A method, device, and storage medium for detecting the working status of a hydraulic motor.

Abstract

This application provides a method, device, and storage medium for detecting the operating status of a hydraulic motor, relating to the field of fault detection technology. The method includes: deploying a sensor cluster at key nodes of the hydraulic motor to construct a multi-dimensional monitoring dataset; calculating the displacement data of the hydraulic motor and the kinematic viscosity data of the hydraulic oil based on the multi-dimensional monitoring dataset to obtain a full-domain monitoring dataset; calculating the actual leakage of the hydraulic motor based on the full-domain monitoring dataset and generating an operating mode switching command; receiving the load monitoring command or shutdown maintenance command, initiating multi-modal fault diagnosis, and obtaining a set of hydraulic motor fault causes using a Bayesian network model; and determining a maintenance plan based on the hydraulic motor fault cause set. This solves the technical problems in existing technologies, such as lag, lack of effective preventative detection measures, inability to promptly detect potential problems in the hydraulic motor, and the resulting high maintenance costs and difficulty in equipment recovery.

Description

Technical Field

[0001] This application relates to the field of fault detection technology, and in particular to a method, device and storage medium for detecting the working status of a hydraulic motor. Background Technology

[0002] In the shipbuilding industry, hydraulic motors, with their superior performance in power output and torque transmission, have become an indispensable component for the normal operation of many ship equipment. For example, in key equipment such as the ship's steering gear system, cargo handling machine system, and anchor winch system, the stable operation of hydraulic motors directly affects the realization of important functions such as ship maneuverability, cargo loading and unloading efficiency, and berthing safety. Due to the crucial role of hydraulic motors in ship operation, ensuring their healthy and stable operation is essential. To achieve this, regular inspection of hydraulic motors is necessary. However, current inspection technologies for ship hydraulic motors have many shortcomings. During use, there is a lack of effective means to monitor operating parameters and status. Often, hydraulic motors are only inspected when abnormal noises or malfunctions occur. This reactive approach carries significant risks, because by the time obvious abnormalities appear, the hydraulic motor may already be internally damaged or even destroyed. Once this happens, even if the cause of the fault can be identified, repair work faces numerous difficulties. On the one hand, repair costs are usually very high, which undoubtedly increases the operating costs of the ship; on the other hand, in some cases, the damage to the hydraulic motor may be irreparable and can only be replaced. This not only increases the cost of equipment replacement, but may also affect the normal operation of the ship. From the perspective of equipment management and maintenance costs, it takes a high price to restore the normal function of the equipment.

[0003] In summary, existing technologies for hydraulic motor testing exhibit significant lag, lacking effective preventative testing measures and failing to promptly identify potential problems. This can lead to high maintenance costs and difficulties in equipment restoration. Therefore, a solution is needed to address these issues. Summary of the Invention

[0004] This disclosure provides a method, device, and storage medium for detecting the working status of a hydraulic motor, in order to solve the technical problems in the prior art, such as lag, lack of effective preventive detection measures, inability to detect potential problems of hydraulic motors in a timely manner, and high maintenance costs and difficulty in equipment recovery.

[0005] According to a first aspect of this disclosure, a method for detecting the operating state of a hydraulic motor is provided, comprising:

[0006] A sensor cluster is deployed at key nodes of the hydraulic motor to acquire multi-dimensional state parameters of the hydraulic motor in real time. The multi-dimensional state parameters of the hydraulic motor are preprocessed to construct a multi-dimensional monitoring dataset, which includes flow data, speed data, pressure data, and temperature data.

[0007] Based on the multidimensional monitoring dataset, the displacement data of the hydraulic motor and the kinematic viscosity data of the hydraulic oil are calculated to obtain the full-domain monitoring dataset;

[0008] The actual leakage of the hydraulic motor is calculated based on the full-domain monitoring dataset. The hydraulic motor status is assessed based on the actual leakage of the hydraulic motor, and an operating mode switching command is generated. The operating mode switching command includes a continuous operation command, a load monitoring command, and a shutdown maintenance command.

[0009] Upon receiving the load monitoring command or shutdown maintenance command, initiate multimodal fault diagnosis and use a Bayesian network model to obtain the set of causes of hydraulic motor faults;

[0010] A repair plan is determined based on the set of causes of hydraulic motor failure. The repair plan includes fault location information, a spare parts list, a process route diagram, and an estimated repair time.

[0011] Furthermore, a multidimensional monitoring dataset is constructed, including:

[0012] The inlet and outlet pipelines, shaft end, and housing drain port of the hydraulic motor are selected as key nodes for deploying a sensor cluster, which includes a flow sensor, a speed sensor, a pressure sensor, and a temperature sensor.

[0013] The sensor cluster continuously collects flow rate, speed, pressure, and temperature data of the hydraulic motor during operation to obtain multi-dimensional state parameters of the hydraulic motor. The IEEE 1588 precision time protocol is used to ensure the synchronous capture of the multi-dimensional state parameters of the hydraulic motor.

[0014] The isolated forest algorithm was used to remove outliers from the multidimensional state parameters of the hydraulic motor, and cubic spline interpolation was used to fill in the missing data segments, reconstruct the continuous signal, and build a multidimensional monitoring dataset.

[0015] Furthermore, obtain the full-domain monitoring dataset, including:

[0016] The displacement data of the hydraulic motor is obtained based on the flow rate data and speed data in the multidimensional monitoring dataset;

[0017] A hydraulic oil viscosity-temperature relationship calculation model is set according to the type of hydraulic oil, and the specific formula is as follows:

[0018] ISO VG32: η=144-3.2T;

[0019] ISO VG48: η=107.5-1.25T;

[0020] ISO VG68: η=159-2.4T;

[0021] Among them, ISO VG32, ISO VG48, and ISO VG68 are three commonly used hydraulic oil categories for hydraulic motors. η represents the kinematic viscosity of the hydraulic oil, and T represents the temperature data.

[0022] Input the temperature data from the multidimensional monitoring dataset to obtain the kinematic viscosity data of the corresponding hydraulic oil;

[0023] By integrating the displacement data, kinematic viscosity data, and multidimensional monitoring dataset, a comprehensive monitoring dataset is constructed.

[0024] Furthermore, generate operating mode switching instructions, including:

[0025] The actual leakage of the hydraulic motor is calculated based on the aforementioned full-domain monitoring dataset, using the following formula:

[0026] ;

[0027] Among them, Q L X represents the actual leakage of the hydraulic motor, X represents the leakage coefficient with a specific value of 0.36, Q represents the displacement data of the hydraulic motor, P represents the pressure data of the hydraulic motor, and η represents the kinematic viscosity data of the hydraulic oil.

[0028] The actual leakage of the hydraulic motor is compared with the leakage baseline to establish a three-level evaluation system. The leakage baseline includes a mild leakage baseline M and a severe leakage baseline N, wherein the mild leakage baseline M is 3%Q / 100bar and the severe leakage baseline N is 5%Q / 100bar.

