CN114548175A - Working machine health state identification method and device and working machine - Google Patents

Working machine health state identification method and device and working machine Download PDF

Info

Publication number
CN114548175A
CN114548175A CN202210169923.4A CN202210169923A CN114548175A CN 114548175 A CN114548175 A CN 114548175A CN 202210169923 A CN202210169923 A CN 202210169923A CN 114548175 A CN114548175 A CN 114548175A
Authority
CN
China
Prior art keywords
fault
identified
machine
health
probability
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.)
Withdrawn
Application number
CN202210169923.4A
Other languages
Chinese (zh)
Inventor
朱超
陈磊
孙强
闫鑫
罗建华
袁爱进
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.)
Shanghai Huaxing Digital Technology Co Ltd
Original Assignee
Shanghai Huaxing Digital Technology 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 Shanghai Huaxing Digital Technology Co Ltd filed Critical Shanghai Huaxing Digital Technology Co Ltd
Priority to CN202210169923.4A priority Critical patent/CN114548175A/en
Publication of CN114548175A publication Critical patent/CN114548175A/en
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/08—Feature extraction
    • 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
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M99/00—Subject matter not provided for in other groups of this subclass
    • G01M99/005—Testing of complete machines, e.g. washing-machines or mobile phones
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions

Abstract

The invention provides a working machine health state identification method and device and a working machine. In addition, the method simultaneously introduces the fault state identification model and the normal state identification model, improves the accuracy of the health state identification result, is beneficial to timely diagnosis of the fault of the operation machine to be identified, and ensures the safe operation of the operation machine. In addition, due to the introduction of the unhealthy probability of the work machine to be identified, the performance of the work machine to be identified can be evaluated more comprehensively and accurately, namely, the fault degree of the work machine to be identified can be quantized, and the work machine to be identified which does not reach the fault state but deviates from the optimal performance state can be evaluated.

