CN117911012B - Equipment health management system based on equipment ecological detection and running state evaluation - Google Patents

Equipment health management system based on equipment ecological detection and running state evaluation Download PDF

Info

Publication number
CN117911012B
CN117911012B CN202410315747.XA CN202410315747A CN117911012B CN 117911012 B CN117911012 B CN 117911012B CN 202410315747 A CN202410315747 A CN 202410315747A CN 117911012 B CN117911012 B CN 117911012B
Authority
CN
China
Prior art keywords
monitoring
equipment
data
module
seasonal
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202410315747.XA
Other languages
Chinese (zh)
Other versions
CN117911012A (en
Inventor
赵建普
唐鹏
吴曦
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chengdu Siyue Intelligent Equipment Co ltd
Original Assignee
Chengdu Siyue Intelligent Equipment Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chengdu Siyue Intelligent Equipment Co ltd filed Critical Chengdu Siyue Intelligent Equipment Co ltd
Priority to CN202410315747.XA priority Critical patent/CN117911012B/en
Publication of CN117911012A publication Critical patent/CN117911012A/en
Application granted granted Critical
Publication of CN117911012B publication Critical patent/CN117911012B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D21/00—Measuring or testing not otherwise provided for
    • G01D21/02—Measuring two or more variables by means not covered by a single other subclass
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/24323—Tree-organised classifiers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/27—Regression, e.g. linear or logistic regression
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/20—Ensemble learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06314—Calendaring for a resource
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/20—Administration of product repair or maintenance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2123/00—Data types
    • G06F2123/02—Data types in the time domain, e.g. time-series data

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Human Resources & Organizations (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Software Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Health & Medical Sciences (AREA)
  • Marketing (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • General Business, Economics & Management (AREA)
  • Biomedical Technology (AREA)
  • Tourism & Hospitality (AREA)
  • General Health & Medical Sciences (AREA)
  • Quality & Reliability (AREA)
  • Molecular Biology (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Medical Informatics (AREA)
  • Educational Administration (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

The invention relates to the technical field of equipment management, in particular to an equipment health management system based on equipment ecological detection and running state evaluation, which comprises an integrated monitoring module, an environment adaptability adjusting module and a control analysis module, wherein the integrated monitoring module integrates the functions of torque monitoring, load monitoring, vibration monitoring, cleanliness monitoring, part input time monitoring, roller mileage monitoring, speed safety monitoring and temperature monitoring, and monitors the equipment state in real time; the environment adaptability adjusting module automatically adjusts the operation parameters of the equipment according to the change of the state of the real-time monitoring equipment; the control analysis module receives the data collected by the integrated monitoring module, analyzes and processes the data, and analyzes the data by adopting the data analysis and processing. The invention not only improves the operation efficiency and reliability of industrial equipment, but also reduces the operation and maintenance cost for enterprises, and brings a new solution to the field of equipment health management.

Description

Equipment health management system based on equipment ecological detection and running state evaluation
Technical Field
The invention relates to the technical field of equipment management, in particular to an equipment health management system based on equipment ecological detection and running state evaluation.
Background
Under the trend of the time of industrial production automation and intellectualization, in order to ensure the healthy use of equipment, ensure the smooth completion of a production plan and ensure the early discovery of problem points, the equipment health management system is invented, aiming at reducing the use burden of equipment engineers, facilitating the overall management and control of users, carrying out the fine management on equipment conditions, controlling the equipment state, knowing important operation parameters of equipment in real time and making important guidance for the prevention and inspection and maintenance of the problem points of the equipment.
The existing system has various equipment types, various equipment control systems and mechanism designs, various functions and huge parameters, mass equipment needs to be analyzed, common points and important control parameters of the equipment are found, key parameters affecting the operation of the equipment are analyzed, corresponding management schemes are made, commonalities are strived for, and the system can adapt to various different equipment through simple adjustment, but is difficult to predict future fault maintenance.
Disclosure of Invention
Based on the above purpose, the invention provides a device health management system based on device ecological detection and running state evaluation.
The equipment health management system based on equipment ecological detection and running state evaluation comprises an integrated monitoring module, an environment adaptability adjusting module and a control analysis module, wherein,
The integrated monitoring module integrates the functions of torque monitoring, load monitoring, vibration monitoring, cleanliness monitoring, part input time monitoring, roller mileage monitoring, speed safety monitoring and temperature monitoring, and monitors the equipment state in real time;
the environment adaptability adjusting module automatically adjusts equipment operation parameters including cooling system strength, lubricating oil supply quantity or operation speed according to the change of the state of the real-time monitoring equipment so as to adapt to different environmental conditions;
the control analysis module receives the data collected by the integrated monitoring module, performs data analysis and processing, and automatically adjusts equipment setting or sends maintenance and fault early warning according to the data analysis result.
Further, the integrated monitoring module specifically includes:
Torque monitoring submodule: real-time torque data of an operating mechanism of the monitoring equipment;
load monitoring submodule: the load normal range is set according to the mechanical characteristics;
Vibration monitoring submodule: the running conditions of the motor, the speed reducer and the transmission mechanism are monitored through the sensor;
cleanliness monitoring submodule: the use environment is monitored in real time through an atmosphere monitoring sensor, and particulate matter data in the environment is measured;
and a part input time monitoring sub-module: judging through program logic, and recording material input time;
roller mileage monitoring sub-module: the method comprises the steps of identifying the movement of a main body, recording the movement distance of equipment through a program, accumulating the movement distance in a database, and displaying the movement distance in monitoring equipment;
Speed safety monitoring submodule: limiting the movement range through a limit sensor;
Temperature monitoring submodule: and judging whether the running states of the electric control cabinet and the motor are normal or not through temperature sensor and shaft running monitoring.
Further, the environmental adaptation adjustment module specifically includes:
Torque monitoring and adjusting submodule: when the abnormal torque is detected, the power output is automatically adjusted to prevent overload or insufficient load of the equipment and ensure that the equipment is in an optimal working state;
Load monitoring and adjusting submodule: according to the real-time load data, adjusting the power distribution and the operation speed of the equipment to cope with different workloads and optimize the energy utilization efficiency;
Vibration monitoring and adjusting submodule: when abnormal vibration is detected, automatically adjusting the fastening degree of the mechanical component or changing the operation mode so as to reduce the influence of vibration on equipment;
Cleanliness monitoring and adjusting submodule: when the environmental cleanliness is reduced, the working frequency of the filtering system is increased, or a cleaning program is started to maintain the cleanliness of the equipment and the environment thereof;
And a part input time monitoring and adjusting sub-module: if the material input time is too long or too short, automatically adjusting the feeding speed or notifying an operator to adjust so as to maintain the continuity and efficiency of the production flow;
Roller mileage monitoring and adjusting sub-module: when the movement distance of the roller reaches a specific mileage, prompting maintenance or automatically adjusting the roller pressure so as to optimize the movement efficiency and reduce the abrasion;
Speed safety monitoring and adjusting submodule: based on real-time monitoring of the running speed of the equipment, automatically adjusting the speed limit or starting a safety mechanism to prevent potential risks caused by too high speed;
temperature monitoring and adjusting submodule: the cooling system is automatically adjusted or the equipment operation mode is adjusted according to the temperature change when the equipment is operated so as to prevent overheating or efficiency reduction caused by over-low temperature.
Further, the data analysis and processing in the control analysis module includes anomaly detection, dynamic trend analysis, and predictive maintenance.
Furthermore, the abnormality detection not only identifies the abnormality according to a preset threshold value, but also automatically learns and identifies the normal mode and the abnormal mode of the equipment operation by adopting a machine learning algorithm, and identifies the mode deviation, the abnormality detection specifically adopts an isolation forest algorithm, the normal mode and the abnormal mode of the equipment operation are automatically learned and identified by utilizing the isolation forest algorithm, the isolation forest algorithm isolates the observed value by randomly selecting the characteristic and randomly selecting the segmentation value of the characteristic, and the isolation forest algorithm is expressed as follows:
Is provided with For data set,/>For/>One sample of the tree is an isolated tree, and for each tree, the anomaly score/>The calculation formula of (2) is as follows:
；