[0029] When 0 L If the leakage is ≤N, the hydraulic motor is determined to be in a first-level health condition, and a continuous operation command is generated. The current leakage is within the allowable fluctuation range, the hydraulic motor is in good condition, and the regular testing cycle continues.

[0030] When N L If the value is ≤M, the hydraulic motor is determined to be in a level 2 warning state, and a load monitoring command is generated. At this time, the leakage of the hydraulic motor exceeds the safety margin and needs to be reduced to 85% of the rated power for operation. Simultaneously, multi-mode fault diagnosis is initiated.

[0031] When Q L >M indicates that the hydraulic motor is in a level 3 fault state, and a shutdown and maintenance command is generated. At this time, the hydraulic motor is leaking severely and needs to be stopped immediately for multi-modal fault diagnosis. ​​

[0032] Furthermore, upon receiving the load monitoring command or shutdown maintenance command, multimodal fault diagnosis is initiated, and a Bayesian network model is used to obtain the set of causes of hydraulic motor faults, including:

[0033] Obtain the full-domain monitoring dataset, divide the full-domain monitoring dataset, and regard the data collected at the last 10 time points as real-time monitoring data, and the remaining data as historical monitoring data;

[0034] Based on the historical monitoring data, a Bayesian network model is constructed;

[0035] Based on the real-time monitoring data, a real-time status feature group is obtained, which includes the real-time actual leakage amount, real-time leakage rate, real-time pressure decay gradient, real-time kinematic viscosity change residual, and real-time temperature anomaly index.

[0036] Input the instantaneous state feature set into the Bayesian network model to obtain the set of causes of hydraulic motor failure.

[0037] Furthermore, a Bayesian network model is constructed, including:

[0038] Multimodal data feature extraction is performed on the historical monitoring data to obtain historical leakage state feature group and historical oil state feature group. The historical leakage state feature group includes historical actual leakage amount, historical leakage rate, and historical pressure decay gradient. The historical oil state feature group includes historical kinematic viscosity change residual and historical temperature anomaly index.

[0039] By analyzing the historical leakage state characteristic group, internal and external leakage diagnosis of the hydraulic motor is performed to obtain the set of internal and external leakage causes.

[0040] By analyzing the historical oil condition characteristic group, hydraulic oil deterioration diagnosis is performed to obtain the set of oil deterioration causes.

[0041] Using the historical leakage state characteristic group and the historical oil state characteristic group as observed variables, and the latent variables of the internal and external leakage cause set and the oil deterioration cause set, a Bayesian network model is constructed.

[0042] According to a second aspect of this disclosure, a hydraulic motor operating status detection device is provided, comprising:

[0043] A multidimensional monitoring dataset acquisition module is used to deploy a sensor cluster at key nodes of the hydraulic motor, acquire multidimensional state parameters of the hydraulic motor in real time, preprocess the multidimensional state parameters of the hydraulic motor, and construct a multidimensional monitoring dataset, which includes flow data, speed data, pressure data, and temperature data.

[0044] A full-domain monitoring dataset construction module is used to calculate the displacement data of the hydraulic motor and the kinematic viscosity data of the hydraulic oil based on the multi-dimensional monitoring dataset to obtain a full-domain monitoring dataset.

[0045] The status assessment and instruction generation module is used to calculate the actual leakage of the hydraulic motor based on the full-domain monitoring dataset, assess the status of the hydraulic motor based on the actual leakage, and generate operating mode switching instructions, including continuous operation instructions, load monitoring instructions, and shutdown maintenance instructions.

[0046] The fault diagnosis module is used to receive the load monitoring command or shutdown maintenance command, initiate multimodal fault diagnosis, and use a Bayesian network model to obtain the fault cause set of the hydraulic motor.

[0047] The maintenance plan determination module is used to determine a maintenance plan based on the set of causes of hydraulic motor failure. The maintenance plan includes fault location information, spare parts list, process route diagram and estimated maintenance time.

[0048] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the steps of the above-described hydraulic motor operating state detection method.

[0049] One or more technical solutions provided in this disclosure have at least the following technical effects or advantages:

[0050] A sensor cluster is deployed at key nodes of the hydraulic motor to acquire multi-dimensional state parameters of the hydraulic motor in real time. These parameters are preprocessed to construct a multi-dimensional monitoring dataset, which includes flow rate, speed, pressure, and temperature data. Based on this dataset, the hydraulic motor's displacement and the kinematic viscosity of the hydraulic oil are calculated to obtain a full-domain monitoring dataset. The actual leakage of the hydraulic motor is calculated based on this dataset, and the motor's state is assessed accordingly. Operating mode switching commands are generated, including continuous operation, load monitoring, and shutdown / maintenance commands. Upon receiving a load monitoring or shutdown / maintenance command, multi-modal fault diagnosis is initiated, and a Bayesian network model is used to obtain a set of hydraulic motor fault causes. A maintenance plan is determined based on this set of causes, including fault location information, a spare parts list, a process route diagram, and an estimated maintenance time. This approach solves the technical problems of existing technologies, such as delays, lack of effective preventative detection measures, inability to promptly identify potential problems in the hydraulic motor, and the resulting high maintenance costs and difficulty in equipment recovery. This achieves the technical effect of effectively monitoring hydraulic motors, promptly identifying problems, and reducing equipment maintenance costs.

[0051] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0053] Figure 1 A flowchart illustrating a method for detecting the working state of a hydraulic motor, as provided in an embodiment of this application;

[0054] Figure 2 This is a schematic diagram of a hydraulic motor operating status detection device provided in an embodiment of this application.

[0055] Figure labeling: 11 Multidimensional monitoring dataset acquisition module, 12 Global monitoring dataset construction module, 13 Status assessment and instruction generation module, 14 Fault diagnosis module, 15 Maintenance plan determination module. Detailed Implementation

[0056] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0057] Example 1: A method for detecting the working status of a hydraulic motor provided in this disclosure, which is referred to below. Figure 1 The methods include:

[0058] S1: Deploy a sensor cluster at key nodes of the hydraulic motor to acquire multi-dimensional state parameters of the hydraulic motor in real time. Preprocess the multi-dimensional state parameters of the hydraulic motor to construct a multi-dimensional monitoring dataset, which includes flow data, speed data, pressure data, and temperature data.

[0059] Specifically, based on the working principle and structural characteristics of the hydraulic motor, the inlet and outlet pipelines, shaft end, and housing drain port are identified as key monitoring nodes. A high-precision sensor cluster consisting of flow, speed, pressure, and temperature sensors is deployed to construct a distributed data acquisition network. A time synchronization mechanism for the sensor cluster is set, using the IEEE 1588 precision time protocol to achieve sub-microsecond clock synchronization, resolving data time misalignment issues caused by differences in internal sensor clocks. The sensor cluster's acquisition frequency is set to 1Hz. Each sensor acquires raw data at this frequency, generating multiple initial data sequences. An isolated forest algorithm is run on each initial data sequence to identify and remove outliers. The initial data sequences after anomaly detection and interpolation repair are categorized and integrated, using initial data sequences of the same category as the integration object, and the time axis as the alignment standard. The average value is taken to obtain the final data, constructing a multi-dimensional monitoring dataset for condition monitoring and fault diagnosis.

