CN112668243A - Motor filter screen blockage early warning method and device for rail train and related equipment - Google Patents
Motor filter screen blockage early warning method and device for rail train and related equipment Download PDFInfo
- Publication number
- CN112668243A CN112668243A CN202110008428.0A CN202110008428A CN112668243A CN 112668243 A CN112668243 A CN 112668243A CN 202110008428 A CN202110008428 A CN 202110008428A CN 112668243 A CN112668243 A CN 112668243A
- Authority
- CN
- China
- Prior art keywords
- motor
- early warning
- traction
- filter screen
- model
- 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.)
- Pending
Links
Images
Landscapes
- Electric Propulsion And Braking For Vehicles (AREA)
Abstract
The application discloses motor filter screen jam early warning method, device, equipment, system and computer readable storage medium of rail train, this method includes: acquiring a sampling value of a state parameter of a traction motor; the state parameters comprise motor temperature parameters and non-motor temperature parameters; calling an expert rule early warning model to calculate a rule early warning estimation value of the traction motor; the rule early warning estimation value is calculated and generated based on the checking result of the consistency of the state parameters of the traction motor and other traction motors; judging whether the value of the rule early warning estimation value is within a preset fault interval or not; if yes, judging that the motor filter screen blockage fault occurs in the traction motor and generating a blockage early warning instruction; if not, calling a machine learning early warning model to judge the blockage of the motor filter screen of the traction motor based on the matching of the non-motor temperature parameter and the motor temperature parameter. The application effectively improves the accuracy, the detection efficiency and the real-time performance of the early warning result of the blockage of the motor filter screen, and ensures the operation safety of the motor.
Description
Technical Field
The application relates to the technical field of fault detection, in particular to a motor filter screen blockage early warning method, device, equipment and system of a rail train and a computer readable storage medium.
Background
The traction motor is used as a power source of the rail transit vehicle, and plays a vital role in train safety. The motor filter screen is a very important link in the motor cooling system, can effectively filter the dust that has in the air-cooled system, if the motor filter screen accumulation dust is too much to take place to block up, must directly influence the radiating effect of motor, and the state and the performance of motor and inside device are greatly influenced, make traction motor have the potential safety hazard.
In practical application, the rail transit vehicle traction motor has poor working environment and different filter screen specifications, so that the frequency of motor filter screen blockage is very high, and the actual maintenance period is long. In order to improve the safety and reliability of the operation of the traction motor of the subway vehicle, the early warning and diagnosis of the blockage of the motor filter screen are very significant. In the related art, the motor temperature is often compared with a preset threshold value, so that the early warning accuracy of the motor temperature is greatly dependent on the setting reasonableness of the preset threshold value. Or a scheme of manual checking at regular intervals is adopted in the related technology, so that the real-time performance is poor, the execution efficiency is low, and manpower and material resources are wasted.
In view of the above, it is an important need for those skilled in the art to provide a solution to the above technical problems.
Disclosure of Invention
The application aims to provide a motor filter screen blockage early warning method, device, equipment, system and computer readable storage medium for a rail train, so that the accuracy and reliability of a motor filter screen blockage early warning result are effectively improved, the blockage detection efficiency and real-time performance are improved, and the motor operation safety is guaranteed.
In order to solve the technical problem, the application discloses a motor filter screen of rail train blocks up early warning method for the first aspect, include:
acquiring a sampling value of a state parameter of a traction motor in a rail train; the state parameters comprise motor temperature parameters and non-motor temperature parameters;
calling a preset expert rule early warning model to calculate a rule early warning estimation value of the traction motor; the rule early warning estimation value is calculated and generated based on a check result of the consistency of the state parameters of the traction motor and other traction motors;
judging whether the value of the rule early warning estimation value is within a preset fault interval or not;
if yes, judging that the motor filter screen blockage fault occurs in the traction motor, and generating a blockage early warning instruction;
if not, calling a preset machine learning early warning model to judge whether the motor filter screen is blocked by the traction motor; and the judgment result of the machine learning early warning model is calculated and generated based on the matching of the non-motor temperature parameter and the motor temperature parameter.
Optionally, the step of calling a preset expert rule early warning model to calculate a rule early warning estimation value of the traction motor includes:
determining characteristic parameters corresponding to the type of the traction motor from various state parameters according to a preset characteristic parameter corresponding relation;
respectively calculating consistency check values of the traction motor and other traction motors under the same working condition aiming at each characteristic parameter of the traction motor;
and carrying out weighted summation on the consistency check values of all the characteristic parameters of the traction motor so as to calculate the rule early warning estimation value of the traction motor.
Optionally, the calculating, for each characteristic parameter of the traction motor, a consistency check value of the traction motor under the same working condition includes:
respectively calculating consistency check values of each characteristic parameter of the traction motor and other traction motors under the same working condition based on a consistency check formula; the consistency check formula is as follows:
wherein, UiThe consistency check value of the ith characteristic parameter of the traction motor is obtained; y isiSampling values of the ith characteristic parameter of the traction motor; mu.s(other)iThe sampling average value of the ith characteristic parameter of other traction motors is obtained; y is(other)ijSampling values of ith characteristic parameters of jth other traction motors; k is the total number of traction motors.
Optionally, for the target model of traction motor, the determining process of the characteristic parameter correspondence includes:
acquiring sample data of the state parameters of the target model traction motor;
respectively calculating the association degree of each non-motor temperature parameter and the motor temperature parameter of the target type traction motor;
and determining the motor temperature parameters and a first preset number of non-motor temperature parameters with the highest degree of association as the characteristic parameters corresponding to the target model traction motor.
Optionally, the calculating the association degree between each non-motor temperature parameter of the target model traction motor and the motor temperature parameter respectively includes:
respectively calculating the association degrees of the non-motor temperature parameters and the motor temperature parameters of the traction motor of the target model according to an association degree calculation formula; the correlation calculation formula is as follows:
wherein, aiSampling values of non-motor temperature parameters of the target type traction motor; biFor the target model of traction motorThe sampled value of the motor temperature parameter; m is the sampling data quantity;
is aiAverage value of (d);
is b isiAverage value of (a).
Optionally, after the calculating the association degree between each non-motor temperature parameter and the motor temperature parameter of the target model traction motor respectively, the method further includes:
and determining corresponding weights for the characteristic parameters according to the relevance, so as to perform weighted summation calculation based on the weights when the step of performing weighted summation on the consistency check values of the characteristic parameters of the traction motor is performed.
Optionally, the training process of the machine learning early warning model includes:
acquiring sample data of state parameters when the motor filter screen is not blocked;
carrying out normalization processing on the sample data;
determining a network structure of a neural network model;
and training the neural network model by taking the non-motor temperature parameter in the sample data as input and the motor temperature parameter as output so as to generate the machine learning early warning model.
Optionally, call predetermined machine learning early warning model and carry out motor filter screen jam judgement to this traction motor, include:
determining a group of sampling values of the state parameters as a target data group according to a plurality of groups of sampling values of the state parameters of the traction motor;
carrying out normalization processing on the target data set;
inputting the target data group into the machine learning early warning model so as to calculate and generate an estimated value of the motor temperature according to the non-motor temperature parameter in the target data group;
calculating the accumulated deviation of the estimated value of the motor temperature and the sampling value of the motor temperature in the target data group;
calculating a model early warning estimation value according to the accumulated deviation and the duration time of the accumulated deviation; the model early warning estimation value is respectively in positive correlation change with the accumulated deviation and the duration;
judging whether the magnitude of the model early warning estimation value is larger than a preset fault threshold value or not;
if yes, judging that the motor filter screen blockage fault occurs in the traction motor, and generating a blockage early warning instruction.
Optionally, the determining, according to the plurality of groups of sampling values of the state parameters of the traction motor, a group of sampling values of the state parameters as a target data group includes:
respectively calculating a motor temperature mean value and a motor temperature standard deviation in each group of state parameter sampling values of the traction motor;
taking the second preset quantity group data with the lowest motor temperature mean value as a data group to be selected;
and determining a group of data with the minimum standard deviation of the motor temperature in the data group to be selected as the target data group.
Optionally, after it is determined that the motor filter screen blockage fault occurs in the traction motor, the method further includes:
weighting and summing the rule early warning estimation value and the model early warning estimation value to calculate a comprehensive early warning estimation value of the traction motor;
and determining the fault early warning grade corresponding to the size of the comprehensive early warning estimation value according to a preset grade corresponding relation table.
Optionally, the non-motor temperature parameter comprises:
ambient temperature, actual traction, actual braking force, motor speed, motor current.
In a second aspect, the application discloses rail train's motor filter screen blocks up early warning device includes:
the parameter acquisition module is used for acquiring a sampling value of a state parameter of a traction motor in the rail train; the state parameters comprise motor temperature parameters and non-motor temperature parameters;
the rule early warning module is used for calling a preset expert rule early warning model to calculate a rule early warning estimation value of the traction motor; the rule early warning estimation value is calculated and generated based on a check result of the consistency of the state parameters of the traction motor and other traction motors;
the rule judging module is used for judging whether the value of the rule early warning estimation value is within a preset fault interval or not;
the model early warning module is used for calling a preset machine learning early warning model to judge the blockage of the motor filter screen of the traction motor after judging that the value of the rule early warning estimation value is not within a preset fault interval; the judgment result of the machine learning early warning model is calculated and generated based on the matching of the non-motor temperature parameter and the motor temperature parameter;
the model training module is used for pre-training and generating the machine learning early warning model;
and the fault early warning module is used for generating a blockage early warning instruction after the rule judgment module judges that the value of the rule early warning estimation value is within a preset fault interval or the model early warning module judges that the motor filter screen is blocked.
Optionally, the rule early warning module is specifically configured to:
determining characteristic parameters corresponding to the type of the traction motor from various state parameters according to a preset characteristic parameter corresponding relation; respectively calculating consistency check values of the traction motor and other traction motors under the same working condition aiming at each characteristic parameter of the traction motor; and carrying out weighted summation on the consistency check values of all the characteristic parameters of the traction motor so as to calculate the rule early warning estimation value of the traction motor.
Optionally, the rule early warning module is specifically configured to:
respectively calculating consistency check values of each characteristic parameter of the traction motor and other traction motors under the same working condition based on a consistency check formula; the consistency check formula is as follows:
wherein, UiThe consistency check value of the ith characteristic parameter of the traction motor is obtained; y isiSampling values of the ith characteristic parameter of the traction motor; mu.s(other)iThe sampling average value of the ith characteristic parameter of other traction motors is obtained; y is(other)ijSampling values of ith characteristic parameters of jth other traction motors; k is the total number of traction motors.
Optionally, the rule pre-warning module is further configured to:
acquiring sample data of the state parameters of the target model traction motor; respectively calculating the association degree of each non-motor temperature parameter and the motor temperature parameter of the target type traction motor; and determining the motor temperature parameters and a first preset number of non-motor temperature parameters with the highest degree of association as the characteristic parameters corresponding to the target model traction motor.
Optionally, the rule early warning module is specifically configured to:
respectively calculating the association degrees of the non-motor temperature parameters and the motor temperature parameters of the traction motor of the target model according to an association degree calculation formula; the correlation calculation formula is as follows:
wherein, aiSampling values of non-motor temperature parameters of the target type traction motor; biSampling values of motor temperature parameters of the target type traction motor; m is the sampling data quantity;
is aiAverage value of (d);
is b isiAverage value of (a).
Optionally, the rule pre-warning module is further configured to:
and determining corresponding weights for the characteristic parameters according to the relevance, so as to perform weighted summation calculation based on the weights when the step of performing weighted summation on the consistency check values of the characteristic parameters of the traction motor is performed.
Optionally, the model training module is specifically configured to:
acquiring sample data of state parameters when the motor filter screen is not blocked; carrying out normalization processing on the sample data; determining a network structure of a neural network model; and training the neural network model by taking the non-motor temperature parameter in the sample data as input and the motor temperature parameter as output so as to generate the machine learning early warning model.
Optionally, the model early warning module is specifically configured to:
determining a group of sampling values of the state parameters as a target data group according to a plurality of groups of sampling values of the state parameters of the traction motor; carrying out normalization processing on the target data set; inputting the target data group into the machine learning early warning model so as to calculate and generate an estimated value of the motor temperature according to the non-motor temperature parameter in the target data group; calculating the accumulated deviation of the estimated value of the motor temperature and the sampling value of the motor temperature in the target data group; calculating a model early warning estimation value according to the accumulated deviation and the duration time of the accumulated deviation; the model early warning estimation value is respectively in positive correlation change with the accumulated deviation and the duration; judging whether the magnitude of the model early warning estimation value is larger than a preset fault threshold value or not; if yes, judging that the motor filter screen blockage fault occurs in the traction motor, and generating a blockage early warning instruction.
Optionally, the model early warning module is specifically configured to:
respectively calculating a motor temperature mean value and a motor temperature standard deviation in each group of state parameter sampling values of the traction motor; taking the second preset quantity group data with the lowest motor temperature mean value as a data group to be selected; and determining a group of data with the minimum standard deviation of the motor temperature in the data group to be selected as the target data group.
Optionally, the method further comprises:
the comprehensive early warning module is used for weighting and summing the rule early warning estimation value and the model early warning estimation value so as to calculate a comprehensive early warning estimation value of the traction motor; and determining the fault early warning grade corresponding to the size of the comprehensive early warning estimation value according to a preset grade corresponding relation table.
Optionally, the non-motor temperature parameter comprises:
ambient temperature, actual traction, actual braking force, motor speed, motor current.
In a third aspect, the application discloses an early warning device, comprising:
a memory for storing a computer program;
a processor for executing the computer program to implement the steps of any one of the above-mentioned motor screen clogging warning methods for a rail train.
In a fourth aspect, the application discloses a motor filter screen blockage early warning system of a rail train, which comprises a sensor group, ground human-computer interaction equipment and the early warning equipment;
the sensor group is used for sampling each state parameter of each traction motor in the rail train and sending the sampling value to the processor of the early warning equipment; the state parameters comprise motor temperature parameters and non-motor temperature parameters;
and the ground human-computer interaction equipment is in communication connection with the processor of the early warning equipment and is used for receiving and executing the blockage early warning instruction sent by the processor.
In a fifth aspect, the present application further discloses a computer readable storage medium, in which a computer program is stored, and the computer program is used to implement the steps of any one of the above-mentioned motor filter screen blockage warning methods for a rail train when being executed by a processor.
The application provides a rail train's motor filter screen blocks up early warning method includes: acquiring a sampling value of a state parameter of a traction motor in a rail train; the state parameters comprise motor temperature parameters and non-motor temperature parameters; calling a preset expert rule early warning model to calculate a rule early warning estimation value of the traction motor; the rule early warning estimation value is calculated and generated based on a check result of the consistency of the state parameters of the traction motor and other traction motors; judging whether the value of the rule early warning estimation value is within a preset fault interval or not; if yes, judging that the motor filter screen blockage fault occurs in the traction motor, and generating a blockage early warning instruction; if not, calling a preset machine learning early warning model to judge whether the motor filter screen is blocked by the traction motor; and the judgment result of the machine learning early warning model is calculated and generated based on the matching of the non-motor temperature parameter and the motor temperature parameter.
The motor filter screen blockage early warning method, device, equipment and system of the rail train and the computer readable storage medium have the advantages that: the fault detection method and the fault detection device have the advantages that the expert rule early warning model and the machine learning early warning model are successively utilized for fault detection, fault verification is respectively carried out from the two aspects of consistency of state parameters of other traction motors and matching of non-motor temperature parameters and motor temperature parameters, the early warning accuracy of motor filter screen blocking faults can be effectively improved, real-time online detection can be realized, the efficiency and timeliness of fault early warning and processing are improved, and further the safe operation guarantee of a railway train is greatly improved.
Drawings
In order to more clearly illustrate the technical solutions in the prior art and the embodiments of the present application, the drawings that are needed to be used in the description of the prior art and the embodiments of the present application will be briefly described below. Of course, the following description of the drawings related to the embodiments of the present application is only a part of the embodiments of the present application, and it will be obvious to those skilled in the art that other drawings can be obtained from the provided drawings without any creative effort, and the obtained other drawings also belong to the protection scope of the present application.
Fig. 1 is a flowchart of a method for early warning of a blockage of a motor filter screen of a rail train according to an embodiment of the present disclosure;
FIG. 2 is a flowchart of a method for calculating a rule early warning estimation value of a traction motor by using a rule judgment early warning model according to an embodiment of the present application;
fig. 3 is a flowchart of a method for determining a feature parameter correspondence relationship disclosed in an embodiment of the present application;
FIG. 4 is a flowchart of a method for training a machine learning early warning model disclosed in an embodiment of the present application;
FIG. 5 is a flowchart of a method for determining motor filter clogging using a machine learning early warning model according to an embodiment of the present disclosure;
fig. 6 is a motor filter screen blockage warning device of a rail train disclosed in an embodiment of the present application;
fig. 7 is a block diagram of a structure of an early warning device disclosed in an embodiment of the present application.
Detailed Description
The core of the application lies in providing a motor filter screen blockage early warning method, device, equipment, system and computer readable storage medium for the rail train, so that the accuracy and reliability of a motor filter screen blockage early warning result are effectively improved, the blockage detection efficiency and real-time performance are improved, and the motor operation safety is guaranteed.
In order to more clearly and completely describe the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application. 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 application.
The traction motor is used as a power source of the rail transit vehicle, and plays a vital role in train safety. The motor filter screen is a very important link in the motor cooling system, can effectively filter the dust that has in the air-cooled system, if the motor filter screen accumulation dust is too much to take place to block up, must directly influence the radiating effect of motor, and the state and the performance of motor and inside device are greatly influenced, make traction motor have the potential safety hazard.
In practical application, the rail transit vehicle traction motor has poor working environment and different filter screen specifications, so that the frequency of motor filter screen blockage is very high, and the actual maintenance period is long. In order to improve the safety and reliability of the operation of the traction motor of the subway vehicle, the early warning and diagnosis of the blockage of the motor filter screen are very significant. In the related art, the motor temperature is often compared with a preset threshold value, so that the early warning accuracy of the motor temperature is greatly dependent on the setting reasonableness of the preset threshold value. Or a scheme of manual checking at regular intervals is adopted in the related technology, so that the real-time performance is poor, the execution efficiency is low, and manpower and material resources are wasted.
In view of this, this application provides a motor filter screen of rail train blocks up early warning scheme, can effectively solve above-mentioned problem.
Referring to fig. 1, the embodiment of the application discloses a motor filter screen blockage early warning method for a rail train, which mainly comprises the following steps:
s101: acquiring a sampling value of a state parameter of a traction motor in a rail train; the state parameters include a motor temperature parameter and a non-motor temperature parameter.
Specifically, it should be pointed out that this application is carrying out jam fault detection and early warning in-process to the motor filter screen, has gathered track train last traction motor's state parameter at first. It should be noted that the state parameters acquired by the present application are motor filter screen blockage related parameters, that is, related parameters that may be affected by motor filter screen blockage to a certain extent; and, in particular, the state parameter in the present application includes not only the motor temperature but also other non-motor temperature parameters. As a specific example, the non-motor temperature parameters may specifically include the following five parameters:
ambient temperature, actual traction, actual braking force, motor speed, motor current.
It should be further noted that a plurality of traction motors are usually arranged on the rail train, and therefore, motor filter screen blockage fault detection needs to be carried out on each traction motor separately, so that the method and the device need to collect state parameters of each traction motor respectively. For a traction motor, each state parameter needs to be continuously detected in real time, and a large number of sampling values are obtained through continuous detection for many times.
S102: calling a preset expert rule early warning model to calculate a rule early warning estimation value of the traction motor; and calculating and generating the rule early warning estimation value based on the verification result of the consistency of the state parameters of the traction motor and other traction motors.
It should be noted that, in the embodiment of the present application, an expert rule early warning model is first used to determine the motor filter screen blockage fault of the traction motor. Consistency check is carried out on the state parameters of the traction motor and the state parameters of other traction motors, and a consistency check result can be used for calculating a rule early warning estimation value: the worse the consistency with the state parameters of other traction motors, the higher the fault possibility, and the larger the calculated rule early warning estimation value.