Wherein, Is to isolate samples/>, in the treeDesired value of required path length,/>Is the number of samples that are to be taken,Also represents the random tree upper point/>, in the random forestFor measuring the average path length of a dot/>The ease of isolation, the shorter the path length, indicating/>The easier it is to isolate, i.e. the closer to outliers,/>Is a normalization factor for the path length, used to normalize/>Conversion to an anomaly score between 0 and 1, anomaly score/>The closer to 1, indicating sample/>The closer to the outlier.
Further, the dynamic trend analysis includes:
Subtle trends in the performance of the device, including progressively increasing energy consumption or progressively decreasing efficiency, are identified from the long-term collected data, and the dynamic trend analysis is based on historical data and real-time data, with seasonal resolved time series STL algorithms identifying and analyzing trends and seasonal patterns in the time series data.
Further, the time series STL algorithm analyzes the data collected for a long time and the real-time data to identify subtle variation trends of the device performance, and calculates as follows:
Is provided with For observations in the original time series, the time series STL algorithm decomposes into the form:
Wherein/> Representing trend components,/>Representing seasonal components,/>Representing residual components, STL fits trend and seasonal components, respectively, by locally weighted regression, wherein,
Trend componentRepresenting a long-term trend of a time sequence, showing the overall change of the performance of the equipment over time, trend component/>Is calculated as follows:
Initial step from the original time sequence Removing seasonal influence of rough estimation to obtain seasonal data, fitting seasonal components by using local weighted regression to obtain preliminary trend estimation, and gradually refining the trend estimation by iterative adjustment until convergence to a stable trend line;
seasonal assembly Capturing periodic fluctuations in data, seasonal components/>Is calculated as follows:
from the original time series Trend component removal/>Obtaining trending data, carrying out seasonal decomposition on the trending data, calculating an average value in each seasonal period, and smoothing each seasonal average value by using local weighted regression to obtain a smoothed seasonal component;
Residual assembly Including trends and seasonal outside variations, identifying unusual variations or anomalies, residual component/>Is calculated as follows:
By removing trend components and seasonal components from the original time series: information other than trends and seasonality is included for identifying anomalies and aperiodic fluctuations.
Further, the predictive maintenance is based on a predictive model, which is based on deep learning, not only predicts when equipment needs maintenance, but also provides a fault reason and an optimal maintenance strategy, and according to the actual running condition and history maintenance record of the equipment, proposes a customized maintenance plan, and performs predictive maintenance analysis by combining the deep learning model of a convolutional neural network and a long-term and short-term memory network, specifically as follows:
Convolutional neural network part: the device comprises a plurality of convolution layers, wherein each layer is provided with a specific number of convolution kernels and is used for extracting features of different levels, the convolution layers use device sensor data as input, the input comprises torque, temperature and vibration, each convolution layer captures features of different scales by using convolution kernels of different sizes, each convolution layer is followed by a pooling layer, the pooling layer is used for reducing the space dimension of the features and reducing the calculation amount, and a ReLU function is used as an activation function of the convolution layers;
long-term memory network LSTM layer: the LSTM layer comprises a plurality of LSTM units, each LSTM unit has a specific number of neurons, the LSTM layer is of a multi-layer structure so as to enhance the learning ability of a model, and an activation function of the LSTM layer uses a tanh function;
Model parameters and calculations were as follows:
convolution layer: for each convolution layer, the computation is expressed as: wherein/> Representing convolution operations,/>And/>Weights and offsets, respectively, of the convolution kernels are adjusted for different sensor data characteristics to optimize feature extraction,/>Is a feature map representing key features extracted from the original equipment data;
LSTM layer: the computation of LSTM cells includes forgetting gates, input gates, cell state updates and output gates, the computation of each part involving a weight matrix, bias terms and activation functions, the output of LSTM cells Expressed as: wherein/> Is the activation value of the output gate,/>Is a cell state;
the final output consists of one or more fully connected layers, which convert the output of the LSTM unit into a final prediction result, including the probability that the device needs maintenance, the probability of the type of failure.
Furthermore, the control analysis module has self-learning capability, and analysis and prediction accuracy is improved along with the time and accumulation of data.
Further, the system also comprises a user interface, wherein the user interface displays equipment monitoring data, analysis results and maintenance suggestions.
The invention has the beneficial effects that:
According to the invention, the subtle change of the equipment state can be more accurately identified by utilizing various sensor data in the integrated monitoring module and combining with the advanced CNN-LSTM model in the control analysis module, so that the maintenance requirement can be predicted in time, the prediction accuracy is improved, and the unexpected fault and the downtime of the equipment are greatly reduced.
According to the invention, through deep analysis of the equipment operation data, the system can rapidly identify potential fault sources and reasons for performance degradation, and the efficiency and effectiveness of maintenance work are greatly improved through real-time diagnosis and automatic adjustment.
According to the invention, through continuous performance monitoring and periodic maintenance, the system is beneficial to preventing serious mechanical damage, prolonging the service life of equipment, improving the overall operation reliability, and the introduction of the environment adaptability adjusting module enables the system to automatically adjust equipment parameters according to different operation conditions and environmental factors, so that the system can maintain optimal performance in changeable working environments.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only of the invention and that other drawings can be obtained from them without inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of a system functional module according to an embodiment of the present invention.
Detailed Description
The present invention will be further described in detail with reference to specific embodiments in order to make the objects, technical solutions and advantages of the present invention more apparent.
