CN117574689A - Power state evaluation method and system based on BiGRU - Google Patents
Power state evaluation method and system based on BiGRU Download PDFInfo
- Publication number
- CN117574689A CN117574689A CN202410056355.6A CN202410056355A CN117574689A CN 117574689 A CN117574689 A CN 117574689A CN 202410056355 A CN202410056355 A CN 202410056355A CN 117574689 A CN117574689 A CN 117574689A
- Authority
- CN
- China
- Prior art keywords
- power supply
- training
- biglu
- output voltage
- power state
- 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
- 238000011156 evaluation Methods 0.000 title claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 61
- 238000013135 deep learning Methods 0.000 claims abstract description 28
- 238000013210 evaluation model Methods 0.000 claims abstract description 16
- 230000006872 improvement Effects 0.000 claims abstract description 16
- 238000000034 method Methods 0.000 claims abstract description 15
- 238000012795 verification Methods 0.000 claims abstract description 15
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 9
- 238000004088 simulation Methods 0.000 claims description 39
- 238000004590 computer program Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 9
- 210000002569 neuron Anatomy 0.000 claims description 6
- 230000007246 mechanism Effects 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 3
- 239000003990 capacitor Substances 0.000 claims description 2
- 238000007599 discharging Methods 0.000 claims description 2
- 230000007774 longterm Effects 0.000 abstract description 6
- 238000003745 diagnosis Methods 0.000 abstract description 5
- 238000012545 processing Methods 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000000342 Monte Carlo simulation Methods 0.000 description 1
- 230000008033 biological extinction Effects 0.000 description 1
- 208000028659 discharge Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
-
- 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
Abstract
The invention discloses a power state evaluation method and a system based on BiGRU, and relates to the technical field of fault diagnosis of analog circuit electronic devices; the method comprises the following steps: acquiring training data; the training data comprise output voltage data of a power supply sample and fault types corresponding to the output voltage data of the power supply sample; dividing the training data into a training set and a verification set; constructing a deep learning network model based on BiGRU improvement; training and verifying the deep learning network model by adopting a gradient descent algorithm based on the training set and the verification set to obtain a power state evaluation model; and inputting the output voltage data of the power supply to be detected into the power supply state evaluation model to obtain the fault type of the power supply to be detected. According to the invention, by introducing the deep learning network model based on BiGRU improvement, the long-term dependency relationship in the time sequence data can be effectively captured, and the accurate assessment of the power supply state is realized.
Description
Technical Field
The invention relates to the technical field of fault diagnosis of analog circuit electronic devices, in particular to a power state evaluation method and system based on BiGRU.
Background
Along with the rapid development of electronic technology, the application of electronic systems is more and more widespread, and the power supply is used as an important module for maintaining the operation of the electronic systems, so that the state evaluation is carried out on the electronic systems, and the electronic systems have important significance for guaranteeing the reliable operation of the systems. The conventional fault diagnosis state evaluation method needs to rely on a great deal of experience and complex signal processing technology.
In recent years, the rapid development of deep learning technology provides a new solution for power state evaluation. Deep learning is a machine learning method based on a neural network, and has the advantages of automatic learning characteristics and pattern recognition capability. Among them, a model based on a recurrent neural network (Recurrent Neural Network, RNN) is widely used for modeling and prediction tasks of time series data. However, the conventional RNN model is prone to problems of gradient extinction or gradient explosion when dealing with long-term dependencies.
Disclosure of Invention
The invention aims to provide a power state evaluation method and a system based on BiGRU, which can effectively capture long-term dependency in time sequence data and realize accurate evaluation of power state by introducing a deep learning network model based on BiGRU improvement.
In order to achieve the above object, the present invention provides the following solutions:
a biglu-based power state evaluation method, the biglu-based power state evaluation method comprising:
acquiring training data; the training data comprise output voltage data of a power supply sample and fault types corresponding to the output voltage data of the power supply sample;
dividing the training data into a training set and a verification set;
constructing a deep learning network model based on BiGRU improvement;
training and verifying the deep learning network model by adopting a gradient descent algorithm based on the training set and the verification set to obtain a power state evaluation model;
and inputting the output voltage data of the power supply to be detected into the power supply state evaluation model to obtain the fault type of the power supply to be detected.