[0060] S2: Calculate the displacement data of the hydraulic motor and the kinematic viscosity data of the hydraulic oil based on the multidimensional monitoring dataset to obtain the full-domain monitoring dataset;

[0061] Specifically, flow rate and rotational speed data are obtained from a multidimensional monitoring dataset, and hydraulic motor displacement data is calculated using formulas. Hydraulic oil kinematic viscosity data is obtained through temperature data, and a viscosity-temperature relationship calculation model is constructed to determine the corresponding viscosity-temperature relationship calculation formulas for the three types of hydraulic oil. Because hydraulic oil viscosity exhibits a lag in response to temperature, a first-order inertial element formula with a lag time constant is introduced to correct the kinematic viscosity data during the hydraulic motor's start-up and heating phase. Initially, the hydraulic oil kinematic viscosity does not respond to temperature changes, but it approaches an ideal value over time. Finally, the multidimensional monitoring dataset is expanded based on the obtained displacement and kinematic viscosity data. Timestamps are used as indexes to ensure synchronization of all data sampling times. Two new columns are added to the original dataset to store displacement and kinematic viscosity data respectively, constructing a comprehensive monitoring dataset to lay the data foundation for subsequent leakage calculations.

[0062] S3: Calculate the actual leakage of the hydraulic motor based on the full-domain monitoring dataset, assess the hydraulic motor status based on the actual leakage of the hydraulic motor, and generate an operating mode switching instruction, which includes a continuous operation instruction, a load monitoring instruction, and a shutdown maintenance instruction.

[0063] Specifically, the actual leakage of the hydraulic motor is calculated by combining displacement, pressure, and kinematic viscosity data from the comprehensive monitoring dataset, and then compared with the leakage baseline to determine the current status of the hydraulic motor. The leakage baseline is mainly divided into a minor leakage baseline and a major leakage baseline, representing the short-term fluctuation range and the boundary that may lead to irreversible damage, respectively. When the actual leakage is less than or equal to the minor leakage baseline, the hydraulic motor is operating healthily, indicating that the equipment is in good condition. At this time, the leakage is normal and has no substantial impact on equipment performance and safety, generating a continuous operation command. If the leakage is close to the minor leakage baseline but not exceeding the limit, the sampling frequency needs to be increased to prevent exceeding the baseline and causing missed faults. When the actual leakage exceeds the minor leakage baseline but does not reach the major leakage baseline, it is determined that the hydraulic motor has an early abnormality, generating a load monitoring command. Although irreversible damage has not yet been reached, the safety margin has been exceeded, requiring derated operation, and multimodal fault diagnosis is initiated to find the fault source. When the actual leakage exceeds the major leakage baseline, the hydraulic motor has severely failed, and continued operation will damage the equipment. At this point, the leakage reaches a dangerous level. The power source must be cut off to avoid serious failure. A shutdown and maintenance command is generated, the power supply is cut off, the locking device is activated, multi-modal diagnostics is used to locate the fault point, and maintenance-related information is pushed based on the results.

[0064] S4: Receive the load monitoring command or shutdown maintenance command, start multimodal fault diagnosis, and use a Bayesian network model to obtain the set of causes of hydraulic motor faults;

[0065] Specifically, upon receiving load monitoring commands or shutdown / maintenance commands, multimodal fault diagnosis is initiated, and the cause of hydraulic motor failure is determined based on existing monitoring data. A full-domain monitoring dataset is acquired and divided, with the data collected at the last 10 time points considered as real-time monitoring data and the remaining data as historical monitoring data. Historical monitoring data of the hydraulic motor is acquired, and a historical monitoring dataset is constructed. Multimodal feature extraction is performed on the historical monitoring dataset to generate historical leakage state feature groups and historical oil state feature groups. Through analysis and processing of these historical leakage state feature groups and historical oil state feature groups, the diagnosis of internal and external leakage and oil degradation of the hydraulic motor is completed, obtaining sets of internal and external leakage causes and oil degradation causes. A Bayesian network model for hydraulic motor fault diagnosis is constructed using the historical leakage state feature groups, historical oil state feature groups, internal and external leakage cause sets, and oil degradation cause sets. Real-time state feature groups are extracted from real-time monitoring data and input into the Bayesian network model to calculate the probability distribution of fault causes. The top three combinations of fault causes are output as the hydraulic motor fault cause set.

[0066] S5: Determine a repair plan based on the set of causes of hydraulic motor failure. The repair plan includes fault location information, spare parts list, process route diagram and estimated repair time.

[0067] Specifically, the process involves acquiring a set of causes for hydraulic motor failures and determining a repair plan based on the detailed causes within that set. The repair plan is a comprehensive and systematic plan designed to provide precise and effective solutions for various potential hydraulic motor failures. It encompasses several important aspects, such as fault location information, spare parts lists, process route diagrams, and estimated repair time. The core of generating a repair plan lies in the fact that after determining the fault location, it's not enough to merely locate it superficially; a detailed inspection of the fault location is essential. This inspection process requires the use of specialized testing equipment and methods to comprehensively and deeply analyze the specific condition of the fault location, such as detecting the degree of wear, corrosion, and component deformation—all possible manifestations of the failure. Based on the results of the detailed inspection of the fault location, and leveraging extensive experience and professional knowledge, targeted repair plans are developed for these faulty areas. The design of the process route is a crucial step in the repair planning. This requires planning a scientific, reasonable, efficient, and feasible repair process based on factors such as the equipment's structural characteristics, the type of failure, and the difficulty of repair. Simultaneously, estimating the repair time is also an indispensable part of the repair plan. This requires comprehensive consideration of factors such as the complexity of the fault, the ease of the repair operation, and the skill level of the repair personnel. Finally, all of the above should be integrated to generate a report, including accurate fault location information, a complete spare parts list, a detailed process route diagram, and a reasonable estimated repair time. This allows staff to have a comprehensive understanding of the repair plan, providing accurate guidance for subsequent repair work and strong data support for subsequent fault detection.

[0068] Taking piston cylinder wear failure as an example, if the hydraulic motor failure is mainly due to internal leakage caused by piston cylinder wear, the fault location is the piston cylinder part of the hydraulic motor. In this case, a thorough inspection of the wear condition of the cylinder's inner wall is necessary. Using endoscopes or other testing equipment, the wear can be precisely located and its extent determined. Once the wear level is confirmed, a spare parts list should be compiled. If the cylinder is a monolithic structure, the entire cylinder should be prepared as a spare. If it is a detachable structure, only the worn cylinder barrel or piston component should be prepared, depending on the situation. A process route should be planned, specifically including: removing the hydraulic motor from the equipment according to operating procedures; disassembling step-by-step down to the piston cylinder component; using specialized tools to disassemble the old piston cylinder and cleaning the installation area; strictly controlling the sealing and installation accuracy when installing the new piston cylinder; performing necessary adjustments such as pressure and flow tests after installation; and reinstalling the hydraulic motor back into the equipment and performing overall debugging. Given the complexity of disassembling and installing the plunger cylinder block and the requirements for precision adjustment, the estimated maintenance time is 8 to 12 hours. The actual time will be affected by the complexity of the equipment structure and the skill level of the maintenance personnel.