It is to be noted that unless otherwise defined, technical or scientific terms used herein should be taken in a general sense as understood by one of ordinary skill in the art to which the present invention belongs. The terms "first," "second," and the like, as used herein, do not denote any order, quantity, or importance, but rather are used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that elements or items preceding the word are included in the element or item listed after the word and equivalents thereof, but does not exclude other elements or items. The terms "connected" or "connected," and the like, are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", etc. are used merely to indicate relative positional relationships, which may also be changed when the absolute position of the object to be described is changed.
As shown in fig. 1, the device health management system based on device ecological detection and running state evaluation comprises an integrated monitoring module, an environment adaptability adjusting module and a control analysis module, and the specific scheme is as follows:
And (3) an integrated monitoring module: the module integrates the functions of torque monitoring, load monitoring, vibration monitoring, cleanliness monitoring, part input time monitoring, roller mileage monitoring, speed safety monitoring and temperature monitoring, and can monitor equipment states such as mechanical torque, operation load, vibration level, environmental cleanliness, material input time, moving distance, operation speed and equipment temperature in real time from multiple aspects, so that comprehensive equipment health management is ensured.
And the environment adaptability adjusting module is used for: and automatically adjusting equipment operation parameters such as cooling system strength, lubricating oil supply amount or operation speed according to the monitored environmental change so as to adapt to different environmental conditions, and improve the stability and service life of equipment.
And a control analysis module: the module is responsible for receiving the data collected by the integrated monitoring module and performing deep data analysis and processing, including but not limited to anomaly detection, trend analysis and predictive maintenance, and can automatically adjust equipment setting or send maintenance and fault early warning according to the data analysis result, so that efficient and safe operation of the equipment is ensured.
The integrated monitoring module specifically comprises:
torque monitoring submodule: the torque real-time data of the running mechanism of the monitoring equipment are used for timely checking according to the occurrence position if abnormal data appear, so that accidents are avoided;
load monitoring submodule: the device is used for monitoring the real-time state of the running load of the device, setting the normal load range according to the mechanical characteristics, and automatically alarming the device to prompt related personnel to check when the data are abnormal;
Vibration monitoring submodule: the running conditions of the motor, the speed reducer, the transmission mechanism and other moving parts are monitored through the sensor, if the vibration is too large and exceeds a preset range, the mechanism is possibly loose or other faults, and the faults need to be checked in time;
Cleanliness monitoring submodule: the use environment is monitored in real time through the atmosphere monitoring sensor, the particulate matter data in the environment is measured, the data exceeds the range, the corresponding monitoring sensor alarms, corresponding signals are fed back, equipment wear or pollution sources in the space are considered, and timely comparison and investigation are needed;
And a part input time monitoring sub-module: judging through program logic, recording the material input time, and sending an alarm when the material is not flowed out after overtime;
roller mileage monitoring sub-module: the method comprises the steps of identifying the movement of a main body, recording the movement distance of equipment through a program, accumulating in a database, displaying on monitoring equipment, and alarming and prompting according to different grades when the mileage reaches an early warning value;
speed safety monitoring submodule: the running of the movement equipment is dangerous, the movement range is required to be strictly limited, and the movement range is limited by a limit sensor;
temperature monitoring submodule: judging whether the operation states of the electric control cabinet and the motor are normal or not through a temperature sensor and shaft operation monitoring, if the operation temperature exceeds a preset range, giving an alarm, wherein the equipment state is overheat, the program control equipment stops running, the shaft movement stops, the alarm is released after the normal temperature is recovered, and the program resumes the equipment operation; when the temperature monitoring exceeds a normal value, judging that the equipment is short-circuited, stopping the program control equipment, and alarming to prompt personnel to check.
The environment adaptability adjusting module specifically comprises:
Torque monitoring and adjusting submodule: when the torque abnormality is detected, the power output is automatically adjusted to prevent overload or insufficient load of the equipment and ensure that the equipment is in an optimal working state. When the torque monitoring finds that the moment of the equipment is smaller than the set value but the mechanism can operate, an alarm for judging that the mechanism is likely to slip is sent out, manual intervention is performed after the alarm, manual reverse rotation is performed, the belt state and the moment are observed, if the tension belt can be normally relaxed, the normal rotation is observed again. If the equipment is recovered to be normal, the equipment continues to operate and continuously monitors, and if the equipment cannot be recovered, personnel are required to further disassemble and check.
Load monitoring and adjusting submodule: according to the real-time load data, the power distribution and the operation speed of the equipment are adjusted so as to cope with different workloads and optimize the energy utilization efficiency. The load level is found to exceed the normal value, and the load level is divided into warning and fault according to different setting ranges, wherein the warning is that the load exceeds the maximum value of the setting range by 10%, and at the moment, the equipment continues to operate, but the personnel is prompted to pay attention to the load; if the load exceeds the set maximum value by more than 25%, the equipment stops running, a fault alarm is sent, and a person is prompted to check the corresponding position, so that the equipment possibly gets stuck or interferes.
Vibration monitoring and adjusting submodule: upon detection of an abnormal vibration, the degree of tightening of the mechanical components is automatically adjusted or the mode of operation is changed to reduce the impact of the vibration on the device.
Cleanliness monitoring and adjusting submodule: when the environmental cleanliness is reduced, the operating frequency of the filtration system is increased, or a cleaning procedure is initiated to maintain the cleanliness of the equipment and its environment.
And a part input time monitoring and adjusting sub-module: if the material input time is too long or too short, the feeding speed is automatically adjusted or an operator is informed to adjust, so that the continuity and the efficiency of the production flow are maintained.