Optionally, the acquiring training data specifically includes:
setting up a simulation model of a power supply sample in Pspice simulation software;
analyzing the simulation model to obtain key devices;
the parameters of key devices of the simulation model are adjusted for multiple times, different faults are injected into the simulation model, and simulation models with multiple fault types are obtained;
discharging the simulation model of each fault type to obtain output voltage data of the simulation model of each fault type;
the training data is constructed based on the output voltage data of the simulation model for each fault type.
Optionally, analyzing the simulation model to obtain a key device, which specifically includes:
parameter adjustment is carried out on a basic device of the simulation model, and a device affecting the output of a power supply sample is determined; the basic device comprises a capacitor, a resistor and an inductor;
and outputting the device which influences the output of the power supply sample as the key device.
Optionally, the biglu model includes a biglu-based improved feature extraction module, an attention mechanism, a fully connected layer, and a Softmax layer connected in sequence.
Optionally, the feature extraction module based on the BiGRU improvement comprises a BiGRU layer and a GRU layer which are sequentially connected;
the biglu layer and the GRU layer each include a plurality of GRU neurons.
A power state evaluation system based on biglu, which is applied to the above power state evaluation method based on biglu, the power state evaluation system based on biglu includes:
the acquisition module is used for acquiring training data; the training data comprise output voltage data of a power supply sample and fault types corresponding to the output voltage data of the power supply sample;
the classification module is used for dividing the training data into a training set and a verification set;
the construction module is used for constructing a deep learning network model based on BiGRU improvement;
the training module is used for training and verifying the deep learning network model by adopting a gradient descent algorithm based on the training set and the verification set to obtain a power state evaluation model;
and the prediction module is used for inputting the output voltage data of the power supply to be detected into the power supply state evaluation model to obtain the fault type of the power supply to be detected.
An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the biglu-based power state assessment method as described above when the computer program is executed.
A computer readable storage medium having stored thereon a computer program which when executed implements a biglu-based power state assessment method as described above.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention discloses a power state evaluation method and a system based on BiGRU, wherein the method comprises the following steps: acquiring training data; the training data comprise output voltage data of a power supply sample and fault types corresponding to the output voltage data of the power supply sample; dividing the training data into a training set and a verification set; constructing a deep learning network model based on BiGRU improvement; training and verifying the deep learning network model by adopting a gradient descent algorithm based on the training set and the verification set to obtain a power state evaluation model; and inputting the output voltage data of the power supply to be detected into the power supply state evaluation model to obtain the fault type of the power supply to be detected. According to the invention, by introducing the deep learning network model based on BiGRU improvement, the long-term dependency relationship in the time sequence data can be effectively captured, and the accurate assessment of the power supply state is realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a BiGRU-based power state evaluation method in an embodiment of the invention;
FIG. 2 is a schematic diagram of the internal architecture of a GRU neuron according to the present invention;
FIG. 3 is a schematic diagram of a power state estimation model according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention aims to provide a power state evaluation method and a system based on BiGRU, which can effectively capture long-term dependency in time sequence data and realize accurate evaluation of power state by introducing a deep learning network model based on BiGRU improvement.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Example 1
As shown in fig. 1, the power state evaluation method based on biglu of the present invention includes:
step 101: acquiring training data; the training data comprises output voltage data of the power supply sample and fault types corresponding to the output voltage data of the power supply sample.
Step 102: the training data is divided into a training set and a validation set.
Step 103: and constructing a deep learning network model based on BiGRU improvement.
Step 104: and training and verifying the deep learning network model by adopting a gradient descent algorithm based on the training set and the verification set to obtain a power state evaluation model.
In specific implementation, 200 sets of training data acquired under each fault are divided, wherein 150 sets of data are used as training sets, and 50 sets of data are used as test sets;
and carrying out sliding window processing on the training set, and dividing each group of data to ensure that the length of each group of data accords with the input length of the network.
Encoding fault tagsN is the number of fault types.
And inputting the training set subjected to sliding window processing and the corresponding fault codes into a deep learning network model, optimizing parameters of the deep learning network model through a gradient descent algorithm, and improving the accuracy of predicting the fault labels of the input data by the deep learning network model.
And after the deep learning network model is trained, sliding window processing is carried out on the test set data, the test set data and the coded corresponding label are input into the model, and the effectiveness of the diagnosis deep learning network model is verified.