[0069] Furthermore, step S1 of this application also includes:

[0070] The inlet and outlet pipelines, shaft end, and housing drain port of the hydraulic motor are selected as key nodes for deploying a sensor cluster, which includes a flow sensor, a speed sensor, a pressure sensor, and a temperature sensor.

[0071] The sensor cluster continuously collects flow rate, speed, pressure, and temperature data of the hydraulic motor during operation to obtain multi-dimensional state parameters of the hydraulic motor. The IEEE 1588 precision time protocol is used to ensure the synchronous capture of the multi-dimensional state parameters of the hydraulic motor.

[0072] The isolated forest algorithm was used to remove outliers from the multidimensional state parameters of the hydraulic motor, and cubic spline interpolation was used to fill in the missing data segments, reconstruct the continuous signal, and build a multidimensional monitoring dataset.

[0073] Specifically, based on the working principle and structural characteristics of the hydraulic motor, the inlet and outlet pipelines, shaft end, and housing drain port are identified as key monitoring nodes. A high-precision sensor cluster is deployed to construct a distributed data acquisition network. The sensor cluster includes flow sensors, speed sensors, pressure sensors, and temperature sensors. The flow sensor is a pipeline induction electromagnetic flow meter, installed in the inlet and outlet pipelines of the hydraulic motor. The flow sensor integrates a ring clamp mounting base and embeds a non-contact Hall effect sensor, which can capture the flow waveform in real time, achieving accurate monitoring of the hydraulic motor's flow. The speed sensor is a photoelectric speed sensor, installed at the shaft end of the hydraulic motor. The photoelectric speed sensor obtains the hydraulic motor's speed information by detecting changes in the optical signal at the shaft end. The pressure sensor is a silicon piezoresistive pressure sensor, sealed with an O-ring on the pressure measuring manifold of the inlet and outlet pipelines. The temperature sensor is a platinum resistance temperature sensor, embedded in the heat-conducting groove on the outer wall of the housing drain port pipeline.

[0074] A time synchronization mechanism for the sensor cluster is established, employing the IEEE 1588 precision time protocol to achieve sub-microsecond clock synchronization, resolving data time misalignment issues caused by internal clock differences among different sensors. The IEEE 1588 precision time protocol is a protocol for achieving high-precision time synchronization over Ethernet. Its working principle involves establishing a master-slave hierarchy, setting master and slave clocks internally in the central controller and within the sensors respectively. The master clock sends a synchronization signal to all sensors every second, and the sensors immediately calibrate their own clocks upon receiving the signal. Through this method, each sensor can calibrate its local clock offset, thereby achieving time and frequency synchronization and ensuring that the time alignment error of flow rate, rotational speed, pressure, and temperature data is less than 10 μs.

[0075] The sensor cluster acquisition frequency is set to 1Hz based on the operating characteristics of the hydraulic motor and the requirements for fault detection. Although the flow rate and temperature change relatively slowly during the operation of the hydraulic motor, all data must be aligned on the time axis in this application to lay the data foundation for subsequent calculations. Therefore, to avoid data gaps, the acquisition frequency of all data must be consistent with the pressure and flow rate data with the highest acquisition frequency.

[0076] Each sensor continuously acquires raw data according to its acquisition frequency. The acquired raw data is then arranged chronologically to generate initial data sequences for each sensor. Since the sensor cluster includes not only four types of sensors, but also a variety of sensors within each type, the initial data sequences specifically include multiple sets of data, such as Initial Flow Data Sequence 1, Initial Flow Data Sequence 2, Initial Rotation Speed ​​Data Sequence 1, Initial Pressure Data Sequence 1, Initial Pressure Data Sequence 2, Initial Temperature Data Sequence 1, and Initial Temperature Data Sequence 2. An isolated forest algorithm is run separately for each sequence to identify and remove outliers. Specific operations include: first, setting the time window length and sliding step size based on data characteristics; for example, a 10-second window for pressure data with a 5-second sliding step size, achieving 50% overlap. This operation avoids memory overflow caused by processing large amounts of data at once. Finally, the initial data sequences acquired by each sensor are standardized separately, converting them into a standardized format to eliminate the impact of dimensional differences on the algorithm and enhance its robustness. Secondly, an isolated forest model is independently constructed for each initial data sequence. During model construction, 200 isolated trees are trained in parallel. 256 samples are randomly selected from each tree to construct decision paths, and the maximum depth of the trees is limited to prevent overfitting. This process randomly partitions the feature space, ensuring that normal data points require longer paths to be isolated, while outliers, significantly deviating from the group distribution, can be quickly identified at shallower nodes. Thirdly, based on the constructed isolated forest model, the system quantifies the anomaly probability for each data point. All isolated trees are traversed, and the average path length required to isolate the target data point is calculated: normal data, due to its dense distribution, typically requires multiple node splits for isolation, resulting in longer paths; while outliers, far from the main data group, only require a few splits, significantly shortening the path. The average path length is calculated, and outlier data is determined based on this average path length. Finally, data cleaning is performed. Data points identified as outliers are directly marked as invalid values ​​(NaN), and their location information is preserved in the time series. Simultaneously, a cubic spline interpolation algorithm is initiated to reconstruct missing segments using preceding and following normal data. For example, when signal interference causes errors in pressure data acquisition, the system automatically retrieves normal pressure data after marking all five abnormal data points, generating a smooth transition curve to ensure the temporal continuity of subsequent analysis. Detailed logs are generated for all processing steps, including the number of abnormal data points and the repair method, providing a traceable data foundation for equipment health management.

[0077] The initial data sequences, after anomaly detection and interpolation repair, are categorized and integrated. Initial data sequences of the same category are used as the integration objects, and the time axis is used as the alignment standard. The average value of all sensor-recorded data at that time is taken as the final obtained data. For example, at 10:00, the pressure data collected by initial pressure data sequence one and initial pressure data sequence two are 19.8 MPa and 20.0 MPa respectively, then the final pressure data is 19.9 MPa. Based on this method, the integrated flow rate data, rotational speed data, pressure data, and temperature data are obtained respectively, constructing a multi-dimensional monitoring dataset for condition monitoring and fault diagnosis.

[0078] Furthermore, step S2 of this application also includes:

[0079] The displacement data of the hydraulic motor is obtained based on the flow rate data and speed data in the multidimensional monitoring dataset;

[0080] A hydraulic oil viscosity-temperature relationship calculation model is set according to the type of hydraulic oil, and the specific formula is as follows:

[0081] ISO VG32: η=144-3.2T;

[0082] ISO VG48: η=107.5-1.25T;

[0083] ISO VG68: η=159-2.4T;

[0084] Among them, ISO VG32, ISO VG48, and ISO VG68 are three commonly used hydraulic oil categories for hydraulic motors. η represents the kinematic viscosity of the hydraulic oil, and T represents the temperature data.