Roller mileage monitoring and adjusting sub-module: when the roller movement distance reaches a specific mileage, maintenance is prompted or roller pressure is automatically adjusted to optimize movement efficiency and reduce wear.
Speed safety monitoring and adjusting submodule: based on real-time monitoring of the device operating speed, the speed limit is automatically adjusted or a safety mechanism is activated to prevent potential risks caused by excessive speed. On the one hand, the operation area of the monitoring equipment is matched with the corresponding speed setting, acceleration and deceleration are carried out in a safe range, on the other hand, when the equipment is powered off, the program automatically controls each shaft to be started in a braking mode, the mechanism is prevented from sliding off or cannot be controlled, and further when the equipment is detected to stop, the shaft still moves, and at the moment, the braking system is immediately involved.
Temperature monitoring and adjusting submodule: the cooling system is automatically adjusted or the equipment operation mode is adjusted according to the temperature change when the equipment is operated so as to prevent overheating or efficiency reduction caused by over-low temperature.
The module can ensure the optimal performance of the equipment under various environments and operating conditions, reduce energy consumption and maintenance cost, and improve the overall reliability and service life of the equipment.
The control analysis module specifically comprises:
Advanced anomaly detection: not only is the abnormality identified according to a preset threshold value, but also a normal mode and an abnormal mode of the equipment operation are automatically learned and identified by adopting a machine learning algorithm, subtle mode deviations can be identified, even if the deviations are not considered as abnormalities in the traditional method, an isolation forest algorithm is adopted for advanced abnormality detection, the normal mode and the abnormal mode of the equipment operation are automatically learned and identified by utilizing the isolation forest algorithm, the isolation forest algorithm isolates an observed value by randomly selecting a characteristic and randomly selecting a segmentation value of the characteristic, abnormal data points are generally easier to isolate, and therefore fewer steps are required to isolate, and the method is calculated as follows:
Is provided with For data set,/>For/>One sample of the tree in the isolated forest algorithm is an isolated tree (itere), for each tree, anomaly score/>The calculation formula of (2) is as follows: /(I)；
Wherein,Is to isolate samples/>, in the treeDesired value of required path length,/>Is the number of samples that are to be taken,Is the standardized tree height, in the above calculation formula,/>Representing points on random trees in random forest/>Is the average path length of the metric point/>Key indicator of isolated difficulty, shorter path length, indicating/>The easier it is to isolate, i.e. the more likely it is an outlier,/>Is a normalization factor of the path length under normal conditions, used forConversion to an anomaly score between 0 and 1, anomaly score/>The closer to 1, indicating sample/>The more likely it is an anomaly.
Number of samples: In the present invention, the number of samples/>If hundreds of records are generated per device per day, as determined by the monitored data volume of the device, daily or weekly data may be selected as a sample set.
Number of trees: the number of trees should be sufficiently large to ensure stability and reliability of the results, and in the context of equipment health monitoring, between 100 and 500 trees may be provided to achieve good anomaly detection performance.
Subsample size: the sub-sample size used for each tree may be determined based on the characteristics of the device and the diversity of the data, and may be set to a small proportion (256 or 512) of the training data to ensure diversity between trees.
Path length: The path length is the number of steps to reach a leaf node during isolation, and reflects the degree of deviation of a data point from a normal operation mode in the context of device monitoring data.
Normalization factor: A normalization factor is used to convert the path length to an anomaly score between 0 and 1, this factor being dependent on the number of samples/>And the characteristics of the data are adjusted according to the specific equipment monitoring scene.
Dynamic trend analysis: the subtle trend of the performance of the device is identified by the data collected for a long time, including the gradually increased energy consumption or the gradually reduced efficiency, and the trend analysis is based on historical data and also combined with real-time data to provide more accurate and timely insight, and particularly, the trend and seasonal pattern in the time series data are identified and analyzed by using the seasonal decomposition time series STL, so that the data are particularly suitable for processing the performance monitoring data of the device, because the data usually have obvious trend and periodicity, the dynamic trend analysis uses STL algorithm to analyze the data collected for a long time and the real-time data so as to identify the subtle trend of the performance of the device, and the method is calculated as follows: is provided withAs observations in the original time series, the basic form of STL decomposition is:
Wherein/> Representing trend components,/>Representing seasonal components,/>Representing the residual components, STL fits trend and seasonal components, respectively, by local weighted regression (Loess), wherein,
Trend componentRepresenting a long-term trend of a time sequence, the overall change condition of the performance of the equipment along with time can be shown, and trend components/>Is calculated as follows:
Initial step from the original time sequence Removing seasonal influence of rough estimation to obtain seasonal data, fitting the seasonal data by using local weighted regression (Loess) to obtain preliminary trend estimation, and gradually refining the trend estimation by iterative adjustment until convergence to a stable trend line;
seasonal assembly Capturing periodic fluctuations in data, such as periodic usage patterns of the device during the day or week, seasonal components/>Is calculated as follows:
from the original time series Trend component removal/>Obtaining trending data, carrying out seasonal decomposition on the trending data, calculating average values in each seasonal period, and smoothing the seasonal average values by using local weighted regression to obtain smoothed seasonal components;
Residual assembly Including trends and out-of-season changes, helping to identify unusual changes or anomalies, residual components/>Is calculated as follows:
the residual component is obtained by removing the trend component and seasonal component from the original time series: this component contains information other than trends and seasonality that helps identify anomalies and non-periodic fluctuations.
Trend componentParameter definition:
window length: the window length of the trend component determines the degree of smoothness of the trend line, and in the device health monitoring, the window length may be adjusted according to the device operation period, and for continuously operating devices, longer window lengths may be selected to capture smoother trend lines.
