CN106682685A - Microwave heating temperature field distribution characteristic deep learning-based local temperature variation anomaly detection method - Google Patents
Microwave heating temperature field distribution characteristic deep learning-based local temperature variation anomaly detection method Download PDFInfo
- Publication number
- CN106682685A CN106682685A CN201611106593.5A CN201611106593A CN106682685A CN 106682685 A CN106682685 A CN 106682685A CN 201611106593 A CN201611106593 A CN 201611106593A CN 106682685 A CN106682685 A CN 106682685A
- Authority
- CN
- China
- Prior art keywords
- data
- microwave heating
- length
- layer
- temperature
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Control Of High-Frequency Heating Circuits (AREA)
Abstract
The invention discloses a microwave heating temperature field distribution characteristic deep learning-based local temperature variation anomaly detection method. According to the method, a convolutional neural network and automatic coding-combined method is used to learn the multi-dimensional big data of the temperature field distribution of a microwave heating process, deep logical relationships between the data are found out, the structure of the data is learned, and therefore, characteristics having expression ability stronger than that of input can be obtained; and the Isolation Forest algorithm is adopted to perform anomaly detection. With the microwave heating temperature field distribution characteristic deep learning-based local temperature variation anomaly detection method of the invention, the local overheating or even thermal runaway of a heating medium in the microwave heating process due to the change of the dielectric coefficient and thermal conductivity of the heating medium with temperature rise due to the coupling of a complex time-varying electromagnetic field and a temperature field can be detected reliably, and therefore, timey processing can be performed, and security accidents can be avoided.
Description
Technical field
The present invention relates to microwave heating control technology.
Background technology
Microwave heating, substantially using the energy feature of microwave.Microwave energy is directed through medium, and medium interior molecules exist
Absorbing aggravate molecular thermalmotion after microwave energy, so as to temperature rising, reaches the purpose of heating.Relative to other traditional heating
Mode, it has the advantages that, and high efficiency, pollution-free, firing rate be fast, thermal loss is little.Microwave heating is used as a kind of new cleaning
Mode of heating, it undoubtedly has very big using value.
But in microwave heating process, the coupling condition of complicated electromagnetic fiele and temperature field can be related to, be heated
The dielectric coefficient of medium, thermal conductivity can change with the rising of temperature, and the change of these uncertainties may cause matchmaker
Matter the hot-spot even phenomenon of thermal runaway.There is such case, if could not process in time, safe thing can be caused when serious
Therefore.
The content of the invention
Present invention aim to address in microwave heating process, the problem of hot-spot (thermal runaway).
The technical scheme adopted to realize the object of the invention is such, and one kind is special based on microwave heating temperature field distribution
Levy the local temperature variation abnormality detection method of deep learning, it is characterised in that:
The temperature anomaly detection model of microwave heating equipment is obtained by step 1~4:
1) in the microwave heating temperature field of microwave heating equipment, m acquisition node is arranged, n0 moment, obtains every
The temperature data of one acquisition node, forms multidimensional data sample set Cq;
2) the multidimensional data sample set C for obtaining step 1qIt is normalized, obtains data
3) depth characteristic is extracted:
3-1) select a training sample sample S in the multidimensional data of the microwave distribution field after normalizationqAs
Input variable, sets convolution kernel size LkWith step-length Ak, obtain the data set H of Feature Mapping value compositionk。
The size of pond layer 3-2) is set, using the method processing data collection H in maximum pondk, obtain data set Ik,
3-3) by data set IkBring autocoder into, by minimal reconstruction error, reversely repaiied using back-propagation algorithm
The parameter of each layer of positive network, final study is to featureThe step of repeating 3), constantly learns new sample data, by network
Iteration layer by layer, final updated obtains characteristic data set
4) with data setFor the input of Outlier Detection Algorithm Isolation Forest (isolated forest),
Obtain temperature anomaly detection model:X is the input matrix of model;
During work, by step, whether abnormal microwave heating equipment temperature is obtained:
A) adopt and step 1) identical method, in recording m acquisition node of microwave heating equipment, each collection section
The temperature data n0 moment is put, and is saved in matrix E;
B) adopt and step 2) identical method, by matrix E, normalized obtains matrix F;
C) the data sample F that selected data is concentrated sets length L of convolutional layer sliding window as input variableKAnd step
Long AK, extract each line parameter from input variable successively, using the described each line parameter of sliding window traversal, cut
For some data slotsWherein, u isRound up the value of acquisition, k=1,2 ... K, and K is the convolutional network number of plies.
