Predictive maintenance method and system for bridge crane

Abstract

The invention belongs to the technical field of safety maintenance, and particularly relates to a predictive maintenance method and system for a bridge crane. The invention discloses a predictive maintenance method for a bridge crane. The method comprises the following steps: monitoring and collecting state data of the bridge crane in real time through a sensor, preprocessing each piece of data, and converting the data into time sequence data; in combination with a risk assessment and fault prediction model, according to historical data and current data, assessing the remaining service life, the fault occurrence probability and the health degree of each piece of equipment; performing quantitative evaluation on the monitoring result of the equipment by using a multi-parameter weighted scoring principle according to the state judgment evaluation model, and dividing the health state of the equipment according to a scoring result; and the evaluation result is displayed on the HMI in a chart form, and corresponding maintenance measures are sent to maintenance personnel for optimization according to the result. The operation state of the bridge crane is effectively monitored and evaluated, equipment faults are predicted and prevented, and the reliability and safety of the equipment are improved.

Description

Technical Field

[0001] This invention belongs to the field of safety maintenance technology, specifically relating to a predictive maintenance method and system for bridge cranes. Background Technology

[0002] Existing bridge crane monitoring systems fail to provide adequate "health" monitoring for equipment. For hydropower plants, this requires a significant investment of time and effort in operation and maintenance. Therefore, designing a preventative maintenance system and applying it to bridge cranes in hydropower plants can greatly improve production and maintenance efficiency, making this project imperative.

[0003] Predictive maintenance systems can not only improve the reliability and stability of equipment, but also reduce maintenance costs and downtime. This is because predictive maintenance systems can develop personalized maintenance plans based on the actual operating conditions of the equipment, making maintenance more precise and efficient.

[0004] The application of preventative maintenance systems in hydropower plant bridge cranes also features intelligent characteristics. It can achieve remote monitoring and management of equipment through connections with other systems. Where network access permits, you can check the equipment's operating status and perform remote monitoring anytime, anywhere, as long as you have a network connection. This not only improves work flexibility and efficiency but also reduces the waste of human resources.

[0005] Predictive maintenance systems are an inevitable trend in future industrial development. Their advantages include improved equipment reliability and stability, reduced maintenance costs, and shorter downtime. They will help achieve intelligent management of gantry cranes, enhance maintenance efficiency, and improve production productivity. This will enable intelligent monitoring and diagnosis of gantry crane faults, facilitating the rapid identification of the causes of equipment anomalies, improving the efficiency of fault handling, enhancing the reliability of safe equipment operation, and ensuring the automated and intelligent operation of gantry cranes. This will continuously move towards the goal of fully realizing intelligent and green management of hydropower station operation and maintenance systems. Summary of the Invention

[0006] In view of the above-mentioned problems, the present invention is proposed.

[0007] Therefore, the technical problem solved by this invention is: how to effectively monitor and evaluate the operating status of bridge cranes, predict and prevent equipment failures, and improve the reliability and safety of the equipment.

[0008] To address the aforementioned technical problems, this invention provides the following technical solution: Real-time monitoring and collection of bridge crane status data via sensors; preprocessing each data point to convert it into time-series data; combining risk assessment and fault prediction models, evaluating the remaining service life, probability of failure, and health status of each piece of equipment based on historical and current data; quantitatively evaluating the monitoring results of the equipment using a multi-parameter weighted scoring principle based on the status discrimination assessment model, and classifying the health status of the equipment according to the scoring results; displaying the assessment results in chart form on the HMI, and sending corresponding maintenance measures to maintenance personnel for optimization based on the results.

[0009] As a preferred embodiment of the predictive maintenance method for a bridge crane described in this invention, the sensors include a three-axis vibration sensor for the motor, an infrared high-precision temperature sensor for the motor and reducer, and a switching wear sensor for the motor brake.

[0010] The time-series data includes timestamps, sensor readings, and device ID numbers, represented as follows:

[0011]

[0012]

[0013] Where n is the number of devices, m is the number of sensors, and X ij y(t) is the reading of the j-th sensor of the i-th device, Δt is the sampling time interval, and y(t) is the converted vector containing the timestamp and device ID.

[0014] As a preferred embodiment of the predictive maintenance method for bridge cranes described in this invention, the real-time monitoring includes displaying vibration data of each motor device along the x, y, and z axes on the vibration sensor interface. Each motor has its matching alarm value. When the data from the middle x, y, and z axis sensors exceeds a threshold, the corresponding red alarm light on the right illuminates. When there is data that does not exceed the threshold, the corresponding green light below illuminates. When an alarm light illuminates, the corresponding alarm light on the HMI main interface also illuminates simultaneously.

[0015] The vibration alarm value is calculated based on historical operating data and is reset according to the parameters provided by the manufacturer when the equipment is replaced.

[0016] The real-time monitoring also includes an interface for temperature and wear sensors. Alarm value 1 in the upper left corner is the first alarm threshold, and alarm value 2 in the upper right corner is the second threshold. When the temperature of a motor exceeds the first threshold, the corresponding yellow-green alarm light in the middle lights up. When the temperature exceeds the second threshold, the red and yellow alarm lights are lit up at the same time. The middle of the interface is a real-time curve table of temperature data for each motor.

[0017] When the bridge crane starts running, the data is updated every 0.1 seconds;

[0018] The alarm light array on the right side of the temperature and wear sensor interface is a status display for the wear switch. When the wear is normal, the yellow-green light is on. When the brake shoe is worn too much and needs to be replaced, the yellow-green light goes out and the red light comes on.

[0019] As a preferred embodiment of the predictive maintenance method for bridge cranes described in this invention, the risk assessment and failure prediction model includes using a regression method to assess the remaining useful life (RUL) before the next failure occurs, using a classification method to assess the probability of equipment failure in the next 30 cycles, and using an averaging method to calculate equipment health, expressed as:

[0020]

[0021]

[0022]

[0023] Where p(z=1|X) ij (t) represents the probability of equipment failure, β0 and β1 are the parameters of the logistic regression model, α is the threshold for the failure probability, RUL is the remaining useful life of the equipment, z is the predicted failure result of the equipment in the next n steps, and K(X) is the probability of equipment failure in the next n steps. ij (t),X i ) is the Gaussian kernel function, α i and y i is the Lagrange multiplier and class label of the i-th sample, estimated using the SMO algorithm, and b is the intercept of the hyperplane, calculated using the properties of support vectors.

[0024] As a preferred embodiment of the predictive maintenance method for bridge cranes described in this invention, the state discrimination and health assessment model includes a model based on the probability p(z=1|X) of equipment failure. ij (t)), the remaining service life of the equipment RUL, and the wear condition of the equipment W. i The equipment is divided into three states: normal, low-level alarm, and high-level alarm.

[0025] If the probability of equipment failure is less than L 1p The remaining service life of the equipment is greater than L 1R If the wear and tear of the equipment is zero, the equipment is considered to be in normal condition; if the probability of equipment failure is L... 1p and L 2p Between these points, the remaining service life of the equipment is in the range of L. 1R and L 2RIf the probability of equipment failure is between L and L, and the wear level of the equipment is equal to 1, the equipment is classified as a low-level alarm; if the probability of equipment failure is greater than or equal to L... 2p The remaining service life of the equipment is less than or equal to L. 2R If the wear and tear of the equipment is equal to 2, the equipment status is determined to be a high-level alarm.

[0026] Among them, W i This is the data from the wear sensor of the i-th device;

[0027] As a preferred embodiment of the predictive maintenance method for bridge cranes described in this invention, the state discrimination and health assessment model further includes a health assessment expressed as:

[0028]

[0029] Where D is the time series dataset, n is the number of devices, and k i Let p(z=1|X) be the weight of the i-th device. ij (t) represents the probability of equipment failure, RUL represents the remaining useful life of the equipment, and L 1p and L 2p L represents the low-level and high-level alarm thresholds for the probability of equipment failure. 1R and L 2R Low-level and high-level alarm thresholds for the remaining service life of the equipment;

[0030] Based on the health assessment value, the health status is divided into healthy, early sub-health, mid-sub-health, late sub-health, and fault. If the health assessment value is equal to 100, the equipment's health status is healthy, indicating that the equipment is in optimal operating condition and does not require immediate maintenance. If the health assessment value is between 80 and 100, the equipment's health status is early sub-health, and the frequency of safety checks should be increased. If the health assessment value is between 60 and 80, the equipment's health status is mid-sub-health, and the monitoring frequency should be increased, with light maintenance performed. If the health assessment value is between 40 and 60, the equipment's health status is late sub-health, and immediate maintenance should be performed. If the health assessment value is between 0 and 40, the equipment's health status is faulty, a shutdown alarm should be broadcast, and emergency replacement should be performed.

[0031] Another objective of this invention is to provide a system for predictive maintenance of bridge cranes, which can effectively monitor and evaluate the operating status of bridge cranes through risk assessment and fault prediction models and condition discrimination assessment models, predict and prevent equipment failures, and improve the reliability and safety of the equipment.

[0032] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a predictive maintenance system for bridge cranes, comprising a sensor module, a data processing module, a risk assessment module, a status discrimination module, an HMI, and an optimization module;

[0033] The sensor module is used to monitor and collect the status data of the bridge crane in real time, including motor vibration, temperature, and brake wear.

[0034] The data processing module is used to preprocess the collected state data and convert it into time series data for subsequent analysis and evaluation.

[0035] The risk assessment module is used to combine risk assessment and fault prediction models to assess the remaining service life, probability of failure, and health of each device based on historical and current data.