Step 105: and inputting the output voltage data of the power supply to be detected into the power supply state evaluation model to obtain the fault type of the power supply to be detected.
As an embodiment, the acquiring training data specifically includes:
and constructing a simulation model of the power supply sample in Pspice simulation software.
And analyzing the simulation model to obtain the key device.
And (3) carrying out multiple times of adjustment on parameters of key devices of the simulation model, and injecting different faults into the simulation model to obtain the simulation models with various fault types.
In specific implementation, aiming at key devices influencing power supply output, parameters of the key devices are modified on software to inject faults into a simulation model.
And performing discharge treatment on the simulation model of each fault type to obtain output voltage data of the simulation model of each fault type.
200 Monte Carlo simulations were performed for each fault with a critical device parameter drift of 50% as the fault state.
Output voltage data of a simulation model of the Monte Carlo analysis is saved.
And carrying out standardization processing on output data of the simulation model obtained by Monte Carlo analysis: the calculation formula is as follows:
；
wherein,an output voltage value of the simulation model at a certain time i after the normalization processing is represented; />The output voltage value of the simulation model at a certain moment i, and (2)>、/>Is the maximum and minimum of the output voltages of the simulation model in a set of sampled data.
The training data is constructed based on the output voltage data of the simulation model for each fault type.
As an embodiment, the analyzing the simulation model to obtain the key device specifically includes:
parameter adjustment is carried out on a basic device of the simulation model, and a device affecting the output of a power supply sample is determined; the base device includes a capacitance, a resistance, and an inductance. The resistance in the built simulation model is set to be 5% of tolerance, the capacitance tolerance is set to be 10%, the inductance tolerance is set to be 20%, the device value can drift along with the tolerance setting, and the drift is set to be normal distribution so as to simulate the real situation. The tolerance is the range of the adjusted parameter relative to the normal parameter when the parameter is adjusted. For example, if the normal value of the resistance in the simulation model is 10Ω, the adjustment range of the resistance in the simulation model is 9.5-10.5Ω.
And outputting the device which influences the output of the power supply sample as the key device.
As shown in fig. 3, the biglu model includes, as one embodiment, a biglu-based improved feature extraction module, an attention mechanism, a fully connected layer, and a Softmax layer, which are connected in sequence.
As an embodiment, the feature extraction module based on biglu improvement includes a biglu layer and a GRU layer connected in sequence.
As shown in fig. 2, both the biglu layer and the GRU layer include a plurality of GRU neurons.
In a specific implementation, a feature extraction module based on BiGRU improvement is built, and the feature extraction module comprises a BiGRU layer and GRU layers, wherein each layer is composed of GRU neurons. The GRU neurons can find the sequence correlation between samples in time sequence, which can be used for input signalsAnd performing nonlinear operation to obtain the signal time sequence related characteristics. The calculation formula is as follows:
；
wherein:is a hidden state at the moment t; />Is the input at time t; />Is a hidden state at the moment t-1; />、、/>Reset gate, update gate and calculate candidate hidden layer respectively; />Representing a sigmoid activation function, tanh being a hyperbolic tangent activation function,/for>Indicating +.>Corresponding weights, ++>Indicating +.>Corresponding weights, ++>Representing +.>Corresponding weights, ++>Representing +.>Corresponding weights, ++>Representing the computation of the +.>Corresponding weights, ++>Representing the computation of the +.>Corresponding weights, ++>Indicating +.>The corresponding deviation is used to determine the deviation,indicating +.>Corresponding deviation->Representing +.>Corresponding deviation->Representing +.>Corresponding deviation->Representing the computation of the +.>Corresponding deviation->Representing the computation of the +.>Corresponding biasAnd (3) difference. During each unit transfer, +.>For controlling the state before reservation is required. If->=0, thenOnly information of the current input state is contained, and +.>For controlling the amount of information that the hidden state needs to be forgotten at a previous time.
And adding an attention mechanism to the network output part of the feature extraction module, weighting the extracted features, and distributing weight parameters according to the importance degree of the extracted feature information.
And building a fully-connected neural network layer, taking the characteristic information weighted by the attention mechanism as input, and outputting the characteristic information as a probability value corresponding to the fault label for judging a final diagnosis result.
Example 2
A biglu-based power state evaluation system applied to the biglu-based power state evaluation method of embodiment 1, the biglu-based power state evaluation system comprising:
the acquisition module is used for acquiring training data; the training data comprises output voltage data of the power supply sample and fault types corresponding to the output voltage data of the power supply sample.
And the classification module is used for dividing the training data into a training set and a verification set.
And the construction module is used for constructing a deep learning network model based on BiGRU improvement.
And the training module is used for training and verifying the deep learning network model by adopting a gradient descent algorithm based on the training set and the verification set to obtain a power state evaluation model.
And the prediction module is used for inputting the output voltage data of the power supply to be detected into the power supply state evaluation model to obtain the fault type of the power supply to be detected.
An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the biglu-based power state assessment method according to embodiment 1 when the computer program is executed.
A computer-readable storage medium having stored thereon a computer program which, when executed, implements the biglu-based power state assessment method according to embodiment 1.
According to the invention, by introducing the deep learning network model based on BiGRU improvement, the long-term dependency relationship in the time sequence data can be effectively captured, and the accurate assessment of the power supply state is realized.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to assist in understanding the methods of the present invention and the core ideas thereof; also, it is within the scope of the present invention to be modified by those of ordinary skill in the art in light of the present teachings. In view of the foregoing, this description should not be construed as limiting the invention.
Claims (8)
1. The power state evaluation method based on the BiGRU is characterized by comprising the following steps of:
acquiring training data; the training data comprise output voltage data of a power supply sample and fault types corresponding to the output voltage data of the power supply sample;
dividing the training data into a training set and a verification set;
constructing a deep learning network model based on BiGRU improvement;
training and verifying the deep learning network model by adopting a gradient descent algorithm based on the training set and the verification set to obtain a power state evaluation model;
and inputting the output voltage data of the power supply to be detected into the power supply state evaluation model to obtain the fault type of the power supply to be detected.
2. The biglu-based power state assessment method according to claim 1, wherein the acquiring training data specifically comprises:
setting up a simulation model of a power supply sample in Pspice simulation software;
analyzing the simulation model to obtain key devices;
the parameters of key devices of the simulation model are adjusted for multiple times, different faults are injected into the simulation model, and simulation models with multiple fault types are obtained;
discharging the simulation model of each fault type to obtain output voltage data of the simulation model of each fault type;
the training data is constructed based on the output voltage data of the simulation model for each fault type.
3. The biglu-based power state assessment method according to claim 2, wherein analyzing the simulation model to obtain key devices specifically comprises:
parameter adjustment is carried out on a basic device of the simulation model, and a device affecting the output of a power supply sample is determined; the basic device comprises a capacitor, a resistor and an inductor;
and outputting the device which influences the output of the power supply sample as the key device.
4. The bigu-based power state assessment method according to claim 1, wherein the bigu model comprises a bigu-improvement-based feature extraction module, an attention mechanism, a full connectivity layer, and a Softmax layer connected in sequence.
5. The biglu-based power state assessment method according to claim 4, wherein the biglu-based improved feature extraction module comprises a biglu layer and a GRU layer connected in sequence;
the biglu layer and the GRU layer each include a plurality of GRU neurons.
6. A biglu-based power state evaluation system, wherein the biglu-based power state evaluation system is applied to the biglu-based power state evaluation method according to any one of claims 1-5, and comprises:
the acquisition module is used for acquiring training data; the training data comprise output voltage data of a power supply sample and fault types corresponding to the output voltage data of the power supply sample;
the classification module is used for dividing the training data into a training set and a verification set;
the construction module is used for constructing a deep learning network model based on BiGRU improvement;
the training module is used for training and verifying the deep learning network model by adopting a gradient descent algorithm based on the training set and the verification set to obtain a power state evaluation model;
and the prediction module is used for inputting the output voltage data of the power supply to be detected into the power supply state evaluation model to obtain the fault type of the power supply to be detected.
7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the biglu-based power state assessment method according to any one of claims 1-5 when the computer program is executed.
8. A computer readable storage medium, wherein a computer program is stored on the storage medium, which when executed implements the biglu-based power state assessment method according to any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410056355.6A CN117574689A (en) | 2024-01-16 | 2024-01-16 | Power state evaluation method and system based on BiGRU |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410056355.6A CN117574689A (en) | 2024-01-16 | 2024-01-16 | Power state evaluation method and system based on BiGRU |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117574689A true CN117574689A (en) | 2024-02-20 |
Family
ID=89886552
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410056355.6A Pending CN117574689A (en) | 2024-01-16 | 2024-01-16 | Power state evaluation method and system based on BiGRU |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117574689A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104793171A (en) * | 2015-04-23 | 2015-07-22 | 广西电网有限责任公司电力科学研究院 | Fault simulation based smart meter fault detection method |
| WO2023231995A1 (en) * | 2022-05-30 | 2023-12-07 | 浙大城市学院 | Transfer-learning-based life prediction and health assessment method for aero-engine |
-
2024
- 2024-01-16 CN CN202410056355.6A patent/CN117574689A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104793171A (en) * | 2015-04-23 | 2015-07-22 | 广西电网有限责任公司电力科学研究院 | Fault simulation based smart meter fault detection method |
| WO2023231995A1 (en) * | 2022-05-30 | 2023-12-07 | 浙大城市学院 | Transfer-learning-based life prediction and health assessment method for aero-engine |
Non-Patent Citations (4)
| Title |
|---|
| 戴伏生: "《基础电子电路设计与实践》", 30 April 2002, pages: 387 - 388 * |
| 杨博文;刘飞;刘侃;王海龙;: "PSpice在电路故障仿真中的应用", 战术导弹技术, no. 04, 15 July 2012 (2012-07-15), pages 120 - 123 * |
| 胡睿: "基于互相关能比熵和ＢｉＧＲＵ－ＧＲＵ的轧机关键零部件早期故障诊断", 《测试与故障诊断》, 22 February 2022 (2022-02-22), pages 95 - 102 * |
| 陈悦然 等: "基于 MCNN-BiGRU-Attention 的轴承故障诊断", 《计算机系统应用》, 21 July 2023 (2023-07-21), pages 1 - 5 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110321603B (en) | Depth calculation model for gas path fault diagnosis of aircraft engine | |
| US20220222409A1 (en) | Method and system for predicting remaining useful life of analog circuit | |
| CN113723007B (en) | Equipment residual life prediction method based on DRSN and sparrow search optimization | |
| CN115545334B (en) | Land utilization type prediction method and device, electronic equipment and storage medium | |
| CN114048468A (en) | Intrusion detection method, intrusion detection model training method, device and medium | |
| CN110956309A (en) | Flow activity prediction method based on CRF and LSTM | |
| CN115587666A (en) | Load prediction method and system based on seasonal trend decomposition and hybrid neural network | |
| Lin et al. | Remaining useful life prediction in prognostics using multi-scale sequence and Long Short-Term Memory network⋆ | |
| CN115308558A (en) | Method and device for predicting service life of CMOS (complementary Metal oxide semiconductor) device, electronic equipment and medium | |
| CN111401473A (en) | Infrared target classification method based on attention mechanism convolutional neural network | |
| Collier et al. | Transfer and marginalize: Explaining away label noise with privileged information | |
| CN114154622A (en) | Algorithm model for traffic operation system flow data acquisition missing completion | |
| CN112766339A (en) | Trajectory recognition model training method and trajectory recognition method | |
| CN117574689A (en) | Power state evaluation method and system based on BiGRU | |
| CN116665798A (en) | Air pollution trend early warning method and related device | |
| CN111783688A (en) | Remote sensing image scene classification method based on convolutional neural network | |
| CN116542701A (en) | Carbon price prediction method and system based on CNN-LSTM combination model | |
| CN116579408A (en) | Model pruning method and system based on redundancy of model structure | |
| CN116432535A (en) | Performance verification error analysis method for electric power metering device considering multiple environmental elements | |
| CN112859034B (en) | Natural environment radar echo amplitude model classification method and device | |
| CN115459982A (en) | Power network false data injection attack detection method | |
| CN111796173B (en) | Partial discharge pattern recognition method, computer device, and storage medium | |
| CN112819067A (en) | Method, device, equipment and storage medium for processing bad data of power distribution network | |
| CN113743670B (en) | GRU model-based circuit fault real-time prediction method and verification circuit | |
| Xie et al. | Unsupervised STDP-based Radioisotope Identification Using Spiking Neural Networks Implemented on SpiNNaker |
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 |