Description

Working machine health state identification method and device and working machine
Technical Field
The invention relates to the technical field of monitoring of working machines, in particular to a working machine health state identification method and device and a working machine.
Background
With the development of internet technology and the improvement of digital infrastructure, it becomes possible to monitor the operation condition and diagnose faults of target equipment by using information uploaded by online equipment. In the work machine industry, it would be of great value to effectively utilize information uploaded by the on-line equipment to identify the monitored status of the work machine and to intervene or predictively maintain the work machine before it has a major failure.
At present, the method for pre-diagnosing faults mainly comprises the following steps: (1) the method is established on an expert knowledge base and logic judgment, is only suitable for equipment with a simple structure, and cannot be suitable for operation machinery with a complex structure. (2) The classifier is used for supervised classification training, and whether equipment fails or not is judged through the classifier obtained through training.
Therefore, it is urgently needed to provide a method for identifying the health state of the working machine, which can comprehensively and accurately evaluate the performance of the working machine.
Disclosure of Invention
The invention provides a method and a device for identifying the health state of a working machine and the working machine, which are used for overcoming the defects in the prior art.
The invention provides a method for identifying the health state of a working machine, which comprises the following steps:
acquiring time sequence detection data characteristics corresponding to all indexes of the operation machine to be identified;
based on a normal state identification model, performing density estimation on the time sequence detection data characteristics corresponding to each index, and determining the unhealthy probability of the operation machine to be identified;
performing density estimation on the time sequence detection data characteristics corresponding to the indexes based on the fault state identification model corresponding to each index, and determining the fault probability of the to-be-identified operation machine under each index;
and identifying the comprehensive health state of the to-be-identified working machine based on the unhealthy probability and the fault probability.
According to the method for identifying the health state of the working machine, the method for identifying the comprehensive health state of the working machine to be identified based on the unhealthy probability and the fault probability comprises the following steps:
determining a comprehensive fault probability of the to-be-identified work machine based on the unhealthy probability and the fault probability;
determining a health status score of the work machine to be identified based on the comprehensive fault probability;
and determining a comprehensive health state identification result of the to-be-identified working machine based on the health state score of the to-be-identified working machine.
According to the work machine health status identification method provided by the invention, the determination of the comprehensive health status identification result of the work machine to be identified based on the health status score of the work machine to be identified comprises the following steps:
determining a plurality of health state scoring threshold intervals of the to-be-identified working machine; wherein, the plurality of health state scoring threshold intervals are all corresponding to different evaluation labels;
and determining a target health state score threshold interval in which the health state score is located in the plurality of health state score threshold intervals, and determining the comprehensive health state identification result based on an evaluation label corresponding to the target health state score threshold interval.
According to the method for identifying the health state of the working machine, the fault state identification model corresponding to each index is constructed on the basis of density estimation of a fault sample database corresponding to each index of the working machine, and the working machine carries fault maintenance record information; the fault sample database carries time sequence fault data characteristics; accordingly, the number of the first and second electrodes,
the fault state identification model corresponding to each index is constructed based on the following method:
and modeling the fault sample database by adopting a density estimation algorithm based on the time sequence fault data characteristics to obtain the fault state identification model.
According to the work machine health state identification method, the density estimation algorithm comprises a parameterized density estimation algorithm of known data distribution and/or a non-parameterized density estimation algorithm of unknown data distribution.
According to the method for identifying the health state of the working machine, the normal state identification model is constructed by modeling the normal sample database by adopting a density estimation algorithm based on the time sequence normal data characteristics carried by the normal sample database of the working machine.
According to the method for identifying the health state of the working machine, the fault sample database is modeled by adopting a density estimation algorithm based on the time sequence fault data characteristics to obtain the fault state identification model, and the method comprises the following steps:
if an overlapping area exists in data distribution between the fault sample database and the normal sample database, cutting the fault sample database by adopting preset cutting parameters;
based on the time sequence fault data characteristics, modeling a result obtained by cutting by adopting a density estimation algorithm to obtain the fault state identification model;
and the preset cutting parameters are obtained based on a calculation mode of the comprehensive fault probability.
According to the method for identifying the health state of the working machine, the step of acquiring the time sequence detection data characteristics corresponding to each index of the working machine to be identified comprises the following steps:
determining detection parameters collected by various sensors installed on the operation machine to be identified;
determining each index based on the detection parameters and the working principle of each system included in the operation machine to be identified;
performing feature extraction on the time sequence detection data acquired by the sensor corresponding to each index to obtain a feature extraction result;
and determining the time sequence detection data characteristics based on the characteristic extraction result.
The present invention also provides a working machine health status recognition apparatus, including:
the characteristic acquisition module is used for acquiring time sequence detection data characteristics corresponding to all indexes of the operation machine to be identified;
the unhealthy probability determining module is used for performing density estimation on the time sequence detection data characteristics corresponding to the indexes based on a normal state identification model to determine the unhealthy probability of the operation machine to be identified;
the fault probability determination module is used for performing density estimation on the time sequence detection data characteristics corresponding to each index based on the fault state identification model corresponding to each index, and determining the fault probability of the operation machine to be identified under each index;
and the state identification module is used for identifying the comprehensive health state of the to-be-identified working machine based on the unhealthy probability and the fault probability.
The present invention also provides a work machine comprising: the work machine health state recognition device described above.
The present invention also provides an electronic device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the above-mentioned work machine health status identification methods when executing the computer program.
The present disclosure also provides a non-transitory computer-readable storage medium having stored thereon a computer program that, when executed by a processor, performs the steps of the work machine health status identification method as described in any of the above.
According to the method, the normal state identification model and the fault state identification model corresponding to each index are introduced, the density estimation construction is carried out on the normal sample database of the working machine and the fault sample database corresponding to each index, and the method and the device can be suitable for the working machine with a complex structure. In addition, the method simultaneously introduces the fault state identification model and the normal state identification model, improves the accuracy of the health state identification result, is beneficial to timely diagnosis of the fault of the operation machine to be identified, and ensures the safe operation of the operation machine. In addition, due to the introduction of the unhealthy probability of the work machine to be identified, the performance of the work machine to be identified can be evaluated more comprehensively and accurately, namely, the fault degree of the work machine to be identified can be quantized, and the work machine to be identified which does not reach the fault state but deviates from the optimal performance state can be evaluated.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a flow chart illustrating a method for identifying a health of a work machine according to the present disclosure;
FIG. 2 is a second flowchart illustrating a method for identifying a health status of a work machine according to the present disclosure;
FIG. 3 is a graphical illustration of a work machine's combined fault probability and health score as a function of work machine duration (man-hours/h) in accordance with the present disclosure;
fig. 4 is a graph showing a change in 5 indexes of the working machine according to the present invention with the working time period (man-hour/h) of the working machine;
FIG. 5 is a schematic diagram illustrating a work machine health status identification apparatus in accordance with the present disclosure;
fig. 6 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Due to the existence of the current fault pre-diagnosis methods: 1) the device can only be suitable for equipment with a simple structure and cannot be suitable for operation machinery with a complex structure; 2) a large number of fault samples are needed, the problem of unbalanced samples easily exists, the accuracy of a judgment result is further reduced, and the like, so that the timely diagnosis of the faults of the operation machine is hindered, and the safe operation of the operation machine is not facilitated. Therefore, the embodiment of the invention provides a working machine health state identification method.
Fig. 1 is a schematic flow chart of a method for identifying a health status of a work machine according to an embodiment of the present invention, as shown in fig. 1, the method includes:
s1, acquiring time sequence detection data characteristics corresponding to each index of the operation machine to be identified;