[0036] The status discrimination module is used to quantitatively evaluate the monitoring results of the equipment according to the status discrimination evaluation model and the multi-parameter weighted scoring principle, and to classify the health status of the equipment according to the scoring results.

[0037] The HMI is used to display the assessment results in chart form. The data includes the equipment's remaining service life, probability of failure, health status, health condition, vibration, temperature, and wear.

[0038] The optimization module is used to send corresponding maintenance measures to maintenance personnel for optimization based on the evaluation results, including maintenance plans, maintenance suggestions, fault diagnosis, and fault troubleshooting.

[0039] The HMI includes the HMI main interface, the predictive maintenance system main interface, the vibration sensor interface, the temperature and wear sensor interface, the alarm chart interface, the historical alarm chart interface, the trend chart interface, the analysis report fault judgment interface, the fault prediction data calculation interface, and the maintenance plan interface.

[0040] The HMI main interface is used to display system information and status, including system startup ID, alarm indicator light, double beam monitoring system button and PDM system button;

[0041] The main interface of the predictive maintenance system is used to display and control the functions of the predictive maintenance system, including vibration sensor button, temperature and wear sensor button, alarm chart button, historical alarm chart button, trend chart button, analysis report fault diagnosis button, fault prediction data calculation button, and maintenance measures button.

[0042] The vibration sensor interface is used to display real-time data and graphs from 4 groups of 12 vibration sensors, as well as alarm values ​​and alarm lights.

[0043] The temperature and wear sensor interface is used to display real-time data and graphs from four temperature sensors and four wear sensors, as well as alarm values ​​and alarm lights.

[0044] The alarm chart interface is used to display a real-time list of any device and any type of alarm information, as well as the cause of the device failure;

[0045] The historical alarm chart interface is used to display a compilation and statistics of historical data, providing data support for subsequent calculations and evaluations;

[0046] The trend chart interface is used to display the data trends of all analog sensor groups. By defining time intervals or zooming in and out of the time axis, the data trends of different time periods can be displayed, providing numerical and chart basis for manual intervention and judgment.

[0047] The analysis report fault judgment interface is used to display most fault phenomena, fault causes and handling methods, and has been coded to provide a coding reference for the fault handling results in the maintenance solution interface.

[0048] The fault prediction data calculation interface is used to display predictive results obtained through three different calculation methods, including averaging, regression and classification, to provide data reference for maintenance personnel.

[0049] A computer device includes a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program to implement the steps of a predictive maintenance method for a bridge crane.

[0050] A computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of a predictive maintenance method for a bridge crane.

[0051] The beneficial effects of this invention are: it enables real-time monitoring and analysis of the status data of bridge cranes, improving the operating efficiency and quality of the equipment; it enables accurate assessment of the remaining service life, failure probability, and health status of bridge cranes, improving the reliability and safety of the equipment; it enables quantitative judgment and visual display of the health status of bridge cranes, improving maintenance effectiveness; and it enables optimization and guidance of maintenance measures for bridge cranes, improving equipment maintenance efficiency and cost savings. Attached Figure Description

[0052] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0053] Figure 1 A flowchart of a predictive maintenance method for a bridge crane provided in one embodiment of the present invention.

[0054] Figure 2 The flowchart below illustrates the algorithm evaluation process for a predictive maintenance method for a bridge crane, as provided in one embodiment of the present invention.

[0055] Figure 3 This is a diagram of the main interface of a predictive maintenance system for a bridge crane, provided as an embodiment of the present invention.

[0056] Figure 4 A vibration sensor interface for a predictive maintenance method for a bridge crane, as provided in one embodiment of the present invention.

[0057] Figure 5 A temperature wear sensor interface is provided for a predictive maintenance method for a bridge crane according to one embodiment of the present invention.

[0058] Figure 6 This is an analysis report fault judgment interface for a predictive maintenance method for bridge cranes provided in one embodiment of the present invention.

[0059] Figure 7 This is a structural diagram of a predictive maintenance system for a bridge crane provided in one embodiment of the present invention. Detailed Implementation

[0060] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0061] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0062] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0063] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0064] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0065] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0066] Example 1

[0067] Reference Figures 1-6 As an embodiment of the present invention, a predictive maintenance method for a bridge crane is provided, comprising:

[0068] S1: Real-time monitoring and collection of bridge crane status data through sensors, and preprocessing of each data point to convert it into time series data;

[0069] Furthermore, the sensors include a three-axis vibration sensor for the motor, an infrared high-precision temperature sensor for the motor and reducer, and a switching wear sensor for the motor brake.

[0070] It should be noted that the motor triaxial vibration sensor is used to measure the vibration amplitude and frequency of the motor; the high-precision infrared temperature sensor for the motor and reducer is used to measure the surface temperature of the motor and reducer; and the switch-type wear sensor for the motor brake is used to detect the wear condition of the motor brake.

[0071] The time-series data includes timestamps, sensor readings, and device ID numbers, represented as follows:

[0072]

[0073]

[0074] Where n is the number of devices, m is the number of sensors, and X ij y(t) is the reading of the j-th sensor of the i-th device, Δt is the sampling time interval, and y(t) is the converted vector containing the timestamp and device ID.

[0075] Furthermore, the real-time monitoring includes displaying vibration data for each motor device along the x, y, and z axes on the vibration sensor interface. Each motor has its matching alarm value. When the data from the middle x, y, and z axis sensors exceeds the threshold, the corresponding red alarm light on the right illuminates. When there is data that does not exceed the threshold, the corresponding green light below illuminates. When an alarm light illuminates, the corresponding alarm light on the HMI main interface also illuminates simultaneously.

[0076] The vibration alarm value is calculated based on historical operating data and is reset according to the parameters provided by the manufacturer when the equipment is replaced.

[0077] The real-time monitoring also includes an interface for temperature and wear sensors. Alarm value 1 in the upper left corner is the first alarm threshold, and alarm value 2 in the upper right corner is the second threshold. When the temperature of a motor exceeds the first threshold, the corresponding yellow-green alarm light in the middle lights up. When the temperature exceeds the second threshold, the red and yellow alarm lights are lit up at the same time. The middle of the interface is a real-time curve table of temperature data for each motor.

[0078] When the bridge crane starts running, the data is updated every 0.1 seconds;

[0079] It should be noted that in the predictive maintenance system for bridge cranes, the choice to update data every 0.1 seconds is based on a comprehensive consideration of the balance between equipment characteristics, technical capabilities, real-time monitoring needs, and cost-effectiveness.

[0080] Bridge cranes encounter rapidly changing loads and environmental conditions during operation, requiring data acquisition systems to react quickly to capture these changes. Faults and performance degradation often develop gradually. Updating data every 0.1 seconds provides sufficient information density to help detect and diagnose these minute changes in a timely manner.

[0081] The data update frequency is limited by the data acquisition capabilities of the device and the data processing capabilities of the backend system. 0.1 seconds is a frequency that is achievable and effective under current technological conditions.

[0082] Too frequent data updates (e.g., every 0.01 seconds) can lead to a dramatic increase in data volume, posing challenges to storage and analysis, while every 0.1 seconds provides a manageable amount of data.

[0083] Higher data collection frequency means greater consumption of hardware and software resources, as well as higher operational costs. A frequency of every 0.1 seconds is an economical and effective choice. Cost control needs to be considered while ensuring monitoring quality. Updating data every 0.1 seconds represents the optimal balance between cost and benefit.

[0084] It should also be noted that this frequency is sufficient to ensure real-time monitoring of the crane's status, allowing operators and maintenance systems to respond promptly to any potential problems. Overly frequent data updates could lead to information overload, impacting the decision-making efficiency of operators and the system. An update frequency of 0.1 seconds ensures the operability and usability of the data.

[0085] In summary, updating the data every 0.1 seconds is a result of considerations regarding the operating environment, technical capabilities, real-time monitoring needs, and cost-effectiveness of the bridge crane. This frequency provides sufficient data density for monitoring and predicting equipment status while avoiding the technical and economic burden of excessively frequent data collection. It is a practical choice that balances real-time monitoring needs with limitations in technical capabilities.

[0086] The alarm light array on the right side of the temperature and wear sensor interface is a status display for the wear switch. When the wear is normal, the yellow-green light is on. When the brake shoe is worn too much and needs to be replaced, the yellow-green light goes out and the red light comes on.

[0087] S2: Combining risk assessment and failure prediction models, based on historical and current data, assess the remaining service life, failure probability, and health of each device;

[0088] Furthermore, the risk assessment and failure prediction model includes using regression methods to assess the remaining useful life (RUL) before the next failure occurs, using classification methods to assess the probability of failure in the next n steps, the probability of a device failing in the next 30 cycles, and using an averaging method to calculate the device health, expressed as:

[0089]

[0090]

[0091]

[0092] Where p(z=1|X) ij(t) represents the probability of equipment failure, β0 and β1 are the parameters of the logistic regression model, α is the threshold for the failure probability, RUL is the remaining useful life of the equipment, z is the predicted failure result of the equipment in the next n steps, and K(X) is the probability of equipment failure in the next n steps. ij (t),X i ) is the Gaussian kernel function, α i and y i is the Lagrange multiplier and class label of the i-th sample, estimated using the SMO algorithm, and b is the intercept of the hyperplane, calculated using the properties of support vectors.

[0093] It should be noted that when it is necessary to change the algorithm, the algorithm needs to be evaluated. The specific algorithm evaluation process is as follows:

[0094] Test Preparation Phase: The test preparation phase includes activities such as application, test type selection, and database updates. Based on the test application submitted by the maintenance department, the test type is first determined, and it is judged whether model training is required. If training is required, it is judged whether the client has its own data. If the client has its own data, the data requirements are judged. If no data is provided, sample sampling is performed. If the client's own data meets the data requirements, it is updated into the database and sample sampling is performed again. If it does not meet the requirements, updated data is allowed for a second judgment. Qualified data is updated into the database. If no training is required, the test phase begins.

[0095] Algorithm Testing Phase: The algorithm testing phase includes activities such as sampling, determining whether to train, deciding whether to disclose the algorithm, interface development, and algorithm evaluation. Algorithms requiring training are sampled and trained; then, a decision is made based on customer needs regarding whether they can be disclosed. Algorithms not requiring training are directly assessed for disclosure. If disclosure is permitted, white-box testing begins, where the algorithm is run and evaluated. If disclosure is not permitted, black-box testing begins, where an interface is developed to integrate the algorithm.

[0096] Algorithm debugging phase: This phase includes activities such as determining whether to debug, updating the algorithm, and issuing reports. If the evaluation results do not meet the maintenance personnel's satisfaction, algorithm debugging and updates are permitted, and the evaluation will be conducted again. The number of updates will not exceed two. Once the satisfaction level is achieved, the model solution will be finalized.

[0097] S3: Based on the status discrimination assessment model, use the multi-parameter weighted scoring principle to quantitatively evaluate the monitoring results of the equipment, and classify the health status of the equipment according to the scoring results;

[0098] Furthermore, the state discrimination and health assessment model includes a probability p(z = 1|X) of equipment failure. ij (t)), the remaining service life of the equipment RUL, and the wear condition of the equipment W. i The equipment is divided into three states: normal, low-level alarm, and high-level alarm.

[0099] If the probability of equipment failure is less than L 1p The remaining service life of the equipment is greater than L 1R If the wear and tear on the equipment is zero, the equipment is considered to be in normal condition; if the probability of equipment failure is 1... 1p and L 2p Between these points, the remaining service life of the equipment is in the range of L. 1R and L 2R If the probability of equipment failure is between L and L, and the wear level of the equipment is equal to 1, the equipment is classified as a low-level alarm; if the probability of equipment failure is greater than or equal to L... 2p The remaining service life of the equipment is less than or equal to L. 2R Furthermore, the wear and tear of the equipment is equal to 2, and the equipment is determined to be in a high-level alarm state.

[0100] Furthermore, the state discrimination and health assessment model also includes a health assessment expressed as:

[0101]

[0102] Where D is the time series dataset, n is the number of devices, and k i Let p(z=1|X) be the weight of the i-th device. ij (t) represents the probability of equipment failure, RUL represents the remaining useful life of the equipment, and L 1p and L 2p L represents the low-level and high-level alarm thresholds for the probability of equipment failure. 1R and L 2R Low-level and high-level alarm thresholds for the remaining service life of the equipment;

[0103] Based on the health assessment value, the health status is divided into healthy, early sub-health, mid-sub-health, late sub-health, and fault. If the health assessment value is equal to 100, the equipment's health status is healthy, indicating that the equipment is in optimal operating condition and does not require immediate maintenance. If the health assessment value is between 80 and 100, the equipment's health status is early sub-health, and the frequency of safety checks should be increased. If the health assessment value is between 60 and 80, the equipment's health status is mid-sub-health, and the monitoring frequency should be increased, with light maintenance performed. If the health assessment value is between 40 and 60, the equipment's health status is late sub-health, and immediate maintenance should be performed. If the health assessment value is between 0 and 40, the equipment's health status is faulty, a shutdown alarm should be broadcast, and emergency replacement should be performed.

[0104] It should be noted that during maintenance activities, the measures taken and the parts replaced should be recorded. After maintenance, the equipment's health should be reassessed to ensure the effectiveness of the maintenance activities. After the maintenance activities are completed, the equipment's health records and historical maintenance data should be updated. The equipment status should be continuously monitored to ensure that any new or unresolved problems are quickly identified and addressed.

[0105] It should also be noted that a feedback loop is needed after maintenance: maintenance results and experience should be fed back into the predictive maintenance model to continuously optimize its accuracy. If maintenance activities lead to a significant improvement in health scores, this data should be used to improve future health assessments. Throughout the process, all equipment-related data should be continuously monitored and used during routine maintenance and emergencies. Using this approach, maintenance teams can take proactive measures before equipment problems actually occur, thereby saving costs, improving efficiency, and reducing downtime.

[0106] S4: Display the evaluation results in chart form on the HMI, and send the corresponding maintenance measures to the maintenance personnel for optimization based on the results.

[0107] Based on the remaining service life, probability of failure, and health status of the equipment, develop maintenance plans and priorities, allocate maintenance personnel and resources, and prevent failures from occurring.