D) with data slotAnd linear filterConvolution is carried out, along with a biasing pbkAs activation letter
Several inputs, wherein activation primitive adopt sigmoid, tanh, or relu, and by activation primitive Feature Mapping value composition is obtained
Data set PHk:
E the pond window for GK, maximum pond data set PH) are sizedk, obtain data set PIk:
F) by data set PIkBring autocoder into, it is reverse successively using back-propagation algorithm by minimal reconstruction error
The each layer parameter of corrective networks, training obtains feature
Using with step 3) identical method learns new data training sample, updates step 3-1) and described in input become
Amount, obtains each layer parameter of network and trains new sample according to previous training sample, and k, convolutional layer sliding window are updated with k+1
Length Lk+1 and step-length Ak+1 it is constant, the length of pond window is also constant.Repeat process 3), final study is to feature
LKAnd AKRespectively kth carries out step D) to E) when, the setting convolutional layer sliding window length of setting and step-length;
GKStep D is carried out for kth) arrive E) when, the pond window size of setting;
G) by eigenmatrixAs step 4) obtained by temperature anomaly detection model:'s
Input matrix, output abnormality or normal.
What deserves to be explained is, the present invention adopts multilamellar convolutional network structure, and from original multi-dimensional data signal depth is extracted
The feature representation of level, then delivers this to Isolation Forest (isolated forest) abnormality detection models to detect exception
Feature.It is that a kind of initial data that efficiently can be gathered from microwave heating process extracts more preferable feature representation, will extracts
Feature bring abnormality detection system into and detected, whether the temperature change that can reliably measure heated medium distribution field deposits
In exception.
Description of the drawings
Fig. 1. method of the present invention flow process
Fig. 2. convolutional neural networks feature extraction schematic diagram
Fig. 3. automatic encoding Feature Mapping structure chart
Fig. 4. adopt the method for the present invention, 700 power microwave data to do abnormality detection (solid dot is abnormal)
Fig. 5. adopt the method for the present invention, 800 power microwave data to do abnormality detection (solid dot is abnormal).
Specific embodiment
With reference to embodiment, the invention will be further described, but should not be construed above-mentioned subject area of the invention only
It is limited to following embodiments.Without departing from the idea case in the present invention described above, according to ordinary skill knowledge and used
With means, various replacements and change are made, all should be included within the scope of the present invention.
A kind of abnormality detection side of the multidimensional big data information of the microwave heating process thermo parameters method based on deep learning
Method,
Build background analysis system and detection means;
The background analysis system is to the data sample C in raw microwave heating processqWhen being analyzed, by step 1
~4 temperature anomaly detection models for obtaining microwave heating equipment:
Comprise the following steps:
1) primary data sample C is obtainedq, q is data sample numbering, q=1,2 ...,
Wherein: For the microwave data of a node, sample CqIn, tool
There is m different node, the data of each node are collected in t, t=1,2 ... n0;
2) by primary data sample CqNormalized, obtainsBuild data
Collection { S1, S2 ... }
3) depth characteristic is extracted:
3-1) the data sample S that selected data is concentratedqAs input variable, size L of convolution kernel is setkWith step-length Ak, according to
It is secondary that each line parameter is extracted from input variable, using the described each line parameter of sliding window traversal, it is cut into some
Data slotWherein, i isRound up the value of acquisition, k=1,2 ... K, and K is the convolutional network number of plies;
3-2) with data slotAnd linear filterConvolution is carried out, along with a biasing bkAs activation primitive
Input variable, by activation primitive obtain Feature Mapping value composition data set Hk, activation primitive therein can adopt
The functions such as sigmoid, tanh, or relu.
The pond window for GK 3-3) is sized, using maximum pond method processing data collection Hk, obtain data set
Ik:
3-4) characteristic data set I will be obtainedkAutocoder is updated to, by minimal reconstruction error, using back propagation
The each layer parameter of the reverse layer-by-layer correction network of algorithm, training obtains feature
Using with step 3) identical method learns new data training sample, and with new samples step 3-1 is updated) described in
Input variable, each layer parameter of network obtained according to previous training sample trains new sample, and k, convolutional layer are updated with k+1
Length Lk+1 and step-length Ak+1 of sliding window is constant, and the length of pond window is also constant.Repeat process 3), final study is arrived
Feature
LKAnd AKRespectively kth carries out step 3-2) arrive 3-3) when, the size and step-length of the setting convolution kernel of setting;GK
Step 3-2 is carried out for kth) arrive 3-3) when, the pond window size of setting;
4) with data setFor the input of Isolation Forest, abnormality score S (X) is obtained,
Obtain temperature anomaly detection model:
The detection means includes data collection section and data analysis component;During work, by step A~G, obtain micro-
Whether Wave heating unit temp is abnormal:
A) data of the data collection section collection heating node are saved in matrix E, and pass to data analysis component;
Wherein:
For the data of a node, in sample E, have
M different node, the parameter of each node is collected at the pt moment, pt=1,2 ... pn0;
B) adopt and step 2) identical method, by data sample E, normalized is obtained
C) the data sample F1 that selected data is concentrated sets length L of convolutional layer sliding window as input variableK
With step-length AK, extract each line parameter from input variable successively, using the described each line parameter of sliding window traversal, will
It is cut into some data slotsWherein, u isRound up the value of acquisition, k=1,2 ... K, and K is convolutional network
The number of plies.
D) with data slotAnd linear filterConvolution is carried out, along with a biasing pbkAs activation letter
Several inputs, wherein activation primitive can obtain Feature Mapping value using sigmoid, tanh, or relu etc. by activation primitive
The data set PH of compositionk:
E the pond window for GK, maximum pond data set PH) are sizedk, obtain data set PIk:
F) by data set PIkBring autocoder into, it is reverse successively using back-propagation algorithm by minimal reconstruction error
The each layer parameter of corrective networks, training obtains feature
Using with step 3) identical method learns new data training sample, updates step 3-1) and described in input variable,
The each layer parameter of network obtained according to previous training sample trains new sample, and k is updated with k+1, convolutional layer sliding window
Length Lk+1 and step-length Ak+1 are constant, and the length of pond window is also constant.Repeat process 3), final study is to feature
LKAnd AKRespectively kth carries out step D) to E) when, the setting convolutional layer sliding window length of setting and step-length;
GKStep D is carried out for kth) arrive E) when, the pond window size of setting;
G) by eigenmatrixAs the temperature anomaly detection model:Input matrix, it is defeated
Go out abnormal or normal.
Claims (1)
1. it is a kind of based on microwave heating temperature field distribution depths of features learn local temperature variation abnormality detection method, its feature
It is:
The temperature anomaly detection model of microwave heating equipment is obtained by step 1~4:
1) in the microwave heating temperature field of the microwave heating equipment, m acquisition node is arranged, n0 moment, obtains every
The temperature data of one acquisition node, forms multidimensional data sample set Cq;
2) the multidimensional data sample set C for obtaining step 1qIt is normalized, obtains data
3) depth characteristic is extracted:
3-1) select a training sample S in the multidimensional data of the microwave distribution field after normalizationqAs input variable,
Setting convolution kernel size LkWith step-length Ak, obtain the data set H of Feature Mapping value compositionk。
The size of pond layer 3-2) is set, using the method processing data collection H in maximum pondk, obtain data set Ik,
3-3) by data set IkAutocoder is brought into, by minimal reconstruction error, using the reverse corrective networks of back-propagation algorithm
The parameter of each layer, final study is to featureThe step of repeating 3), constantly learns new sample data, by network layer by layer
Iteration, final updated obtains characteristic data set
4) with data setFor the input of Outlier Detection Algorithm Isolation Forest (isolated forest),
Obtain temperature anomaly detection model:X is the input matrix of model;
During work, by step, whether abnormal microwave heating equipment temperature is obtained:
A) adopt and step 1) identical method, in recording m acquisition node of microwave heating equipment, each acquisition node exists
The temperature data at n0 moment, and it is saved in matrix E;
B) adopt and step 2) identical method, by matrix E, normalized obtains matrix F;
C) the data sample F that selected data is concentrated sets length L of convolutional layer sliding window as input variableKWith step-length AK,
Each line parameter is extracted from input variable successively, using the described each line parameter of sliding window traversal, if being cut into
Dry data slotWherein, u isRound up the value of acquisition, k=1,2 ... K, and K is the convolutional network number of plies.
D) with data slotAnd linear filterConvolution is carried out, along with a biasing pbkAs activation primitive
Input, wherein activation primitive adopts sigmoid, tanh, or relu, and by activation primitive the data of Feature Mapping value composition are obtained
Collection PHk:
E the pond window for GK, maximum pond data set PH) are sizedk, obtain data set PIk:
F) by data set PIkAutocoder is brought into, by minimal reconstruction error, using the reverse layer-by-layer correction of back-propagation algorithm
The each layer parameter of network, training obtains feature
Using with step 3) identical method learns new data training sample, updates step 3-1) and described in input variable, press
The each layer parameter of network being obtained according to previous training sample and training new sample, k, the length of convolutional layer sliding window are updated with k+1
Degree Lk+1 and step-length Ak+1 are constant, and the length of pond window is also constant.Repeat process 3), final study is to feature
LKAnd AKRespectively kth carries out step D) to E) when, the setting convolutional layer sliding window length of setting and step-length;GKFor
Kth carries out step D) arrive E) when, the pond window size of setting;
G) by eigenmatrixAs step 4) obtained by temperature anomaly detection model:Input square
Battle array, output abnormality or normal.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611106593.5A CN106682685B (en) | 2016-12-06 | 2016-12-06 | Local temperature change abnormity detection method based on microwave heating temperature field distribution characteristic deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611106593.5A CN106682685B (en) | 2016-12-06 | 2016-12-06 | Local temperature change abnormity detection method based on microwave heating temperature field distribution characteristic deep learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106682685A true CN106682685A (en) | 2017-05-17 |
CN106682685B CN106682685B (en) | 2020-05-01 |
Family
ID=58866338
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611106593.5A Active CN106682685B (en) | 2016-12-06 | 2016-12-06 | Local temperature change abnormity detection method based on microwave heating temperature field distribution characteristic deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106682685B (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108182450A (en) * | 2017-12-25 | 2018-06-19 | 电子科技大学 | A kind of airborne Ground Penetrating Radar target identification method based on depth convolutional network |
CN109521725A (en) * | 2017-09-20 | 2019-03-26 | 西门子公司 | The method, apparatus and equipment and machine readable media of detection abnormal data |
CN109655483A (en) * | 2018-12-14 | 2019-04-19 | 四川大学 | A kind of material microstructure defect inspection method based on deep learning algorithm |
WO2020010701A1 (en) * | 2018-07-11 | 2020-01-16 | 平安科技(深圳)有限公司 | Pollutant anomaly monitoring method and system, computer device, and storage medium |
CN110766056A (en) * | 2019-09-27 | 2020-02-07 | 中山大学 | Abnormal image detection method integrating image generation and multi-label classification |
CN108777873B (en) * | 2018-06-04 | 2021-03-02 | 江南大学 | Wireless sensor network abnormal data detection method based on weighted mixed isolated forest |
CN113177290A (en) * | 2021-03-25 | 2021-07-27 | 中国人民解放军军事科学院国防科技创新研究院 | Satellite component temperature field prediction method based on depth proxy model normalization |
CN113253125A (en) * | 2021-05-19 | 2021-08-13 | 北方工业大学 | Information fusion-based lithium iron phosphate battery thermal runaway monitoring method and system |
CN114547970A (en) * | 2022-01-25 | 2022-05-27 | 中国长江三峡集团有限公司 | Intelligent diagnosis method for abnormity of top cover drainage system of hydraulic power plant |
CN111428886B (en) * | 2020-04-10 | 2023-08-04 | 青岛聚好联科技有限公司 | Method and device for adaptively updating deep learning model of fault diagnosis |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070010906A1 (en) * | 2005-07-11 | 2007-01-11 | Tokyo Electron Limited | Apparatus and system for monitoring a substrate processing, program for monitoring the processing and storage medium storing same |
CN104755892A (en) * | 2012-10-26 | 2015-07-01 | 富士通株式会社 | Temperature measuring system and abnormality detecting method |
CN105164508A (en) * | 2013-02-18 | 2015-12-16 | 赛拉诺斯股份有限公司 | Systems and methods for collecting and transmitting assay results |
CN105447512A (en) * | 2015-11-13 | 2016-03-30 | 中国科学院自动化研究所 | Coarse-fine optical surface defect detection method and coarse-fine optical surface defect detection device |
CN105678267A (en) * | 2016-01-08 | 2016-06-15 | 浙江宇视科技有限公司 | Scene recognition method and device |
-
2016
- 2016-12-06 CN CN201611106593.5A patent/CN106682685B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070010906A1 (en) * | 2005-07-11 | 2007-01-11 | Tokyo Electron Limited | Apparatus and system for monitoring a substrate processing, program for monitoring the processing and storage medium storing same |
CN104755892A (en) * | 2012-10-26 | 2015-07-01 | 富士通株式会社 | Temperature measuring system and abnormality detecting method |
CN105164508A (en) * | 2013-02-18 | 2015-12-16 | 赛拉诺斯股份有限公司 | Systems and methods for collecting and transmitting assay results |
CN105447512A (en) * | 2015-11-13 | 2016-03-30 | 中国科学院自动化研究所 | Coarse-fine optical surface defect detection method and coarse-fine optical surface defect detection device |
CN105678267A (en) * | 2016-01-08 | 2016-06-15 | 浙江宇视科技有限公司 | Scene recognition method and device |
Non-Patent Citations (2)
Title |
---|
NARASIMHA PRASAD.ETC: "Monitoring and Extracting Abnormalities in Land Surface Temperature Images for Automatic Identification of Forest Fires", 《2013 EUROPEAN MODELLING SYMPOSIUM》 * |
崔丽艳: "基于图像分析的微波加热温度场监测系统研究与实现", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109521725A (en) * | 2017-09-20 | 2019-03-26 | 西门子公司 | The method, apparatus and equipment and machine readable media of detection abnormal data |
US11379745B2 (en) | 2017-09-20 | 2022-07-05 | Siemens Aktiengesellschaft | Method, apparatus, device and machine-readable medium for detecting abnormal data using an autoencoder |
CN108182450A (en) * | 2017-12-25 | 2018-06-19 | 电子科技大学 | A kind of airborne Ground Penetrating Radar target identification method based on depth convolutional network |
CN108182450B (en) * | 2017-12-25 | 2021-03-30 | 电子科技大学 | Airborne ground penetrating radar target identification method based on deep convolutional network |
CN108777873B (en) * | 2018-06-04 | 2021-03-02 | 江南大学 | Wireless sensor network abnormal data detection method based on weighted mixed isolated forest |
WO2020010701A1 (en) * | 2018-07-11 | 2020-01-16 | 平安科技(深圳)有限公司 | Pollutant anomaly monitoring method and system, computer device, and storage medium |
CN109655483A (en) * | 2018-12-14 | 2019-04-19 | 四川大学 | A kind of material microstructure defect inspection method based on deep learning algorithm |
CN110766056A (en) * | 2019-09-27 | 2020-02-07 | 中山大学 | Abnormal image detection method integrating image generation and multi-label classification |
CN111428886B (en) * | 2020-04-10 | 2023-08-04 | 青岛聚好联科技有限公司 | Method and device for adaptively updating deep learning model of fault diagnosis |
CN113177290A (en) * | 2021-03-25 | 2021-07-27 | 中国人民解放军军事科学院国防科技创新研究院 | Satellite component temperature field prediction method based on depth proxy model normalization |
CN113177290B (en) * | 2021-03-25 | 2023-09-26 | 中国人民解放军军事科学院国防科技创新研究院 | Satellite component temperature field prediction method based on depth agent model normalization |
CN113253125A (en) * | 2021-05-19 | 2021-08-13 | 北方工业大学 | Information fusion-based lithium iron phosphate battery thermal runaway monitoring method and system |
CN113253125B (en) * | 2021-05-19 | 2023-02-17 | 北方工业大学 | Information fusion-based lithium iron phosphate battery thermal runaway monitoring method and system |
CN114547970A (en) * | 2022-01-25 | 2022-05-27 | 中国长江三峡集团有限公司 | Intelligent diagnosis method for abnormity of top cover drainage system of hydraulic power plant |
CN114547970B (en) * | 2022-01-25 | 2024-02-20 | 中国长江三峡集团有限公司 | Intelligent diagnosis method for abnormality of top cover drainage system of hydropower plant |
Also Published As
Publication number | Publication date |
---|---|
CN106682685B (en) | 2020-05-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106682685A (en) | Microwave heating temperature field distribution characteristic deep learning-based local temperature variation anomaly detection method | |
CN109580215A (en) | A kind of wind-powered electricity generation driving unit fault diagnostic method generating confrontation network based on depth | |
CN103927534B (en) | A kind of online visible detection method of coding character based on convolutional neural networks | |
Bahmani et al. | Mass transfer kinetics of eggplant during osmotic dehydration by neural networks | |
CN109446635B (en) | Electric power industrial control attack classification method and system based on machine learning | |
CN110210434A (en) | Pest and disease damage recognition methods and device | |
CN109978079A (en) | A kind of data cleaning method of improved storehouse noise reduction self-encoding encoder | |
CN110532674B (en) | Method for measuring temperature of boiler furnace of coal-fired power plant | |
CN109460621A (en) | Based on the gas turbine blades fault detection method for improving neural network | |
CN104361414B (en) | Power transmission line icing prediction method based on correlation vector machine | |
CN106951783A (en) | A kind of Method for Masquerade Intrusion Detection and device based on deep neural network | |
CN104318305B (en) | Inverter low-frequency noise fault diagnosis method based on wavelets and neural network | |
CN115018021A (en) | Machine room abnormity detection method and device based on graph structure and abnormity attention mechanism | |
Liang et al. | Multi-sensor fusion approach for fire alarm using BP neural network | |
Liu et al. | Fault diagnosis of PV array based on optimised BP neural network by improved adaptive genetic algorithm | |
CN112632680B (en) | Large civil engineering structure water leakage condition reconstruction method based on deep learning | |
CN110289987A (en) | Multi-agent system network resilience appraisal procedure based on representative learning | |
CN115052304A (en) | GCN-LSTM-based industrial sensor network abnormal data detection method | |
Boeri et al. | Predicting the drying kinetics of salted codfish (Gadus Morhua): semi‐empirical, diffusive and neural network models | |
Gombay et al. | Limit theorems for change in linear regression | |
CN102123062A (en) | Network data anomaly detection method based on dendritic cell algorithm | |
CN116316429B (en) | Distribution transformer relay protection method and device based on insulating oil on-line monitoring | |
CN105095923A (en) | Image processing method and device | |
CN108459585A (en) | Power station fan method for diagnosing faults based on sparse locally embedding depth convolutional network | |
CN105004363A (en) | Sensor performance on-line test device and method based on multi-threshold wavelet under strong interference |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |