CN111310965A - Aircraft track prediction method based on LSTM network - Google Patents
Aircraft track prediction method based on LSTM network Download PDFInfo
- Publication number
- CN111310965A CN111310965A CN201911041365.8A CN201911041365A CN111310965A CN 111310965 A CN111310965 A CN 111310965A CN 201911041365 A CN201911041365 A CN 201911041365A CN 111310965 A CN111310965 A CN 111310965A
- Authority
- CN
- China
- Prior art keywords
- data
- lstm
- aircraft
- network
- neural network
- 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
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000012549 training Methods 0.000 claims abstract description 31
- 238000013528 artificial neural network Methods 0.000 claims abstract description 29
- 238000011156 evaluation Methods 0.000 claims abstract description 10
- 238000003062 neural network model Methods 0.000 claims abstract description 9
- 238000010606 normalization Methods 0.000 claims abstract description 9
- 210000002569 neuron Anatomy 0.000 claims abstract description 8
- 238000012545 processing Methods 0.000 claims abstract description 7
- 230000000694 effects Effects 0.000 claims abstract description 6
- 238000012360 testing method Methods 0.000 claims description 7
- 238000007726 management method Methods 0.000 description 6
- 238000013135 deep learning Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000000306 recurrent effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield 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
- G06N3/045—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
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Economics (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses an aircraft track prediction method based on an LSTM network, which comprises the following steps: (1) reading in flight path data of the aircraft, and carrying out normalization processing on the data; (2) establishing an LSTM deep neural network model, and setting input and output of a neural network, step number, LSTM layer number and neuron number of each layer; (3) setting evaluation indexes as RMSE and MAE; (4) training a network model, and debugging related parameters to minimize the values of RMSE and MAE, namely optimizing the network effect; (5) and outputting the predicted flight path of the aircraft through the trained optimal network. The method can be used for quickly and effectively predicting the flight path of the aircraft in a period of time in the future, and effectively improving the robustness and the accuracy of the flight path prediction by applying the LSTM network to the problem of path prediction and training the LSTM path prediction model through a large amount of data.
Description
Technical Field
The invention belongs to the technical field of software of algorithm design, and particularly relates to an aircraft track prediction method based on an LSTM network.
Background
The flight path prediction is an important technology in airspace traffic management, realizes accurate prediction on the flight path of an aircraft, and is a necessary condition for realizing the intellectualization of airspace traffic management. The accurate track prediction can improve the efficiency of airspace traffic management, and the inaccurate track prediction can cause the confusion of airspace traffic management and increase the flight risk coefficient of an aircraft. Therefore, flight path prediction of the aircraft has important research value.
A traditional track prediction model based on a BP neural network realizes multi-dimensional aircraft track characteristic prediction. In general, the BP neural network is selected to be 3 layers, increasing the number of layers of the BP network increases training time of the BP neural network and is prone to overfitting, since the mapping representation capability of the BP neural network is limited, BP parameters are difficult to adjust, and in essence, the BP network easily enables the training result of a loss function to fall into a local optimal point rather than a global optimal point, so that the generalization capability of the algorithm model is limited to a certain extent, and the change in a time sequence cannot be modeled.
Deep learning can realize complex function approximation by learning a deep nonlinear network structure, and has the capability of learning essential characteristics of a data set from a few sample sets. The recurrent neural network RNN is a typical model of deep learning that is good at processing time series data. Taking the prediction of the flight path attitude of the aircraft as an example, the future flight path attitude of the aircraft depends on the flight path attitude value at the historical moment. The RNN performs the same operation on all nodes, with the output at the current time being dependent on the previous calculation. And (3) learning the motion rule of the aircraft at the current and past moments by using an RNN-LSTM recurrent neural network, and then predicting the flight path of the aircraft. The idea not only avoids the complicated modeling process in the traditional algorithm, but also ensures that the established model is in accordance with the actual logic.
The LSTM neural network model can effectively keep longer-time memory, and can fully utilize useful remote information in the flight path sequence data, thereby achieving more accurate flight path prediction effect. The LSTM network model can be trained by means of the GPU, so that the training time is greatly shortened, and a track prediction result can be obtained more quickly. The method not only can predict the position of the aircraft in a future period of time, but also can predict the attitude of the aircraft in the future period of time while ensuring the accuracy and rapidity of flight path prediction, so that related personnel can obtain more information, and airspace traffic management is facilitated.