[0108] Based on the health status of the equipment, provide maintenance suggestions and operation instructions to improve maintenance efficiency and quality and restore the equipment to normal operation;

[0109] Based on the equipment's vibration, temperature, and wear, diagnose the cause and location of the fault, provide troubleshooting and repair methods, and reduce maintenance time and costs;

[0110] The alarm lights on the equipment alert maintenance personnel to any abnormal conditions, prompt them to handle the alarm information promptly, and prevent accidents from occurring.

[0111] Example 2

[0112] The second embodiment of the present invention differs from the first embodiment in that:

[0113] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0114] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0115] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0116] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0117] Example 3

[0118] Reference Figure 7 This is the third embodiment of the present invention, which provides a predictive maintenance system for a bridge crane, including a sensor module, a data processing module, a risk assessment module, a status discrimination module, an HMI, and an optimization module;

[0119] The sensor module is used to monitor and collect the status data of the bridge crane in real time, including motor vibration, temperature, and brake wear.

[0120] The data processing module is used to preprocess the collected state data and convert it into time series data for subsequent analysis and evaluation.

[0121] The risk assessment module is used to combine risk assessment and fault prediction models to assess the remaining service life, probability of failure, and health of each device based on historical and current data.

[0122] The status discrimination module is used to quantitatively evaluate the monitoring results of the equipment according to the status discrimination evaluation model and the multi-parameter weighted scoring principle, and to classify the health status of the equipment according to the scoring results.

[0123] The HMI is used to display the assessment results in chart form. The data includes the equipment's remaining service life, probability of failure, health status, health condition, vibration, temperature, and wear.

[0124] The optimization module is used to send corresponding maintenance measures to maintenance personnel for optimization based on the evaluation results, including maintenance plans, maintenance suggestions, fault diagnosis, and fault troubleshooting.

[0125] The HMI includes the HMI main interface, the predictive maintenance system main interface, the vibration sensor interface, the temperature and wear sensor interface, the alarm chart interface, the historical alarm chart interface, the trend chart interface, the analysis report fault judgment interface, the fault prediction data calculation interface, and the maintenance plan interface.

[0126] The HMI main interface is used to display system information and status, including system startup ID, alarm indicator light, double beam monitoring system button and PDM system button;

[0127] The main interface of the predictive maintenance system is used to display and control the functions of the predictive maintenance system, including vibration sensor button, temperature and wear sensor button, alarm chart button, historical alarm chart button, trend chart button, analysis report fault diagnosis button, fault prediction data calculation button, and maintenance measures button.

[0128] The vibration sensor interface is used to display real-time data and graphs from 4 groups of 12 vibration sensors, as well as alarm values ​​and alarm lights.

[0129] The temperature and wear sensor interface is used to display real-time data and graphs from four temperature sensors and four wear sensors, as well as alarm values ​​and alarm lights.

[0130] The alarm chart interface is used to display a real-time list of any device and any type of alarm information, as well as the cause of the device failure;

[0131] The historical alarm chart interface is used to display a compilation and statistics of historical data, providing data support for subsequent calculations and evaluations;

[0132] The trend chart interface is used to display the data trends of all analog sensor groups. By defining time intervals or zooming in and out of the time axis, the data trends of different time periods can be displayed, providing numerical and chart basis for manual intervention and judgment.

[0133] The analysis report fault judgment interface is used to display most fault phenomena, fault causes and handling methods, and has been coded to provide a coding reference for the fault handling results in the maintenance solution interface.

[0134] The fault prediction data calculation interface is used to display predictive results obtained through three different calculation methods, including averaging, regression and classification, to provide data reference for maintenance personnel.

[0135] Example 4

[0136] The fourth embodiment of the present invention provides a predictive maintenance method for bridge cranes.

[0137] To verify the beneficial effects of the above method, theoretical experiments and practical applications were conducted. Sensors were used to monitor equipment status, regression and classification models were used to predict failure probability and remaining service life, and a multi-parameter weighted scoring principle was used to assess equipment health. The results were compared with traditional maintenance methods. The following is a table of the obtained data:

[0138] Table 1: Equipment Condition Assessment under Predictive Maintenance Methods

[0139]

[0140]

[0141] Table 2: Comparison of Failure Counts between Predictive Maintenance Methods and Traditional Maintenance Methods

[0142] Equipment Number Number of failures under traditional maintenance methods Number of failures under predictive maintenance methods Failure reduction rate 1 2 0 100％ 2 3 1 66.7％ 3 4 2 50％ 4 5 3 40％ 5 6 4 33.3％ 6 7 5 28.6％

[0143] As can be seen from Table 1, predictive maintenance methods can classify equipment into three states—normal, low-level alarm, and high-level alarm—based on the equipment's failure probability, remaining service life, and health status. This allows for proactive maintenance measures to prevent equipment failure or extend its service life.