Fitting degree: the fitness of the trend component determines the fitness of the trend line to the raw data. In device monitoring, an appropriate fit may help identify long-term performance changes.
Seasonal assemblyParameter definition:
seasonal period: this is a key parameter in determining the period of seasonal fluctuations, which for industrial equipment may be set according to the production cycle or a specific mode of operation, such as a 24 hour cycle or a weekly cycle.
Seasonal smoothness: the seasonal smoothness determines the smoothness of the seasonal fluctuations, and higher smoothness suppresses noise, which more clearly shows the seasonal pattern.
Residual assemblyParameter definition:
Rejection threshold: noise or outliers are contained in the residual component, and outliers are excluded from analysis by setting a rejection threshold to more accurately identify real equipment problems.
Predictive maintenance decision support: the control analysis module uses a prediction model based on deep learning to predict when equipment is likely to need maintenance, provides possible fault reasons and an optimal maintenance strategy, and proposes a customized maintenance plan according to the actual running condition and historical maintenance record of the equipment, thereby minimizing downtime and improving maintenance efficiency, and particularly adopts a deep learning model combining a Convolutional Neural Network (CNN) and a long short term memory network (LSTM) to perform predictive maintenance analysis, wherein the method comprises the following steps:
model architecture:
CNN part: the device comprises a plurality of convolution layers, each layer has a specific number of convolution kernels for extracting features of different levels, the convolution layers use device sensor data as input, such as torque, temperature and vibration, each convolution layer uses convolution kernels (e.g. 3x3 and 5x 5) with different sizes to capture features of different scales, each convolution layer is followed by a pooling layer for reducing the space dimension of the features and reducing the calculation amount, and an activation function of the convolution layers uses a ReLU function.
LSTM layer: the LSTM layer contains a series of LSTM cells, each cell having a specific number of neurons, and is a multi-layer structure to enhance learning ability of the model, and an activation function of the LSTM layer uses a tanh function.
Model parameters and calculations:
Convolution layer:
for each convolutional layer, its calculation is expressed as: wherein/> Representing convolution operations,/>And/>Weights and offsets, respectively, of the convolution kernels are adjusted for different sensor data characteristics to optimize feature extraction,/>Is a feature map representing key features extracted from the original equipment data;
LSTM layer:
The computation of the LSTM cells includes four main parts, forget gates, input gates, cell state updates and output gates, with each part's computation involving a weight matrix, bias terms and activation functions.
Output of LSTM cellExpressed as:/>Wherein/>Is the activation value of the output gate,/>Is the cell state.
The final output of the model consists of one or more fully connected layers that convert the output of the LSTM into a final prediction result that includes the probability that the device needs maintenance, the likelihood of a fault type, or other relevant indicators.
CNN part: the number and size of the convolutional layers: depending on the complexity and feature dimensions of the device data, multiple convolution layers may be provided, each layer having a different number of convolution kernels, more convolution layers being required to effectively extract features for devices containing multiple sensor data.
Convolution kernel size: an appropriate convolution kernel size, such as 3x3 or 5x5, is selected based on the characteristics of the device data to effectively capture local features in the data.
Pooling layer: a pooling layer is added after the convolution layer to reduce feature dimensions and computational complexity while maintaining critical information.
LSTM section:
LSTM cell number: depending on the length and complexity of the time series data, a suitable number of LSTM cells are selected, requiring more cells to capture long-term dependencies for long time series and complex data patterns.
LSTM number of layers: the multi-layer LSTM structure enhances the learning ability of the model to time series data, and the selection of the layer number depends on the time dependence and complexity of the data.
Full connection layer and output layer: number of neurons of fully connected layer: according to the target (maintenance requirement, potential fault type) to be predicted.
Output layer: the design into a specific predictive task, such as bi-classification, multi-classification or regression, depends on the specific type of maintenance requirements.
In addition, the module has self-learning capability, and the analysis and prediction accuracy is improved continuously along with the time and the accumulation of data.
Also included is a user interface that presents device monitoring data, analysis results, and maintenance recommendations, specific functions including, but not limited to:
Data visualization: real-time data from the integrated monitoring module, such as torque, temperature, vibration, etc., is presented in the form of charts, dashboards and trend lines.
Abnormality alert and notification: when the control analysis module detects an anomaly or predicts a need for maintenance, the user interface will display an alarm and provide detailed fault information and suggested action schemes.
Maintenance schedule planning: based on predictive maintenance analysis, the user interface provides a maintenance schedule planning tool that helps the user schedule maintenance activities according to system recommendations.
Historical data analysis: the functions of accessing historical monitoring data and maintaining records are provided for long-term performance analysis and trend identification by the user.
Interactive control: allowing the user to directly adjust device parameters or apply system-recommended maintenance measures through the interface.
The design of the user interface aims at improving user experience, enabling equipment operators and maintenance engineers to easily monitor equipment states, rapidly respond to system alarms, and effectively planning maintenance activities, so that the running efficiency and reliability of equipment are improved.
Those of ordinary skill in the art will appreciate that: the discussion of any of the embodiments above is merely exemplary and is not intended to suggest that the scope of the invention is limited to these examples; the technical features of the above embodiments or in the different embodiments may also be combined within the idea of the invention, the steps may be implemented in any order and there are many other variations of the different aspects of the invention as described above, which are not provided in detail for the sake of brevity.
The present invention is intended to embrace all such alternatives, modifications and variances which fall within the broad scope of the appended claims. Therefore, any omission, modification, equivalent replacement, improvement, etc. of the present invention should be included in the scope of the present invention.

