A lubricating oil pump station leakage detection method and system

By splitting the functional structure of the lubricating oil pump station and laying out sensors, real-time monitoring and building an adaptive detector, the problem of untimely response to leakage failures in traditional detection methods is solved, and efficient leakage detection and operation and maintenance of the lubricating oil pump station is achieved.

CN118706340BActive Publication Date: 2024-11-15JIANGSU JIANGHAI LUBRICANTS HYDRAULIC EQUIP CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411205259.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-11-15
Estimated Expiration
2044-08-30

AI Technical Summary

Technical Problem

Traditional lubricating oil pump station leak detection methods are difficult to monitor and identify leakage faults in real time and accurately, resulting in untimely fault response and inefficient operation and maintenance.

Method used

By splitting the functional structure of the lubricating oil pump station, laying sensors to form a monitoring network, collecting data in real time, and performing attribute marking and preprocessing, and building an adaptive detector to identify abnormalities and perform leakage detection.

Benefits of technology

It realizes efficient and accurate leakage detection and fault operation and maintenance of lubricating oil pump stations, ensuring the safe and efficient operation of industrial production.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118706340B_ABST
    Figure CN118706340B_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of leakage detection, and provides a leakage detection method and system for a lubricating oil pump station. The method includes: performing functional structure decomposition, obtaining a set of structural areas, and performing sensor deployment to obtain a monitoring network set; performing operation monitoring based on the monitoring network set, obtaining a set of monitoring data streams, and performing attribute marking on the set of monitoring data streams to obtain a set of attribute data streams; establishing a preprocessing multi-channel, identifying the set of attribute data streams with the preprocessing multi-channel, and obtaining a set of abnormal data streams; constructing a detector, performing detection and analysis on the set of abnormal data streams based on the detector, and performing leakage fault operation and maintenance based on the detection results. The present application solves the technical problem that traditional detection methods are difficult to monitor and identify leakage faults in real time and accurately, resulting in untimely fault response and low operation and maintenance efficiency, and realizes the rapid positioning and operation and maintenance of leakage faults, thereby ensuring the operational reliability of the lubricating oil pump station.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, specifically to the field of leakage detection technology, and in particular to a leakage detection method and system for a lubricating oil pump station. Background Art

[0002] With the continuous development of industrial production and the rapid advancement of technology, the safe and efficient operation of lubricating oil pump stations, as an indispensable part of industrial equipment, plays a vital role in the stability and production efficiency of the entire production line. However, in the actual production process, leakage problems of lubricating oil pump stations often occur, which may not only lead to waste of lubricating oil, but also cause equipment failure, production interruption and even safety accidents, bringing great risks and losses to the normal operation of lubricating oil pump stations. Summary of the invention

[0003] The present application provides a leakage detection method and system for a lubricating oil pump station, aiming to solve the technical problem that traditional detection methods are difficult to monitor and identify leakage faults in real time and accurately, resulting in untimely fault response and low operation and maintenance efficiency.

[0004] In view of the above problems, the present application provides a leakage detection method and system for a lubricating oil pump station.

[0005] The first aspect disclosed in the present application provides a leakage detection method for a lubricating oil pump station, the method comprising: performing functional structural decomposition on a target lubricating oil pump station to obtain a set of structural areas of the lubricating oil pump station, deploying sensors at key parts based on the set of structural areas of the lubricating oil pump station to obtain a set of structural area sensor monitoring networks; performing operation monitoring on the target lubricating oil pump station based on the set of structural area sensor monitoring networks to obtain a set of multi-source structural area operation monitoring data streams, performing attribute marking on the set of multi-source structural area operation monitoring data streams to obtain a set of structural area operation attribute data streams; establishing a data preprocessing multi-channel, mapping the set of structural area operation attribute data streams with the data preprocessing multi-channel for abnormal identification processing to obtain a set of structural area abnormal operation data streams; constructing a lubricating oil pump station leakage adaptive detector, detecting and analyzing the set of structural area abnormal operation data streams based on the lubricating oil pump station leakage adaptive detector to obtain a lubricating oil pump station leakage detection result, and performing leakage fault operation and maintenance on the target lubricating oil pump station according to the lubricating oil pump station leakage detection result.

[0006] Another aspect disclosed in the present application provides a leakage detection system for a lubricating oil pump station, the system comprising: a sensor deployment module, the sensor deployment module is used to perform functional structure decomposition of a target lubricating oil pump station, obtain a set of structural areas of the lubricating oil pump station, deploy sensors in key parts based on the set of structural areas of the lubricating oil pump station, and obtain a set of structural area sensor monitoring networks; an attribute marking module, the attribute marking module is used to perform operation monitoring of the target lubricating oil pump station based on the set of structural area sensor monitoring networks, obtain a set of multi-source structural area operation monitoring data streams, and attribute mark the set of multi-source structural area operation monitoring data streams. Record, obtain the set of structural area operation attribute data streams; an abnormality recognition module, the abnormality recognition module is used to establish a data preprocessing multi-channel, map the structural area operation attribute data stream set with the data preprocessing multi-channel for abnormality recognition processing, and obtain the set of structural area abnormal operation data streams; a fault operation and maintenance module, the fault operation and maintenance module is used to construct a lubricating oil pump station leakage adaptive detector, based on the lubricating oil pump station leakage adaptive detector, the structural area abnormal operation data stream set is detected and analyzed to obtain the lubricating oil pump station leakage detection result, and the target lubricating oil pump station is subjected to leakage fault operation and maintenance according to the lubricating oil pump station leakage detection result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The above-mentioned method for detecting leakage of a lubricating oil pump station is to perform fine functional structural decomposition on the target lubricating oil pump station, divide the entire pump station into several different structural areas, and form a set of structural areas of the lubricating oil pump station. Based on this set, sensors are arranged at key parts to build a sensor monitoring network covering each structural area. This monitoring network will collect and transmit the operation data of each structural area in real time, forming a set of multi-source structural area operation monitoring data streams. In order to facilitate subsequent data analysis and processing, these data streams are attributed, and the operation attributes of the specific structural area represented by each data stream are clarified, so as to obtain a set of structural area operation attribute data streams. In order to identify abnormal situations from these massive operating data, a data preprocessing multi-channel is established, and the operation attribute data stream is mapped with different preprocessing channels for abnormal identification processing, and a set of abnormal operation data streams of the structural area is extracted. After that, a lubricating oil pump station leakage adaptive detector is constructed, and the abnormal operation data stream is analyzed based on this detector, so as to accurately determine whether the lubricating oil pump station has a leak, and generate a leak detection result. Then, according to the leak detection result, the target lubricating oil pump station is timely operated and maintained for leakage faults. Through refined structural decomposition, sensor deployment, multi-source data monitoring, anomaly identification and adaptive detection analysis, the entire process achieves efficient and accurate leakage detection and fault maintenance of lubricating oil pump stations, providing strong guarantees for the safety and efficiency of industrial production.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 The figure is a flow chart of a leakage detection method for a lubricating oil pump station in one embodiment.

[0012] Figure 2 The present invention is a leakage detection system architecture diagram of a lubricating oil pump station in one embodiment.

[0013] Explanation of the reference numerals: sensor deployment module 1, attribute marking module 2, anomaly identification module 3, fault operation and maintenance module 4. DETAILED DESCRIPTION

[0014] The embodiments of the present application provide a leakage detection method and system for a lubricating oil pump station to solve the technical problem that traditional detection methods are difficult to monitor and identify leakage faults in real time and accurately, resulting in untimely fault response and low operation and maintenance efficiency.

[0015] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0016] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0017] Embodiment 1

[0018] like Figure 1 As shown, the present application provides a leakage detection method for a lubricating oil pump station, the method comprising:

[0019] The target lubricating oil pump station is functionally decomposed to obtain a set of structural areas of the lubricating oil pump station. Sensors are arranged at key locations based on the set of structural areas of the lubricating oil pump station to obtain a set of structural area sensor monitoring networks.

[0020] With the rapid development of modern industrial fields, the safety and reliability of lubricating oil pump stations, as a key component to ensure the stable and efficient operation of mechanical equipment, are increasingly valued. However, traditional leak detection methods often make it difficult to monitor all potential leaks in real time and comprehensively, especially in areas with dense equipment and complex operations, where it is easy to miss some hidden leak sources. Although regular maintenance can regularly check and repair leaks, it cannot respond to sudden leaks in real time, resulting in untimely fault response and low operation and maintenance efficiency.

[0021] In an embodiment of the present application, during the maintenance and monitoring of the target lubricating oil pumping station, in order to more effectively identify and manage potential problems, the system terminal first performs functional structural decomposition of the lubricating oil pumping station. This process subdivides the complex pumping station system into several structural areas, each of which has its own specific functions and characteristics. Once these sets of structural areas are obtained, the system terminal identifies key parts according to the specific functions of these structural areas and determines multiple key parts of the structural areas. Subsequently, sensors are deployed in these key parts to form a set of structural area sensor monitoring networks. This set of structural area sensor monitoring networks covers the key parts of the entire lubricating oil pumping station, and can collect operating data of each structural area in real time, such as temperature, pressure, flow, etc., to provide data support for subsequent leakage detection.

[0022] Further, the present application provides the obtained structure area sensor monitoring network set, including:

[0023] Key parts of each structural area in the structural area set of the lubricating oil pump station are identified to determine the key part set of the structural area; comprehensive monitoring analysis is performed based on the key part set of the structural area to obtain a key part sensor layout position parameter set and a key part sensor layout angle parameter set.

[0024] Preferably, when constructing the monitoring network of the lubricating oil pump station, the system terminal analyzes the set of structural areas of the lubricating oil pump station and identifies the key parts of each structural area according to the specific functions of the structural area. These key parts are the parts that have an important impact on the overall operating performance, safety and reliability of the pump station. Taking the oil storage tank structural area as an example, the key monitoring parts include the oil inlet, the oil outlet, etc. The oil inlet and the oil outlet are important interfaces connecting the oil storage tank with the external pipeline, and their operating status directly affects the flow of oil and the working performance of the pump station. Therefore, the layout of the sensor group in these key parts can ensure comprehensive and effective monitoring of the oil storage tank structural area. After determining the set of key parts of the structural area, the system terminal performs a comprehensive monitoring analysis. The purpose of this step is to determine the optimal layout position and angle of the sensor in the key parts to ensure the accuracy and integrity of the monitoring data. Subsequently, for each key part, the system terminal determines the optimal layout position of the sensor. This is based on historical experience and professional knowledge to evaluate the geometric characteristics, flow characteristics and potential leakage points of the key parts. After determining the installation location, the system terminal determines the installation angle of the sensor based on the directionality of the monitored parameter and the measurement principle of the sensor. For example, when monitoring fluid velocity, it is necessary to ensure that the sensor can accurately capture the direction and speed of fluid flow, so the installation angle should be perpendicular to the direction of fluid flow in order to obtain the most accurate measurement data. By comprehensively considering the structural characteristics, operating environment, and the performance requirements of the sensor itself, the system terminal can obtain the sensor installation location parameter set and angle parameter set for each key location, which can ensure the comprehensiveness and effectiveness of the monitoring network.

[0025] The operating function characteristics of the key parts set of the structural area are extracted to obtain the key parts operating function characteristic set and the operating function threshold parameter set; according to the key parts operating function characteristic set and the operating function threshold parameter set, the key parts sensor type parameter set and the key parts sensor specification parameter set are obtained.