[0144] As shown in Table 2, predictive maintenance methods can significantly reduce the number of equipment failures. Compared with traditional maintenance methods, the average number of failures is reduced by 44.4%, which means that the reliability and efficiency of the equipment are improved, while maintenance costs and downtime losses are also reduced.

[0145] Table 3: Equipment Condition Assessment under Predictive Maintenance System

[0146] Equipment Number Failure probability Remaining service life Health state 1 0.01 150 98 healthy 2 0.03 100 90 Early stage of sub-health 3 0.06 70 80 Mid-stage sub-health 4 0.10 50 70 Post-subhealth 5 0.15 30 60 Fault

[0147] As shown in Table 3, predictive maintenance systems can classify equipment into five states—healthy, early sub-healthy, mid-sub-healthy, late sub-healthy, and faulty—based on the equipment's failure probability, remaining service life, and health status. This provides more detailed equipment health status information, helps users make more reasonable maintenance decisions, and improves equipment safety, productivity, and lifecycle value.

[0148] In conclusion, predictive maintenance for bridge cranes is an effective method that can improve equipment reliability and efficiency, and reduce maintenance costs and downtime losses.

[0149] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for predictive maintenance of a bridge crane, characterized in that, The application relates to a real-time monitoring and maintenance system for a bridge crane. Real-time monitoring and collection of state data of the bridge crane through sensors, pre-processing of each piece of data into time series data; Combination of risk assessment and fault prediction models, assessment of the remaining service life, failure probability and health degree of each device according to historical data and current data; Quantitative assessment of the monitoring results of the device according to the state discrimination assessment model using the multi-parameter weighted scoring principle, and division of the health state of the device according to the scoring results; Display of the assessment results in the form of charts on the HMI, and sending of corresponding maintenance measures to maintenance personnel for optimization according to the results.

2. A bridge crane predictive maintenance method as claimed in claim 1, characterized in that: The sensors include motor three-axis coordinate vibration sensors, motor and reducer infrared high-precision temperature sensors and motor brake switch quantity abrasion sensors. The time series data includes a time stamp, a sensor reading and a device ID number, and is expressed as: where n is the number of devices, m is the number of sensors, X ij (t) is the reading of the jth sensor of the ith device, Δt is the sampling time interval, and y(t) is the transformed vector containing the time stamp and device ID number.

3. A bridge crane predictive maintenance method as claimed in claim 2, characterized in that: The real-time monitoring includes display of vibration data of x, y and z axes of each motor device on the vibration sensor interface, each motor has a matched alarm value, when the x, y and z axis sensor data in the middle is greater than the threshold value, the corresponding red alarm lamp on the right is turned on, when there is data and the data is not greater than the threshold value, the corresponding green lamp below is turned on, when a certain alarm lamp is turned on, the corresponding alarm lamp on the HMI main interface is also turned on; The vibration alarm value is calculated according to historical operation data, and is reset according to the parameters provided by the manufacturer when the device is replaced; The real-time monitoring also includes the temperature and abrasion sensor interface, the alarm value 1 in the upper left corner is the first alarm threshold value, the alarm value 2 in the upper right corner is the second threshold value, when the temperature of a certain motor exceeds the first threshold value, the corresponding yellow-green alarm lamp in the middle is turned on, and when the temperature exceeds the second threshold value, the red-yellow alarm lamp is turned on; the middle of the interface is a real-time curve table of the temperature data of each motor; When the bridge crane starts to run, the data is updated every 0.1 second; The alarm lamp column on the right side of the temperature and abrasion sensor interface is the state display of the abrasion switch, when the abrasion is normal, the yellow-green lamp is turned on, when the brake shoe is excessively worn and needs to be replaced, the yellow-green lamp is turned off and the red lamp is turned on.

4. A bridge crane predictive maintenance method as claimed in claim 3, characterized in that: The risk assessment and fault prediction model includes using a regression method to assess the remaining useful life RUL before the next failure, using a classification method to assess the failure probability of the device in the next 30 cycles, and using an average method to calculate the health degree of the device, which is expressed as: where p(z = 1 | X ij (t)) is the probability of the device failing, β0and β1are parameters of the logistic regression model, a is the threshold of the failure probability, RUL is the remaining useful life of the device, z is the prediction result of the device failing in the next n steps, K(X ij (t), X i ) is a Gaussian kernel function, a i and y i are the Lagrange multipliers and the class label of the i-th sample, estimated using the SMO algorithm, b is the intercept of the hyperplane, calculated using the properties of the support vectors.

5. A bridge crane predictive maintenance method as claimed in claim 4, characterized in that: The state discrimination and health degree evaluation model includes a probability p(z=1|X ij (t)) that the equipment fails, a remaining useful life RUL of the equipment, and a wear condition W i The equipment is divided into three states: normal, low-level alarm, and high-level alarm. If the probability of failure of the device is less than L 1p , the remaining useful life of the device is greater than L 1R , and the wear of the device is equal to 0, the state of the device is determined to be normal. If the probability of failure of the device is between L 1p and L 2p , the remaining useful life of the device is between L 1R and L 2R , and the wear of the device is equal to 1, the state of the device is determined to be a low-level alarm; if the probability of failure of the device is greater than or equal to L 2p , the remaining useful life of the device is less than or equal to L 2R , and the wear of the device is equal to 2, the state of the device is determined to be a high-level alarm.

6. A bridge crane predictive maintenance method as claimed in claim 5, characterized in that: The state discrimination and health degree assessment model also includes health degree assessment, which is expressed as: wherein D is a time series dataset, n is the number of devices, k i is the weight of the ith device, p(z=1|X ij (t)) is the probability of device failure, RUL is the remaining useful life of the device, L 1p and L 2p are the low and high level alarm thresholds for the probability of device failure, L 1R and L 2R are the low and high level alarm thresholds for the remaining useful life of the device; According to the health degree evaluation value, the health state is divided into health, early sub-health, middle sub-health, late sub-health and failure; if the health degree evaluation value is equal to 100, the health state of the equipment is health, indicating that the equipment is in the best running state and does not need immediate maintenance; if the health degree evaluation value is between 80 and 100, the health state of the equipment is early sub-health, the number of safety checks is increased; if the health degree evaluation value is between 60 and 80, the health state of the equipment is middle sub-health, the monitoring frequency is increased, and light maintenance is performed; if the health degree evaluation value is between 40 and 60, the health state of the equipment is late sub-health, immediate maintenance is performed; if the health degree evaluation value is between 0 and 40, the health state of the equipment is failure, a shutdown warning is broadcast, and emergency replacement operation is performed.

7. A system for predictive maintenance of a bridge crane according to any one of claims 1 to 6, characterized in that: The system comprises a sensor module, a data processing module, a risk assessment module, a state discrimination module, an HMI and an optimization module; The sensor module is used for real-time monitoring and collecting state data of the bridge crane, including vibration, temperature of the motor and abrasion of the brake; The data processing module is used for pre-processing the collected state data and converting the state data into time series data for subsequent analysis and evaluation; The risk assessment module is used for combining a risk assessment and failure prediction model to evaluate the remaining service life, failure probability and health degree of each equipment according to historical data and current data; The state discrimination module is used for quantitatively evaluating the monitoring result of the equipment according to a state discrimination evaluation model using a multi-parameter weighted scoring principle, and dividing the health state of the equipment according to the scoring result; The HMI is used for displaying the evaluation result in a chart form, including the remaining service life, failure probability, health degree, health state, vibration, temperature and abrasion of the equipment; The optimization module is used for sending corresponding maintenance measures to maintenance personnel for optimization according to the evaluation result, including maintenance plan, maintenance suggestion, failure diagnosis and failure elimination.

8. A system for a predictive maintenance method of a bridge crane as claimed in claim 7, characterized in that: The HMI comprises an HMI main interface, a predictive maintenance system main interface, a vibration sensor interface, a temperature abrasion sensor interface, an alarm chart interface, a historical alarm chart interface, a trend chart interface, an analysis report fault judgment interface, a failure prediction data calculation interface and a maintenance scheme interface; The HMI main interface is used for displaying information and state of the system, including system startup ID, alarm indicator light, double-beam monitoring system button and PDM system button; The predictive maintenance system main interface is used for displaying and controlling functions of the predictive maintenance system, including vibration sensor button, temperature abrasion sensor button, alarm chart button, historical alarm chart button, trend chart button, analysis report fault diagnosis button, failure prediction data calculation button and maintenance measure button; The vibration sensor interface is used for displaying real-time data and a curve graph of the vibration sensor, and alarm value and alarm light; The temperature abrasion sensor interface is used for displaying real-time data and a curve graph of the temperature sensor and the abrasion sensor, and alarm value and alarm light; The alarm chart interface is used for displaying a real-time list of any one equipment and any one alarm information, and a failure cause of the equipment; The historical alarm chart interface is used to display an arrangement and statistics of historical data, and provides data support for subsequent calculation and evaluation; The trend chart interface is used to display data trends of all analog sensor groups, and displays data trends of different time periods by defining a time interval or expanding or reducing a time axis, so as to provide numerical and chart basis for manual intervention judgment; The analysis report fault judgment interface is used to display most of fault phenomena, causes and processing methods, and is coded, so as to provide coding reference for fault processing results in the maintenance scheme interface; The fault prediction data calculation interface is used to display predictive results obtained by three different calculation methods, including average method, regression method and classification method, so as to provide data reference for maintenance personnel. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8. The processor executes the computer program to realize the steps of the method in any one of claims 1 to 6.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the method in any one of claims 1 to 6.

Images