[0085] Input the temperature data from the multidimensional monitoring dataset to obtain the kinematic viscosity data of the corresponding hydraulic oil;

[0086] By integrating the displacement data, kinematic viscosity data, and multidimensional monitoring dataset, a comprehensive monitoring dataset is constructed.

[0087] Specifically, a multidimensional monitoring dataset is obtained, and the flow rate and speed data are extracted from the dataset to calculate the displacement data of the hydraulic motor. The specific formula is as follows: Where D represents displacement, Q represents flow rate, N represents rotational speed, and ω represents volumetric efficiency. Volumetric efficiency is determined by the hydraulic motor model. Due to mechanical friction and hysteresis in actual operating conditions, the effective rotational speed may be slightly reduced; therefore, the displacement data needs to be corrected based on volumetric efficiency. The kinematic viscosity data of the hydraulic oil is obtained from the temperature data in the multidimensional monitoring dataset. A hydraulic oil viscosity-temperature relationship calculation model is constructed, and the calculation formulas for the viscosity-temperature relationship corresponding to the three types of hydraulic oil are determined. It should be noted that in a hydraulic system, the viscosity of the hydraulic oil does not respond instantaneously to temperature. When the temperature changes, the molecular structure adjustment within the oil requires a certain amount of time, causing the viscosity change to lag behind the temperature change. Therefore, during the startup of the hydraulic system, the temperature gradually rises, and the calculated kinematic viscosity data needs to be corrected. However, during the stable operation of the hydraulic motor, since the temperature change is no longer significant, the kinematic viscosity correction is not necessary. For the heating process during hydraulic motor startup, a first-order inertial element is introduced to describe this dynamic process with hysteresis characteristics, thus achieving the correction of the hydraulic oil viscosity. The specific formula is: , where η actual Here, η is the corrected kinematic viscosity data, e is the natural constant, t is time, and μ is the hysteresis time constant, characterizing the degree of hysteresis in the oil viscosity response to temperature changes. When t=0, =1, at which point η actual =0, which means that at the initial moment, the kinematic viscosity of the hydraulic oil has not yet had time to respond to the temperature change. As time t increases, Gradually decrease, η actual It gradually approaches η. When t is large enough, η is approximately equal to 0 at this time. actual Approximately equal to η, meaning that after a sufficiently long period of time, the oil viscosity will eventually reach the ideal value calculated by the model.

[0088] Based on the displacement and kinematic viscosity data calculated using the formulas above, the multidimensional monitoring dataset is expanded. Using timestamps as indexes to ensure strict synchronization of sampling times for all data sequences, two new columns are added to the original multidimensional dataset to store displacement and kinematic viscosity data respectively, constructing a comprehensive monitoring dataset to provide a data foundation for subsequent leakage calculations.

[0089] Furthermore, step S3 of this application also includes:

[0090] The actual leakage of the hydraulic motor is calculated based on the aforementioned full-domain monitoring dataset, using the following formula:

[0091] ;

[0092] Among them, Q LX represents the actual leakage of the hydraulic motor, X represents the leakage coefficient with a specific value of 0.36, Q represents the displacement data of the hydraulic motor, P represents the pressure data of the hydraulic motor, and η represents the kinematic viscosity data of the hydraulic oil.

[0093] The actual leakage of the hydraulic motor is compared with the leakage baseline to establish a three-level evaluation system. The leakage baseline includes a mild leakage baseline M and a severe leakage baseline N, wherein the mild leakage baseline M is 3%Q / 100bar and the severe leakage baseline N is 5%Q / 100bar.

[0094] When 0 L If the leakage is ≤N, the hydraulic motor is determined to be in a first-level health condition, and a continuous operation command is generated. The current leakage is within the allowable fluctuation range, the hydraulic motor is in good condition, and the regular testing cycle continues.

[0095] When N L If the value is ≤M, the hydraulic motor is determined to be in a level 2 warning state, and a load monitoring command is generated. At this time, the leakage of the hydraulic motor exceeds the safety margin and needs to be reduced to 85% of the rated power for operation. Simultaneously, multi-mode fault diagnosis is initiated.

[0096] When Q L >M indicates that the hydraulic motor is in a level 3 fault state, and a shutdown and maintenance command is generated. At this time, the hydraulic motor is leaking severely and needs to be stopped immediately for multi-modal fault diagnosis.

[0097] Specifically, the system acquires displacement, pressure, and kinematic viscosity data from a comprehensive monitoring dataset. This data is then combined with a pre-defined formula derived from extensive experimental data to calculate the actual leakage of the hydraulic motor. A leakage baseline is established, serving as a dynamic reference value to quantify the safe threshold for hydraulic motor leakage. This baseline is divided into a mild leakage baseline (M) and a severe leakage baseline (N), with specific values ​​of M = 3%Q / 100bar and N = 5%Q / 100bar. The mild leakage baseline (M) is defined as the allowable short-term fluctuation range, while the severe leakage baseline (N) is defined as the range that could lead to irreversible damage to the hydraulic motor.

[0098] When the actual leakage is less than or equal to the minor leakage baseline, the hydraulic motor is in a healthy operating state. In this state, the leakage is within the allowable fluctuation range of the equipment design, indicating that the sealing system is intact, the oil viscosity is stable, and there are no significant signs of wear or fatigue. At this point, it is determined that the current leakage has no substantial impact on equipment performance and safety, and there is no need to adjust operating parameters. A continuous operation command is generated to maintain rated power output, the original detection cycle is maintained, and periodic health reports are generated to record trend data. However, if the leakage is extremely close to the minor leakage baseline but not exceeding the limit, the data sampling frequency needs to be increased to prevent exceeding the minor leakage baseline and causing missed fault detection.

[0099] ​​When the actual leakage exceeds the baseline for minor leakage but does not reach the baseline for severe leakage, it indicates an early abnormality in the hydraulic motor, possibly caused by minor wear of internal components, oil contamination, or localized overheating. Immediate intervention is required to prevent further deterioration. At this point, the leakage has exceeded the safety margin but has not yet reached the threshold for irreversible damage. Risk should be reduced by derating the operation, generating a load monitoring command to limit the output power to 85% of the rated value, and reducing system pressure to slow the leakage rate. Simultaneously, multimodal fault diagnosis is initiated to locate potential fault sources. The data sampling frequency is increased to 10 times per second to track the leakage trend in real time.

[0100] When the actual leakage exceeds the severe leakage baseline, it indicates a serious failure of the hydraulic motor, such as excessive wear of the piston cylinder clearance or abnormal scratches on the distributor plate. Continued operation will lead to oil contamination, a sharp drop in efficiency, or even equipment damage. At this point, the leakage is deemed to have reached a dangerous threshold, and the power source must be immediately cut off to avoid catastrophic failure. A shutdown and maintenance command should be generated, the hydraulic pump power supply should be cut off, the safety locking device should be activated to prevent inertial slippage, and the fault point should be located using multimodal fault diagnosis. Based on the diagnostic results, a spare parts list, process route diagram, and estimated maintenance time should be provided.