Disclosure of Invention
The purpose of the invention is: in order to make up for the defects of the BP neural network method, the invention provides an aircraft track prediction method based on an LSTM deep neural network. By using the deep learning method, the accuracy of aircraft track prediction is improved, the problems that the training result of the loss function is easy to fall into a local optimal point and overfitting is easy to happen in the traditional BP method are solved, and useful remote information in track sequence data can be fully utilized.
The technical scheme of the invention is as follows: an aircraft track prediction method based on an LSTM network comprises the following steps:
step 1), acquiring the past time and the current track attitude data of the aircraft, and carrying out normalization processing on the data, thereby eliminating the influence on a prediction result due to overlarge or undersize data values; wherein, the data of the 'past time' is 6 to 8 times of the predicted time;
step 2), establishing an LSTM deep neural network model, setting input and output, step length, LSTM layer number and neuron number of each layer of the neural network, representing the data preprocessed in the step 1) into a vector form, and dividing the data represented into the vector form into training data and test data, wherein the training data accounts for 10% -20% of the training data, and the rest is the test data;
step 3), selecting the evaluation indexes of the network model as RMSE and MAE; wherein, RMSE is the root mean square error, which is the square root of the ratio of the square of the deviation between the predicted value and the true value to the observation time n; MAE is the mean absolute error, which is the average of the absolute values of the deviations of all individual observations from the arithmetic mean.
Step 4), training the LSTM network model by using the training data in the step 2) and the setting of the LSTM network, and debugging network related parameters according to the evaluation indexes in the step 3) to enable the RMSE and MAE values to be minimum, namely, the network effect is optimal;
and 5) outputting the predicted track and the predicted attitude of the aircraft through the optimal LSTM deep neural network trained in the step 4).
Compared with the traditional BP neural network method, the method for predicting the flight path of the aircraft based on the LSTM network has the advantages that: the LSTM neural network model used by the invention can effectively keep longer-time memory and can fully utilize useful remote information in the flight path sequence data, thereby achieving more accurate flight path prediction effect. The LSTM network model can be trained by means of the GPU, so that the training time is greatly shortened, and a track prediction result can be obtained more quickly. The method not only can predict the position of the aircraft in a future period of time, but also can predict the attitude of the aircraft in the future period of time while ensuring the accuracy and rapidity of flight path prediction, so that related personnel can obtain more information, and airspace traffic management is facilitated.
Drawings
FIG. 1 is a schematic flow chart of an aircraft track prediction method based on an LSTM network.
FIG. 2 is a diagram of an LSTM deep neural network architecture.
FIG. 3 is a comparison graph of experimental results of a traditional BP neural network method and an LSTM deep neural network method based on RMSE evaluation indexes.
FIG. 4 is a comparison graph of experimental results of a traditional BP neural network method and an LSTM deep neural network method based on MAE evaluation indexes.
FIG. 5 is a graph of aircraft longitude predictions based on the LSTM network.
FIG. 6 is a diagram of aircraft latitude prediction results based on the LSTM network.
FIG. 7 is a graph of aircraft height prediction results based on an LSTM network.
FIG. 8 is a graph of aircraft pitch angle predictions based on an LSTM network.
FIG. 9 is a graph of aircraft roll angle predictions based on an LSTM network.
FIG. 10 is a graph of aircraft yaw angle predictions based on an LSTM network.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings. The following will describe in detail the embodiment of the present invention with reference to fig. 1, and the specific steps are as follows:
and 2, establishing an LSTM deep neural network model, wherein the structure of the LSTM deep neural network is shown as the attached figure 2. Expressing the data preprocessed in the step 1) into a vector form, and dividing the obtained vector form data into training data and test data, wherein the training data accounts for 10% -20% of the training data, and the rest is the test data. The unit processed by the neural network is a vector,the input of the LSTM neural network adds the concept of time sequence step, the processing unit is expanded from vector to tensor, the transverse operation of the recurrent neural network model can be regarded as accumulation, and the memory neuron of the LSTM can select memory or forget accumulated information to predict the output at a certain moment. In an embodiment of the invention, historical aircraft trajectory characteristics at 6-8 times may be selected as input, i.e., the number of steps corresponding to the input layer step is selected to be 6-8. Different steps have different degrees of influence on the accuracy of the network, the selection of the number of neurons in the hidden layer and the accuracy of the experiment. The number of neurons in the input and output layers of the neural network is determined by the characteristics of the training data. In training, the number of LSTM layer neurons has a large influence on the prediction accuracy of the network. If the number of the nodes is set to be too small, the neural network cannot be well fitted and has the problem of learning regression, training times need to be increased in the training process, and the network precision is low; on the contrary, if the number of nodes is too large, the training time is increased and the overfitting is caused, thereby affecting the performance. For the determination of the number of hidden layer nodes, an empirical formula is usually adopted to determine an estimated value, the estimated value is used as an initial value, trial and error are performed through experiments, and a value with the minimum recognition error after training is selected as the number of hidden layer nodes. The empirical formula for calculating the number M of hidden layer nodes is:wherein m and n are the number of nodes of the input layer and the output layer, respectively, and a is [0,10 ]]Is constant. In the embodiment of the invention, the step of the LSTM neural network model input layer is selected to be 6, which represents a matrix with the input scale of 6 multiplied by 6 and corresponds to the position and the posture of the aircraft at six continuous moments; the number of LSTM layer neurons is 14; the output layer is a vector with 1 time, namely the length of 6, and represents the track characteristic at the time of the next time interval. Setting a target error to be 10^ -6, a learning rate to be 0.06, a dropout to be 0.2 and a maximum iteration number to be 10;
and 3, evaluating an aircraft position and attitude prediction model by adopting a Root Mean Square Error (RMSE) and a Mean Absolute Error (MAE). Root mean square error refers to the expected value of the square of the difference between the observed (observed) and the true (predicted) parameter valuesArithmetic square root, RMSE is mathematically expressed as:the mean absolute error represents the average of the absolute errors of the predictions from the actual in a set of test data. The mathematical expression of MAE is:the two error indexes can evaluate the change degree of data, and the smaller the RMSE and MAE values are, the better the accuracy of the experimental model for predicting the position and the attitude of the aircraft is.
And 4) training the LSTM network model by using the training data and the LSTM network in the step 2) and adopting a back propagation algorithm, and debugging network related parameters according to the evaluation indexes in the step 3) to enable the RMSE and MAE values to be minimum, namely the network effect to be optimal. In the training process, the Batch Normalization is used for standardizing the input airway planning environment information on a sample axis, so that the generalization capability of a neural network model can be effectively improved, a target algorithm can be better fitted, the prediction of an airway or airway points is completed, and the training speed is improved to a certain extent;
and 5) outputting the predicted track and the predicted attitude of the aircraft through the optimal LSTM deep neural network trained in the step 4.
The experimental results of the traditional BP neural network method and the LSTM deep neural network method based on the RMSE evaluation index are shown in the attached drawing 3, the experimental results of the traditional BP neural network method and the LSTM deep neural network method based on the MAE evaluation index are shown in the attached drawing 4, and the prediction results of the longitude, the latitude, the altitude, the pitch angle, the roll angle and the yaw angle of the aircraft based on the LSTM network are respectively shown in the attached drawings 5 to 10.
Claims (2)
1. An aircraft track prediction method based on an LSTM network is characterized in that: the algorithm comprises the following steps:
step 1), acquiring the past time and the current track attitude data of the aircraft, and carrying out normalization processing on the data, thereby eliminating the influence on a prediction result due to overlarge or undersize data values;
step 2), establishing an LSTM deep neural network model, setting input and output, step length, LSTM layer number and neuron number of each layer of the neural network, representing the data preprocessed in the step 1) into a vector form, and dividing the data represented into the vector form into training data and test data, wherein the training data accounts for 10% -20% of the training data, and the rest is the test data;
step 3), selecting the evaluation indexes of the network model as RMSE and MAE;
step 4), training the LSTM network model by using the training data in the step 2) and the setting of the LSTM network, and debugging network related parameters according to the evaluation indexes in the step 3) to enable the RMSE and MAE values to be minimum, namely, the network effect is optimal;
and 5) outputting the predicted track and the predicted attitude of the aircraft through the optimal LSTM deep neural network trained in the step 4).