Claims (4)

1. The equipment health management system based on equipment ecological detection and running state evaluation is characterized by comprising an integrated monitoring module, an environment adaptability adjusting module and a control analysis module, wherein,
The integrated monitoring module integrates functions of torque monitoring, load monitoring, vibration monitoring, cleanliness monitoring, part input time monitoring, roller mileage monitoring, speed safety monitoring and temperature monitoring, and monitors equipment states in real time, and the integrated monitoring module specifically comprises:
Torque monitoring submodule: real-time torque data of an operating mechanism of the monitoring equipment;
load monitoring submodule: the load normal range is set according to the mechanical characteristics;
Vibration monitoring submodule: the running conditions of the motor, the speed reducer and the transmission mechanism are monitored through the sensor;
cleanliness monitoring submodule: the use environment is monitored in real time through an atmosphere monitoring sensor, and particulate matter data in the environment is measured;
and a part input time monitoring sub-module: judging through program logic, and recording material input time;
roller mileage monitoring sub-module: the method comprises the steps of identifying the movement of a main body, recording the movement distance of equipment through a program, accumulating the movement distance in a database, and displaying the movement distance in monitoring equipment;
Speed safety monitoring submodule: limiting the movement range through a limit sensor;
Temperature monitoring submodule: judging whether the running states of the electric control cabinet and the motor are normal or not through temperature sensor and shaft running monitoring;
The environment adaptability adjusting module automatically adjusts equipment operation parameters including cooling system strength, lubricating oil supply quantity or operation speed according to the change of the state of the real-time monitoring equipment;
The control analysis module receives the data collected by the integrated monitoring module, performs data analysis and processing, and automatically adjusts equipment setting or sends maintenance and fault early warning according to the data analysis result, wherein the data analysis and processing comprises anomaly detection, dynamic trend analysis and predictive maintenance;
The abnormal detection not only identifies the abnormality according to a preset threshold value, but also automatically learns and identifies the normal mode and the abnormal mode of the equipment operation by adopting a machine learning algorithm, and identifies the mode deviation, the abnormal detection specifically adopts an isolation forest algorithm, the normal mode and the abnormal mode of the equipment operation are automatically learned and identified by utilizing the isolation forest algorithm, the isolation forest algorithm isolates the observed value by randomly selecting the characteristic and randomly selecting the segmentation value of the characteristic, and the isolation forest algorithm is expressed as follows:
Is provided with For data set,/>For/>One sample of the tree is an isolated tree, and for each tree, the anomaly score/>The calculation formula of (2) is as follows:
；
Wherein, Is to isolate samples/>, in the treeDesired value of required path length,/>Is the number of samples that are to be taken,Also represents the random tree upper point/>, in the random forestFor measuring the average path length of a dot/>The ease of isolation, the shorter the path length, indicating/>The easier it is to isolate, i.e. the closer to outliers,/>Is a normalization factor for the path length, used to normalize/>Conversion to an anomaly score between 0 and 1, anomaly score/>The closer to 1, indicating sample/>The closer to the outlier;
the dynamic trend analysis includes:
identifying subtle variation trends of equipment performance, including gradually increased energy consumption or gradually reduced efficiency, through long-term collected data, wherein the dynamic trend analysis is based on historical data and real-time data, and a seasonal decomposition time sequence STL algorithm is adopted to identify and analyze trends and seasonal patterns in the time sequence data;
The time series STL algorithm analyzes the data collected for a long time and the real-time data to identify subtle variation trends of the equipment performance, and calculates as follows:
Is provided with For observations in the original time series, the time series STL algorithm decomposes into the form:
Wherein/> Representing trend components,/>Representing seasonal components,/>Representing residual components, STL fits trend and seasonal components, respectively, by locally weighted regression, wherein,
Trend componentRepresenting a long-term trend of a time sequence, showing the overall change of the performance of the equipment over time, trend component/>Is calculated as follows:
Initial step from the original time sequence Removing seasonal influence of rough estimation to obtain seasonal data, fitting seasonal components by using local weighted regression to obtain preliminary trend estimation, and gradually refining the trend estimation by iterative adjustment until convergence to a stable trend line;
seasonal assembly Capturing periodic fluctuations in data, seasonal components/>Is calculated as follows:
from the original time series Trend component removal/>Obtaining trending data, carrying out seasonal decomposition on the trending data, calculating an average value in each seasonal period, and smoothing each seasonal average value by using local weighted regression to obtain a smoothed seasonal component;
Residual assembly Including trend and seasonal changes, identifying unusual changes or anomalies, residual componentsIs calculated as follows:
By removing trend components and seasonal components from the original time series: information other than trends and seasonality is included for identifying anomalies and aperiodic fluctuations;
The predictive maintenance is based on a predictive model, the predictive model is based on deep learning, not only predicts when equipment needs maintenance, but also provides a fault reason and an optimal maintenance strategy, and according to the actual running condition and the historical maintenance record of the equipment, a customized maintenance plan is provided, and predictive maintenance analysis is performed by combining with the deep learning model of a convolutional neural network and a long-term and short-term memory network, specifically as follows:
Convolutional neural network part: the device comprises a plurality of convolution layers, wherein each layer is provided with a specific number of convolution kernels and is used for extracting features of different levels, the convolution layers use device sensor data as input, the input comprises torque, temperature and vibration, each convolution layer captures features of different scales by using convolution kernels of different sizes, each convolution layer is followed by a pooling layer, the pooling layer is used for reducing the space dimension of the features and reducing the calculation amount, and a ReLU function is used as an activation function of the convolution layers;
long-term memory network LSTM layer: the LSTM layer comprises a plurality of LSTM units, each LSTM unit has a specific number of neurons, the LSTM layer is of a multi-layer structure so as to enhance the learning ability of a model, and an activation function of the LSTM layer uses a tanh function;
Model parameters and calculations were as follows:
convolution layer: for each convolution layer, the computation is expressed as: wherein/> Representing convolution operations,/>And/>Weights and offsets, respectively, of the convolution kernels are adjusted for different sensor data characteristics to optimize feature extraction,/>Is a feature map representing key features extracted from the original equipment data;
LSTM layer: the computation of the LSTM cells in the LSTM layer comprises forgetting gates, input gates, cell state updating and output gates, the computation of each part involves a weight matrix, bias terms and activation functions, the output of the LSTM cells Expressed as: wherein/> Is the activation value of the output gate,/>Is a cell state;
the final output consists of one or more fully connected layers, which convert the output of the LSTM unit into a final prediction result, including the probability that the device needs maintenance, the probability of the type of failure.
2. The device health management system based on device ecological detection and operation state evaluation according to claim 1, wherein the environmental adaptation adjustment module specifically comprises:
Torque monitoring and adjusting submodule: when the abnormal torque is detected, the power output is automatically adjusted to prevent overload or insufficient load of the equipment and ensure that the equipment is in an optimal working state;
Load monitoring and adjusting submodule: according to the real-time load data, adjusting the power distribution and the operation speed of the equipment to cope with different workloads and optimize the energy utilization efficiency;
Vibration monitoring and adjusting submodule: when abnormal vibration is detected, automatically adjusting the fastening degree of the mechanical component or changing the operation mode so as to reduce the influence of vibration on equipment;
Cleanliness monitoring and adjusting submodule: when the environmental cleanliness is reduced, the working frequency of the filtering system is increased, or a cleaning program is started to maintain the cleanliness of the equipment and the environment thereof;
And a part input time monitoring and adjusting sub-module: if the material input time is too long or too short, automatically adjusting the feeding speed or notifying an operator to adjust so as to maintain the continuity and efficiency of the production flow;
Roller mileage monitoring and adjusting sub-module: when the movement distance of the roller reaches a specific mileage, prompting maintenance or automatically adjusting the roller pressure so as to optimize the movement efficiency and reduce the abrasion;
Speed safety monitoring and adjusting submodule: based on real-time monitoring of the running speed of the equipment, automatically adjusting the speed limit or starting a safety mechanism to prevent potential risks caused by too high speed;
temperature monitoring and adjusting submodule: the cooling system is automatically adjusted or the equipment operation mode is adjusted according to the temperature change when the equipment is operated so as to prevent overheating or efficiency reduction caused by over-low temperature.
3. The device health management system based on device ecological detection and operational state assessment according to claim 2, wherein the control analysis module has self-learning capability, and improves analysis and prediction accuracy over time and accumulation of data.
4. The device health management system based on device ecological detection and operational status assessment of claim 3, further comprising a user interface that presents device monitoring data, analysis results, and maintenance recommendations.
CN202410315747.XA 2024-03-20 2024-03-20 Equipment health management system based on equipment ecological detection and running state evaluation Active CN117911012B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410315747.XA CN117911012B (en) 2024-03-20 2024-03-20 Equipment health management system based on equipment ecological detection and running state evaluation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410315747.XA CN117911012B (en) 2024-03-20 2024-03-20 Equipment health management system based on equipment ecological detection and running state evaluation

Publications (2)

Publication Number Publication Date
CN117911012A CN117911012A (en) 2024-04-19
CN117911012B true CN117911012B (en) 2024-05-28

Family

ID=90685610

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410315747.XA Active CN117911012B (en) 2024-03-20 2024-03-20 Equipment health management system based on equipment ecological detection and running state evaluation

Country Status (1)

Country Link
CN (1) CN117911012B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118228938B (en) * 2024-05-23 2024-07-26 中交二公局东萌工程有限公司 Engineering machinery operation monitoring data acquisition, analysis, evaluation and management system

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10964130B1 (en) * 2018-10-18 2021-03-30 Northrop Grumman Systems Corporation Fleet level prognostics for improved maintenance of vehicles
CN112613646A (en) * 2020-12-08 2021-04-06 上海交通大学烟台信息技术研究院 Equipment state prediction method and system based on multi-dimensional data fusion
CN112904220A (en) * 2020-12-30 2021-06-04 厦门大学 UPS (uninterrupted Power supply) health prediction method and system based on digital twinning and machine learning, electronic equipment and storable medium
CN114897262A (en) * 2022-06-08 2022-08-12 南京国电南自轨道交通工程有限公司 Rail transit equipment fault prediction method based on deep learning
CN115760484A (en) * 2022-12-07 2023-03-07 湖北华中电力科技开发有限责任公司 Method, device and system for improving potential danger identification capability of power distribution station area and storage medium
CN117171661A (en) * 2023-11-03 2023-12-05 山东鸿远新材料科技股份有限公司 Chemical plant sewage treatment monitoring method and system
CN117421582A (en) * 2023-11-08 2024-01-19 浙江正泰中自控制工程有限公司 Equipment health analysis method based on multi-source data driving
CN117575332A (en) * 2024-01-12 2024-02-20 唐山伟仁建筑工程有限公司 Road construction safety monitoring method and system
CN117707027A (en) * 2024-02-05 2024-03-15 国网甘肃省电力公司白银供电公司 Centralized control station one-key sequential control system based on multi-source data fusion

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11774944B2 (en) * 2016-05-09 2023-10-03 Strong Force Iot Portfolio 2016, Llc Methods and systems for the industrial internet of things
EP3789838B1 (en) * 2019-09-09 2024-06-12 Alisea S.r.l. Systems and methods for artificial intelligence-based maintenance of an air conditioning system
US20230062971A1 (en) * 2021-09-01 2023-03-02 General Electric Company Apparatus, systems, and methods for monitoring, analyzing, and adjusting additive machine and build health and configuration

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10964130B1 (en) * 2018-10-18 2021-03-30 Northrop Grumman Systems Corporation Fleet level prognostics for improved maintenance of vehicles
CN112613646A (en) * 2020-12-08 2021-04-06 上海交通大学烟台信息技术研究院 Equipment state prediction method and system based on multi-dimensional data fusion
CN112904220A (en) * 2020-12-30 2021-06-04 厦门大学 UPS (uninterrupted Power supply) health prediction method and system based on digital twinning and machine learning, electronic equipment and storable medium
CN114897262A (en) * 2022-06-08 2022-08-12 南京国电南自轨道交通工程有限公司 Rail transit equipment fault prediction method based on deep learning
CN115760484A (en) * 2022-12-07 2023-03-07 湖北华中电力科技开发有限责任公司 Method, device and system for improving potential danger identification capability of power distribution station area and storage medium
CN117171661A (en) * 2023-11-03 2023-12-05 山东鸿远新材料科技股份有限公司 Chemical plant sewage treatment monitoring method and system
CN117421582A (en) * 2023-11-08 2024-01-19 浙江正泰中自控制工程有限公司 Equipment health analysis method based on multi-source data driving
CN117575332A (en) * 2024-01-12 2024-02-20 唐山伟仁建筑工程有限公司 Road construction safety monitoring method and system
CN117707027A (en) * 2024-02-05 2024-03-15 国网甘肃省电力公司白银供电公司 Centralized control station one-key sequential control system based on multi-source data fusion

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
Artificial intelligence and machine learning in energy conversion and management;Konstantinos Mira, 等;《Energies》;20231125;第16卷(第23期);第1-36页 *
Deep learning-based robust actuator fault detection and isolation scheme for highly redundant multirotor UAVs;Yisak Debele, 等;《Drones》;20230701;第7卷(第7期);第1-26页 *
基于小样本学习的工业故障诊断方法研究;岳昌源;《中国优秀硕士学位论文全文数据库信息科技辑》;20240315(第3期);第I140-303页 *
工业时序数据异常检测方法研究及应用;陈俊熙;《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》;20221015(第10期);第C029-188页 *
数据驱动的设备状态评估与故障诊断方法研究;王鹏凯;《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》;20240215(第2期);第C042-1030页 *

Also Published As

Publication number Publication date
CN117911012A (en) 2024-04-19

Similar Documents

Publication Publication Date Title
CN117911012B (en) Equipment health management system based on equipment ecological detection and running state evaluation
US11403160B2 (en) Fault predicting system and fault prediction method
CN109240244B (en) Data-driven equipment running state health degree analysis method and system
CN108829933A (en) A kind of method of the predictive maintenance and health control of semiconductor manufacturing facility
US6871160B2 (en) Intelligent condition-based engine/equipment management system
US20040049715A1 (en) Computer networked intelligent condition-based engine/equipment management system
KR102618023B1 (en) Failure prediction diagnosis system and method through pattern analysis according to failure type
CN113093693A (en) Online fault diagnosis method for operation state of coal mining machine
CN113063611B (en) Equipment monitoring management method and system
US20230131828A1 (en) Predictive maintenance system and method for intelligent manufacturing equipment
CN114066274A (en) Equipment performance early warning method and device
CN117761444B (en) Method and system for monitoring service life of surge protector
CN115526375A (en) Space flight equipment predictive maintenance system
KR20200084401A (en) Air Compressor Bidirectional Fault Diagnosis and Remote Control System Utilizing Big Data
Wang Key techniques in intelligent predictive maintenance (IPdM)–a framework of intelligent faults diagnosis and prognosis system (IFDaPS)
Calabrese et al. An event based machine learning framework for predictive maintenance in industry 4.0
CN109784127B (en) Equipment health state early warning method and system thereof
Lughofer et al. Prologue: Predictive maintenance in dynamic systems
CN118015812A (en) Equipment abnormality alarm method
CN117111446B (en) Fuzzy PID control optimization method for magnetic suspension flywheel motor
WO2020253950A1 (en) Monitoring method, predicting method, monitoring system and computer program
CN113723827A (en) Subway electromechanical equipment operation risk diagnosis and operation and maintenance control method and system
CN116674380A (en) Electric driving method and system suitable for automobile industry and intelligent manufacturing
CN111766514A (en) Data analysis method for equipment detection points
Liu et al. A fault early warning and health status rating method for ensuring safe operation of rotating equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant