CN112067289A - Motor shaft and transmission shaft abnormal vibration early warning algorithm based on neural network - Google Patents
Motor shaft and transmission shaft abnormal vibration early warning algorithm based on neural network Download PDFInfo
- Publication number
- CN112067289A CN112067289A CN202010846935.7A CN202010846935A CN112067289A CN 112067289 A CN112067289 A CN 112067289A CN 202010846935 A CN202010846935 A CN 202010846935A CN 112067289 A CN112067289 A CN 112067289A
- Authority
- CN
- China
- Prior art keywords
- neural network
- network model
- data
- training
- motor shaft
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/02—Gearings; Transmission mechanisms
- G01M13/028—Acoustic or vibration analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Acoustics & Sound (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention relates to an abnormal vibration early warning algorithm of a motor shaft and a transmission shaft based on a neural network, which is mainly technically characterized in that: classifying and sorting long-term historical operating data of the test bed to obtain an offline training set; establishing a BP neural network model; determining parameters of a BP neural network model; training a BP neural network model by using an offline data set until a stopping criterion is reached, and obtaining an initial BP neural network model; training a BP neural network model on line to obtain a trained BP neural network model; and the BP neural network model calculates the vibration value under the current working condition in real time and compares the vibration value with the vibration value measured by the sensor in real time, and the system automatically warns when any monitored motor shaft or transmission shaft generates abnormal vibration. The invention can accurately detect the abnormal vibration of the motor shaft and the rotating shaft through the neural network model and pre-alarm in time, can reduce the accident probability to the maximum extent, and opens up a new way for the abnormal vibration fault diagnosis of the motor shaft system.
Description
Technical Field
The invention belongs to the technical field of test equipment, and particularly relates to an abnormal vibration early warning algorithm for a motor shaft and a transmission shaft based on a neural network.
Background
The transmission test bed plays an important role in test equipment, the motor is indispensable in the transmission test bed, and the abnormal vibration of the motor shaft often causes huge test accidents or property loss.
The motor shaft is used as a rotating apparatus, and abnormal vibration is caused due to unbalance of a rotor, misalignment of a shaft system, vibration of an oil film, bending, cracking, loosening and the like. However, the vibration signal magnitude can also change under different rotating speed and torque working conditions in the operation process, and the vibration has a normal vibration range under different working conditions. Since the normal vibration range is related to the motor speed and torque, but the relationship is not linear, it is difficult to find a function to represent the relationship, which brings certain difficulty to identify the abnormal vibration.
The abnormal vibration condition of the motor shaft is easy to be ignored during the operation of the system. The system runs under abnormal vibration for a long time, so that the service life of the motor is shortened for a light person, and serious accidents such as shaft breakage and the like can occur for a serious person. Therefore, how to accurately detect the abnormal vibration of the motor shaft and the transmission shaft and send out early warning is a problem which needs to be solved urgently at present.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, provides an abnormal vibration early warning algorithm for a motor shaft and a transmission shaft based on a neural network, and can accurately detect the abnormal vibration of the motor shaft and the transmission shaft and send out early warning information.
The technical problem to be solved by the invention is realized by adopting the following technical scheme:
an abnormal vibration early warning algorithm for a motor shaft and a transmission shaft based on a neural network comprises the following steps:
step 4, training a BP neural network model by using an offline data set until a stop criterion is reached, and obtaining an initial BP neural network model;
step 5, training a BP neural network model on line by using the motor shaft and transmission shaft data collected by the test bed to obtain the trained BP neural network model;
and 6, in the running process of the system, calculating the vibration value under the current working condition in real time by the BP neural network model, comparing the vibration value with the vibration value measured by the sensor in real time, and automatically early warning the system when any monitored motor shaft or transmission shaft generates abnormal vibration.
The long-term historical data of the test bed comprises motor torque, rotating speed and XYZ-axis vibration values, and the data are obtained by measuring from a sensor of a test bed data acquisition system.
The step 1 is to carry out normalization processing on the data by adopting a Min-Max conversion method.
The BP neural network model established in the step 2 is a three-layer BP neural network model, the rotating speed and the torque in the training set are used as the input of the BP neural network model, and the XYZ-axis vibration value is used as the output of the BP neural network model.
The parameters of the BP neural network model comprise the number of neurons in an input layer, the number of neurons in a hidden layer, the number of neurons in an output layer, an activation function, a learning rate, a training mode and a stopping criterion, wherein:
the number of neurons in the input layer is 2, and the neurons comprise rotating speed and torque;
the number of cryptomelane neurons is between 15 and 21;
the number of neurons in the output layer is 3, and the neurons comprise XYZ-axis vibration values;
the termination criteria are: up to 200000 times or the precision up to 1 × 10-23;
The activation function is:where a is the tilt parameter, then when a → ∞ is thenv is a function argument;
learning rate: a learning rate of the decay type is used, in which the initial setting is 0.1, the decay rate is 0.9, and the decay rate is 50.
The method for training the BP neural network model in the steps 4 and 5 comprises the following steps:
initializing: randomly selecting a weight value and a threshold value according to uniform distribution with the average value of 0;
a training sample is prepared: presenting one round of the training samples to a BP neural network model, and sequentially performing forward calculation and backward calculation on the training samples according to the third step and the fourth step;
calculating in the forward direction: error signals can be calculated through forward calculation;
fourthly, reverse calculation: and calculating a weight modifier according to the error signal, and modifying the last calculated network weight.
Carrying out iterative processing: and providing the presented new sample for the BP neural network model, and repeatedly calculating the update weight according to the third step and the fourth step until the stop criterion is met.
The specific implementation method of the step 5 is as follows: acquiring motor shaft and transmission shaft data acquired by a test bed, training a BP neural network model on line, updating the network according to a new round of acquired data every time in a cycle until outlet fault data are obtained, and acquiring the trained BP neural network model; or performing a new round of off-line training of batch data when the model is updated regularly to obtain the trained BP neural network model.
And 6, in the process of comparing the vibration values, judging according to a preset early warning sensitivity dead zone, and automatically early warning when the comparison difference value exceeds the early warning sensitivity dead zone.
The invention has the advantages and positive effects that:
1. the method adopts a three-layer BP neural network model to calculate and obtain the normal vibration value range under any rotating speed and torque, compares the obtained normal vibration range of the motor under any working condition with the real-time measurement value of the sensor, considers that abnormal vibration exists when the normal vibration range is exceeded, and gives an alarm in advance in time, can reduce the accident probability to the maximum extent, has ultrahigh capabilities of memory, information processing and fault identification, overcomes the technical bottleneck of the traditional diagnosis method, and opens up a new way for the abnormal vibration fault diagnosis of the motor shafting.
2. The BP neural network model can be updated regularly, the offline learning method is adopted to label data in batches to train the neural network, and the online learning has higher real-time performance and the network model is updated along with the time so that the network model always keeps stronger nonlinear fitting capability and better robustness.
Drawings
FIG. 1 is a process flow diagram of the present invention;
FIG. 2 is a graph of error convergence during model training according to the present invention;
FIG. 3 is a graph showing the identification of a failure test data according to the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
An abnormal vibration early warning algorithm for a motor shaft and a transmission shaft based on a neural network is shown in figure 1 and comprises the following steps:
The long-term historical data of the test bed comprises motor torque, rotating speed and XYZ-axis vibration values, and the data are obtained by measuring from a sensor of a test bed data acquisition system.
In this step, the original data x is processed1,x2,x3…xnThe normalization (using Min-Max conversion) process was performed as follows:
obtaining dimensionless data: y is1,y2,y3…yn。
In the step, dimensionless data obtained by processing historical operation data is used as an off-line training set for training the neural network, wherein the rotating speed and the torque are used as the input of the neural network, and the XYZ-axis vibration value is used as the output of the neural network.
And 2, establishing a three-layer BP neural network model for training an off-line training set.
The invention uses a three-layer BP neural network (multi-layer feedforward neural network trained according to an error back propagation algorithm) model to train training set data, and error information output by a neuron j of the BP neural network when iterating to an nth training example is defined as follows:
ej(n)=dj(n)-yj(n)
defining the error energy transient for neuron j as:
the sum of the error energies of all neurons of the output layer is thus defined as:
assuming that N is the total number of samples in the training set, the average of the error energies can be obtained:
in the above formula, the first and second carbon atoms are,avthe cost function is a measure of learning ability, and the weight parameter is finally adjusted through the learning of the networkTo obtainavA minimization is achieved.
Synaptic weight omega of back propagation algorithmji(n) applying a weight modifier Δ ωji(n) which is proportional to (n) vs. ωji(n) partial derivatives of (n). This gradient is expressed according to the differential chain rule as:
in the above formula, ωjiCorrection value Δ ω of (n)ji(n) is defined by the rule of delta:bringing it into the formula:
Δωji(n)=ηj(n)yi(n)
here, the local gradient ηIs the learning efficiency, the minus sign in the formula indicates that the weight change direction which makes (n) fall is sought in the weight space.
And 3, determining parameters of the BP neural network model, including the number of input layer neurons, the number of hidden layer neurons, the number of output layer neurons, an activation function, a learning rate, a training mode and a stopping criterion.
In this step, the parameters of the BP neural network model are defined as follows:
number of input layer neurons: the number of input variables is determined by the number of input variables, which is 2, and the number of input layer neurons is 2, because the number of input variables is the rotational speed and the torque, respectively.
Number of hidden layer neurons: experiments prove that the model precision is better when the number of hidden layer neurons is between 15 and 21.
The number of neurons in the output layer was 3: XYZ axes vibration values.
Network training termination condition: up to 200000 times or the precision up to 1 × 10-23。
Activation function: selectingBecause the Sigmoid function is a strictly increasing function and the Sigmoid function has a good balance between linear and non-linear behavior, a in the activation function is a tilt parameter, which is the case when a → ∞v is a function argument.
Learning rate: the attenuation type learning rate is adopted, the initial setting is 0.1, the attenuation rate is 0.9, and the attenuation rate is 50 (namely, the attenuation is carried out once after 50 iterations).
Step 4, off-line training of the BP neural network model: and (3) training the BP neural network model by using the offline data set obtained in the step (1) until a stopping criterion is reached, and obtaining an initial BP neural network model.
FIG. 2 is a series updating method of error convergence curve and back propagation algorithm using weights in training process, which is realized mainly by training samplesPerforming cyclic training, wherein the specific method comprises the following steps:
(1) initializing, and randomly selecting a weight value and a threshold value according to uniform distribution with the average value of 0.
(2) And (4) training samples, presenting one round of the training samples to the network, and sequentially performing the forward and reverse calculations of the steps (3) and (4) on the training samples.
(3) Forward calculation, by which an error signal e can be calculatedj(n)。
(4) And (4) performing reverse calculation, calculating a weight modifier according to the error signal, and modifying the last calculated network weight.
(5) And (4) iteration, presenting a new sample to the network, and repeatedly calculating the updated weight value according to the steps (3) and (4) until the stop criterion is met.
Step 5, training a BP neural network model on line: and (3) normalizing the motor shaft and transmission shaft data (rotating speed and matrix signals) acquired by the test bed, performing online training on the BP neural network model, and updating the network according to the acquired data of a new round every time in a circulation mode until outlet fault data are obtained, so as to obtain the trained BP neural network model. Or performing a new round of off-line training of batch data when the model is updated regularly to obtain the trained BP neural network model.
After an initial BP neural network model is established, network weight updating is carried out on data collected by a circulation system each time, an updating algorithm is similar to offline batch training, except that an error function is not sample set error accumulation any more, but a current single sample model error, corresponding calculation is carried out according to the current single sample, and the network weight is updated according to a back propagation algorithm.
And 6, in the running process of the system, calculating the vibration value under the current working condition in real time by the BP neural network model, comparing the vibration value with the vibration value measured by the sensor in real time, and automatically warning and prompting the system when any monitored motor shaft or transmission shaft vibrates abnormally.
In the step, in order to avoid false alarm action caused by numerical value fluctuation in a normal range, a sensitivity dead zone of early warning is set, and if the contrast difference exceeds the sensitivity dead zone in the contrast process, automatic early warning is carried out.
Fig. 3 shows the measured data of abnormal vibration of the X-axis of the power motor in the primary test process, after the working condition changes, the measured data of the X-axis vibration sensor and the model calculation data have large deviation, and an alarm is given according to the setting of the program, so that slight faults are found after the shutdown inspection, and larger test accidents are avoided.
It should be emphasized that the embodiments described herein are illustrative rather than restrictive, and thus the present invention is not limited to the embodiments described in the detailed description, but also includes other embodiments that can be derived from the technical solutions of the present invention by those skilled in the art.
Claims (8)
1. The utility model provides a motor shaft and transmission shaft abnormal vibration early warning algorithm based on neural network which characterized in that: the method comprises the following steps:
step 1, classifying and sorting long-term historical operating data of a test bed, marking abnormal fault type data and normal operating data, carrying out normalization processing on the data to obtain dimensionless data, and taking the dimensionless data as an offline training set;
step 2, establishing a BP neural network model for training an off-line training set;
step 3, determining parameters of the BP neural network model;
step 4, training a BP neural network model by using an offline data set until a stop criterion is reached, and obtaining an initial BP neural network model;
step 5, training a BP neural network model on line by using the motor shaft and transmission shaft data collected by the test bed to obtain the trained BP neural network model;
and 6, in the running process of the system, calculating the vibration value under the current working condition in real time by the BP neural network model, comparing the vibration value with the vibration value measured by the sensor in real time, and automatically early warning the system when any monitored motor shaft or transmission shaft generates abnormal vibration.
2. The motor shaft and transmission shaft abnormal vibration early warning algorithm based on the neural network as claimed in claim 1, wherein: the long-term historical data of the test bed comprises motor torque, rotating speed and XYZ-axis vibration values, and the data are obtained by measuring from a sensor of a test bed data acquisition system.
3. The motor shaft and transmission shaft abnormal vibration early warning algorithm based on the neural network as claimed in claim 1, wherein: the step 1 is to carry out normalization processing on the data by adopting a Min-Max conversion method.
4. The motor shaft and transmission shaft abnormal vibration early warning algorithm based on the neural network as claimed in claim 1, wherein: the BP neural network model established in the step 2 is a three-layer BP neural network model, the rotating speed and the torque in the training set are used as the input of the BP neural network model, and the XYZ-axis vibration value is used as the output of the BP neural network model.
5. The motor shaft and transmission shaft abnormal vibration early warning algorithm based on the neural network as claimed in claim 1, wherein: the parameters of the BP neural network model comprise the number of neurons in an input layer, the number of neurons in a hidden layer, the number of neurons in an output layer, an activation function, a learning rate, a training mode and a stopping criterion, wherein:
the number of neurons in the input layer is 2, and the neurons comprise rotating speed and torque;
the number of cryptomelane neurons is between 15 and 21;
the number of neurons in the output layer is 3, and the neurons comprise XYZ-axis vibration values;
the termination criteria are: up to 200000 times or the precision up to 1 × 10-23;
The activation function is:where a is the tilt parameter, then when a → ∞ is thenv is a function argument;
learning rate: a learning rate of the decay type is used, in which the initial setting is 0.1, the decay rate is 0.9, and the decay rate is 50.
6. The motor shaft and transmission shaft abnormal vibration early warning algorithm based on the neural network as claimed in claim 1, wherein: the method for training the BP neural network model in the steps 4 and 5 comprises the following steps:
initializing: randomly selecting a weight value and a threshold value according to uniform distribution with the average value of 0;
a training sample is prepared: presenting one round of the training samples to a BP neural network model, and sequentially performing forward calculation and backward calculation on the training samples according to the third step and the fourth step;
calculating in the forward direction: error signals can be calculated through forward calculation;
fourthly, reverse calculation: and calculating a weight modifier according to the error signal, and modifying the last calculated network weight.
Carrying out iterative processing: and providing the presented new sample for the BP neural network model, and repeatedly calculating the update weight according to the third step and the fourth step until the stop criterion is met.
7. The motor shaft and transmission shaft abnormal vibration early warning algorithm based on the neural network as claimed in claim 1, wherein: the specific implementation method of the step 5 is as follows: acquiring motor shaft and transmission shaft data acquired by a test bed, training a BP neural network model on line, updating the network according to a new round of acquired data every time in a cycle until outlet fault data are obtained, and acquiring the trained BP neural network model; or performing a new round of off-line training of batch data when the model is updated regularly to obtain the trained BP neural network model.
8. The motor shaft and transmission shaft abnormal vibration early warning algorithm based on the neural network as claimed in claim 1, wherein: and 6, in the process of comparing the vibration values, judging according to a preset early warning sensitivity dead zone, and automatically early warning when the comparison difference value exceeds the early warning sensitivity dead zone.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010846935.7A CN112067289A (en) | 2020-08-21 | 2020-08-21 | Motor shaft and transmission shaft abnormal vibration early warning algorithm based on neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010846935.7A CN112067289A (en) | 2020-08-21 | 2020-08-21 | Motor shaft and transmission shaft abnormal vibration early warning algorithm based on neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112067289A true CN112067289A (en) | 2020-12-11 |
Family
ID=73662523
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010846935.7A Pending CN112067289A (en) | 2020-08-21 | 2020-08-21 | Motor shaft and transmission shaft abnormal vibration early warning algorithm based on neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112067289A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112434755A (en) * | 2020-12-15 | 2021-03-02 | 电子科技大学 | Data anomaly sensing method based on heterogeneous system |
CN113988202A (en) * | 2021-11-04 | 2022-01-28 | 季华实验室 | Mechanical arm abnormal vibration detection method based on deep learning |
Citations (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102848266A (en) * | 2012-09-29 | 2013-01-02 | 西南交通大学 | Machine tool spindle accuracy prediction method |
US20130204808A1 (en) * | 2012-02-08 | 2013-08-08 | General Electric Company | Fault Prediction of Monitored Assets |
CN103454113A (en) * | 2013-09-16 | 2013-12-18 | 中国人民解放军国防科学技术大学 | Method for monitoring health of rotary machine suitable for working condition changing condition |
CN103773900A (en) * | 2013-12-30 | 2014-05-07 | 镇江市高等专科学校 | Solid state fermentation control method based on neural network and particle swarm algorithm |
CN103793746A (en) * | 2014-02-19 | 2014-05-14 | 上海交通大学 | Method for identifying parameters of coal-fired power plant boiler superheater model |
CN104156422A (en) * | 2014-08-06 | 2014-11-19 | 辽宁工程技术大学 | Gas concentration real-time prediction method based on dynamic neural network |
CN106768262A (en) * | 2016-11-11 | 2017-05-31 | 华北电力大学(保定) | A kind of transformer online monitoring method based on surface vibration signals analysis |
CN107036819A (en) * | 2017-05-02 | 2017-08-11 | 大唐东北电力试验研究所有限公司 | The Turbo-generator Set remote oscillation fault diagnosis method and system of multi-parameter amendment |
CN107977508A (en) * | 2017-11-29 | 2018-05-01 | 北京优利康达科技股份有限公司 | A kind of dynamo bearing failure prediction method |
CN108760266A (en) * | 2018-05-31 | 2018-11-06 | 西安交通大学 | The virtual degeneration index building method of mechanical key component based on learning distance metric |
CN109255441A (en) * | 2018-10-18 | 2019-01-22 | 西安电子科技大学 | Spacecraft fault diagnosis method based on artificial intelligence |
CN109376413A (en) * | 2018-10-08 | 2019-02-22 | 四川大学 | Line neural network method for diagnosing faults based on data-driven |
US20190095781A1 (en) * | 2017-09-23 | 2019-03-28 | Nanoprecise Sci Corp. | System and method for automated fault diagnosis and prognosis for rotating equipment |
CN109657989A (en) * | 2018-12-20 | 2019-04-19 | 南京航空航天大学 | Helicopter high-speed overload input stage health state evaluation method |
CN109670145A (en) * | 2018-12-12 | 2019-04-23 | 中国航空工业集团公司上海航空测控技术研究所 | Helicopter Main Reducer vibration performance threshold setting method based on observer |
CN109818964A (en) * | 2019-02-01 | 2019-05-28 | 长沙市智为信息技术有限公司 | A kind of ddos attack detection method, device, equipment and storage medium |
CN109861942A (en) * | 2019-01-22 | 2019-06-07 | 东南大学 | A kind of on-line study method of artificial intelligence auxiliary OFDM receiver |
CN110008586A (en) * | 2019-04-03 | 2019-07-12 | 中国水利水电科学研究院 | A kind of gate safety evaluation method |
CN110097209A (en) * | 2019-03-26 | 2019-08-06 | 朗坤智慧科技股份有限公司 | A kind of equipment deterioration analysis method based on parameter residual error |
CN110263767A (en) * | 2019-07-12 | 2019-09-20 | 南京工业大学 | In conjunction with the intelligent Rotating Shaft Fault method of compressed data acquisition and deep learning |
CN110570013A (en) * | 2019-08-06 | 2019-12-13 | 山东省科学院海洋仪器仪表研究所 | Single-station online wave period data prediction diagnosis method |
CN110766143A (en) * | 2019-10-31 | 2020-02-07 | 上海埃威航空电子有限公司 | Equipment fault intelligent diagnosis method based on artificial neural network |
CN111094927A (en) * | 2017-09-26 | 2020-05-01 | 舍弗勒技术股份两合公司 | Bearing fault diagnosis method and device, readable storage medium and electronic equipment |
CN111275007A (en) * | 2020-02-24 | 2020-06-12 | 山东大学 | Bearing fault diagnosis method and system based on multi-scale information fusion |
-
2020
- 2020-08-21 CN CN202010846935.7A patent/CN112067289A/en active Pending
Patent Citations (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130204808A1 (en) * | 2012-02-08 | 2013-08-08 | General Electric Company | Fault Prediction of Monitored Assets |
CN102848266A (en) * | 2012-09-29 | 2013-01-02 | 西南交通大学 | Machine tool spindle accuracy prediction method |
CN103454113A (en) * | 2013-09-16 | 2013-12-18 | 中国人民解放军国防科学技术大学 | Method for monitoring health of rotary machine suitable for working condition changing condition |
CN103773900A (en) * | 2013-12-30 | 2014-05-07 | 镇江市高等专科学校 | Solid state fermentation control method based on neural network and particle swarm algorithm |
CN103793746A (en) * | 2014-02-19 | 2014-05-14 | 上海交通大学 | Method for identifying parameters of coal-fired power plant boiler superheater model |
CN104156422A (en) * | 2014-08-06 | 2014-11-19 | 辽宁工程技术大学 | Gas concentration real-time prediction method based on dynamic neural network |
CN106768262A (en) * | 2016-11-11 | 2017-05-31 | 华北电力大学(保定) | A kind of transformer online monitoring method based on surface vibration signals analysis |
CN107036819A (en) * | 2017-05-02 | 2017-08-11 | 大唐东北电力试验研究所有限公司 | The Turbo-generator Set remote oscillation fault diagnosis method and system of multi-parameter amendment |
US20190095781A1 (en) * | 2017-09-23 | 2019-03-28 | Nanoprecise Sci Corp. | System and method for automated fault diagnosis and prognosis for rotating equipment |
CN111094927A (en) * | 2017-09-26 | 2020-05-01 | 舍弗勒技术股份两合公司 | Bearing fault diagnosis method and device, readable storage medium and electronic equipment |
CN107977508A (en) * | 2017-11-29 | 2018-05-01 | 北京优利康达科技股份有限公司 | A kind of dynamo bearing failure prediction method |
CN108760266A (en) * | 2018-05-31 | 2018-11-06 | 西安交通大学 | The virtual degeneration index building method of mechanical key component based on learning distance metric |
CN109376413A (en) * | 2018-10-08 | 2019-02-22 | 四川大学 | Line neural network method for diagnosing faults based on data-driven |
CN109255441A (en) * | 2018-10-18 | 2019-01-22 | 西安电子科技大学 | Spacecraft fault diagnosis method based on artificial intelligence |
CN109670145A (en) * | 2018-12-12 | 2019-04-23 | 中国航空工业集团公司上海航空测控技术研究所 | Helicopter Main Reducer vibration performance threshold setting method based on observer |
CN109657989A (en) * | 2018-12-20 | 2019-04-19 | 南京航空航天大学 | Helicopter high-speed overload input stage health state evaluation method |
CN109861942A (en) * | 2019-01-22 | 2019-06-07 | 东南大学 | A kind of on-line study method of artificial intelligence auxiliary OFDM receiver |
CN109818964A (en) * | 2019-02-01 | 2019-05-28 | 长沙市智为信息技术有限公司 | A kind of ddos attack detection method, device, equipment and storage medium |
CN110097209A (en) * | 2019-03-26 | 2019-08-06 | 朗坤智慧科技股份有限公司 | A kind of equipment deterioration analysis method based on parameter residual error |
CN110008586A (en) * | 2019-04-03 | 2019-07-12 | 中国水利水电科学研究院 | A kind of gate safety evaluation method |
CN110263767A (en) * | 2019-07-12 | 2019-09-20 | 南京工业大学 | In conjunction with the intelligent Rotating Shaft Fault method of compressed data acquisition and deep learning |
CN110570013A (en) * | 2019-08-06 | 2019-12-13 | 山东省科学院海洋仪器仪表研究所 | Single-station online wave period data prediction diagnosis method |
CN110766143A (en) * | 2019-10-31 | 2020-02-07 | 上海埃威航空电子有限公司 | Equipment fault intelligent diagnosis method based on artificial neural network |
CN111275007A (en) * | 2020-02-24 | 2020-06-12 | 山东大学 | Bearing fault diagnosis method and system based on multi-scale information fusion |
Non-Patent Citations (2)
Title |
---|
中国环境监测总站 著: "《环境空气质量预报预警方法技术指南(第二版)》", 30 June 2017 * |
李中等: "基于广义回归神经网络的变压器表面振动基频幅值计算", 《高电压技术》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112434755A (en) * | 2020-12-15 | 2021-03-02 | 电子科技大学 | Data anomaly sensing method based on heterogeneous system |
CN113988202A (en) * | 2021-11-04 | 2022-01-28 | 季华实验室 | Mechanical arm abnormal vibration detection method based on deep learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114813105B (en) | Gear box fault early warning method and system based on working condition similarity evaluation | |
CN111922095A (en) | Vibration diagnosis method for abnormal torsional vibration fault of roller of cold rolling mill | |
WO2022134495A1 (en) | Deep learning-based device anomaly detection method and system, and computer medium | |
CN110006552B (en) | Method for detecting abnormal temperature of equipment | |
CN112067289A (en) | Motor shaft and transmission shaft abnormal vibration early warning algorithm based on neural network | |
CN116067432B (en) | Escalator variable working condition fault diagnosis method | |
CN110570013B (en) | Single-station online wave period data prediction diagnosis method | |
CN116292367B (en) | Power plant fan system abnormal condition detection method based on one-dimensional convolution | |
CN112284440A (en) | Sensor data deviation self-adaptive correction method | |
CN115859077A (en) | Multi-feature fusion motor small sample fault diagnosis method under variable working conditions | |
CN114091525A (en) | Rolling bearing degradation trend prediction method | |
CN115982663A (en) | Fault early warning system of new forms of energy fan equipment | |
CN115496188A (en) | Coal mill fault early warning method based on deep learning convolutional neural network | |
CN108416386A (en) | A kind of method and system judged extremely for Hydropower Unit bearing temperature | |
Samhouri et al. | An intelligent machine condition monitoring system using time-based analysis: neuro-fuzzy versus neural network | |
CN112418306B (en) | Gas turbine compressor fault early warning method based on LSTM-SVM | |
CN117556347A (en) | Power equipment fault prediction and health management method based on industrial big data | |
CN110889646B (en) | Method for checking correctness of operating parameters of generator set adapting to working condition change | |
US11339763B2 (en) | Method for windmill farm monitoring | |
Karadayi et al. | Fault-related Alarm Detection of a Wind Turbine SCADA System | |
CN116204825A (en) | Production line equipment fault detection method based on data driving | |
Mei et al. | A novel fusion prognostic approach for the prediction of the remaining useful life of a lithiumion battery | |
TWI762101B (en) | Abnormal Diagnosis Device and Abnormal Diagnosis Program | |
CN114151291B (en) | Early fault monitoring method for wind turbine generator | |
CN114936520B (en) | Steam turbine fault early warning analysis method based on improved MSET |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20201211 |