2. The LSTM network-based aircraft trajectory prediction method of claim 1, wherein: the data of the past time is 6-8 times of the predicted time.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911041365.8A CN111310965A (en) | 2019-10-30 | 2019-10-30 | Aircraft track prediction method based on LSTM network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911041365.8A CN111310965A (en) | 2019-10-30 | 2019-10-30 | Aircraft track prediction method based on LSTM network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111310965A true CN111310965A (en) | 2020-06-19 |
Family
ID=71161794
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911041365.8A Pending CN111310965A (en) | 2019-10-30 | 2019-10-30 | Aircraft track prediction method based on LSTM network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111310965A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111897353A (en) * | 2020-07-08 | 2020-11-06 | 西北工业大学 | Aircraft maneuvering trajectory prediction method based on GRU |
CN112115550A (en) * | 2020-09-13 | 2020-12-22 | 西北工业大学 | Aircraft maneuvering trajectory prediction method based on Mogrifier-BiGRU |
CN112612001A (en) * | 2020-11-30 | 2021-04-06 | 天津光电通信技术有限公司 | Track prediction method and device based on BP neural network algorithm |
CN112859898A (en) * | 2021-01-18 | 2021-05-28 | 中山大学 | Aircraft trajectory prediction method based on two-channel bidirectional neural network |
CN112948969A (en) * | 2021-03-01 | 2021-06-11 | 哈尔滨工程大学 | Ship rolling prediction method based on LSTMC hybrid network |
CN113099474A (en) * | 2021-04-08 | 2021-07-09 | 湖南国科轩宇信息科技有限公司 | Method and device for predicting short-term traffic demand of satellite internet user terminal |
CN113671161A (en) * | 2021-07-13 | 2021-11-19 | 郑州大学 | Unmanned aerial vehicle pavement disease detection method based on LSTM neural network algorithm |
CN113673149A (en) * | 2021-08-02 | 2021-11-19 | 南京航空航天大学 | Unmanned aerial vehicle decision evaluation method in complex environment |
CN114756977A (en) * | 2022-06-16 | 2022-07-15 | 成都飞机工业(集团)有限责任公司 | Method, device and equipment for predicting boring cutter yield of intersection hole of airplane and storage medium |
CN115310677A (en) * | 2022-07-18 | 2022-11-08 | 四川大学 | Flight path prediction method and device based on binary coded representation and multi-classification |
CN115935192A (en) * | 2023-01-10 | 2023-04-07 | 中国民用航空飞行学院 | Flight training data prediction method based on incremental online learning framework |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018161722A1 (en) * | 2017-03-08 | 2018-09-13 | 深圳市景程信息科技有限公司 | Power load forecasting method based on long short-term memory neural network |
CN108764560A (en) * | 2018-05-22 | 2018-11-06 | 电子科技大学 | Aircraft scene trajectory predictions method based on shot and long term Memory Neural Networks |
CN109508812A (en) * | 2018-10-09 | 2019-03-22 | 南京航空航天大学 | A kind of aircraft Trajectory Prediction method based on profound memory network |
-
2019
- 2019-10-30 CN CN201911041365.8A patent/CN111310965A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018161722A1 (en) * | 2017-03-08 | 2018-09-13 | 深圳市景程信息科技有限公司 | Power load forecasting method based on long short-term memory neural network |
CN108764560A (en) * | 2018-05-22 | 2018-11-06 | 电子科技大学 | Aircraft scene trajectory predictions method based on shot and long term Memory Neural Networks |
CN109508812A (en) * | 2018-10-09 | 2019-03-22 | 南京航空航天大学 | A kind of aircraft Trajectory Prediction method based on profound memory network |
Non-Patent Citations (2)
Title |
---|
张振兴等: "空战飞行对敌目标逼近航迹预测仿真", 《空军工程大学学报(自然科学版)》 * |
权波等: "基于LSTM的船舶航迹预测模型", 《计算机科学》 * |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111897353B (en) * | 2020-07-08 | 2022-08-02 | 西北工业大学 | Aircraft maneuvering trajectory prediction method based on GRU |
CN111897353A (en) * | 2020-07-08 | 2020-11-06 | 西北工业大学 | Aircraft maneuvering trajectory prediction method based on GRU |
CN112115550A (en) * | 2020-09-13 | 2020-12-22 | 西北工业大学 | Aircraft maneuvering trajectory prediction method based on Mogrifier-BiGRU |
CN112612001A (en) * | 2020-11-30 | 2021-04-06 | 天津光电通信技术有限公司 | Track prediction method and device based on BP neural network algorithm |
CN112859898A (en) * | 2021-01-18 | 2021-05-28 | 中山大学 | Aircraft trajectory prediction method based on two-channel bidirectional neural network |
CN112859898B (en) * | 2021-01-18 | 2022-03-22 | 中山大学 | Aircraft trajectory prediction method based on two-channel bidirectional neural network |
CN112948969A (en) * | 2021-03-01 | 2021-06-11 | 哈尔滨工程大学 | Ship rolling prediction method based on LSTMC hybrid network |
CN113099474A (en) * | 2021-04-08 | 2021-07-09 | 湖南国科轩宇信息科技有限公司 | Method and device for predicting short-term traffic demand of satellite internet user terminal |
CN113099474B (en) * | 2021-04-08 | 2022-08-12 | 湖南国科轩宇信息科技有限公司 | Method and device for predicting short-term traffic demand of satellite internet user terminal |
CN113671161A (en) * | 2021-07-13 | 2021-11-19 | 郑州大学 | Unmanned aerial vehicle pavement disease detection method based on LSTM neural network algorithm |
CN113673149A (en) * | 2021-08-02 | 2021-11-19 | 南京航空航天大学 | Unmanned aerial vehicle decision evaluation method in complex environment |
CN114756977A (en) * | 2022-06-16 | 2022-07-15 | 成都飞机工业(集团)有限责任公司 | Method, device and equipment for predicting boring cutter yield of intersection hole of airplane and storage medium |
CN114756977B (en) * | 2022-06-16 | 2022-10-25 | 成都飞机工业(集团)有限责任公司 | Method, device and equipment for predicting boring cutter yield of intersection hole of airplane and storage medium |
CN115310677A (en) * | 2022-07-18 | 2022-11-08 | 四川大学 | Flight path prediction method and device based on binary coded representation and multi-classification |
CN115935192A (en) * | 2023-01-10 | 2023-04-07 | 中国民用航空飞行学院 | Flight training data prediction method based on incremental online learning framework |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111310965A (en) | Aircraft track prediction method based on LSTM network | |
CN108280551B (en) | Photovoltaic power generation power prediction method utilizing long-term and short-term memory network | |
US11593611B2 (en) | Neural network cooperation | |
CN106022521B (en) | Short-term load prediction method of distributed BP neural network based on Hadoop architecture | |
CN111563706A (en) | Multivariable logistics freight volume prediction method based on LSTM network | |
CN106448151B (en) | Short-term traffic flow prediction method | |
CN110245801A (en) | A kind of Methods of electric load forecasting and system based on combination mining model | |
US11080586B2 (en) | Neural network reinforcement learning | |
Suryo et al. | Improved time series prediction using LSTM neural network for smart agriculture application | |
Kadir et al. | Wheat yield prediction: Artificial neural network based approach | |
Dada et al. | Artificial neural network models for rainfall prediction | |
CN111723523A (en) | Estuary residual water level prediction method based on cascade neural network | |
CN115271186B (en) | Reservoir water level prediction and early warning method based on delay factor and PSO RNN Attention model | |
Yang et al. | LSTM-based deep learning model for civil aircraft position and attitude prediction approach | |
Deng et al. | Visibility Forecast for Airport Operations by LSTM Neural Network. | |
CN113128769A (en) | Intelligent flight delay prediction method based on deep learning | |
CN110738363B (en) | Photovoltaic power generation power prediction method | |
CN116960962A (en) | Mid-long term area load prediction method for cross-area data fusion | |
Xu et al. | Residual autoencoder-LSTM for city region vehicle emission pollution prediction | |
CN115062764B (en) | Intelligent illuminance adjustment and environmental parameter Internet of things big data system | |
CN115600498A (en) | Wind speed forecast correction method based on artificial neural network | |
Kumarasiri et al. | Rainfall forecasting: an artificial neural network approach | |
CN114444763A (en) | Wind power prediction method based on AFSA-GNN | |
CN108960406B (en) | MEMS gyroscope random error prediction method based on BFO wavelet neural network | |
Landge et al. | Radar Vision-Weather Forecasting Using CNN-LSTM |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200619 |