s2, based on the normal state recognition model, performing density estimation on the time sequence detection data characteristics corresponding to each index, and determining the unhealthy probability of the operation machine to be recognized;
s3, performing density estimation on the time sequence detection data characteristics of each index based on the fault state identification model corresponding to each index, and determining the fault probability of the to-be-identified working machine under each index;
and S4, identifying the comprehensive health state of the work machine to be identified based on the unhealthy probability and the fault probability.
Specifically, in the method for identifying the health status of the working machine provided in the embodiment of the present invention, the execution subject is a working machine health status identification device, the device may be configured in a server, and the server may be a local server or a cloud server, where the local server may specifically be a computer, a tablet computer, and the like, and the method is not particularly limited in the embodiment of the present invention.
It will be appreciated that the work machine may include: at least one of a drilling machine, an excavating machine, a loading machine, a carrier machine, a municipal machine, a crusher, and a vehicle driven by a driver. An excavating machine is a working machine for excavating a mine. A loading machine is a working machine for loading cargo into a carrier machine. The loading machine includes at least one of a hydraulic excavator, an electric excavator, and a wheel loader. The carrier machine is a working machine for carrying cargo. Municipal machines are working machines used for the landscaping of urban roads, such as sweepers, watering lorries, and dust suction trucks. The crusher is a working machine that crushes earth and stones input from a carrier machine. In embodiments of the present disclosure, work machine state of health identification may be understood as a work machine performance assessment.
Step S1 is executed first to obtain the time-series detection data characteristic corresponding to each index of the work machine to be identified, which may be the work machine whose health status needs to be determined at the present time. Due to the complex structure of the work machine to be identified, a plurality of systems are usually provided, such as a hydraulic system, a power system, a heat dissipation system and the like, and the performance of each system is represented by a corresponding index. The individual indicators are therefore understood to mean all indicators relating to the individual system performance of the work machine to be identified, which are used to characterize the health state of the work machine to be identified.
It can be understood that each index may be a related parameter of the work machine to be identified, which is directly acquired by a sensor, or may be a parameter obtained by performing a certain conversion process on the related parameter, and is not specifically limited herein.
Here, the time-series detection data characteristic corresponding to each index refers to a characteristic of time-series detection data of the work machine to be identified, which is acquired by a sensor corresponding to each index, and the time-series detection data corresponding to a certain index may be acquired by arranging detection data of the work machine to be identified, which is acquired by the sensor corresponding to the index within a preset time period, in order of acquisition time. The length of the preset time period may be set as needed, and may be set to a time period from a time when the work machine starts to be put into use to a current time, for example.
The time series detection data characteristics corresponding to each index may be obtained by performing feature extraction on the time series detection data corresponding to each index, and may include time domain characteristics and frequency domain characteristics, and the time domain characteristics may include statistical characteristics of extremum, slope, variance, standard deviation, and the like of the time series detection data, which is not specifically limited herein.
Then, step S2 is executed to perform density estimation on the time-series detection data characteristics corresponding to each index of the work machine to be recognized through the normal state recognition model, and determine the unhealthy probability of the work machine to be recognized.
It can be understood that the normal state identification model is used for predicting the health degree of the to-be-identified working machine in the corresponding index when the to-be-identified working machine is in the normal state, so as to obtain the unhealthy probability of the to-be-identified working machine in the corresponding index when the to-be-identified working machine is in the normal state.
The function realization of the normal state recognition model is realized by carrying out density estimation on the time sequence detection data characteristics corresponding to each index of the operation machine to be recognized. Here, the normal state recognition model may be constructed by density estimation on a normal sample database of the work machine.
It will be appreciated that the type of work machine used to construct the health recognition model is the same as the type of work machine to be recognized, e.g., if the work machine to be recognized is an excavator, then the work machine is also an excavator. The work machine may be the work machine to be identified, or may be another work machine of the same type as the work machine to be identified. When the operating machine is the operating machine to be identified, the data contained in the normal sample database is the time sequence sample detection data acquired by each sensor of the operating machine to be identified in the historical normal state. When the working machine is other working machines, the normal sample database contains time sequence sample detection data acquired by each sensor of the same type of working machine in a normal state. Furthermore, the normal state identification model for determining the unhealthy probability of the work machine to be identified can be constructed by density estimation through a normal sample database of the work machine.
After the normal state recognition model is built, the time sequence detection data characteristics corresponding to all indexes of the operation machine to be recognized can be input into the normal state recognition model, and the unhealthy probability of the operation machine to be recognized is output through the normal state recognition model.
Then, step S3 is executed to perform density estimation on the time-series detection data characteristics corresponding to each index of the work machine to be identified through the fault state identification model corresponding to each index, and determine the fault probability of the work machine to be identified under each index.
It can be understood that each index of the to-be-identified work machine corresponds to a fault state identification model, and the fault state identification model is used for predicting the fault degree of the to-be-identified work machine under the corresponding index to obtain the fault probability of the to-be-identified work machine under the corresponding index.
The fault state identification model is functionally realized by carrying out density estimation on time sequence detection data characteristics corresponding to corresponding indexes of the to-be-identified operation machine. Here, the fault state identification model may be constructed by density estimation of a fault sample database corresponding to the respective index of the work machine. The relevant data of the work machine during the full life cycle is known and carries maintenance record information during the full life cycle. The maintenance record information may include information such as a fault type and index data at the time of maintenance, and is not particularly limited herein.
It is understood that the type of work machine used to construct the fault status identification model is the same as the type of work machine to be identified, e.g., the work machine to be identified is an excavator, and the work machine is also an excavator. The work machine may be the work machine to be identified, or may be another work machine of the same type as the work machine to be identified. When the operation machine used for constructing the fault state identification model is the operation machine to be identified, the data contained in the fault sample database is time sequence sample detection data acquired by each sensor of the operation machine to be identified in the historical fault state. When the working machine is other working machines, the fault sample database contains time sequence sample detection data acquired by each sensor of the same type of working machine in a fault state. Furthermore, the fault state identification model for determining the fault probability of the work machine to be identified can be constructed by density estimation through a fault sample database of the work machine.
The type of the working machine used for constructing the fault state identification model is the same as that of the working machine to be identified, so that the indexes related to the working machine and the working machine to be identified are the same, each index of the working machine corresponds to one fault sample database, and the fault sample database comprises time sequence sample detection data acquired by each sensor corresponding to each index of the working machine. Furthermore, a fault state identification model for determining the fault probability of the to-be-identified working machine under each index can be constructed by estimating the density of a fault sample database corresponding to the corresponding index of the working machine sample. Here, the density estimation of the fault sample database can be realized by adopting a density estimation algorithm, and then a fault state identification model is constructed and obtained.
After the fault state identification model corresponding to each index is constructed, the time sequence detection data characteristics corresponding to each index can be input into the fault state identification model corresponding to the index, and the fault probability of the to-be-identified operation machine under the index is output through the fault state identification model corresponding to the index.
Finally, step S4 is executed to identify the comprehensive health status of the work machine to be identified according to the unhealthy probability of the work machine to be identified and the failure probability of the work machine to be identified under each index. Here, the fault probabilities of the work machine to be recognized under each index may be preliminarily fused, and then the result of the preliminary fusion and the unhealthy probability may be secondarily fused, and the obtained secondary fusion result may be used to represent the comprehensive health state of the work machine to be recognized, thereby realizing the recognition of the comprehensive health state of the work machine to be recognized.
It will be appreciated that the overall health of the work machine to be identified may include one of a healthy, sub-healthy, faulty, and severely faulty condition. For example, the work machine to be identified is in a sub-health state, which indicates that the work machine to be identified does not have an obvious fault appearance, and has no serious abnormality in characteristics and indexes, but the comprehensive health state can reflect that the performance of the work machine to be identified is degraded. By giving a refined comprehensive health state recognition result, the comprehensive health state of the operation machine to be recognized can be described qualitatively, so that a user can intuitively feel the condition of the operation machine to be recognized, and the user can make further decisions during use.