S103: judging whether the value of the rule early warning estimation value is within a preset fault interval or not; if yes, entering S104; if not, the process proceeds to S105.
In order to pre-judge whether the motor filter screen blockage fault occurs to the traction motor by utilizing the rule early warning estimation value, the application sets a preset fault interval of the rule early warning estimation value. For example, with IruleIndicating the rule early warning estimation value, setting [ TthAnd + ∞) is a preset fault interval, then when I isrule>TthAnd in time, the traction motor can be judged to have the motor filter screen blockage fault.
S104: and judging the motor filter screen blockage fault of the traction motor, and generating a blockage early warning instruction.
S105: calling a preset machine learning early warning model to judge whether the motor filter screen is blocked or not on the traction motor; and calculating and generating a judgment result of the machine learning early warning model based on the matching of the non-motor temperature parameter and the motor temperature parameter.
It should be noted that the machine learning early warning model is also trained in advance in the present application. For guaranteeing the accuracy of the early warning result, even if the regular early warning estimation value of the traction motor is not taken within the preset fault interval, the motor filter screen blockage fault can not be immediately judged, and the machine learning early warning model can be further utilized to carry out more precise and accurate judgment.
Further, considering that the accuracy of the rule early warning model is weaker than that of the machine learning early warning model, the threshold T of the preset fault interval of the rule early warning estimation valuethMay be specifically set to a large value so that when I is satisfiedrule>TthWhen the traction motor is used, the traction motor can be considered to have a serious motor filter screen blockage fault. When I isrule<TthAnd then, the machine learning early warning model can be utilized to detect the fault.
The motor filter screen of the rail train that this application embodiment provided blocks up early warning method includes: acquiring sampling values of state parameters of traction motors in a rail train; the state parameters comprise motor temperature parameters and non-motor temperature parameters; calling a preset expert rule early warning model to calculate a rule early warning estimation value of the traction motor aiming at each traction motor; the rule early warning estimation value is calculated and generated based on the checking result of the consistency of the state parameters of the traction motor and other traction motors; judging whether a traction motor with a regular early warning estimation value within a preset fault interval exists or not; if yes, judging that the corresponding traction motor has a motor filter screen blockage fault, and generating a blockage early warning instruction; if not, calling a preset machine learning early warning model to judge the blockage of the motor filter screen; and calculating and generating a judgment result of the machine learning early warning model based on the matching of the non-motor temperature parameter and the motor temperature parameter.
Therefore, the motor filter screen blockage early warning method for the rail train provided by the application utilizes the expert rule early warning model and the machine learning early warning model to detect faults, and carries out fault check from two aspects of consistency of state parameters of other traction motors and matching of non-motor temperature parameters and motor temperature parameters respectively, has strong applicability, is not only suitable for traction motors with various specifications, but also is not influenced by various working conditions and environmental temperatures, can effectively improve early warning accuracy of motor filter screen blockage faults, can realize real-time online detection, improves efficiency and timeliness of fault early warning and treatment, and further greatly improves safe operation guarantee of the rail train.
As a specific embodiment, on the basis of the above content, the process of calculating the rule early warning estimation value of the traction motor by calling the preset expert rule early warning model according to the motor filter screen blockage early warning method of the rail train provided by the embodiment of the present application may specifically refer to fig. 2:
s201: and determining characteristic parameters corresponding to the type of the traction motor from various state parameters according to the preset characteristic parameter corresponding relation.
Wherein, various traction motor models and corresponding characteristic parameters are recorded in the characteristic parameter corresponding relation. A person skilled in the art can set several state parameters for each type of traction motor as characteristic parameters participating in the consistency check on the basis of experience or experimental tests in advance.
S202: and respectively calculating consistency check values of the traction motor and other traction motors under the same working condition aiming at each characteristic parameter of the traction motor.
As a specific embodiment, when calculating the consistency check value of each characteristic parameter of the traction motor with respect to other traction motors under the same working condition, the following consistency check formula may be specifically adopted:
wherein, UiThe consistency check value of the ith characteristic parameter of the traction motor is obtained; y isiSampling values of the ith characteristic parameter of the traction motor; mu.s(other)iThe sampling average value of the ith characteristic parameter of other traction motors is obtained; y is(other)ijSampling values of ith characteristic parameters of jth other traction motors; k is the total number of traction motors.
S203: and carrying out weighted summation on the consistency check values of all the characteristic parameters of the traction motor so as to calculate the rule early warning estimation value of the traction motor.
Referring to fig. 3, fig. 3 is a flowchart of a method for determining a feature parameter corresponding relationship disclosed in the embodiment of the present application.
On the basis of the above, as a specific embodiment, for the target model of traction motor, the characteristic parameter corresponding relationship used in step S201 can be determined by the process shown in fig. 3:
s301: and acquiring sample data of the state parameters of the traction motor of the target model.
S302: and respectively calculating the association degree of each non-motor temperature parameter and the motor temperature parameter of the traction motor of the target model.
As a specific embodiment, when the association degrees of the non-motor temperature parameters and the motor temperature parameters of the target model traction motor are respectively calculated, the following association degree calculation formula may be specifically adopted:
wherein, aiSampling values of non-motor temperature parameters of a target model traction motor; biSampling values of motor temperature parameters of a target model traction motor; m is the sampling data quantity;
is aiAverage value of (d);
is b isiAverage value of (a).
S303: and determining the motor temperature parameters and the first preset number of non-motor temperature parameters with the highest relevance as the characteristic parameters corresponding to the target model traction motor.
For example, if the first preset number is 3, the 3 non-motor temperature parameters with the highest correlation degree with the motor temperature and the motor temperature parameter may be determined together as the characteristic parameters corresponding to the target model traction motor.
As a specific embodiment, on the basis of the above, after the association degree between each non-motor temperature parameter and the motor temperature parameter of the traction motor of the target model is calculated in step S302, a corresponding weight may be further determined for each characteristic parameter according to the association degree, so that when step S203 is performed, i.e., when the consistency check values of each characteristic parameter of the traction motor are weighted and summed, weighted and summed calculation is performed based on the weights.
It is easily understood that the greater the degree of association, the greater the corresponding weight value may be.
Referring to fig. 4, fig. 4 is a flowchart of a method for training a machine learning early warning model according to an embodiment of the present disclosure.
As a specific embodiment, on the basis of the above content, the training process of the machine learning early warning model of the motor filter screen blockage early warning method for the rail train provided by the embodiment of the present application may specifically refer to fig. 4:
s401: and acquiring sample data of the state parameters when the motor filter screen is not blocked.
For example, at ambient temperature y1iActual tractive effort y2iActual braking force y3iMotor speed y4iMotor current y5iMotor temperature y6iFor example, the 6 state parameters are obtained by constructing a state parameter vector set from a large number of sample data of the state parameters of the traction motor:
wherein i is 1, 2, …, n; n +1 is the total amount of data for each state parameter.
S402: and carrying out normalization processing on the sample data.
For example, each parameter data is dimensionless by using a normalization method as follows:
wherein y' is normalized data, and y is data before normalization; [ min (x), max (x) ] is a preset normalization interval.
Further, as a specific embodiment, after normalization is performed, considering that the data length n +1 is generally large and is inconvenient to be directly and completely input to the neural network model for training, data can be converted, a certain window length s is selected, and sample data with the original length of n +1 is grouped, wherein the length of each group of data is s:
s403: a network structure of the neural network model is determined.
A machine Learning early warning model is constructed by selecting a circular neural network, and the network layer number, neuron number of each layer, input and output dimension and other structural parameters of the neural network and important parameters such as Window _ size, Batch _ size, Learning _ rate, Loss, Steps and the like are determined.
S404: with non-motor temperature parameter (Y) in the sample data1_in、…、Ym_in) As input, and with a corresponding motor temperature parameter (Y)1_out、…、Ym_out) For output, the neural network model is trained to generate a machine learning early warning model.
Referring to fig. 5, fig. 5 is a flowchart of a method for determining motor filter screen clogging by using a machine learning early warning model according to an embodiment of the present application.
As a specific embodiment, the method for early warning of motor filter screen blockage of a rail train provided in the embodiment of the present application may specifically include, on the basis of the foregoing contents, the following steps when a preset machine learning early warning model is called to determine that a motor filter screen is blocked for the traction motor:
s501: and determining a group of sampling values of the state parameters as a target data group according to a plurality of groups of sampling values of the state parameters of the traction motor.
Specifically, similar to the training of the machine learning early warning model, the same window length s can be selected, original sample data is divided into multiple groups of data, and the length of each group of data is s.
S502: and carrying out normalization processing on the target data set.
Specifically, the process of normalization is also similar to the normalization process when training the machine learning early warning model.
S503: and inputting the target data group into a machine learning early warning model so as to calculate and generate an estimated value of the motor temperature according to the non-motor temperature parameters in the target data group.
The machine learning early warning model trained by the application is trained on the basis of sample data without motor filter screen blockage faults, so that a reasonable estimated value of the motor temperature can be calculated according to input non-motor temperature parameters. Therefore, if the motor filter screen of the traction motor is blocked at the moment, the input sampling value of the motor temperature is deviated from the estimated value of the motor temperature calculated by the machine learning early warning model.
S504: and calculating the accumulated deviation of the estimated value of the motor temperature and the sampling value of the motor temperature in the target data set.
S505: calculating a model early warning estimation value according to the accumulated deviation and the duration time of the accumulated deviation; the model early warning estimation value is respectively in positive correlation change with the accumulated deviation and the duration.
Because the input data is not a single data but a group of data, the processing process of the machine learning early warning model is a continuous process, and in the process, the deviation between the estimated value of the motor temperature and the sampling value is accumulated continuously. Based on the accumulated deviation and the occurrence duration, the embodiment of the present application may calculate to obtain the model early warning estimation value, for example, a method of weighted calculation may be specifically adopted.
S506: judging whether the magnitude of the early warning estimation value of the model is larger than a preset fault threshold value or not; if yes, the process proceeds to S507.
S507: and judging the motor filter screen blockage fault of the traction motor, and generating a blockage early warning instruction.
As a specific embodiment, when determining a set of sampling values of the state parameter as the target data set according to a plurality of sets of sampling values of the state parameter of the traction motor in step S501, the method may specifically include the following steps:
respectively calculating a motor temperature mean value and a motor temperature standard deviation in each group of state parameter sampling values of the traction motor;
taking the second preset quantity group data with the lowest motor temperature mean value as a data group to be selected;
and determining a group of data with the minimum standard deviation of the motor temperature in the data group to be selected as a target data group.
Therefore, the data of the group with the low mean value and the small standard deviation of the motor temperature is selected, and the data of the group with the least possibility of filter screen blockage is selected from the data of the group of the traction motor. If the filter screen of the traction motor is judged to be blocked based on the group of data, the fact that the filter screen of the traction motor is blocked can be fully determined.
As a specific embodiment, the method for early warning of motor filter screen blockage of a rail train according to the embodiment of the present application, on the basis of the foregoing, further includes, after determining that a motor filter screen blockage fault occurs in the traction motor:
carrying out weighted summation on the regular early warning estimation value and the model early warning estimation value to calculate a comprehensive early warning estimation value of the traction motor;
and determining the fault early warning grade corresponding to the size of the comprehensive early warning estimation value according to a preset grade corresponding relation table.
Specifically, the rule early warning estimation value is recorded as IruleSetting the weight as Wrule(ii) a Record model early warning estimate as value ImodelSetting the weight as Wmodel(ii) a The calculation of the comprehensive early warning estimation value Z is as follows:
Z＝Wrule·Irule+Wmodel·Imodel。
referring to fig. 6 shows, the embodiment of the application discloses motor filter screen of rail train blocks up early warning device mainly includes:
the parameter acquisition module 601 is used for acquiring a sampling value of a state parameter of a traction motor in a rail train; the state parameters comprise motor temperature parameters and non-motor temperature parameters;
the rule early warning module 602 is used for calling a preset expert rule early warning model to calculate a rule early warning estimation value of the traction motor; the rule early warning estimation value is calculated and generated based on the checking result of the consistency of the state parameters of the traction motor and other traction motors;
a rule judging module 603, configured to judge whether a value of the rule early warning estimation value is within a preset fault interval;
the model early warning module 604 is used for calling a preset machine learning early warning model to judge whether the motor filter screen of the traction motor is blocked after judging that the early warning estimation value of the rule is not within a preset fault interval; the judgment result of the machine learning early warning model is calculated and generated based on the matching of the non-motor temperature parameter and the motor temperature parameter;
a model training module 605, configured to pre-train to generate a machine learning early warning model;
and the fault early warning module 606 is configured to generate a blockage early warning instruction after the rule judgment module 603 determines that the value of the rule early warning estimation value is within a preset fault interval, or after the model early warning module 604 determines that a motor filter screen blockage fault occurs.
Therefore, the motor filter screen blockage early warning device for the rail train disclosed by the embodiment of the application utilizes the expert rule early warning model and the machine learning early warning model to carry out fault detection, and carries out fault verification from two aspects of state parameter consistency with other traction motors and matching of non-motor temperature parameters and motor temperature parameters respectively, has strong applicability, is not only suitable for traction motors with various specifications, but also is not influenced by various working conditions and environment temperatures, can effectively improve the early warning accuracy of motor filter screen blockage faults, can carry out real-time online detection, improves the efficiency and timeliness of fault early warning and treatment, and further greatly improves the safe operation guarantee of the rail train.
As a specific embodiment, the motor filter screen blockage early warning device of the rail train disclosed in the embodiment of the present application is based on the above contents, and the rule early warning module 602 is specifically configured to:
determining characteristic parameters corresponding to the type of the traction motor from various state parameters according to a preset characteristic parameter corresponding relation; respectively calculating consistency check values of the traction motor and other traction motors under the same working condition aiming at each characteristic parameter of the traction motor; and carrying out weighted summation on the consistency check values of all the characteristic parameters of the traction motor so as to calculate the rule early warning estimation value of the traction motor.
As a specific embodiment, the motor filter screen blockage early warning device of the rail train disclosed in the embodiment of the present application is based on the above contents, and the rule early warning module 602 is specifically configured to:
respectively calculating consistency check values of each characteristic parameter of the traction motor and other traction motors under the same working condition based on a consistency check formula; the consistency check formula is as follows:
wherein, UiThe consistency check value of the ith characteristic parameter of the traction motor is obtained; y isiSampling values of the ith characteristic parameter of the traction motor; mu.s(other)iThe sampling average value of the ith characteristic parameter of other traction motors is obtained; y is(other)ijSampling values of ith characteristic parameters of jth other traction motors; k is the total number of traction motors.
As a specific embodiment, the motor filter screen blockage early warning device of the rail train disclosed in the embodiment of the present application is based on the above contents, and the rule early warning module 602 is further configured to:
acquiring sample data of state parameters of a target model traction motor; respectively calculating the association degree of each non-motor temperature parameter and the motor temperature parameter of the traction motor of the target model; and determining the motor temperature parameters and the first preset number of non-motor temperature parameters with the highest relevance as the characteristic parameters corresponding to the target model traction motor.
As a specific embodiment, the motor filter screen blockage early warning device of the rail train disclosed in the embodiment of the present application is based on the above contents, and the rule early warning module 602 is specifically configured to:
respectively calculating the association degree of each non-motor temperature parameter of the traction motor of the target model and the motor temperature parameter according to an association degree calculation formula; the correlation calculation formula is as follows:
wherein, aiSampling values of non-motor temperature parameters of a target model traction motor; biSampling values of motor temperature parameters of a target model traction motor; m is the sampling data quantity;
is aiAverage value of (d);
is b isiAverage value of (a).
As a specific embodiment, the motor filter screen blockage early warning device of the rail train disclosed in the embodiment of the present application is based on the above contents, and the rule early warning module 602 is further configured to:
and determining corresponding weights for the characteristic parameters according to the relevance, so that when the step of weighting and summing the consistency check values of the characteristic parameters of the traction motor is executed, weighting and summing calculation is carried out based on the weights.
As a specific embodiment, the motor filter screen blockage warning device of the rail train disclosed in the embodiment of the present application is based on the above contents, and the model training module 605 is specifically configured to:
acquiring sample data of state parameters when the motor filter screen is not blocked; carrying out normalization processing on the sample data; determining a network structure of a neural network model; and training the neural network model by taking the non-motor temperature parameter in the sample data as input and the motor temperature parameter as output so as to generate a machine learning early warning model.
As a specific embodiment, the motor filter screen blockage warning device of the rail train disclosed in the embodiment of the present application is based on the above contents, and the model warning module 604 is specifically configured to:
determining a group of sampling values of the state parameters as a target data group according to a plurality of groups of sampling values of the state parameters of the traction motor; carrying out normalization processing on the target data group; inputting the target data set into a machine learning early warning model so as to calculate and generate an estimated value of the motor temperature according to the non-motor temperature parameters in the target data set; calculating the accumulated deviation of the estimated value of the motor temperature and the sampling value of the motor temperature in the target data set; calculating a model early warning estimation value according to the accumulated deviation and the duration time of the accumulated deviation; the model early warning estimation value is respectively in positive correlation change with the accumulated deviation and the duration; judging whether the magnitude of the early warning estimation value of the model is larger than a preset fault threshold value or not; if yes, judging that the motor filter screen blockage fault occurs in the traction motor, and generating a blockage early warning instruction.
As a specific embodiment, the motor filter screen blockage warning device of the rail train disclosed in the embodiment of the present application is based on the above contents, and the model warning module 604 is specifically configured to:
respectively calculating a motor temperature mean value and a motor temperature standard deviation in each group of state parameter sampling values of the traction motor; taking the second preset quantity group data with the lowest motor temperature mean value as a data group to be selected; and determining a group of data with the minimum standard deviation of the motor temperature in the data group to be selected as a target data group.
As a specific embodiment, the motor filter screen of the rail train that this application embodiment discloses blocks up early warning device still includes on the basis of above-mentioned content:
the comprehensive early warning module is used for weighting and summing the regular early warning estimation value and the model early warning estimation value so as to calculate the comprehensive early warning estimation value of the traction motor; and determining the fault early warning grade corresponding to the size of the comprehensive early warning estimation value according to a preset grade corresponding relation table.
As a specific embodiment, the motor filter screen of the rail train that this application embodiment discloses blocks up early warning device on the basis of above-mentioned content, non-motor temperature parameter includes:
ambient temperature, actual traction, actual braking force, motor speed, motor current.
For specific contents of the motor filter screen blockage early warning device for the rail train, reference may be made to the detailed description of the motor filter screen blockage early warning method for the rail train, and details are not repeated here.
Referring to fig. 7, an embodiment of the present application discloses an early warning device, including:
a memory 701 for storing a computer program;
a processor 702 for executing a computer program to implement the steps of the method for motor screen congestion warning for a rail train as any one of the above.
Further, the application also discloses a motor filter screen blockage early warning system of the rail train, which comprises a sensor group, ground human-computer interaction equipment and the early warning equipment;
the sensor group is used for sampling each state parameter of each traction motor in the rail train and sending the sampling value to the processor of the early warning equipment; the state parameters comprise motor temperature parameters and non-motor temperature parameters;
and the ground human-computer interaction equipment is in communication connection with the processor of the early warning equipment and is used for receiving and executing the blockage early warning instruction sent by the processor.
Further, the embodiment of the application also discloses a computer readable storage medium, in which a computer program is stored, and the computer program is used for implementing the steps of the motor filter screen blockage early warning method for the rail train as any one of the above methods when being executed by a processor.
For the specific contents of the motor filter screen blockage warning system, the warning device and the computer readable storage medium of the rail train, reference may be made to the detailed description of the motor filter screen blockage warning method of the rail train, which is not repeated herein.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the equipment disclosed by the embodiment, the description is relatively simple because the equipment corresponds to the method disclosed by the embodiment, and the relevant parts can be referred to the method part for description.
It is further noted that, throughout this document, relational terms such as "first" and "second" are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The technical solutions provided by the present application are described in detail above. The principles and embodiments of the present application are explained herein using specific examples, which are provided only to help understand the method and the core idea of the present application. It should be noted that, for those skilled in the art, without departing from the principle of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall into the protection scope of the present application.
Claims (15)
1. The utility model provides a motor filter screen of rail train blocks up early warning method which characterized in that includes:
acquiring a sampling value of a state parameter of a traction motor in a rail train; the state parameters comprise motor temperature parameters and non-motor temperature parameters;
calling a preset expert rule early warning model to calculate a rule early warning estimation value of the traction motor; the rule early warning estimation value is calculated and generated based on a check result of the consistency of the state parameters of the traction motor and other traction motors;
judging whether the value of the rule early warning estimation value is within a preset fault interval or not;
if yes, judging that the motor filter screen blockage fault occurs in the traction motor, and generating a blockage early warning instruction;
if not, calling a preset machine learning early warning model to judge whether the motor filter screen is blocked by the traction motor; and the judgment result of the machine learning early warning model is calculated and generated based on the matching of the non-motor temperature parameter and the motor temperature parameter.
2. The motor filter screen blockage early warning method according to claim 1, wherein the step of calling a preset expert rule early warning model to calculate a rule early warning estimation value of the traction motor comprises the following steps:
determining characteristic parameters corresponding to the type of the traction motor from various state parameters according to a preset characteristic parameter corresponding relation;
respectively calculating consistency check values of the traction motor and other traction motors under the same working condition aiming at each characteristic parameter of the traction motor;
and carrying out weighted summation on the consistency check values of all the characteristic parameters of the traction motor so as to calculate the rule early warning estimation value of the traction motor.
3. The motor filter screen blockage early warning method according to claim 2, wherein the step of calculating consistency check values of the traction motor and other traction motors under the same working condition aiming at each characteristic parameter of the traction motor comprises the following steps:
respectively calculating consistency check values of each characteristic parameter of the traction motor and other traction motors under the same working condition based on a consistency check formula; the consistency check formula is as follows:
wherein, UiThe consistency check value of the ith characteristic parameter of the traction motor is obtained; y isiSampling values of the ith characteristic parameter of the traction motor; mu.s(other)iThe sampling average value of the ith characteristic parameter of other traction motors is obtained; y is(other)ijSampling values of ith characteristic parameters of jth other traction motors; k is the total number of traction motors.
4. The motor filter screen blockage warning method according to claim 3, wherein for a target model traction motor, the determination process of the characteristic parameter correspondence relationship comprises the following steps:
acquiring sample data of the state parameters of the target model traction motor;
respectively calculating the association degree of each non-motor temperature parameter and the motor temperature parameter of the target type traction motor;
and determining the motor temperature parameters and a first preset number of non-motor temperature parameters with the highest degree of association as the characteristic parameters corresponding to the target model traction motor.
5. The motor filter screen blockage warning method according to claim 4, wherein the step of respectively calculating the correlation degree of each non-motor temperature parameter and the motor temperature parameter of the target type traction motor comprises the following steps:
respectively calculating the association degrees of the non-motor temperature parameters and the motor temperature parameters of the traction motor of the target model according to an association degree calculation formula; the correlation calculation formula is as follows:
wherein, aiSampling values of non-motor temperature parameters of the target type traction motor; biSampling values of motor temperature parameters of the target type traction motor;m is the sampling data quantity; a is aiAverage value of (d); b is biAverage value of (a).
6. The motor filter screen blockage warning method according to claim 4, further comprising, after the calculating the correlation degree between each non-motor temperature parameter and the motor temperature parameter of the target model traction motor respectively:
and determining corresponding weights for the characteristic parameters according to the relevance, so as to perform weighted summation calculation based on the weights when the step of performing weighted summation on the consistency check values of the characteristic parameters of the traction motor is performed.
7. The motor filter screen blockage warning method according to claim 1, wherein the training process of the machine learning warning model comprises the following steps:
acquiring sample data of state parameters when the motor filter screen is not blocked;
carrying out normalization processing on the sample data;
determining a network structure of a neural network model;
and training the neural network model by taking the non-motor temperature parameter in the sample data as input and the motor temperature parameter as output so as to generate the machine learning early warning model.
8. The motor filter screen blockage warning method according to claim 7, wherein the step of calling a preset machine learning warning model to judge the motor filter screen blockage of the traction motor comprises the following steps:
determining a group of sampling values of the state parameters as a target data group according to a plurality of groups of sampling values of the state parameters of the traction motor;
carrying out normalization processing on the target data set;
inputting the target data group into the machine learning early warning model so as to calculate and generate an estimated value of the motor temperature according to the non-motor temperature parameter in the target data group;
calculating the accumulated deviation of the estimated value of the motor temperature and the sampling value of the motor temperature in the target data group;
calculating a model early warning estimation value according to the accumulated deviation and the duration time of the accumulated deviation; the model early warning estimation value is respectively in positive correlation change with the accumulated deviation and the duration;
judging whether the magnitude of the model early warning estimation value is larger than a preset fault threshold value or not;
if yes, judging that the motor filter screen blockage fault occurs in the traction motor, and generating a blockage early warning instruction.
9. The motor filter screen blockage warning method according to claim 8, wherein the step of determining a group of sampling values of the state parameters as a target data group according to a plurality of groups of sampling values of the state parameters of the traction motor comprises the following steps:
respectively calculating a motor temperature mean value and a motor temperature standard deviation in each group of state parameter sampling values of the traction motor;
taking the second preset quantity group data with the lowest motor temperature mean value as a data group to be selected;
and determining a group of data with the minimum standard deviation of the motor temperature in the data group to be selected as the target data group.
10. The motor filter screen blockage warning method according to claim 9, further comprising, after determining that the motor filter screen blockage failure occurs in the traction motor:
weighting and summing the rule early warning estimation value and the model early warning estimation value to calculate a comprehensive early warning estimation value of the traction motor;
and determining the fault early warning grade corresponding to the size of the comprehensive early warning estimation value according to a preset grade corresponding relation table.
11. The motor screen plugging warning method according to any one of claims 1 to 10, wherein the non-motor temperature parameter comprises:
ambient temperature, actual traction, actual braking force, motor speed, motor current.
12. The utility model provides a rail train's motor filter screen blocks up early warning device which characterized in that includes:
the parameter acquisition module is used for acquiring a sampling value of a state parameter of a traction motor in the rail train; the state parameters comprise motor temperature parameters and non-motor temperature parameters;
the rule early warning module is used for calling a preset expert rule early warning model to calculate a rule early warning estimation value of the traction motor; the rule early warning estimation value is calculated and generated based on a check result of the consistency of the state parameters of the traction motor and other traction motors;
the rule judging module is used for judging whether the value of the rule early warning estimation value is within a preset fault interval or not;
the model early warning module is used for calling a preset machine learning early warning model to judge the blockage of the motor filter screen of the traction motor after judging that the value of the rule early warning estimation value is not within a preset fault interval; the judgment result of the machine learning early warning model is calculated and generated based on the matching of the non-motor temperature parameter and the motor temperature parameter;
the model training module is used for pre-training and generating the machine learning early warning model;
and the fault early warning module is used for generating a blockage early warning instruction after the rule judgment module judges that the value of the rule early warning estimation value is within a preset fault interval or the model early warning module judges that the motor filter screen is blocked.
13. An early warning device, comprising:
a memory for storing a computer program;
a processor for executing said computer program to carry out the steps of the method of motor screen congestion warning for a rail train according to any of claims 1 to 11.
14. A motor filter screen blockage warning system for a rail train, comprising a sensor group, a ground human-computer interaction device and the warning device of claim 13;
the sensor group is used for sampling each state parameter of each traction motor in the rail train and sending the sampling value to the processor of the early warning equipment; the state parameters comprise motor temperature parameters and non-motor temperature parameters;
and the ground human-computer interaction equipment is in communication connection with the processor of the early warning equipment and is used for receiving and executing the blockage early warning instruction sent by the processor.
15. A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, is adapted to carry out the steps of the motor screen clogging warning method according to any one of claims 1 to 11.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110008428.0A CN112668243A (en) | 2021-01-05 | 2021-01-05 | Motor filter screen blockage early warning method and device for rail train and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110008428.0A CN112668243A (en) | 2021-01-05 | 2021-01-05 | Motor filter screen blockage early warning method and device for rail train and related equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112668243A true CN112668243A (en) | 2021-04-16 |
Family
ID=75412964
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110008428.0A Pending CN112668243A (en) | 2021-01-05 | 2021-01-05 | Motor filter screen blockage early warning method and device for rail train and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112668243A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110515935A (en) * | 2019-08-29 | 2019-11-29 | 新誉轨道交通科技有限公司 | A kind of dirty stifled failure prediction method of train air-conditioning strainer, device, equipment and system |
| CN111024416A (en) * | 2018-10-09 | 2020-04-17 | 株洲中车时代电气股份有限公司 | Fault diagnosis method and system for train traction system |
| CN111237134A (en) * | 2020-01-14 | 2020-06-05 | 上海电力大学 | Offshore double-fed wind driven generator fault diagnosis method based on GRA-LSTM-stacking model |
| US20200248641A1 (en) * | 2019-02-01 | 2020-08-06 | Toyota Jidosha Kabushiki Kaisha | Control device of internal combustion engine, in-vehicle electronic control unit, machine learning system, control method of internal combustion engine, manufacturing method of electronic control unit, and output parameter calculation device |
-
2021
- 2021-01-05 CN CN202110008428.0A patent/CN112668243A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111024416A (en) * | 2018-10-09 | 2020-04-17 | 株洲中车时代电气股份有限公司 | Fault diagnosis method and system for train traction system |
| US20200248641A1 (en) * | 2019-02-01 | 2020-08-06 | Toyota Jidosha Kabushiki Kaisha | Control device of internal combustion engine, in-vehicle electronic control unit, machine learning system, control method of internal combustion engine, manufacturing method of electronic control unit, and output parameter calculation device |
| CN110515935A (en) * | 2019-08-29 | 2019-11-29 | 新誉轨道交通科技有限公司 | A kind of dirty stifled failure prediction method of train air-conditioning strainer, device, equipment and system |
| CN111237134A (en) * | 2020-01-14 | 2020-06-05 | 上海电力大学 | Offshore double-fed wind driven generator fault diagnosis method based on GRA-LSTM-stacking model |
Non-Patent Citations (2)
| Title |
|---|
| ZHANG, S 等: "Measuring method of Radial Run Out of High Speed Railway Wheelset", 《2020 10TH INSTITUTE OF ELECTRICAL AND ELECTRONICS ENGINEERS INTERNATIONAL CONFERENCE ON CYBER TECHNOLOGY IN AUTOMATION, CONTROL, AND INTELLIGENT SYSTEMS (IEEE-CYBER 2020)》, pages 390 - 393 * |
| 刘可安 等: "城轨车辆关键部件状态修技术与系统解决方案", 《机车电传动》, no. 4, pages 1 - 7 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105300692B (en) | A kind of bearing failure diagnosis and Forecasting Methodology based on expanded Kalman filtration algorithm | |
| CN104392390A (en) | TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) model-based intelligent substation secondary equipment evaluation method | |
| CN109544399B (en) | Power transmission equipment state evaluation method and device based on multi-source heterogeneous data | |
| CN109856515A (en) | A kind of direct current cables state of insulation judgment method and system | |
| CN117092578B (en) | Wire harness conduction intelligent detection system based on data acquisition and processing | |
| CN104698343A (en) | Method and system for judging power grid faults based on historical recording data | |
| DE102018115155A1 (en) | Method for predicting plant data and device using it | |
| CN117454114B (en) | Subway tunnel tunneling blasting vibration safety monitoring device based on multi-point location distribution | |
| CN113027658B (en) | Real-time state evaluation method for water turbine runner and application thereof | |
| CN113340353A (en) | Monitoring method, equipment and medium for power transmission line | |
| CN116126645A (en) | Machine room stability monitoring method and system based on digital twinning | |
| CN106157616A (en) | A kind of magnitude of traffic flow short-term prediction device | |
| CN108414016A (en) | A kind of sewage network monitoring system based on big data technology | |
| CN109285331B (en) | Power cable temperature early warning system based on data analysis and temperature prediction | |
| CN111222968A (en) | Enterprise tax risk management and control method and system | |
| CN113934536A (en) | Data acquisition method facing edge calculation | |
| CN105741184B (en) | Transformer state evaluation method and device | |
| CN117010597B (en) | Smart city security monitoring cloud platform based on Internet of things | |
| CN117034194B (en) | Nuclear power instrument equipment operation and maintenance management system, method and equipment based on artificial intelligence | |
| CN113253051A (en) | Cloud computing-based smart grid operation and maintenance operation system safety online real-time monitoring method and device and storage medium | |
| CN117191147A (en) | Flood discharge dam water level monitoring and early warning method and system | |
| CN112668243A (en) | Motor filter screen blockage early warning method and device for rail train and related equipment | |
| CN113934862B (en) | Community security risk prediction method, device, electronic equipment and medium | |
| CN114500075A (en) | User abnormal behavior detection method and device, electronic equipment and storage medium | |
| CN115520741A (en) | Elevator operation monitoring and early warning method and system based on neural network and storage medium |
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 |