[0101] Furthermore, step S4 of this application also includes:

[0102] Obtain the full-domain monitoring dataset, divide the full-domain monitoring dataset, and regard the data collected at the last 10 time points as real-time monitoring data, and the remaining data as historical monitoring data;

[0103] Based on the historical monitoring data, a Bayesian network model is constructed;

[0104] Based on the real-time monitoring data, a real-time status feature group is obtained, which includes the real-time actual leakage amount, real-time leakage rate, real-time pressure decay gradient, real-time kinematic viscosity change residual, and real-time temperature anomaly index.

[0105] Input the instantaneous state feature set into the Bayesian network model to obtain the set of causes of hydraulic motor failure.

[0106] Specifically, a Bayesian network model is constructed based on historical monitoring data and fault detection records of the hydraulic motor. Real-time monitoring data is obtained from the full-domain monitoring dataset, and key features are extracted from this data to construct a real-time state feature group, which serves as the input data for the Bayesian network. The data in the real-time state feature group are obtained in the same way as the corresponding data in the historical leakage state feature group and the historical oil state feature group.

[0107] The data from the instantaneous state feature set is input into a Bayesian network model to calculate the probability distribution of fault causes. Each data point in the instantaneous state feature set is mapped to a leaf node of the Bayesian network, activating the associated fault hypothesis. The posterior probability distribution is calculated using a connection tree algorithm, and the top three fault causes by probability are output. These fault causes are then combined into a set of fault causes for the hydraulic motor.

[0108] Furthermore, step S4 of this application also includes:

[0109] Multimodal data feature extraction is performed on the historical monitoring data to obtain historical leakage state feature group and historical oil state feature group. The historical leakage state feature group includes historical actual leakage amount, historical leakage rate, and historical pressure decay gradient. The historical oil state feature group includes historical kinematic viscosity change residual and historical temperature anomaly index.

[0110] By analyzing the historical leakage state characteristic group, internal and external leakage diagnosis of the hydraulic motor is performed to obtain the set of internal and external leakage causes.

[0111] By analyzing the historical oil condition characteristic group, hydraulic oil deterioration diagnosis is performed to obtain the set of oil deterioration causes.

[0112] Using the historical leakage state characteristic group and the historical oil state characteristic group as observed variables, and the latent variables of the internal and external leakage cause set and the oil deterioration cause set, a Bayesian network model is constructed.

[0113] Specifically, historical monitoring data of hydraulic motors are obtained through a comprehensive monitoring dataset, which is then used to construct a historical monitoring dataset. This dataset includes historical flow data, historical speed data, historical pressure data, historical temperature data, historical displacement data, and historical kinematic viscosity data. Multimodal feature extraction is performed on the historical monitoring dataset to obtain historical actual leakage, historical leakage rate, historical pressure decay gradient, historical kinematic viscosity change residual, and historical temperature anomaly index.

[0114] Specifically, the historical actual leakage amount Q leak The calculation method is consistent with the actual leakage of the hydraulic motor mentioned above. The methods for obtaining the remaining data are as follows:

[0115] The formula for calculating the historical leakage rate is: Where V represents the historical leakage rate, ΔQ Leak The difference in actual leakage represents the value of the actual leakage between the two collection times, and ΔT represents the time difference.

[0116] The formula for calculating the historical pressure decay gradient is as follows: Where B represents the historical pressure decay gradient, ΔP represents the pressure difference, and ΔT represents the time difference;

[0117] By fitting the long-term trend line of historical kinematic viscosity data using the least squares method, a linear equation is constructed. Calculate the residual of historical kinematic viscosity changes, where y represents the fitted historical kinematic viscosity data, a is the slope, x is the index value of the historical kinematic viscosity data after being sorted by timestamp, and b is the intercept, representing the initial fitted historical kinematic viscosity data. Calculate the mean of x and y based on all currently available historical kinematic viscosity data. and The slope 'a' of the equation is calculated using two mean data points, and the specific formula is as follows: Where xi represents the i-th index value, and yi represents the i-th historical kinematic viscosity data. This represents the mean of all index values. This represents the mean of all historical kinematic viscosity data. Based on the obtained... , The intercept b of the linear equation is derived from data a. Based on the constructed linear equation, the fitted historical kinematic viscosity data can be calculated, and the historical kinematic viscosity change residual C can be calculated accordingly, specifically expressed as C=η-y, where η is the original historical kinematic viscosity data and y is the fitted historical kinematic viscosity data.

[0118] The formula for calculating the historical temperature anomaly index is as follows: T anomaly T represents the total number of temperature anomalies, where i is the index value of the historical temperature data after sorting by timestamp, n is the total number of historical temperature data, and T is the index value of the historical temperature data. i T represents the temperature value corresponding to each sampling time. threshold This represents a preset safe temperature threshold, such as 90℃, I(T) i >T threshold The function is an exponential function used to determine whether the current temperature exceeds the threshold. If it does, the value is 1; otherwise, it is 0.

[0119] The historical leakage state characteristic group is formed by integrating three types of data: historical actual leakage volume, historical leakage rate, and historical pressure decay gradient. The historical kinematic viscosity change residual and historical temperature anomaly index are formed as the historical oil state characteristic group.

[0120] By analyzing historical leakage state characteristic groups, internal and external leakage diagnoses of hydraulic motors are performed, and the causes of leakage failures are statistically analyzed. When Q... leak A sustained flow rate greater than 5 L / min, with minimal fluctuations in volume (V) and a slow decrease in volume (B), indicates internal leakage. Possible causes include piston cylinder wear, distributor plate scratches, and fatigue deformation of the slipper. When Q... leakA sudden increase, with a peak instantaneous V greater than 0.1 L / s and a rapid decrease in B, indicates an external leak. Specific causes include loose pipe joints, aging or damaged O-rings, etc. All causes of internal and external leaks are integrated into a set of internal and external leak contributing factors.

[0121] By analyzing the historical hydraulic fluid state characteristic set, hydraulic fluid degradation is diagnosed, and all causes of degradation are statistically analyzed. The residuals of historical kinematic viscosity changes and historical temperature anomaly indices are analyzed to determine the type of hydraulic fluid failure. When the residual is a persistent positive residual and T... anomaly When the value is greater than 10, it is determined that the hydraulic oil has undergone oxidation and deterioration. In this case, oxidation products increase the oil viscosity, leading to an increase in the actual viscosity. When the residual exhibits high-frequency, drastic fluctuations and T... anomaly A value less than 3 indicates a water contamination fault in the hydraulic oil. In this case, water causes localized temperature measurement distortion, leading to unstable theoretical calculations. When the residual exhibits periodic negative spikes, it indicates a particulate contamination fault. In this case, particles accumulate, clogging the sensor and resulting in lower temperature measurements. When the residual trend changes from positive to negative, the additive is depleted or ineffective, altering the viscosity-temperature characteristics and causing the theoretical model to gradually fail. All causes of oil degradation are integrated into a set of oil degradation contributing factors.

[0122] A Bayesian network model for hydraulic motor fault diagnosis is constructed based on historical leakage state characteristic groups, historical oil state characteristic groups, internal and external leakage cause sets, and oil deterioration cause sets.

[0123] Historical actual leakage, historical leakage rate, historical pressure decay gradient, historical kinematic viscosity change residual, and historical temperature anomaly index are set as observed variables (leaf nodes). Internal leakage, external leakage, and oil deterioration are intermediate variables. Piston cylinder wear, distributor plate scratches, slipper fatigue deformation, pipeline joint loosening, O-ring aging or breakage, oxidation deterioration, moisture contamination, and additive depletion or failure are implicit variables (root nodes).

[0124] Based on the previously determined causal relationships between each observed variable and intermediate and latent variables, a directed acyclic graph is constructed, for example, Q. leak Continuous flow rate greater than 5L / min → internal leakage → piston cylinder wear, Q leak Sustained flow exceeding 5 L / min → internal leakage → distributor plate scratches. When constructing the directed acyclic graph (DAG), the parent node of each node should contain all its direct causes, while ensuring there are no cycles in the graph. Based on historical maintenance and inspection data, a conditional probability table is established, defining the conditional probability of each node under its parent node's state. For parent node combinations not appearing in the data, Laplace smoothing is used to avoid zero-probability issues, assuming each conditional probability occurs at least once. A Bayesian network model is constructed using the DAG and the conditional probability table.

[0125] Example 2: Based on the same inventive concept as the hydraulic motor operating status detection method in the foregoing examples, this application also provides a hydraulic motor operating status detection device. Please refer to the appendix. Figure 2 The device includes:

[0126] The multidimensional monitoring dataset acquisition module 11 is used to deploy a sensor cluster at the key nodes of the hydraulic motor, acquire multidimensional state parameters of the hydraulic motor in real time, preprocess the multidimensional state parameters of the hydraulic motor, and construct a multidimensional monitoring dataset, which includes flow data, speed data, pressure data, and temperature data.

[0127] The full-domain monitoring dataset construction module 12 is used to calculate the displacement data of the hydraulic motor and the kinematic viscosity data of the hydraulic oil based on the multi-dimensional monitoring dataset to obtain the full-domain monitoring dataset.

[0128] The status assessment and instruction generation module 13 is used to calculate the actual leakage of the hydraulic motor based on the full-domain monitoring dataset, assess the status of the hydraulic motor based on the actual leakage, and generate operating mode switching instructions. These operating mode switching instructions include continuous operation instructions, load monitoring instructions, and shutdown maintenance instructions. The assessment of the hydraulic motor status and generation of operating mode switching instructions further includes:

[0129] The actual leakage of the hydraulic motor is calculated based on the aforementioned full-domain monitoring dataset, using the following formula:

[0130] ;

[0131] Among them, Q L X represents the actual leakage of the hydraulic motor, X represents the leakage coefficient with a specific value of 0.36, Q represents the displacement data of the hydraulic motor, P represents the pressure data of the hydraulic motor, and η represents the kinematic viscosity data of the hydraulic oil.

[0132] The actual leakage of the hydraulic motor is compared with the leakage baseline to establish a three-level evaluation system. The leakage baseline includes a mild leakage baseline M and a severe leakage baseline N, wherein the mild leakage baseline M is 3%Q / 100bar and the severe leakage baseline N is 5%Q / 100bar.

[0133] When 0 L If the leakage is ≤N, the hydraulic motor is determined to be in a first-level health condition, and a continuous operation command is generated. The current leakage is within the allowable fluctuation range, the hydraulic motor is in good condition, and the regular testing cycle continues.

[0134] When N L ​​If the value is ≤M, the hydraulic motor is determined to be in a level 2 warning state, and a load monitoring command is generated. At this time, the leakage of the hydraulic motor exceeds the safety margin and needs to be reduced to 85% of the rated power for operation. Simultaneously, multi-mode fault diagnosis is initiated.

[0135] When Q L >M indicates that the hydraulic motor is in a level 3 fault state and generates a shutdown and maintenance command. At this time, the hydraulic motor is leaking severely and needs to be stopped immediately for multi-modal fault diagnosis.

[0136] Fault diagnosis module 14 is used to receive the load monitoring command or shutdown maintenance command, start multimodal fault diagnosis, and use Bayesian network model to obtain the fault cause set of hydraulic motor.

[0137] The maintenance plan determination module 15 is used to determine a maintenance plan based on the set of causes of hydraulic motor failure. The maintenance plan includes fault location information, spare parts list, process route diagram and estimated maintenance time.

[0138] Based on the same inventive concept, the embodiments also provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of a hydraulic motor operating state detection method as described in the above embodiments.

[0139] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0140] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for detecting the working status of a hydraulic motor, characterized in that, The method includes: A sensor cluster is deployed at key nodes of the hydraulic motor to acquire multi-dimensional state parameters of the hydraulic motor in real time. The multi-dimensional state parameters of the hydraulic motor are preprocessed to construct a multi-dimensional monitoring dataset, which includes flow data, speed data, pressure data, and temperature data. Based on the multidimensional monitoring dataset, the displacement data of the hydraulic motor and the kinematic viscosity data of the hydraulic oil are calculated to obtain the full-domain monitoring dataset; The actual leakage of the hydraulic motor is calculated based on the comprehensive monitoring dataset. The hydraulic motor status is assessed based on the actual leakage, and a mode switching command is generated. This mode switching command includes a continuous operation command, a load monitoring command, and a shutdown / maintenance command. Assessing the hydraulic motor status and generating the mode switching command further includes: The actual leakage of the hydraulic motor is calculated based on the aforementioned full-domain monitoring dataset, using the following formula: ； Among them, Q L X represents the actual leakage of the hydraulic motor, X represents the leakage coefficient with a specific value of 0.36, Q represents the displacement data of the hydraulic motor, P represents the pressure data of the hydraulic motor, and η represents the kinematic viscosity data of the hydraulic oil. The actual leakage of the hydraulic motor is compared with the leakage baseline to establish a three-level evaluation system. The leakage baseline includes a mild leakage baseline M and a severe leakage baseline N, wherein the mild leakage baseline M is 3%Q / 100bar and the severe leakage baseline N is 5%Q / 100bar. When 0 L If the leakage is ≤N, the hydraulic motor is determined to be in a first-level health condition, and a continuous operation command is generated. The current leakage is within the allowable fluctuation range, the hydraulic motor is in good condition, and the regular testing cycle continues.​ When N L If the value is ≤M, the hydraulic motor is determined to be in a level 2 warning state, and a load monitoring command is generated. At this time, the leakage of the hydraulic motor exceeds the safety margin and needs to be reduced to 85% of the rated power for operation. Simultaneously, multi-mode fault diagnosis is initiated.​ When Q L >M indicates that the hydraulic motor is in a level 3 fault state and generates a shutdown and maintenance command. At this time, the hydraulic motor is leaking severely and needs to be stopped immediately for multi-modal fault diagnosis. Upon receiving the load monitoring command or shutdown maintenance command, initiate multimodal fault diagnosis and use a Bayesian network model to obtain the set of causes of hydraulic motor faults; A repair plan is determined based on the set of causes of hydraulic motor failure. The repair plan includes fault location information, a spare parts list, a process route diagram, and an estimated repair time.

2. The method for detecting the working status of a hydraulic motor as described in claim 1, characterized in that, Construct a multidimensional monitoring dataset, including: The inlet and outlet pipelines, shaft end, and housing drain port of the hydraulic motor are selected as key nodes for deploying a sensor cluster, which includes a flow sensor, a speed sensor, a pressure sensor, and a temperature sensor. The sensor cluster continuously collects flow rate, speed, pressure, and temperature data of the hydraulic motor during operation to obtain multi-dimensional state parameters of the hydraulic motor. The IEEE 1588 precision time protocol is used to ensure the synchronous capture of the multi-dimensional state parameters of the hydraulic motor. The isolated forest algorithm was used to remove outliers from the multidimensional state parameters of the hydraulic motor, and cubic spline interpolation was used to fill in the missing data segments, reconstruct the continuous signal, and build a multidimensional monitoring dataset.

3. The method for detecting the working status of a hydraulic motor as described in claim 1, characterized in that, Obtain the full-domain monitoring dataset, including: The displacement data of the hydraulic motor is obtained based on the flow rate data and speed data in the multidimensional monitoring dataset; A hydraulic oil viscosity-temperature relationship calculation model is set according to the type of hydraulic oil, and the specific formula is as follows: ISO VG32: η=144-3.2T; ISO VG48: η=107.5-1.25T; ISO VG68: η=159-2.4T; Among them, ISO VG32, ISO VG48, and ISO VG68 are three commonly used hydraulic oil categories for hydraulic motors. η represents the kinematic viscosity of the hydraulic oil, and T represents the temperature data. Input the temperature data from the multidimensional monitoring dataset to obtain the kinematic viscosity data of the corresponding hydraulic oil; By integrating the displacement data, kinematic viscosity data, and multidimensional monitoring dataset, a comprehensive monitoring dataset is constructed.

4. The method for detecting the working status of a hydraulic motor as described in claim 1, characterized in that, Upon receiving the load monitoring command or shutdown maintenance command, initiate multimodal fault diagnosis, and utilize a Bayesian network model to obtain the set of causes of hydraulic motor faults, including: Obtain the full-domain monitoring dataset, divide the full-domain monitoring dataset, and regard the data collected at the last 10 time points as real-time monitoring data, and the remaining data as historical monitoring data; Based on the historical monitoring data, a Bayesian network model is constructed; Based on the real-time monitoring data, a real-time status feature group is obtained, which includes the real-time actual leakage amount, real-time leakage rate, real-time pressure decay gradient, real-time kinematic viscosity change residual, and real-time temperature anomaly index. Input the instantaneous state feature set into the Bayesian network model to obtain the set of causes of hydraulic motor failure.

5. The method for detecting the working status of a hydraulic motor as described in claim 4, characterized in that, Constructing a Bayesian network model includes: Multimodal data feature extraction is performed on the historical monitoring data to obtain historical leakage state feature group and historical oil state feature group. The historical leakage state feature group includes historical actual leakage amount, historical leakage rate, and historical pressure decay gradient. The historical oil state feature group includes historical kinematic viscosity change residual and historical temperature anomaly index. By analyzing the historical leakage state characteristic group, internal and external leakage diagnosis of the hydraulic motor is performed to obtain the set of internal and external leakage causes. By analyzing the historical oil condition characteristic group, hydraulic oil deterioration diagnosis is performed to obtain the set of oil deterioration causes. Using the historical leakage state characteristic group and the historical oil state characteristic group as observed variables, and the latent variables of the internal and external leakage cause set and the oil deterioration cause set, a Bayesian network model is constructed.

6. A hydraulic motor operating status detection device, characterized in that, The device is used to implement the hydraulic motor operating status detection method according to any one of claims 1-5, the device comprising: A multidimensional monitoring dataset acquisition module is used to deploy a sensor cluster at key nodes of the hydraulic motor, acquire multidimensional state parameters of the hydraulic motor in real time, preprocess the multidimensional state parameters of the hydraulic motor, and construct a multidimensional monitoring dataset, which includes flow data, speed data, pressure data, and temperature data. A full-domain monitoring dataset construction module is used to calculate the displacement data of the hydraulic motor and the kinematic viscosity data of the hydraulic oil based on the multi-dimensional monitoring dataset to obtain a full-domain monitoring dataset. A status assessment and instruction generation module is used to calculate the actual leakage of the hydraulic motor based on the full-domain monitoring dataset, assess the status of the hydraulic motor based on the actual leakage, and generate operating mode switching instructions. These operating mode switching instructions include continuous operation instructions, load monitoring instructions, and shutdown maintenance instructions. The process of assessing the hydraulic motor status and generating operating mode switching instructions further includes: The actual leakage of the hydraulic motor is calculated based on the aforementioned full-domain monitoring dataset, using the following formula: ； Among them, Q L X represents the actual leakage of the hydraulic motor, X represents the leakage coefficient with a specific value of 0.36, Q represents the displacement data of the hydraulic motor, P represents the pressure data of the hydraulic motor, and η represents the kinematic viscosity data of the hydraulic oil. The actual leakage of the hydraulic motor is compared with the leakage baseline to establish a three-level evaluation system. The leakage baseline includes a mild leakage baseline M and a severe leakage baseline N, wherein the mild leakage baseline M is 3%Q / 100bar and the severe leakage baseline N is 5%Q / 100bar. When 0 L If the leakage is ≤N, the hydraulic motor is determined to be in a first-level health condition, and a continuous operation command is generated. The current leakage is within the allowable fluctuation range, the hydraulic motor is in good condition, and the regular testing cycle continues.​ When N L If the value is ≤M, the hydraulic motor is determined to be in a level 2 warning state, and a load monitoring command is generated. At this time, the leakage of the hydraulic motor exceeds the safety margin and needs to be reduced to 85% of the rated power for operation. Simultaneously, multi-mode fault diagnosis is initiated.​ When Q L >M indicates that the hydraulic motor is in a level 3 fault state and generates a shutdown and maintenance command. At this time, the hydraulic motor is leaking severely and needs to be stopped immediately for multi-modal fault diagnosis. The fault diagnosis module is used to receive the load monitoring command or shutdown maintenance command, initiate multimodal fault diagnosis, and use a Bayesian network model to obtain the fault cause set of the hydraulic motor. The maintenance plan determination module is used to determine a maintenance plan based on the set of causes of hydraulic motor failure. The maintenance plan includes fault location information, spare parts list, process route diagram and estimated maintenance time.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements a method for detecting the working status of a hydraulic motor as described in any one of claims 1-5.

Images