The method for identifying the health state of the operating machine comprises the following steps of firstly, acquiring time sequence detection data characteristics corresponding to all indexes of the operating machine to be identified; then based on a normal state identification model, performing density estimation on the time sequence detection data characteristics corresponding to each index, and determining the unhealthy probability of the operation machine to be identified; performing density estimation on the time sequence detection data characteristics corresponding to each index based on the fault state identification model corresponding to each index, and determining the fault probability of the operation machine to be identified under each index; and identifying the comprehensive health state of the to-be-identified working machine based on the unhealthy probability and the fault probability. The method introduces a normal state identification model and a fault state identification model corresponding to each index, and is suitable for the operation machinery to be identified with a complex structure by carrying out density estimation construction on a normal sample database of the operation machinery and a fault sample database corresponding to each index. In addition, the method simultaneously introduces a fault state identification model and a normal state identification model, improves the accuracy of the health state identification result, is beneficial to timely diagnosis of the fault of the operation machine to be identified, and ensures the safe operation of the operation machine. In addition, due to the introduction of the unhealthy probability of the operation machine to be identified, the performance of the operation machine to be identified can be evaluated more comprehensively and accurately, namely, the fault degree of the operation machine to be identified can be quantified, and the operation machine to be identified which does not reach the fault state but deviates from the optimal performance state can be evaluated.
In addition, the method is based on historical maintenance record information, a fault sample database and a fault state identification model are respectively established for different indexes, and multi-dimensional characteristics related to faults can be extracted, so that the capability of capturing known fault types is effectively improved, the change trend of the data characteristics can be detected from the time sequence corresponding to each index of the operation machine to be identified, the health state of the operation machine to be identified can be identified in an absolute numerical value, and the capability of identifying faults is effectively improved.
On the basis of the above embodiment, the method for identifying the health state of the working machine according to the embodiment of the present invention, where the identifying the comprehensive health state of the working machine to be identified based on the unhealthy probability and the fault probability includes:
determining a comprehensive fault probability of the to-be-identified work machine based on the unhealthy probability and the fault probability;
determining a health status score of the work machine to be identified based on the comprehensive fault probability;
and determining a comprehensive health state identification result of the to-be-identified working machine based on the health state score of the to-be-identified working machine.
Specifically, in the embodiment of the present invention, when the comprehensive health state of the work machine to be identified is identified according to the unhealthy probability of the work machine to be identified and the fault probability of the work machine to be identified under each index, the comprehensive fault probability of the work machine to be identified may be determined according to the unhealthy probability of the work machine to be identified and the fault probability of the work machine to be identified under each index. For example, the probability of failure of the work machine to be identified under each index may be represented as PfiI is the ith index of the operation machine to be identified, the value of i is 1,2, the diameter of m is a constant value, and the type of the operation machine to be identified is determined. The fault probability of the operation machine to be identified under each index can be fused by the function to obtain an initial fusion result, which can be expressed as Pf＝Fun(Pf1,Pf2,..Pfm). The fusion mode may be a weighted summation mode, or may be other modes, and is not limited in particular here.
And then carrying out secondary fusion on the initial fusion result and the unhealthy probability of the operation machine to be identified to obtain the comprehensive fault probability. For example, the probability of unhealthy work machines to be identified may be represented as PnThen the composite failure probability can be expressed as P ═ f (P)f，Pnθ), θ is a fusion parameter, which may be a constant. The second-order fusion may be a weighted summation, and θ may be used as a weight coefficient of the initial fusion result, or may be used in other manners, which is not specifically limited herein.
Thereafter, a health status score of the work machine to be identified may be determined by integrating the failure probabilities. The higher the composite fault probability is, the lower the health status score is, and the health score can be obtained by establishing a linear or nonlinear mapping relationship between the composite fault probability and the health score. In addition, the present invention may be implemented by other methods, and is not limited in detail herein.
And finally, according to the health state score of the operation machine to be identified, the comprehensive health state identification result of the operation machine to be identified can be determined. Here, the correspondence between each health status score and the integrated health status recognition result may be determined in advance, and then the corresponding integrated health status recognition result may be determined by the correspondence. It is understood that the integrated health status recognition result is the integrated health status of the work machine to be recognized, which is recognized by the work machine health status recognition method provided in the embodiment of the present invention, and may include one of a healthy status, a sub-healthy status, a faulty status, a minor fault, a medium fault, and a serious fault, which is not limited herein.
In the embodiment of the invention, the comprehensive fault probability for representing the health state of the operation machine to be identified can be determined through two times of fusion actions, the health state score is further determined through the comprehensive fault probability, the comprehensive health state identification result of the operation machine to be identified is determined through the health state score, and the accuracy of the comprehensive health state identification result can be ensured.
On the basis of the above embodiment, the method for identifying the health state of the working machine according to the embodiment of the present invention, where the determination of the comprehensive health state identification result of the working machine to be identified based on the health state score of the working machine to be identified includes:
determining a plurality of health state scoring threshold intervals of the to-be-identified working machine; wherein, the plurality of health state scoring threshold intervals are all corresponding to different evaluation labels;
and determining a target health state score threshold interval in which the health state score is located in the plurality of health state score threshold intervals, and determining the comprehensive health state identification result based on an evaluation label corresponding to the target health state score threshold interval.
Specifically, in the embodiment of the present invention, when determining the health status identification result of the work machine to be identified, a plurality of health status score threshold intervals in which the work machine to be identified is located may be determined first. Different evaluation labels are corresponding to all the health state scoring threshold value intervals. The evaluation label may be one of the states of health, sub-health, fault, minor fault, medium fault, and major fault shown in the above embodiments.
For example, the health status score threshold interval may include: the evaluation labels corresponding to the health state scoring threshold intervals are respectively healthy, sub-healthy, faulty and serious faults, wherein the evaluation labels are greater than or equal to 80 points, less than 80 points and greater than or equal to 60 points, less than 60 points and greater than or equal to 40 points and less than 40 points.
Then, a target health state scoring threshold interval in which the health state score of the to-be-identified working machine is located in each health state scoring threshold interval can be determined, and a comprehensive health state identification result of the to-be-identified working machine is determined through an evaluation label corresponding to the target health state scoring threshold interval. For example, if the health score of the work machine to be identified is 65 points, the target health score threshold interval is less than 80 points and not less than 60 points, and the comprehensive health status identification result of the work machine to be identified is a sub-health status.
In the embodiment of the invention, a plurality of health state scoring threshold intervals of the operating machine to be identified are introduced, and the comprehensive health state identification result of the operating machine to be identified is determined by judging the target health state scoring threshold interval in which the health state scoring of the operating machine to be identified is positioned, so that the identification efficiency and the identification accuracy can be improved.
On the basis of the above embodiment, in the method for identifying the health state of the working machine provided in the embodiment of the present invention, the fault state identification model corresponding to each index is constructed based on density estimation of a fault sample database corresponding to each index of the working machine, and the working machine carries fault maintenance record information; the fault sample database carries time sequence fault data characteristics; accordingly, the number of the first and second electrodes,
the fault state identification model corresponding to each index is constructed based on the following method:
and modeling the fault sample database by adopting a density estimation algorithm based on the time sequence fault data characteristics to obtain the fault state identification model.
Specifically, in the embodiment of the present invention, the fault sample database carries a time series fault data feature, that is, a feature of time series fault data acquired by a sensor corresponding to a corresponding index when the working machine has a fault under the corresponding index, where the time series fault data is acquired by arranging fault data according to the acquisition time.
Furthermore, when the fault state identification model corresponding to each index is constructed, the fault sample database can be modeled by adopting a density estimation algorithm through the time sequence fault data characteristics, and the fault state identification model can be obtained. The density estimation algorithm may adopt a conventional density estimation algorithm, and is not limited in particular here.
In the embodiment of the invention, the density estimation algorithm is adopted to realize the construction of the fault state identification model, and the time sequence fault data characteristics carried by the fault sample database are utilized, so that the obtained fault state identification model is more accurate and has better performance.
On the basis of the above embodiment, in the method for identifying the health state of the working machine provided in the embodiment of the present invention, the normal state identification model is constructed by modeling the normal sample database by using a density estimation algorithm based on the time sequence normal data characteristics carried by the normal sample database of the working machine.
Specifically, in the embodiment of the present invention, the normal sample database carries a time sequence normal data feature, that is, a feature of time sequence normal data acquired by a sensor corresponding to each index of the operation machine in a historical normal state, where the time sequence normal data is acquired by arranging normal data according to the acquisition time.
Furthermore, when the normal state identification model is constructed, the normal sample database can be modeled by adopting a density estimation algorithm through the time sequence normal data characteristics, and the normal state identification model can be obtained. The density estimation algorithm may adopt a conventional density estimation algorithm, and is not limited in particular here.
In the embodiment of the invention, the density estimation algorithm is adopted to realize the construction of the normal state identification model, and the time sequence normal data characteristics carried by the normal sample database are utilized, so that the obtained normal state identification model is more accurate and has better performance.
Based on the above embodiments, the method for identifying the health state of the working machine provided in the embodiments of the present invention includes a parameterized density estimation algorithm of a known data distribution and/or a non-parameterized density estimation algorithm of an unknown data distribution.
In particular, in the embodiment of the present invention, the density estimation algorithm used may include a parameterized density estimation algorithm of a known data distribution and/or a non-parameterized density estimation algorithm of an unknown data distribution. That is, for data with known data distribution in the fault sample database or the normal sample database, a parameterized density estimation algorithm with known data distribution may be adopted, and for data with unknown data distribution in the fault sample database or the normal sample database, a non-parameterized density estimation algorithm with unknown data distribution may be adopted.
The parameterized density estimation algorithm of the known data distribution may include a maximum likelihood estimation algorithm, and the unparameterized density estimation algorithm of the unknown data distribution may include a kernel density estimation algorithm, which is not limited herein.
In the embodiment of the invention, a specific density estimation algorithm is provided, the modeling of a normal sample database or a fault sample database of various data distributions can be met, and the smooth implementation of the modeling is ensured.
On the basis of the foregoing embodiment, the method for identifying a health state of a working machine according to an embodiment of the present invention, which models the fault sample database by using a density estimation algorithm based on the time series fault data characteristics to obtain the fault state identification model, includes:
if an overlapping area exists in data distribution between the fault sample database and the normal sample database, cutting the fault sample database by adopting preset cutting parameters;
based on the time sequence fault data characteristics, modeling a result obtained by cutting by adopting a density estimation algorithm to obtain the fault state identification model;
and the preset cutting parameters are obtained based on a calculation mode of the comprehensive fault probability.
Specifically, in the embodiment of the present invention, when the fault state identification model is constructed, if there is an overlapping area in data distribution between the fault sample database and the normal sample database, the fault sample database may be cut by using preset cutting parameters. The preset cutting parameters may include a preset cutting size and a preset cutting mode.
Here, the preset cutting parameter is determined by a calculation method of the total failure probability, and can be obtained by the calculation method of the total failure probability. For example, the preset cutting parameter may be θ in the above-described embodiment.
Therefore, the result obtained by cutting can be modeled by adopting a density estimation algorithm through the time sequence fault data characteristics, and a fault state identification model is obtained.
In the embodiment of the invention, when an overlapping area exists between the fault sample database and the normal sample database in the data distribution, the fault sample database can be cut, the preset cutting parameters depend on the calculation method of the comprehensive fault probability, and the accuracy and consistency of the identification result of the health state of the working machine can be improved.
On the basis of the foregoing embodiment, the method for identifying a health status of a working machine according to an embodiment of the present invention, where the acquiring of a time-series detection data characteristic corresponding to each index of the working machine to be identified includes:
determining detection parameters collected by various sensors installed on the operation machine to be identified;
determining each index based on the detection parameters and the working principle of each system included in the operation machine to be identified;
performing feature extraction on the time sequence detection data acquired by the sensor corresponding to each index to obtain a feature extraction result;
and determining the time sequence detection data characteristics based on the characteristic extraction result.
Specifically, in the embodiment of the present invention, when acquiring the time sequence detection data characteristics corresponding to each index of the work machine to be identified, detection parameters acquired by each sensor installed on the work machine to be identified may be determined, each sensor may include a temperature sensor, a pressure sensor, and the like, and the detection parameters may include a water temperature, an oil temperature, a hydraulic cylinder pressure, and the like. And then, according to the working principle of each system contained in the operation machine to be identified, the detection parameters are processed by combining with a theoretical formula followed by each system to form each index of the operation machine to be identified.
It is understood that, when determining each index, the index may be implemented by using the same type of working machine as the working machine to be identified, and is not limited specifically here.
Then, feature extraction can be performed on the time sequence detection data acquired by the sensor corresponding to each index, so that a feature extraction result is obtained. For example, the feature extraction result may be a time domain feature, a frequency domain feature, and the like, and the time domain feature may include statistical features such as an extremum, a slope, a variance, and a standard deviation of the time series detection data, and is not limited specifically herein.
And finally, determining the time sequence detection data characteristics corresponding to each index according to the characteristic extraction result. Here, the feature extraction result may be directly used as the time-series detection data feature corresponding to each index, or the feature extraction result may be filtered, for example, by a correlation coefficient between features, and the time-series detection data feature corresponding to each index may be determined from the result obtained by the filtering.
In the embodiment of the invention, when the time sequence detection data characteristics corresponding to each index of the operation machine to be identified are obtained, each index is determined by combining the working principle of each system contained in the operation machine to be identified, and the time sequence detection data collected by the sensor corresponding to each index is further subjected to characteristic extraction to obtain the characteristic extraction result, so that the time sequence detection data characteristics corresponding to each index of the operation machine to be identified are obtained, and therefore, a characteristic basis can be provided for the realization of the operation machine health state identification method.
Fig. 2 is a schematic diagram of a complete flow of a method for identifying a health status of a work machine according to an embodiment of the present invention, as shown in fig. 2, the method includes:
1) and determining each index based on the working principle of the operation machine to be identified.
2) And determining a normal sample database and a fault sample database based on the maintenance record information.
3) And extracting the characteristics of the data in the normal sample database and the fault sample database to respectively obtain the time sequence normal data characteristics and the time sequence fault data characteristics.
4) And if an overlapping area exists between the fault sample database and the normal sample database in the data distribution, cutting the fault sample database by adopting preset cutting parameters.
5) And modeling by adopting a density estimation algorithm through a fault sample database corresponding to each index to obtain each fault state identification model, and determining the fault probability of the operation machine to be identified under each index through each fault state identification model.
6) And modeling by adopting a density estimation algorithm through a normal sample database to obtain a normal state identification model, and determining the unhealthy probability of the operation machine to be identified through the normal state identification model.
It should be noted that, the execution of step 5) and step 6) are independent from each other, and the execution order of the two steps may be adjusted as needed, for example, step 5) may be executed first and then step 6) may be executed, or step 6) may be executed first and then step 5) may be executed, and steps 5) and 6) may also be executed at the same time, which is not limited specifically herein.
7) And determining the comprehensive fault probability of the operation machine to be identified according to the unhealthy probability and the fault probability of the operation machine to be identified under each index.
8) And determining the health state score of the operation machine to be identified according to the comprehensive fault probability, and determining the comprehensive health state identification result of the operation machine to be identified according to the health state score.
