters at the current moment into the online hydraulic model to generate prediction data of the current operating state.

[0097] S24. Obtain the observation information at the current moment, calculate the difference between the observation information and the predicted data, and identify outliers that exceed the expected range.

[0098] S25. Calculate the residuals between the observed data and the predicted data, and extract the observation features and dependency features based on the difference analysis and residual calculation.

[0099] For example, in lubricating hydraulic equipment, the parameters to be observed include pressure, temperature, and flow rate.

[0100] Current observation information:

[0101] Observed pressure: 150 bar. Observed temperature: 80℃. Observed flow rate: 20 L / min.

[0102] Predictive data from the online hydraulic model:

[0103] Predicted pressure: 140 bar. Predicted temperature: 78°C. Predicted flow rate: 22 L / min.

[0104] Variance analysis and residual calculation:

[0105] Pressure difference, 10 bar. Temperature difference, °C. Flow rate difference, -2 L / min.

[0106] Comparative analysis revealed that:

[0107] A large residual for pressure indicates that the actual pressure is higher than the predicted value. A negative residual for flow indicates that the actual flow is lower than the predicted value.

[0108] The key observational features were high pressure and low flow rate, suggesting a blockage or reduced pump efficiency in the system.

[0109] Further analysis revealed that the decrease in flow rate was primarily dependent on the increase in pressure, which was caused by the increase in internal resistance of the system.

[0110] S3. Make pre-decision based on observation characteristics and dependency characteristics to determine potential operation and maintenance solutions.

[0111] In a further embodiment, the process of making pre-decision based on observation features and dependency features to determine potential operation and maintenance solutions includes the following steps:

[0112] S31. Extract key features from observational and dependent features, such as constant values ​​(mean, maximum, minimum) and trends (slope, fluctuation range) of pressure, temperature, flow rate, etc.

[0113] S32. Based on the extracted key features, use machine learning algorithms to classify possible failure modes. According to past experience and equipment working principles, define possible failure modes, such as leakage, blockage, wear, etc.

[0114] S33. For the identified fault modes, analyze the potential causes by combining expert knowledge and historical data, such as equipment leakage, pipeline blockage, sensor failure, improper parameter settings, etc.

[0115] S34. Based on the failure mode and potential causes, use matching algorithms (such as similarity calculation, rule matching, etc.) to match suitable potential operation and maintenance solutions from the existing operation and maintenance solution library, including replacing parts, adjusting parameters, adding lubricating oil, etc.

[0116] In a further embodiment, classifying fault modes using machine learning algorithms based on extracted key features includes the following steps:

[0117] S321. Collect the observed features and dependency features of normal operation and various failure modes in history, as well as the corresponding key features.

[0118] S322. Mark the key features in S321 to indicate the operating status (normal or specific fault mode) of the lubrication hydraulic equipment at that time.

[0119] S323. The labeled dataset is divided into training set and test set by randomization.

[0120] S324. Select Support Vector Machine as the fault classification model, determine the kernel function to be used, train the Support Vector Machine model using the training set and test set, and evaluate and adjust the Support Vector Machine model.

[0121] S325. Given new observation features and dependent features, use a well-trained and evaluated support vector machine model to predict the current state of the lubrication hydraulic equipment, including fault mode or normal state.

[0122] S4. Based on the minimum spanning tree algorithm, a hierarchical communication network is generated to realize selective directed communication. Through the hierarchical communication network, observation information and intention information are shared.

[0123] In a further embodiment, generating a hierarchical communication network based on the minimum spanning tree algorithm to achieve selective directed communication includes the following steps:

[0124] S41. Treat each key component in the lubrication hydraulic equipment as a node in the network, and establish edges between nodes according to the dependencies and functional connections between key components to obtain the initial graph.

[0125] S42. Based on the changes in flow and pressure between key components or other key parameters, set the weights of each edge to obtain a weighted device hierarchy network diagram.

[0126] S43. Determine the number of cluster partitions and use a clustering algorithm to cluster the nodes to obtain the clustering results; select a suitable clustering algorithm, such as the K-means clustering algorithm, to perform node clustering. If there are few nodes, clustering may not be performed.

[0127] S44. Using the nodes in the clustering results as nodes in the communication network, the minimum spanning tree algorithm is used to find the minimum spanning tree in the clustering network, and a hierarchical communication network based on the minimum spanning tree is obtained.

[0128] S45. Define directed communication rules and configure the network to enable selective directed communication.

[0129] Example:

[0130] The nodes include pumps, main valves, auxiliary valves, coolers, pressure sensor 1, pressure sensor 2, temperature sensor, and flow sensor.

[0131] The edges include: Pump → Main Valve, weight 1; Pump → Auxiliary Valve, weight 1; Pump → Cooler, weight 1; Main Valve → Pressure Sensor 1, weight 0.5; Auxiliary Valve → Pressure Sensor 2, weight 0.5; Cooler → Temperature Sensor, weight 0.5; Pump → Flow Sensor, weight 1.

[0132] Construct the minimum spanning tree:

[0133] Pump.

[0134] ├──Main valve.

[0135] │ └── Pressure sensor 1.

[0136] ├──Auxiliary valve.

[0137] │ └── Pressure sensor 2.

[0138] ├──Cooler.

[0139] │ └── Temperature sensor.

[0140] └── Flow sensor.

[0141] As can be seen from the minimum spanning tree above, ensuring that each node in the tree is connected to the pump forms an effective hierarchical communication network.

[0142] Directed communication needs to be established between nodes.

[0143] Pump → Main valve. Pump → Auxiliary valve. Pump → Cooler. Main valve → Pressure sensor 1. Auxiliary valve → Pressure sensor 2. Cooler → Temperature sensor. Pump → Flow sensor.

[0144] Define communication rules.

[0145] Pump → Main Valve: Transmits pump speed and flow rate settings, once per second.

[0146] Pump → Auxiliary Valve: Transmits pump speed and flow rate settings, once per second.

[0147] Pump → Cooler: Transmits pump speed and cooling demand once per second.

[0148] Main valve → Pressure sensor 1: Transmits pressure data once per second.

[0149] Auxiliary valve → Pressure sensor 2: Transmits pressure data once per second.

[0150] Cooler → Temperature sensor: Transmits temperature data once per second.

[0151] Pump → Flow sensor: Transmits flow data once per second.

[0152] In a further embodiment, determining the number of clustering partitions and using a clustering algorithm to cluster the nodes to obtain the clustering results includes the following steps:

[0153] S431. Use the elbow rule to determine the number of cluster partitions.

[0154] S432. Randomly select k nodes as initial cluster centers, and these centers will serve as the reference points for the initial clustering.

[0155] S433. Use Euclidean distance to calculate the distance from each node to each cluster center. The shorter the distance, the closer the node is to the center.

[0156] S434. Based on the calculated distance, assign each node to the nearest cluster.

[0157] S435. Calculate the new center of each cluster, that is, calculate the mean of all nodes in the cluster and use it as the new cluster center.

[0158] S436. Repeat steps S433-S435 until the preset number of iterations is reached or the cluster centers no longer change significantly.

[0159] In a further embodiment, the nodes in the clustering results are used as nodes in the communication network. The minimum spanning tree algorithm is used to find the minimum spanning tree in the clustering network to obtain a hierarchical communication network based on the minimum spanning tree, which includes the following steps:

[0160] S441. Treat all nodes in each cluster partition as independent nodes in the network.

[0161] S442. Construct an undirected graph, where the edges of the undirected graph represent the connections between each independent node, and assign weights to each edge to obtain a weighted undirected graph; the weights between clustered nodes are based on the connection strength between the original set of nodes they represent, which can be the sum or average of the weights of the edges between the original nodes.

[0162] S443. Based on a weighted undirected graph, execute the minimum spanning tree algorithm, select the edge with the minimum weight to gradually build the tree until all nodes are included.

[0163] It should be noted that the minimum spanning tree algorithm includes:

[0164] Prim's algorithm: Starting from any node, it gradually expands to all nodes, adding nodes that connect to the existing tree and edges with the smallest weights at each step.

[0165] Kruskal's algorithm: Sort all edges by weight from smallest to largest, select the edge with the smallest weight that does not form a cycle, and gradually build the tree.

[0166] S444. Based on the structure of the minimum spanning tree, determine the hierarchical relationship between nodes and identify the root node in the minimum spanning tree, which is usually the node connected by the edge with the smallest weight.

[0167] S445. Determine the paths from the root node to all other nodes to form a hierarchical structure.

[0168] S5. Generate several joint decision schemes based on potential operation and maintenance schemes, optimize the generated joint decision schemes using a hierarchical communication network, evaluate all joint decision schemes using a state action value function, and generate the final operation and maintenance scheme based on the evaluation results.

[0169] In a further embodiment, several joint decision-making schemes are generated based on potential operation and maintenance schemes, and the generated joint decision-making schemes are optimized using a hierarchical communication network. A state-action value function is then used to evaluate all joint decision-making schemes. Based on the evaluation results, the final operation and maintenance scheme is generated, including the following steps:

[0170] S51. Combine different potential operation and maintenance solutions according to the equipment components and operation types they affect to generate several joint decision-making solutions.

[0171] S52. Use hierarchical relationship networks to optimize the joint decision-making scheme to ensure efficient information transfer and collaborative operation between different equipment components.

[0172] S53. Use the state-action value function (Q function) to evaluate each joint decision scheme and calculate the state-action value (Q value).

[0173] S54. Sort all joint decision schemes according to the state action values ​​and select the optimal scheme.

[0174] For example, potential maintenance solution A: Adjust pump speed to increase flow rate. Potential maintenance solution B: Adjust main valve to optimize main system pressure. Potential maintenance solution C: Adjust auxiliary valve to optimize auxiliary system pressure. Potential maintenance solution D: Clean the cooler to reduce system temperature.

[0175] Joint decision-making option 1: Option A + Option B + Option D.

[0176] Joint decision-making option 2: Option A + Option C + Option D.

[0177] Joint decision-making option 1:

[0178] Pump → Main valve → Cooler → Pressure sensor 1, Temperature sensor.

[0179] Coordinated operation: After increasing the pump speed, immediately adjust the pressure through the main valve and simultaneously clean the cooler to ensure the temperature drops.

[0180] Joint decision-making option 2:

[0181] Pump → Auxiliary valve → Cooler → Pressure sensor 2, Temperature sensor.

[0182] Coordinated operation: After increasing the pump speed, immediately adjust the pressure through the auxiliary valve and simultaneously clean the cooler to ensure that the temperature drops.

[0183] In a further embodiment, each joint decision scheme is evaluated using a state-action value function. Calculating the state-action value includes the following steps:

[0184] S531. Define the status of the operation and maintenance plan, including equipment operating status, resource usage, historical maintenance records, etc., and define operation and maintenance actions, such as replacing parts, adjusting configurations, restarting services, etc.

[0185] S532. Initialize a state-action value table for each potential operation and maintenance solution, record the Q-value of each state-action pair, and design a reward function to evaluate the immediate effect and long-term impact of each action. Rewards can be based on factors such as system performance improvement, cost savings, and risk reduction.

[0186] S533. Execute the maintenance actions in each joint decision-making scheme and obtain new status information, that is, obtain the actual impact of the maintenance actions on the lubrication hydraulic equipment.

[0187] S534. Based on the new state information, the acquisition of rewards, and the prediction of future returns, update the state action values ​​of each joint decision-making scheme.

[0188] Among them, the acquisition of the new state: after the action is performed, the environment will return a new state, which is the starting point for the next update process.

[0189] Reward Acquisition: An immediate reward is given based on the action performed. This reward is typically calculated based on performance metrics of the new state, reflecting the direct result of the action.

[0190] Predicting future returns: The discount factor and the maximum expected Q value of the new state are used together to predict future returns. The discount factor determines the importance of future rewards relative to current rewards, while the maximum expected Q value represents the maximum expected return among all possible next actions in the new state.

[0191] In a further embodiment, the formula for the reward function is:

[0192] ;

[0193] The formula for updating the state action value is:

[0194] ;

[0195] In the formula, R imd ( t () indicates at a point in time t Instant rewards; P t Indicates a point in time t Actual measured values ​​of performance indicators P ref Indicates the reference or ideal value of a performance indicator. C R represents the cost of implementing the operation and maintenance plan, and R represents the cost at a given point in time. t The risk indicators take into account factors such as failure probability, potential downtime, and safety risks. R long It represents long-term returns and takes into account the expectation of future rewards; w 1 represents the weighting coefficient of the performance indicator. w 2 represents the weighting coefficient for operation and maintenance costs. w 3 indicates the weighting coefficient of the risk indicator;

[0196] Q old (S , A This indicates that maintenance actions are being taken. A Previous status S Q value under, Q new ( S , A () represents the updated Q value; γ This represents the discount factor, used to calculate the current value of future rewards. α Indicates the learning rate; Q ( S next , A next ) indicates the new state S next The following maintenance actions A next The expected Q value; max Anext Q ( S next , A next ) indicates the new state S next The maximum Q value among all maintenance actions.

[0197] For example, set the following parameters.

[0198] Performance metrics: The current performance metrics of the device (e.g., flow rate), ranging from 0 to 100.

[0199] Reference value for performance indicators: The performance indicator value under ideal conditions is set to 90.

[0200] Operation and maintenance costs: The cost of performing each operation and maintenance action, for example, adjusting the pump speed costs 10, and cleaning the cooler costs 20.

[0201] Risk indicators: Based on factors such as failure probability, potential downtime, or safety risks, set to a value between 0 and 1, for example, 0.05.

[0202] Long-term return: Expected future reward, which can be set as a fixed value or an estimate based on some predictive model, such as 100.

[0203] Weighting coefficient w 1, w 2, w 3: Set them to 1, 2, 1 respectively.

[0204] Q old ( S , AThe value is set to 50. The discount factor is set to 0.9. The learning rate is set to 0.1. The maximum expected Q value in the new state is 80.

[0205] At a certain point in time t The actual performance index of the equipment is 85. The maintenance actions in Joint Decision Scheme 1 were performed, including adjusting the pump speed and cleaning the cooler, with a total cost of 30. The updated Q value is 62.65.

[0206] All joint decision-making schemes are ranked according to their Q values, and the optimal scheme, i.e., the scheme with the highest Q value, is selected.

[0207] like Figure 2 As shown, according to another embodiment of the present invention, a lubrication hydraulic equipment operation and maintenance system based on an online hydraulic model is also provided. The lubrication hydraulic equipment operation and maintenance system based on an online hydraulic model includes an equipment data acquisition module 1, a feature acquisition module 2, a potential solution generation module 3, a network generation module 4, and a final solution generation module 5; wherein, the equipment data acquisition module 1, the feature acquisition module 2, the potential solution generation module 3, the network generation module 4, and the final solution generation module 5 are connected in sequence.

[0208] Equipment data acquisition module 1 is used to acquire structural data and operating principle data of lubrication hydraulic equipment, identify key components and their interrelationships, and obtain structural diagrams and operating principle descriptions.

[0209] Feature acquisition module 2 is used to construct an online hydraulic model based on the structure diagram and operating principle description, acquire the observation information at the current moment, compare it with the prediction data of the online hydraulic model, and generate observation features and dependency features.

[0210] Potential solution generation module 3 is used to make pre-decision based on observation features and dependency features to determine potential operation and maintenance solutions.

[0211] Network generation module 4 is used to generate hierarchical communication networks based on the minimum spanning tree algorithm to achieve selective directed communication.

[0212] The final solution generation module 5 is used to generate several joint decision-making schemes based on potential operation and maintenance schemes, optimize the generated joint decision-making schemes using a hierarchical communication network, evaluate all joint decision-making schemes using a state action value function, and generate the final operation and maintenance scheme based on the evaluation results.

[0213] In summary, this invention provides a lubrication hydraulic equipment operation and maintenance method and system based on an online hydraulic model. By acquiring the structural diagram and operating principle description of the lubrication hydraulic equipment, it provides the basic data for subsequent steps and constructs an online hydraulic model, which can accurately simulate the equipment's operating status and predict possible performance changes and failures. Pre-decision is made using observed and dependent features, ensuring sufficient basis for pre-decision and reducing blindness and subjectivity. Developing potential operation and maintenance plans in advance helps to better plan maintenance work. This invention generates a hierarchical communication network using the minimum spanning tree algorithm, ensuring the shortest and most efficient communication path, reducing information transmission delays and resource consumption, facilitating hierarchical processing and distribution of information, reducing communication conflicts and information redundancy. Furthermore, the joint decision-making scheme generated through the optimization of the hierarchical communication network ensures that the decision scheme can be efficiently transmitted and implemented during execution, reducing execution errors and delays. Finally, a state-action value function is used for quantitative evaluation, ensuring the scientific nature and reliability of the decision and reducing biases caused by subjective judgment. The final generated operation and maintenance scheme, after multiple evaluations and optimizations, has high feasibility and execution efficiency, contributing to achieving optimal maintenance results and equipment performance.

[0214] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for the operation and maintenance of lubrication hydraulic equipment based on an online hydraulic model, characterized in that, The operation and maintenance method for lubrication hydraulic equipment based on the online hydraulic model includes the following steps: S1. Obtain structural data and operating principle data of the lubrication hydraulic equipment, identify key components and their interrelationships, and obtain structural diagrams and operating principle descriptions; S2. Based on the structural diagram and operating principle description, construct an online hydraulic model, obtain the observation information at the current moment, compare it with the prediction data of the online hydraulic model, and generate observation features and dependency features. S3. Make pre-decision based on observation characteristics and dependency characteristics to determine potential operation and maintenance solutions; S4. Based on the minimum spanning tree algorithm, generate a hierarchical communication network to achieve selective directed communication; S5. Generate several joint decision schemes based on potential operation and maintenance schemes, optimize the generated joint decision schemes using a hierarchical communication network, evaluate all joint decision schemes using a state action value function, and generate the final operation and maintenance scheme based on the evaluation results. The method for generating a hierarchical communication network based on the minimum spanning tree algorithm to achieve selective directed communication includes the following steps: S41. Treat each key component in the lubrication hydraulic equipment as a node in the network, and establish edges between nodes according to the dependencies and functional connections between key components to obtain the initial graph; S42. Set the weights of each edge to obtain a weighted device hierarchy network diagram; S43. Determine the number of cluster partitions and use a clustering algorithm to cluster the nodes to obtain the clustering results; S44. Using the nodes in the clustering results as nodes in the communication network, the minimum spanning tree algorithm is used to find the minimum spanning tree in the clustering network, and a hierarchical communication network based on the minimum spanning tree is obtained. S45. Define directed communication rules and configure the network to enable selective directed communication.

2. The operation and maintenance method for lubrication hydraulic equipment based on an online hydraulic model according to claim 1, characterized in that, The process of constructing an online hydraulic model, acquiring current-moment observation information, comparing it with the prediction data of the online hydraulic model, and generating observation features and dependency features includes the following steps: S21. Based on the structural diagram and operating principle description, obtain the internal fluid flow of the lubricating hydraulic equipment and construct a fluid dynamics model; S22. Construct a thermodynamic model based on the thermodynamic characteristics of the lubricating hydraulic equipment; S23. Integrate the fluid dynamics model and the thermodynamics model to form an online hydraulic model, and input the initial conditions and parameters at the current moment into the online hydraulic model to generate prediction data of the current operating state. S24. Obtain the observation information at the current moment, calculate the difference between the observation information and the predicted data, and identify outliers that exceed the expected range; S25. Calculate the residuals between the observed data and the predicted data, and extract the observation features and dependency features based on the difference analysis and residual calculation.

3. The method for operation and maintenance of lubrication hydraulic equipment based on an online hydraulic model according to claim 2, characterized in that, The pre-decision-making process based on observational and dependency features to determine potential operation and maintenance solutions includes the following steps: S31. Extract key features from observation features and dependency features; S32. Based on the extracted key features, use machine learning algorithms to classify fault modes; S33. Analyze the potential causes of the identified fault modes; S34. Based on the failure mode and potential causes, match suitable potential operation and maintenance solutions from the existing operation and maintenance solution library.

4. The method for operation and maintenance of lubrication hydraulic equipment based on an online hydraulic model according to claim 3, characterized in that, The classification of fault modes based on extracted key features using machine learning algorithms includes the following steps: S321. Collect observational features and dependency features, as well as corresponding key features, from historical normal operation and various failure modes. S322. Mark the key features in S321 to indicate the operating status of the lubrication hydraulic equipment at that time; S323. Divide the labeled dataset into a training set and a test set; S324. Train the support vector machine model using the training and test sets, and evaluate and adjust the support vector machine model. S325. Given new observation features and dependent features, use a well-trained and evaluated support vector machine model to predict the current state of the lubrication hydraulic equipment.

5. The operation and maintenance method for lubrication hydraulic equipment based on an online hydraulic model according to claim 4, characterized in that, The process of determining the number of clustering partitions and using a clustering algorithm to cluster the nodes to obtain the clustering results includes the following steps: S431. Use the elbow rule to determine the number of cluster partitions; S432. Randomly select k nodes as initial cluster centers; S433. Use Euclidean distance to calculate the distance from each node to each cluster center; S434. Based on the calculated distance, assign each node to the nearest cluster; S435. Calculate the new centers for each cluster; S436. Repeat steps S433-S435 until the preset number of iterations is reached or the cluster centers no longer change significantly.

6. The method for operation and maintenance of lubrication hydraulic equipment based on an online hydraulic model according to claim 5, characterized in that, The process of using nodes from the clustering results as nodes in the communication network, and then using the minimum spanning tree algorithm to find the minimum spanning tree in the clustering network to obtain a hierarchical communication network based on the minimum spanning tree includes the following steps: S441. Treat all nodes in each cluster partition as independent nodes in the network; S442. Construct an undirected graph, where the edges of the undirected graph represent the connections between independent nodes, and assign weights to each edge to obtain a weighted undirected graph. S443. Based on a weighted undirected graph and executing the minimum spanning tree algorithm, the tree is gradually constructed by selecting the edge with the minimum weight until all nodes are included. S444. Based on the structure of the minimum spanning tree, determine the hierarchical relationship between nodes and identify the root node in the minimum spanning tree; S445. Determine the paths from the root node to all other nodes to form a hierarchical structure.

7. The operation and maintenance method for lubrication hydraulic equipment based on an online hydraulic model according to claim 6, characterized in that, The process of generating several joint decision-making schemes based on potential operation and maintenance schemes, optimizing the generated joint decision-making schemes using a hierarchical communication network, evaluating all joint decision-making schemes using a state-action value function, and generating the final operation and maintenance scheme based on the evaluation results includes the following steps: S51. Combine different potential operation and maintenance solutions to generate several joint decision-making solutions; S52. Optimize the generated joint decision-making scheme using a hierarchical relationship network; S53. Use the state-action value function to evaluate each joint decision-making scheme and calculate the state-action value; S54. Sort all joint decision schemes according to the state action values ​​and select the optimal scheme.

8. The method for operation and maintenance of lubrication hydraulic equipment based on an online hydraulic model according to claim 7, characterized in that, The evaluation of each joint decision-making scheme using the state-action value function, and the calculation of the state-action value, includes the following steps: S531. Define the status and operation and maintenance actions of the operation and maintenance plan; S532. Initialize the status action value table for each potential operation and maintenance solution, and design the reward function; S533, Execute the operation and maintenance actions in each joint decision-making scheme and obtain new status information; S534. Based on the new state information, the acquisition of rewards, and the prediction of future returns, update the state action values ​​of each joint decision-making scheme.

9. A method for operation and maintenance of lubrication hydraulic equipment based on an online hydraulic model according to claim 8, characterized in that, The formula for the reward function is: ； The formula for updating the state action value is: ； In the formula, R imd ( t () indicates at a point in time t Instant rewards; P t Indicates a point in time t Actual measured values ​​of performance indicators P ref Indicates a reference value for the performance indicator. C R represents the cost of implementing the operation and maintenance plan, and R represents the cost at a given point in time. t Risk indicators R long Indicates long-term returns; w 1 represents the weighting coefficient of the performance indicator. w 2 represents the weighting coefficient for operation and maintenance costs. w 3 indicates the weighting coefficient of the risk indicator; Q old ( S , A This indicates that maintenance actions are being taken. A Previous status S Q value under, Q new ( S , A () represents the updated Q value; γ Indicates the discount factor. α Indicates the learning rate; Q ( S next , A next ) indicates the new state S next The following maintenance actions A next The expected Q value; max Anext Q ( S next , A next ) indicates the new state S next The maximum Q value among all maintenance actions.

10. A lubrication hydraulic equipment operation and maintenance system based on an online hydraulic model, used to implement the lubrication hydraulic equipment operation and maintenance method based on an online hydraulic model as described in any one of claims 1-9, characterized in that, The online hydraulic model-based lubrication hydraulic equipment operation and maintenance system includes an equipment data acquisition module, a feature acquisition module, a potential solution generation module, a network generation module, and a final solution generation module. The device data acquisition module, the feature acquisition module, the potential solution generation module, the network generation module, and the final solution generation module are sequentially connected. The equipment data acquisition module is used to acquire structural data and operating principle data of the lubrication hydraulic equipment, identify key components and their interrelationships, and obtain structural diagrams and operating principle descriptions. The feature acquisition module is used to construct an online hydraulic model based on the structure diagram and operating principle description, acquire the observation information at the current moment, compare it with the prediction data of the online hydraulic model, and generate observation features and dependency features. The potential solution generation module is used to make pre-decision based on observation features and dependency features to determine potential operation and maintenance solutions. The network generation module is used to generate a hierarchical communication network based on the minimum spanning tree algorithm to achieve selective directed communication. The final solution generation module is used to generate several joint decision-making schemes based on potential operation and maintenance schemes, optimize the generated joint decision-making schemes using a hierarchical communication network, evaluate all joint decision-making schemes using a state action value function, and generate the final operation and maintenance scheme based on the evaluation results.

Images