[0026] Preferably, when constructing the monitoring network of the lubricating oil pump station, in order to ensure the accuracy and effectiveness of the monitoring, the system terminal extracts the operating function characteristics of the key parts of the structural area. The main purpose of this step is to understand the behavioral characteristics of the key parts under normal operation and potential fault conditions, so as to provide a basis for the subsequent sensor selection and configuration. First, the system terminal extracts the operating function characteristics of the key parts of the structural area, and obtains the operating function characteristic set by analyzing the specific role and function of these key parts in the overall operation of the pump station. These characteristics may include temperature, pressure, flow, vibration, etc., which can reflect the working state and performance of the key parts. Subsequently, based on these operating function characteristics, the system terminal determines the operating function threshold parameter set of each key part. These threshold parameters are obtained based on the analysis of the normal operating data and potential fault data of the key parts, and are used to define the parameter range of the key parts under different working conditions. For example, for the temperature monitoring of the oil inlet, a normal temperature range and a high temperature alarm threshold are set. After obtaining the key part operating function characteristic set and the operating function threshold parameter set, the system terminal determines the key part sensor type parameter set and the key part sensor specification parameter set based on this information. According to the operating function type of key parts, such as the need to monitor temperature and pressure data at the oil inlet, the system terminal will select the appropriate sensor type, such as temperature sensor and pressure sensor. At the same time, according to the distribution threshold of each operating function data, the specification parameters of the sensor, such as measurement range, accuracy, response time, etc., are determined to ensure that the selected sensor can meet the monitoring requirements.

[0027] Based on the key part sensor layout position parameter set and the key part sensor layout angle parameter set, as well as the key part sensor type parameter set and the key part sensor specification parameter set, determine the key part sensor layout parameter set; based on the key part sensor layout parameter set, key part sensors are laid out on the lubricating oil pump station structural area set to obtain the structural area sensor monitoring network set.

[0028] Preferably, when constructing the monitoring network of the lubricating oil pump station, the system terminal determines the specific locations where the sensors should be installed and their installation angles based on the obtained key parts sensor layout position parameter set and angle parameter set to ensure that the working status information of the key parts can be captured in the best way. Subsequently, according to the operating function characteristics of the key parts and the corresponding threshold parameters, the system terminal selects the appropriate sensor type and determines their specification parameters. In this way, the system terminal obtains a key part sensor layout parameter set, which contains detailed information on the sensors required for all key parts. Afterwards, based on this key part sensor layout parameter set, the system terminal performs actual sensor layout work in the structural area set of the lubricating oil pump station. These sensors will form a structural area sensor monitoring network set, collect and transmit the operating data of key parts in real time, and provide important support for the safe and efficient operation of the pump station.

[0029] Based on the structural area sensor monitoring network set, the target lubricating oil pump station is operated and monitored, a multi-source structural area operation monitoring data stream set is obtained, and the multi-source structural area operation monitoring data stream set is attribute-labeled to obtain a structural area operation attribute data stream set.

[0030] In one embodiment, after the lubricating oil pump station structural area sensor monitoring network is built, the system terminal performs real-time operation monitoring of the target lubricating oil pump station based on the lubricating oil pump station structural area sensor monitoring network. This monitoring network will continuously collect sensor data from various key parts to form a multi-source structural area operation monitoring data stream set. These data stream sets contain various types of information, such as temperature, pressure, flow, etc., which reflect the operating status of different parts of the pump station. In order to make these data easier to understand and use, the system terminal attributes the multi-source structural area operation monitoring data stream set, that is, to clarify the specific attributes and meanings represented by each data stream. After attribute marking, the system terminal obtains a structural area operation attribute data stream set. Each data stream in this set has a clear attribute label, such as oil inlet temperature, oil outlet pressure, etc., so that the system terminal can quickly and accurately identify and understand the operating status of each part of the pump station.

[0031] A data preprocessing multi-channel is established, and the structure area operation attribute data stream set is mapped with the data preprocessing multi-channel for abnormal identification processing to obtain the structure area abnormal operation data stream set.

[0032] In one embodiment, in the data processing process, in order to ensure the accuracy and effectiveness of the set of operating attribute data streams obtained from the structural area sensor monitoring network, the system terminal establishes multiple data preprocessing channels. Each data preprocessing channel is configured with a data structured format preprocessing node and an operating data anomaly identification node. Among them, the data structured format preprocessing node is obtained by parsing the set of structural area operating attribute data streams through the preprocessing step, and the operating data anomaly identification node is obtained by training the set of structural area operating attribute data streams. Each data preprocessing channel is responsible for processing a certain feature in the data stream, that is, a performance indicator. For example, one channel focuses on identifying abnormal fluctuations in the temperature data stream, while another channel may focus on whether the pressure data stream exceeds the preset safety range. Through the nodes running in these channels, the system terminal can quickly and accurately find abnormal points in the data. These abnormal points represent potential problems or failures in a part of the lubricating oil pump station, such as equipment overheating, excessive pressure or abnormal flow. Subsequently, the system terminal summarizes the abnormal points identified in all channels into a set of abnormal operating data streams in the structural area. This set provides the system terminal with important information about the operating status of the pump station, helping to timely discover problems such as leakage.

[0033] Furthermore, the present application provides the method of establishing multiple channels for data preprocessing, and the method further includes:

[0034] Obtain a set of lubricating oil pump station performance indicators, wherein the set of lubricating oil pump station performance indicators includes pressure stability, temperature, liquid level change, vibration and noise, lubricating oil quality and operating energy consumption; and construct a data preprocessing channel set based on the set of lubricating oil pump station performance indicators.

[0035] Preferably, in the detection of the lubricating oil pump station, the performance index is an important parameter to measure its operating status and performance. These performance index sets cover multiple aspects such as pressure stability, temperature, liquid level change, vibration and noise, lubricating oil quality and operating energy consumption. These indicators can fully reflect the overall performance of the pump station and the working status of each component. In order to accurately analyze and evaluate these performance indicators, the system terminal establishes a data preprocessing channel set. This set contains the framework of multiple data preprocessing channels, and each data preprocessing channel framework has an input layer, a data processing layer and an output layer. The input layer has an input limit for each channel, that is, a corresponding performance indicator, and this input limit only allows the data corresponding to this performance indicator to enter the channel. The data processing layer is used for subsequent configuration of data structured format preprocessing nodes and operation data abnormality identification nodes to perform abnormal identification of performance indicators. The output layer outputs the identification results of the data processing layer to generate the abnormal operation data stream of the performance indicator. By constructing such a data preprocessing channel set, targeted data processing and analysis can be performed on each performance indicator, providing accurate and reliable data support for performance evaluation and fault diagnosis of the lubricating oil pump station.

[0036] The preprocessing step analysis and abnormal data recognition training are respectively performed on the set of structural area operation attribute data streams to obtain data structured format preprocessing nodes and operation data abnormality recognition nodes; based on the data structured format preprocessing nodes and the operation data abnormality recognition nodes, the data preprocessing channel set is configured with processing nodes to establish the data preprocessing multi-channel.

[0037] Preferably, in the process of processing the operation data of the lubricating oil pump station, the system terminal performs preprocessing step analysis and abnormal data identification training on the operation attribute data streams generated by different structural areas. The preprocessing step analysis is to ensure the accuracy and reliability of the data, and the abnormal data identification training is to accurately perform mapping abnormal identification processing. The system terminal generates a data structured format preprocessing node by structured classification of the operation attribute data stream set of the structural area, and performs program analysis according to the classification results. Then, abnormal operation data identification training is performed according to the classification results to obtain the operation data abnormal identification node. These nodes are key components in the data processing flow, and they can process and judge the input data. Subsequently, based on these preprocessing nodes and abnormal identification nodes, each channel structure in the determined data preprocessing channel set is configured. Corresponding preprocessing nodes and abnormal identification nodes are allocated in the data processing layer of each channel to ensure that the data flowing through the channel can be correctly processed and judged. Afterwards, by configuring the data preprocessing channel set of the processing node, the system terminal establishes a data preprocessing multi-channel. The data preprocessing multi-channel can simultaneously process multiple data streams of different sources and different types, and perform real-time and accurate analysis and judgment on them. This not only improves the efficiency and accuracy of data processing, but also provides strong technical support for leak detection in lubricating oil pumping stations.

[0038] Furthermore, the present application provides the method of obtaining the data structured format preprocessing node and the running data anomaly identification node, and further includes:

[0039] The structure area operation attribute data flow set is structuredly classified to obtain an operation data structured format attribute set; based on the operation data structured format attribute set, preprocessing steps are respectively parsed to obtain a data structured format preprocessing program set.

[0040] Optionally, when processing the data stream of the operation attribute of the structural area, the system terminal first performs structured classification. The purpose of this step is to classify the operation data with different properties from different structural areas, such as sensor data such as temperature, pressure, flow, etc., so as to carry out targeted processing later. Through structured classification, the system terminal obtains a set of operation data structured format attributes, which contains attribute information of various types of data, such as data type, data format, data range, etc. This information is crucial for the subsequent preprocessing step analysis. Subsequently, the system terminal determines which preprocessing steps need to be performed based on the attribute information of various types of data recorded in the set of operation data structured format attributes, and searches from the data processing operator library based on the judgment result to screen out the corresponding data processing operator. The data processing operator library is pre-built, and it stores data processing operators pre-written by professionals. Each data processing operator represents a preprocessing program, such as a data cleaning program, a data filtering program, etc. Taking data type as an example, when the data type is character, the system terminal determines based on historical experience that such data needs to be preprocessed, such as removing invalid characters and converting to a standard format. Therefore, the corresponding operator is retrieved from the data processing operator library. By analyzing the preprocessing steps of different attribute information, the system terminal can obtain a set of data structured format preprocessing programs. Each program in this set is targeted at a specific type of operating data, has clear preprocessing steps and algorithms, and can ensure that the data is of high quality and accuracy in the subsequent analysis and decision-making process.

[0041] The data structured format preprocessing node is generated according to the data structured format preprocessing program set; abnormal operation data recognition training is performed on the structure area operation attribute data flow set to obtain the operation data abnormality recognition node.

[0042] Optionally, the system terminal generates corresponding data structured format preprocessing nodes based on the determined preprocessing program set. These nodes are key components in the data processing process. They are responsible for performing specific tasks in the preprocessing program set to ensure that the data reaches the required format and quality before subsequent analysis. At the same time, in order to detect potential faults or problems in a timely manner, the system terminal performs abnormal operation data recognition training on the structural area operation attribute data stream set. This training process uses the standard set of structural area label data to learn how to identify data that does not conform to the normal pattern or exceeds the preset threshold, that is, abnormal operation data. Through training, the system terminal will obtain the operation data anomaly recognition node. This node is another key component in data monitoring and analysis. It can perform anomaly detection on the data stream in real time and generate a set of abnormal operation data streams in the structural area, which provides a strong guarantee for leak detection in the oil pump station.

[0043] Furthermore, the present application provides the method for obtaining the operation data abnormality identification node, and the method further includes:

[0044] The structure area operation attribute data stream set is source labeled and classified to obtain a structure area operation data source label set; based on the structure area operation data source label set, the lubricating oil pump station performance indicator set is standardized and quantified to generate a structure area label data standard set.

[0045] Optionally, when training to obtain the operation data anomaly identification node, the system terminal first classifies the source labeling of the structure area operation attribute data stream set. This means that the operation data is labeled accordingly according to its source, such as the oil pump outlet, oil filter device, etc., to form a structure area operation data source label set. This is done because the same type of operation data may come from different monitoring points, and these data from different sources may have different standards for anomaly identification and processing. Subsequently, the system terminal associates each source label with the corresponding performance indicator according to the structure area operation data source label set. This step is to clarify which performance indicator data comes from which specific monitoring point. After that, the maximum and minimum values ​​of the corresponding data of each indicator are extracted, and then the extreme value difference of each indicator is calculated by subtracting the minimum value from the maximum value of each indicator. Then, the corresponding minimum value is subtracted from the data of each indicator in turn, and then the ratio is calculated with the extreme value difference to generate a standard data of the indicator, and this process is repeated until all data in the lubricating oil pump station performance indicator set are standardized and quantified to generate a standard set of structure area label data. In this way, the accuracy and efficiency of lubricating oil pump station performance evaluation can be ensured, while also facilitating unified management and processing of operating data from different sources.

[0046] According to the structural area label data standard set, the structural area operation attribute data stream set is trained for abnormal operation data recognition to obtain an operation data attribute label abnormal discriminator set; based on the lubricating oil pump station performance indicator set, the operation data attribute label abnormal discriminator set is labeled and classified to obtain the operation data abnormal identification node.

[0047] Optionally, for the structural area operation data of the lubricating oil pump station, the system terminal performs recognition training of abnormal operation data based on the standard set of structural area label data that has been obtained. Since the data under different source labels have their own unique abnormal operation standards, a strategy of performing abnormal recognition training on each label data is adopted. This step ensures the pertinence and accuracy of abnormal recognition training. Specifically, the system terminal divides the data set of the structural area operation attribute data stream according to the standard set of structural area label data to obtain multiple subsets, and the data of each subset corresponds to a specific source label and performance indicator. Subsequently, the system terminal initializes multiple isolation forest models according to the isolation forest, and sets the parameters of the isolation forest model according to the size and complexity of the data set, such as the number of trees, the maximum number of samples, etc. The isolation forest isolates data points by randomly constructing decision trees, and the outliers have a shorter path length. After that, the system terminal uses the labeled data subset to train the isolation forest model. During the training process, the isolation forest model learns the normal pattern of the data under each source label and generates an abnormal score threshold to distinguish normal data from abnormal data. For each source label, the trained isolation forest model is encapsulated into an anomaly discriminator. These discriminators can calculate the anomaly score and determine whether the data under a specific label is abnormal. Then, the system terminal combines all the anomaly discriminators into a set, namely the operation data attribute label anomaly discriminator set. This set contains multiple anomaly discriminators for different source labels, which can fully cover the anomaly identification requirements of the structural area operation data. Finally, the system terminal marks and classifies the trained operation data attribute label anomaly discriminator set based on the performance indicator set of the lubricating oil pump station. The purpose of this step is to associate the discriminator with the specific performance indicator so that the anomaly discriminator related to the specific performance indicator can be quickly located in the actual application. Through the above process, the system terminal finally obtains the operation data anomaly identification node. These nodes will be configured in the data processing layer of the corresponding data preprocessing channel, which can monitor the structural area operation data of the lubricating oil pump station in real time and quickly identify the abnormal data using the corresponding anomaly discriminator. This not only improves the efficiency and accuracy of anomaly detection, but also provides strong support for subsequent leakage warning.

[0048] A lubricating oil pump station leakage adaptive detector is constructed, and based on the lubricating oil pump station leakage adaptive detector, the abnormal operation data stream set of the structural area is detected and analyzed to obtain the lubricating oil pump station leakage detection result, and leakage fault operation and maintenance of the target lubricating oil pump station is performed according to the lubricating oil pump station leakage detection result.

[0049] In one embodiment, the system terminal obtains a leakage performance classification detection data set based on the performance indicator set of the lubricating oil pump station, and constructs an adaptive detector model based on this data set. This model can identify abnormal data that deviates from the normal operating mode, and judge whether there is a leakage risk based on this. Subsequently, the system terminal uses this adaptive detector to perform real-time or regular detection and analysis on the abnormal operation data stream set of the structural area, and can quickly identify abnormal data and further analyze whether these anomalies are caused by leakage faults. After that, the detector will output the leakage detection results of the lubricating oil pump station. These results include information such as the location, degree and possible causes of the leakage, providing a reference for operators. Then, based on the leakage detection results, the operator can perform corresponding leakage fault operation and maintenance work on the target lubricating oil pump station. This includes measures such as emergency shutdown, isolation of fault areas, replacement of damaged parts or adjustment of operating parameters to ensure the safe and stable operation of the pump station.

[0050] Furthermore, the present application provides the method for constructing a lubricating oil pump station leakage adaptive detector, the method also includes:

[0051] A lubricating oil pump station leakage detection data set is collected and acquired, and the lubricating oil pump station leakage detection data set is classified and labeled according to the lubricating oil pump station performance indicator set to obtain a leakage performance classification detection data set; a deep neural network structure is used to perform equal weight layer detection training on the leakage performance classification detection data set to obtain a leakage branch detector set.

[0052] Optionally, in order to build an adaptive leakage detector for a lubricating oil pump station, the system terminal collects leakage detection data of the lubricating oil pump station under different operating conditions. These data contain various performance indicators of the lubricating oil pump station, such as pressure, flow, temperature, etc., which can reflect the operating status and potential problems of the pump station. Once the lubricating oil pump station leakage detection data set is obtained, the system terminal classifies and identifies the data according to the performance indicator set of the lubricating oil pump station. The purpose of this step is to divide the data set into different categories, each category representing a performance indicator range. In this way, the system terminal obtains a leakage performance classification detection data set, which contains all performance data related to leakage. Subsequently, the system terminal divides the leakage performance classification detection data set into a training set and a validation set, and then designs a deep neural network model according to the actual problem, and determines the number of hidden layers, the number of neurons in each layer, and the activation function and other hyperparameters. Afterwards, in the deep neural network model, the system terminal designs an equal weight layer. The role of this layer is to give equal weight to each category of data to ensure that the model can process all types of data equally during the training process. If there is a class imbalance problem in the dataset, that is, the number of samples in some categories is much larger than that in other categories, it is necessary to achieve equal weighting by adjusting the weight of each category in the loss function. This is achieved by setting different weight factors for samples of different categories to ensure that the model treats each type of data equally during the training process. The choice of weight factors is determined based on the ratio of the number of class samples. Then, the deep neural network model is trained using the training set. During the training process, the model will learn how to extract features from the input data and perform leakage detection based on these features. Then, the weights and bias terms of the model are continuously adjusted through the back propagation algorithm and the gradient descent method to minimize the loss function and improve the accuracy of the model. During the training process, the system terminal uses the validation set to monitor the performance of the model and adjust the hyperparameters or improve the model as needed. After the model training and optimization are completed, the system terminal generates a corresponding number of leakage branch detectors according to the number of categories of the leakage performance classification detection dataset. Each leakage branch detector corresponds to a specific leakage performance category and has the ability to perform leakage detection on the data of that category. Through this process, the system terminal obtains a set of leakage branch detectors. This collection includes multiple detectors, each of which is dedicated to leak detection for a specific performance indicator or operating status, determining whether there are signs of leaks to minimize the impact of leaks on pump station operations.

[0053] A multivariate decision analysis is performed on the leakage branch detector set to determine the branch detector decision coefficient information; based on the branch detector decision coefficient information, a weighted fusion is performed on the leakage branch detector set to construct the lubricating oil pump station leakage adaptive detector.

[0054] Optionally, after obtaining the set of leakage branch detectors, the system terminal performs a comprehensive analysis of the output results of these branch detectors to determine the final leakage detection result. First, the system terminal performs a multivariate decision analysis on the set of leakage branch detectors. The purpose of this step is to evaluate the decision-making ability and credibility of each branch detector in a specific situation. By collecting the output results of the branch detectors under different conditions and combining the known leakage conditions, the decision coefficient information of each branch detector is calculated through the decision coefficient fusion function. These decision coefficients reflect the accuracy and reliability of the detector in judging whether there is a leak under specific conditions. After determining the decision coefficient information of the branch detector, the system terminal performs weighted fusion. Weighted fusion is a method of combining the output results of multiple detectors to form a comprehensive decision. By assigning a weight corresponding to its decision coefficient to each branch detector, it can be ensured that in the final decision, more reliable and accurate detectors will have a greater influence. Through weighted fusion, the advantages of multiple branch detectors can be concentrated, while reducing the false alarm or missed alarm problems that may exist in a single detector. Finally, the system terminal built an adaptive leakage detector for lubricating oil pumping stations, which can make a comprehensive judgment based on the output results of multiple branch detectors and combined with the decision coefficient information, so as to more accurately identify the leakage of the lubricating oil pumping station.

[0055] Furthermore, the present application provides the method for determining the branch detector decision coefficient information, and the method further includes:

[0056] Performing a critical evaluation on each performance indicator in the lubricating oil pump station performance indicator set to obtain critical factor information of the performance indicator; verifying the leakage branch detector set to obtain branch detector accuracy information;

[0057] Optionally, in the process of building an adaptive detector for leaks in a lubricating oil pumping station, the system terminal first performs a critical evaluation on a set of lubricating oil pumping station performance indicators to identify which performance indicators have a higher importance or influence on leak detection. This involves analyzing the historical data of each performance indicator to understand their correlation, sensitivity, and reliability with leak events. Through this process, performance indicator critical factor information can be generated, which reflects the relative importance of each performance indicator in leak detection. Subsequently, the system terminal uses historical leak data and historical normal data to verify the set of leak branch detectors and generate multiple verification results. After that, these verification results are compared with the actual labels, the correct verification result data is counted, and the statistical results are compared with the total number of verification results to obtain the accuracy information of the branch detectors, so as to understand their performance in leak detection.

[0058] Construct the decision coefficient blending function: ,in, Identify the decision coefficient of each leakage branch detector, is the key factor information of performance indicators, Identify branch detector accuracy information, along with increases with the increase of .

[0059] Optionally, in the process of building a lubricating oil pump station leakage adaptive detector, in order to integrate the output results of different leakage branch detectors, the system terminal defines a decision coefficient blending function. This function is designed to assign different weights according to the importance of the performance indicators of each branch detector and the accuracy of the detector, thereby forming a more accurate and reliable lubricating oil pump station leakage adaptive detector. The function is as follows:

[0060] ;

[0061] in, Identifies the decision coefficient of each leakage branch detector, which reflects the weight of each branch detector in the final leakage detection decision. The larger the decision coefficient, the more important the branch detector is in the fusion decision. The critical factor information of the performance index is obtained after the critical evaluation of each performance index in the lubricating oil pump station performance index set. The critical factor information reflects the relative importance of different performance indicators in leak detection. The more critical the performance index is, the larger the corresponding factor value is. Identify the branch detector accuracy information, which is the result obtained by verifying the set of leakage branch detectors, and reflects the accuracy of each branch detector in identifying leakage events. The higher the accuracy, the better the performance of the detector. When the product of the criticality factor of the performance indicator of the branch detector and the accuracy increases, the decision coefficient of the branch detector will also increase accordingly. This means that if the performance indicator corresponding to a branch detector is critical and its accuracy is also high, then its weight in the final leakage detection decision will be greater. Through such a fusion function design, it can be ensured that those branch detectors that are both critical and accurate play a greater role in the final lubricating oil pump station leakage adaptive detector, thereby improving the accuracy and reliability of the entire detection.

[0062] The decision coefficients of the leakage branch detector set are calculated based on the decision coefficient blending function to determine the decision coefficient information of the branch detector.

[0063] Optionally, the system terminal uses the constructed decision coefficient fusion function to calculate the decision coefficient of each detector in the leakage branch detector set. By calculating the decision coefficient of each branch detector, the system terminal obtains a set of decision coefficient information. This set of information not only reflects the relative importance of each detector in leakage detection, but also provides a basis for subsequent weighted fusion. Based on these decision coefficients, a more accurate and reliable adaptive detector for lubricating oil pump station leakage can be constructed to achieve rapid and accurate identification of leakage events.

[0064] In summary, the embodiments of the present application have at least the following technical effects:

[0065] The embodiment of the present application identifies key parts and deploys sensors through detailed functional structure decomposition to form a monitoring network covering the entire pump station. Subsequently, the network collects operating data in real time and performs abnormal identification processing through preprocessing multiple channels. Based on the performance indicators of the lubricating oil pump station, the data preprocessing process ensures the accuracy and validity of the data. Afterwards, a leakage branch detector is trained using a deep neural network structure, and the decision coefficient of each detector is determined through multivariate decision analysis to achieve the construction of a leakage adaptive detector. Then, based on this leakage adaptive detector, the lubricating oil pump station is subjected to real-time leakage fault detection. Once a leak is found, corresponding operation and maintenance measures are taken immediately to ensure the stable operation of the pump station. These technical effects jointly solve the technical problem that traditional detection methods are difficult to monitor and identify leakage faults in real time and accurately, resulting in untimely fault response and low operation and maintenance efficiency, and achieve rapid positioning and operation and maintenance of leakage faults, ensuring the operational reliability of the lubricating oil pump station.

[0066] Embodiment 2

[0067] Based on the same inventive concept as the leakage detection method of a lubricating oil pump station in the aforementioned embodiment, Figure 2 As shown, the present application provides a leakage detection system for a lubricating oil pump station, the system comprising:

[0068] Sensor deployment module 1: The sensor deployment module 1 is used to decompose the functional structure of the target lubricating oil pumping station, obtain a lubricating oil pumping station structural area set, deploy sensors in key parts based on the lubricating oil pumping station structural area set, and obtain a structural area sensor monitoring network set.

[0069] Attribute marking module 2: The attribute marking module 2 is used to monitor the operation of the target lubricating oil pump station based on the structural area sensor monitoring network set, obtain a multi-source structural area operation monitoring data stream set, perform attribute marking on the multi-source structural area operation monitoring data stream set, and obtain a structural area operation attribute data stream set.

[0070] Abnormal identification module 3: The abnormal identification module 3 is used to establish a data preprocessing multi-channel, map the structural area operation attribute data stream set with the data preprocessing multi-channel to perform abnormal identification processing, and obtain the structural area abnormal operation data stream set.

[0071] Fault operation and maintenance module 4: The fault operation and maintenance module 4 is used to construct a lubricating oil pump station leakage adaptive detector, detect and analyze the abnormal operation data stream set of the structural area based on the lubricating oil pump station leakage adaptive detector, obtain the lubricating oil pump station leakage detection result, and perform leakage fault operation and maintenance on the target lubricating oil pump station according to the lubricating oil pump station leakage detection result.

[0072] Furthermore, the sensor deployment module 1 is also used to execute the following method:

[0073] Identify key parts of each structural area in the structural area set of the lubricating oil pump station to determine the key part set of the structural area; perform comprehensive monitoring analysis based on the key part set of the structural area to obtain a key part sensor layout position parameter set and a key part sensor layout angle parameter set; extract operation function characteristics of the key part set of the structural area to obtain a key part operation function characteristic set and an operation function threshold parameter set; obtain a key part sensor type parameter set and a key part sensor specification parameter set based on the key part operation function characteristic set and the operation function threshold parameter set; determine the key part sensor layout parameter set based on the key part sensor layout position parameter set and the key part sensor layout angle parameter set, as well as the key part sensor type parameter set and the key part sensor specification parameter set; layout sensors at key parts of the structural area set of the lubricating oil pump station based on the key part sensor layout parameter set to obtain the structural area sensor monitoring network set.

[0074] Furthermore, the abnormality identification module 3 is also used to execute the following method:

[0075] Obtain a set of performance indicators of a lubricating oil pump station, wherein the set of performance indicators of the lubricating oil pump station includes pressure stability, temperature, liquid level change, vibration and noise, lubricating oil quality and operating energy consumption; construct a set of data preprocessing channels according to the set of performance indicators of the lubricating oil pump station; perform preprocessing step analysis and abnormal data recognition training on the set of structural area operating attribute data streams, respectively, to obtain a data structured format preprocessing node and an operating data abnormality recognition node; based on the data structured format preprocessing node and the operating data abnormality recognition node, perform processing node configuration on the data preprocessing channel set to establish the data preprocessing multi-channel.

[0076] Furthermore, the abnormality identification module 3 is also used to execute the following method:

[0077] Perform structured classification on the set of structural area operation attribute data streams to obtain a set of operation data structured format attributes; perform preprocessing step analysis based on the set of operation data structured format attributes to obtain a set of data structured format preprocessing programs; generate the data structured format preprocessing node according to the set of data structured format preprocessing programs; perform abnormal operation data recognition training on the set of structural area operation attribute data streams to obtain the operation data abnormality recognition node.

[0078] Furthermore, the abnormality identification module 3 is also used to execute the following method:

[0079] The structure area operation attribute data stream set is source labeled and classified to obtain the structure area operation data source label set; the lubricating oil pump station performance indicator set is standardized and quantified based on the structure area operation data source label set to generate a structure area label data standard set; the structure area operation attribute data stream set is trained for abnormal operation data recognition according to the structure area label data standard set to obtain an operation data attribute label abnormal discriminator set; the operation data attribute label abnormal discriminator set is labeled and classified based on the lubricating oil pump station performance indicator set to obtain the operation data abnormal recognition node.

[0080] Furthermore, the abnormality identification module 3 is also used to execute the following method:

[0081] Acquire a lubricating oil pump station leakage detection data set, classify and identify the lubricating oil pump station leakage detection data set according to the lubricating oil pump station performance indicator set, and obtain a leakage performance classification detection data set; use a deep neural network structure to perform equal weight layer detection training on the leakage performance classification detection data set to obtain a leakage branch detector set; perform multivariate decision analysis on the leakage branch detector set to determine the branch detector decision coefficient information; perform weighted fusion on the leakage branch detector set based on the branch detector decision coefficient information to construct the lubricating oil pump station leakage adaptive detector.

[0082] Furthermore, the abnormality identification module 3 is also used to execute the following method:

[0083] Perform a critical evaluation on each performance indicator in the lubricating oil pump station performance indicator set to obtain the critical factor information of the performance indicator; verify the leakage branch detector set to obtain the branch detector accuracy information; and construct a decision coefficient fusion function: ,in, Identify the decision coefficient of each leakage branch detector, is the key factor information of performance indicators, Identify branch detector accuracy information, along with increases with the increase of ; based on the decision coefficient blending function, the decision coefficient of the leakage branch detector set is calculated to determine the decision coefficient information of the branch detector.

[0084] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0085] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0086] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for detecting leakage of a lubricating oil pump station, characterized in that: The method comprises: Decomposing the target lubricating oil pump station by functional structure to obtain a lubricating oil pump station structural area set, deploying sensors at key locations based on the lubricating oil pump station structural area set, and obtaining a structural area sensor monitoring network set; Based on the structure area sensor monitoring network set, the target lubricating oil pump station is operated and monitored, a multi-source structure area operation monitoring data stream set is obtained, and attribute marking is performed on the multi-source structure area operation monitoring data stream set to obtain a structure area operation attribute data stream set; Establishing a data preprocessing multi-channel, mapping the structure area operation attribute data stream set with the data preprocessing multi-channel for abnormal identification processing, and obtaining the structure area abnormal operation data stream set; A lubricating oil pump station leakage adaptive detector is constructed, and based on the lubricating oil pump station leakage adaptive detector, the abnormal operation data stream set of the structural area is detected and analyzed to obtain the lubricating oil pump station leakage detection result, and leakage fault operation and maintenance of the target lubricating oil pump station is performed according to the lubricating oil pump station leakage detection result.

2. The method according to claim 1, characterized in that The obtaining of the structure area sensor monitoring network set comprises: Identifying key parts of each structural area in the set of structural areas of the lubricating oil pump station to determine a set of key parts of the structural areas; Performing comprehensive monitoring analysis on the key parts set of the structural area to obtain a key parts sensor layout position parameter set and a key parts sensor layout angle parameter set; Extracting operation function characteristics of the key parts set of the structural area to obtain the operation function characteristic set of the key parts and the operation function threshold parameter set, wherein the operation function threshold parameter set of each key part is determined according to the operation function characteristics, and the operation function threshold parameter is obtained based on the analysis of the normal operation data and potential fault data of the key parts; According to the key part operation function feature set and the operation function threshold parameter set, obtain a key part sensor type parameter set and a key part sensor specification parameter set; Determine a key position sensor layout parameter set based on the key position sensor layout position parameter set and the key position sensor layout angle parameter set, as well as the key position sensor type parameter set and the key position sensor specification parameter set; Based on the key position sensor layout parameter set, key position sensors are deployed on the structural area set of the lubricating oil pump station to obtain the structural area sensor monitoring network set.

3. The method according to claim 1, characterized in that The step of establishing multiple channels for data preprocessing comprises: Obtaining a set of lubricating oil pump station performance indicators, wherein the set of lubricating oil pump station performance indicators includes pressure stability, temperature, liquid level change, vibration and noise, lubricating oil quality, and operating energy consumption; Constructing a data preprocessing channel set according to the lubricating oil pump station performance indicator set; The set of the structural area operation attribute data streams is respectively subjected to preprocessing step analysis and abnormal data identification training to obtain a data structured format preprocessing node and an operation data abnormality identification node; Based on the data structured format preprocessing node and the running data anomaly identification node, the data preprocessing channel set is configured with processing nodes to establish the data preprocessing multi-channel.

4. The method according to claim 3, characterized in that The obtaining of the data structured format preprocessing node and the running data anomaly identification node includes: Performing structured classification on the set of operation attribute data streams of the structural area to obtain a set of operation data structured format attributes; Based on the set of structured format attributes of the running data, preprocessing steps are respectively parsed to obtain a set of structured format preprocessing programs for the data; Generating the data structured format preprocessing node according to the data structured format preprocessing program set; Abnormal operation data recognition training is performed on the set of operation attribute data streams of the structural area to obtain the abnormal operation data recognition node.

5. The method according to claim 4, characterized in that The obtaining of the operation data abnormality identification node includes: Performing source label classification on the structure area operation attribute data stream set to obtain a structure area operation data source label set; Based on the structure area operation data source label set, the lubricating oil pump station performance indicator set is standardized and quantified to generate a structure area label data standard set; Perform abnormal operation data recognition training on the structure area operation attribute data stream set according to the structure area label data standard set to obtain an operation data attribute label abnormal discriminator set; The operation data attribute label anomaly discriminator set is labeled and classified based on the lubricating oil pump station performance indicator set to obtain the operation data anomaly identification node.

6. The method according to claim 3, characterized in that The lubricating oil pump station leakage adaptive detector is constructed, comprising: Acquire a lubricating oil pump station leakage detection data set, classify and identify the lubricating oil pump station leakage detection data set according to the lubricating oil pump station performance indicator set, and obtain a leakage performance classification detection data set; Using a deep neural network structure, the leakage performance classification detection data set is respectively divided into equally weighted layers for detection training to obtain a set of leakage branch detectors; Performing a multivariate decision analysis on the set of leakage branch detectors to determine decision coefficient information of the branch detectors; The leakage branch detector set is weightedly fused based on the decision coefficient information of the branch detector to construct the lubricating oil pump station leakage adaptive detector.

7. The method according to claim 6, characterized in that The determining of the branch detector decision coefficient information includes: Performing a critical evaluation on each performance indicator in the lubricating oil pump station performance indicator set to obtain critical factor information of the performance indicator; Verifying the set of leakage branch detectors to obtain branch detector accuracy information; Construct the decision coefficient blending function: ,in, Identify the decision coefficient of each leakage branch detector, is the key factor information of performance indicators, Identify branch detector accuracy information, along with increases with the increase of The decision coefficients of the leakage branch detector set are calculated based on the decision coefficient blending function to determine the decision coefficient information of the branch detector.

8. A leakage detection system for a lubricating oil pump station, characterized in that: The system is used to implement the leakage detection method of a lubricating oil pump station according to any one of claims 1 to 7, comprising: Sensor deployment module: Decompose the functional structure of the target lubricating oil pump station to obtain a set of structural areas of the lubricating oil pump station, deploy sensors at key locations based on the set of structural areas of the lubricating oil pump station, and obtain a set of structural area sensor monitoring networks; Attribute marking module: performing operation monitoring of the target lubricating oil pump station based on the structural area sensor monitoring network set, obtaining a multi-source structural area operation monitoring data stream set, performing attribute marking on the multi-source structural area operation monitoring data stream set, and obtaining a structural area operation attribute data stream set; Anomaly identification module: establish a data preprocessing multi-channel, map the structure area operation attribute data stream set with the data preprocessing multi-channel to perform anomaly identification processing, and obtain the structure area abnormal operation data stream set; Fault operation and maintenance module: construct a lubricating oil pump station leakage adaptive detector, detect and analyze the abnormal operation data stream set of the structural area based on the lubricating oil pump station leakage adaptive detector, obtain the lubricating oil pump station leakage detection result, and perform leakage fault operation and maintenance on the target lubricating oil pump station according to the lubricating oil pump station leakage detection result.

Citation Information

Patent Citations

  • Pump station operation abnormality diagnosis method and system based on artificial intelligence

    CN108345282A

  • Water supply device of water intake pumping station and use method of water supply device

    CN114197572A

  • Early warning method for detecting leakage of hydraulic lubricating system

    CN117848596A

  • Method and apparatus for diagnosis of abnormality in leak limitation-type seal device

    JP1999271167A

  • Pipeline leak detection system

    US20090308140A1