Fig. 3 is a graph showing a change in the total failure probability and the state of health score of a certain work machine according to the operating time length (man-hour/h) of the work machine. Curve 1 is a change curve of the health status score with the operating duration (man-hour/h) of the work machine, and curve 2 is a change curve of the comprehensive failure probability with the operating duration (man-hour/h) of the work machine. As can be seen from fig. 3, the combined failure probability is inversely related to the trend of the health score.
Fig. 4 is a graph showing changes in 5 indexes of the work machine in fig. 3 according to the operation time period (man-hour/h) of the work machine. As can be seen from fig. 4, 5 indexes change at 2800 man-hours, one of the indexes rises, the other falls, and the other 3 indexes are relatively stable. As can be seen in connection with fig. 3, the health status score curve decreases after 2800 man hours, and the overall failure probability increases. The maintenance record of the operation machine shows that the machine 3400 man-hour equipment engine oil nozzle damages, leads to the engine to hold the car, fall the speed, troubleshooting after changing the oil nozzle. The algorithm detects the fault in advance, forecasts the fault in advance and accurately evaluates the health state of the equipment.
As shown in fig. 5, on the basis of the above embodiment, an embodiment of the present invention provides a work machine health status recognition apparatus, including:
the characteristic acquisition module 51 is used for acquiring time sequence detection data characteristics corresponding to all indexes of the operation machine to be identified;
the unhealthy probability determining module 52 is configured to perform density estimation on the time sequence detection data features corresponding to the respective indicators based on a normal state identification model, and determine an unhealthy probability of the work machine to be identified;
a failure probability determination module 53, configured to perform density estimation on the time sequence detection data features corresponding to each index based on the failure state identification model corresponding to each index, and determine a failure probability of the work machine to be identified under each index;
and the state identification module 54 is used for identifying the comprehensive health state of the work machine to be identified based on the unhealthy probability and the fault probability.
On the basis of the foregoing embodiment, in the working machine health status identification apparatus provided in the embodiment of the present invention, the status identification module is specifically configured to:
determining the comprehensive fault probability of the operation machine to be identified based on the unhealthy probability and the fault probability of the operation machine to be identified under each index;
determining a health status score of the work machine to be identified based on the comprehensive fault probability;
and determining a comprehensive health state identification result of the to-be-identified working machine based on the health state score of the to-be-identified working machine.
On the basis of the foregoing embodiment, in the working machine health status identification apparatus provided in the embodiment of the present invention, the status identification module is further configured to:
determining a plurality of health state scoring threshold intervals of the to-be-identified working machine; wherein, the plurality of health state scoring threshold intervals are all corresponding to different evaluation labels;
and determining a target health state score threshold interval in which the health state score is located in the plurality of health state score threshold intervals, and determining the comprehensive health state identification result based on an evaluation label corresponding to the target health state score threshold interval.
On the basis of the above embodiment, in the operating machine health status identification device provided in the embodiment of the present invention, the fault status identification model corresponding to each index is constructed based on density estimation performed on a fault sample database corresponding to each index of the operating machine, and the operating machine carries fault maintenance record information; the fault sample database carries time sequence fault data characteristics; accordingly, the number of the first and second electrodes,
the fault state identification model building module is used for:
and modeling the fault sample database by adopting a density estimation algorithm based on the time sequence fault data characteristics to obtain the fault state identification model.
On the basis of the above embodiments, the work machine health status identification device provided in the embodiments of the present invention includes a density estimation algorithm including a parameterized density estimation algorithm of a known data distribution and/or an unparameterized density estimation algorithm of an unknown data distribution.
On the basis of the above embodiment, the work machine health status identification device provided in the embodiment of the present invention further includes a normal status identification model building module, configured to:
and obtaining the normal state identification model by adopting a density estimation algorithm based on the time sequence normal data characteristics carried by a normal sample database of the working machine.
On the basis of the foregoing embodiment, in the work machine health status identification apparatus provided in the embodiment of the present invention, the fault status identification model building module is configured to:
if an overlapping area exists in data distribution between the fault sample database and the normal sample database, cutting the fault sample database by adopting preset cutting parameters;
based on the time sequence fault data characteristics, modeling a result obtained by cutting by adopting a density estimation algorithm to obtain the fault state identification model;
and the preset cutting parameters are obtained based on a calculation mode of the comprehensive fault probability.
On the basis of the foregoing embodiment, in the working machine health status identification apparatus provided in the embodiment of the present invention, the feature acquisition module is specifically configured to:
determining detection parameters collected by various sensors installed on the operation machine to be identified;
determining each index based on the detection parameters and the working principle of each system included in the operation machine to be identified;
performing feature extraction on the time sequence detection data acquired by the sensor corresponding to each index to obtain a feature extraction result;
and determining the time sequence detection data characteristics based on the characteristic extraction result.
Specifically, the functions of the modules in the working machine health status identification apparatus provided in the embodiment of the present invention correspond to the operation flows of the steps in the embodiments of the methods one to one, and the achieved effects are also consistent, for which reference is made to the embodiments described above, which are not described in detail in the embodiments of the present invention.
On the basis of the above embodiments, an embodiment of the present invention provides a working machine including: the working machine health status recognition device provided in each of the above embodiments is configured to provide the working machine with a function of recognizing the health status of the working machine.
Fig. 6 illustrates a physical structure diagram of an electronic device, which may include, as shown in fig. 6: a processor (processor)610, a communication Interface (Communications Interface)620, a memory (memory)630 and a communication bus 660, wherein the processor 610, the communication Interface 620 and the memory 630 communicate with each other via the communication bus 640. Processor 610 may invoke logic instructions in memory 630 to perform the method of work machine health identification provided in the various embodiments described above, the method comprising: acquiring time sequence detection data characteristics corresponding to all indexes of the operation machine to be identified; based on a normal state identification model, performing density estimation on the time sequence detection data characteristics corresponding to each index, and determining the unhealthy probability of the operation machine to be identified; performing density estimation on the time sequence detection data characteristics corresponding to the indexes based on the fault state identification model corresponding to each index, and determining the fault probability of the to-be-identified operation machine under each index; and identifying the comprehensive health state of the to-be-identified working machine based on the unhealthy probability and the fault probability.
In addition, the logic instructions in the memory 630 may be implemented in the form of software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, the present disclosure also provides a computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions that, when executed by a computer, enable the computer to perform a method for identifying a health status of a work machine, the method comprising: acquiring time sequence detection data characteristics corresponding to all indexes of the operation machine to be identified; based on a normal state identification model, performing density estimation on the time sequence detection data characteristics corresponding to each index, and determining the unhealthy probability of the operation machine to be identified; performing density estimation on the time sequence detection data characteristics corresponding to the indexes based on the fault state identification model corresponding to each index, and determining the fault probability of the to-be-identified operation machine under each index; and identifying the comprehensive health state of the to-be-identified working machine based on the unhealthy probability and the fault probability.
In yet another aspect, the present disclosure also provides a non-transitory computer-readable storage medium having stored thereon a computer program that, when executed by a processor, is implemented to perform a method of work machine health identification provided in the above embodiments, the method comprising: acquiring time sequence detection data characteristics corresponding to all indexes of the operation machine to be identified; based on a normal state identification model, performing density estimation on the time sequence detection data characteristics corresponding to each index, and determining the unhealthy probability of the operation machine to be identified; performing density estimation on the time sequence detection data characteristics corresponding to the indexes based on the fault state identification model corresponding to each index, and determining the fault probability of the to-be-identified operation machine under each index; and identifying the comprehensive health state of the to-be-identified working machine based on the unhealthy probability and the fault probability.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on the understanding, the above technical solutions substantially or otherwise contributing to the prior art may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the various embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (10)

1. A work machine state of health recognition method, comprising:
acquiring time sequence detection data characteristics corresponding to all indexes of the operation machine to be identified;
based on a normal state identification model, performing density estimation on the time sequence detection data characteristics corresponding to each index, and determining the unhealthy probability of the operation machine to be identified;
performing density estimation on the time sequence detection data characteristics corresponding to the indexes based on the fault state identification model corresponding to each index, and determining the fault probability of the to-be-identified operation machine under each index;
and identifying the comprehensive health state of the to-be-identified working machine based on the unhealthy probability and the fault probability.
2. The work machine state of health identification method of claim 1, wherein identifying the integrated state of health of the work machine to be identified based on the probability of unhealthy and the probability of failure comprises:
determining a comprehensive fault probability of the to-be-identified work machine based on the unhealthy probability and the fault probability;
determining a health status score of the work machine to be identified based on the comprehensive fault probability;
and determining a comprehensive health state identification result of the to-be-identified working machine based on the health state score of the to-be-identified working machine.
3. The work machine state of health identification method of claim 2, wherein determining the integrated state of health identification result for the work machine to be identified based on the state of health score for the work machine to be identified comprises:
determining a plurality of health state scoring threshold intervals of the to-be-identified working machine; wherein, the plurality of health state scoring threshold intervals are all corresponding to different evaluation labels;
and determining a target health state score threshold interval in which the health state score is located in the plurality of health state score threshold intervals, and determining the comprehensive health state identification result based on an evaluation label corresponding to the target health state score threshold interval.
4. The method according to claim 1, wherein the fault state identification model corresponding to each index is constructed based on density estimation of a fault sample database corresponding to each index of a working machine, and the working machine carries fault maintenance record information; the fault sample database carries time sequence fault data characteristics; accordingly, the number of the first and second electrodes,
the fault state identification model corresponding to each index is constructed based on the following method:
and modeling the fault sample database by adopting a density estimation algorithm based on the time sequence fault data characteristics to obtain the fault state identification model.
5. The work machine state of health identification method of claim 4, wherein the density estimation algorithm comprises a parameterized density estimation algorithm of a known data distribution and/or an unparameterized density estimation algorithm of an unknown data distribution.
6. The work machine health status identification method of claim 4, wherein the normal status identification model is constructed by modeling a normal sample database of the work machine using a density estimation algorithm based on time series normal data characteristics carried by the normal sample database.
7. The work machine state of health identification method of claim 6, wherein said modeling said fault sample database using a density estimation algorithm based on said time series fault data characteristics to obtain said fault state identification model comprises:
if an overlapping area exists in data distribution between the fault sample database and the normal sample database, cutting the fault sample database by adopting preset cutting parameters;
based on the time sequence fault data characteristics, modeling a result obtained by cutting by adopting a density estimation algorithm to obtain the fault state identification model;
and the preset cutting parameters are obtained based on a calculation mode of the comprehensive fault probability.
8. The method for identifying the health status of the working machine according to any one of claims 1 to 7, wherein the acquiring the time-series detection data characteristic corresponding to each index of the working machine to be identified includes:
determining detection parameters collected by various sensors installed on the operation machine to be identified;
determining each index based on the detection parameters and the working principle of each system included in the operation machine to be identified;
performing feature extraction on the time sequence detection data acquired by the sensor corresponding to each index to obtain a feature extraction result;
and determining the time sequence detection data characteristics based on the characteristic extraction result.
9. A work machine state of health recognition apparatus, comprising:
the characteristic acquisition module is used for acquiring time sequence detection data characteristics corresponding to all indexes of the operation machine to be identified;
the unhealthy probability determining module is used for performing density estimation on the time sequence detection data characteristics corresponding to the indexes based on a normal state identification model to determine the unhealthy probability of the operation machine to be identified;
the fault probability determination module is used for performing density estimation on the time sequence detection data characteristics corresponding to each index based on the fault state identification model corresponding to each index, and determining the fault probability of the operation machine to be identified under each index;
and the state identification module is used for identifying the comprehensive health state of the to-be-identified working machine based on the unhealthy probability and the fault probability.
10. A work machine, comprising: the work machine state of health identification device of claim 9.
CN202210169923.4A 2022-02-23 2022-02-23 Working machine health state identification method and device and working machine Withdrawn CN114548175A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210169923.4A CN114548175A (en) 2022-02-23 2022-02-23 Working machine health state identification method and device and working machine

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210169923.4A CN114548175A (en) 2022-02-23 2022-02-23 Working machine health state identification method and device and working machine

Publications (1)

Publication Number Publication Date
CN114548175A true CN114548175A (en) 2022-05-27

Family

ID=81677289

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210169923.4A Withdrawn CN114548175A (en) 2022-02-23 2022-02-23 Working machine health state identification method and device and working machine

Country Status (1)

Country Link
CN (1) CN114548175A (en)

Similar Documents

Publication Publication Date Title
CN111241154B (en) Storage battery fault early warning method and system based on big data
JP4175296B2 (en) Construction machine data processing apparatus and construction machine data processing method
US7464063B2 (en) Information processor, state judging unit and diagnostic unit, information processing method, state judging method and diagnosing method
CN113467420B (en) Method and device for detecting zone controller fault
CN106407589B (en) Fan state evaluation and prediction method and system
CN112254972B (en) Excavator oil temperature early warning method and device, server and excavator
CN110388315A (en) Oil transfer pump fault recognition method, apparatus and system based on Multi-source Information Fusion
CN115358155A (en) Power big data abnormity early warning method, device, equipment and readable storage medium
CN113902233A (en) Vehicle safety early warning method, big data platform device, vehicle-mounted terminal and vehicle
CN116665421A (en) Early warning processing method and device for mechanical equipment and computer readable storage medium
CN111858680B (en) System and method for rapidly detecting satellite telemetry time sequence data abnormity in real time
CN116292241A (en) Fault diagnosis early warning method and system for oil delivery pump unit
CN109960232A (en) The method that the selection method of leading auxiliary parameter and plant maintenance diagnose in advance
CN116125958A (en) Intelligent factory fault diagnosis and decision-making system based on digital twinning
CN116880454A (en) Intelligent diagnosis system and method for vehicle faults
CN117196066A (en) Intelligent operation and maintenance information analysis model
CN114462820A (en) Bearing state monitoring and health management system performance testing and optimizing method and system
CN111062827B (en) Engineering supervision method based on artificial intelligence mode
CN112524077A (en) Method, device and system for detecting fan fault
CN114548175A (en) Working machine health state identification method and device and working machine
CN111934903A (en) Docker container fault intelligent prediction method based on time sequence evolution genes
CN115186007A (en) Airborne data identification real-time display method and system for monitoring and reminding
CN114662977A (en) Method and system for detecting abnormity of motion state of offshore drilling platform and electronic equipment
CN114897225A (en) Accident prediction method and device for drilling operation, electronic device and storage medium
CN113685166A (en) Drilling accident early warning method and system

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
WW01 Invention patent application withdrawn after publication

Application publication date: 20220527

WW01 Invention patent application withdrawn after publication