CN109376903B - PM2.5 concentration value prediction method based on game neural network - Google Patents
PM2.5 concentration value prediction method based on game neural network Download PDFInfo
- Publication number
- CN109376903B CN109376903B CN201811050495.3A CN201811050495A CN109376903B CN 109376903 B CN109376903 B CN 109376903B CN 201811050495 A CN201811050495 A CN 201811050495A CN 109376903 B CN109376903 B CN 109376903B
- Authority
- CN
- China
- Prior art keywords
- data
- network
- layer
- output
- 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.)
- Active
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 60
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000004088 simulation Methods 0.000 claims abstract description 21
- 230000006870 function Effects 0.000 claims description 33
- 230000008569 process Effects 0.000 claims description 11
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 210000002569 neuron Anatomy 0.000 claims description 6
- 230000033228 biological regulation Effects 0.000 claims description 3
- 230000000306 recurrent effect Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 abstract description 3
- 238000010219 correlation analysis Methods 0.000 abstract description 3
- 238000009825 accumulation Methods 0.000 description 3
- 238000003915 air pollution Methods 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 239000013618 particulate matter Substances 0.000 description 2
- 241000219000 Populus Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 231100000331 toxic Toxicity 0.000 description 1
- 230000002588 toxic effect Effects 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
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/061—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using biological neurons, e.g. biological neurons connected to an integrated circuit
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Molecular Biology (AREA)
- Neurology (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A PM2.5 concentration value prediction method based on a game neural network comprises the following steps: step 1, collecting original data, wherein the original data comprises PM2.5 concentration value historical data, PM2.5 concentration value index historical data and meteorological historical data; step 2, generating simulation data by adopting a generating network; step 3, judging the authenticity of the simulation data by adopting a discrimination network; and 4, predicting the PM2.5 concentration value by adopting a game neural network. Besides carrying out nonlinear correlation analysis on PM2.5 concentration value historical data, PM2.5 concentration value related index historical data and meteorological historical data, the invention also introduces a generation network to mix original data with noise to output simulation data, sends the simulation data to a discrimination network to be discriminated, carries out iterative iteration and adjustment according to discrimination results, and can accurately describe the time change rule of the PM2.5 concentration value without presetting a target model for a small data set.
Description
Technical Field
The invention relates to the technical field of prediction of air particulate matter PM2.5 concentration values, in particular to a PM2.5 concentration value prediction method based on a game neural network.
Background
PM2.5 refers to particulate matters with the diameter less than or equal to 2.5 microns in the atmosphere, is rich in a large amount of toxic and harmful substances, has long retention time in the atmosphere and long conveying distance, so that the influence on human health and atmospheric environment quality is greater, and another influence, namely dust haze weather, is caused when PM2.5 exceeds the standard. Air pollution is now the focus of attention, and among air pollution indexes, the PM2.5 concentration value is a symbolic detection index for measuring air quality. Nowadays, prediction of the concentration value of PM2.5 in a future time period according to historical data becomes a research problem with strong academic significance and application value.
To solve the above problem, zhang yi et al in the paper "PM 2.5 prediction model based on neural network" performs the concentration value prediction of PM2.5 by selecting a neural network method. Wangming et al used a BP artificial neural network model in the paper "urban PM2.5 concentration spatial prediction based on BP artificial neural network" to predict spatial variation of PM2.5 concentration in the air of a research area. Zhengyi et al put forward a regional PM2.5 daily average prediction method based on a deep belief network in a thesis PM2.5 prediction based on the deep belief network. Yangyun et al, in a paper "PM 2.5 mass concentration prediction research in air", propose a prediction method of BP neural network optimized by using genetic algorithm to realize the prediction of PM2.5 concentration value. In a paper "fuzzy neural network PM2.5 concentration prediction based on improved PSO", Martiancheng et al adopts an improved PSO optimized fuzzy neural network, and a particle swarm algorithm and the fuzzy neural network are fused to predict the change rule of PM2.5 particulate matter concentration. Poplar and the like propose a PM2.5 mass concentration prediction method based on a T-S fuzzy neural network in a paper PM2.5 mass concentration prediction based on the T-S model fuzzy neural network. Su Ying et al in the patent PM2.5 concentration prediction method based on unscented Kalman neural network provide a PM2.5 concentration prediction method based on unscented Kalman neural network.
Through literature research and analysis, currently proposed methods for predicting PM2.5 concentration values all use a neural network as a core architecture to perform nonlinear regression analysis on PM2.5 concentration values and other related indexes (such as AQI, PM10, NO2, CO, SO2 and O3). The neural network model comprises ANN, DNN, FNN, BPNN and the like, and a mixing method after optimization by combining genetic algorithm, random forest and other optimization algorithms. However, through research and study of documents, a large amount of historical data needs to be accumulated in the existing PM2.5 concentration value neural network prediction method for training, the final prediction accuracy has a certain relation with the size of original sample data, and for a small sample set without a large amount of data accumulation, the existing neural network prediction method loses its advantages. Secondly, the existing neural network prediction models define known models to train data, that is, the distribution type of a learning target is established, and the actual work mainly includes learning and adjusting specific parameters of the distribution.
Disclosure of Invention
In order to overcome the defects that the existing PM2.5 concentration value prediction mode cannot train a small data set and a target model needs to be defined in advance, the invention introduces a generation network to mix original data and noise to output simulation data besides carrying out nonlinear correlation analysis on PM2.5 concentration value historical data, PM2.5 concentration value related index historical data and meteorological historical data, sends the simulation data to a discrimination network for discrimination, and carries out repeated iteration and adjustment according to a discrimination result, thereby providing a PM2.5 concentration value prediction method based on a game neural network, which aims at the small data set, does not need to preset the target model and can accurately describe the time change rule of the PM2.5 concentration value.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a PM2.5 concentration value prediction method based on a game neural network comprises the following steps:
step 2, generating simulation data by adopting a generating network, wherein the process is as follows:
step 2.1, creating a three-layer neural network comprising an input layer, a hidden layer and an output layer, and setting the number of nodes of the hidden layer and the output layer, wherein the number of the nodes of the hidden layer adopts an empirical formula to give an estimated value, and the empirical formula is as follows:
in the above formula, a and b are the numbers of neurons of the input layer and the output layer, respectively, and c is a constant between [0 and 10 ];
2.2, respectively setting the dimensionality of data of an input layer and an output layer, training functions, connection functions and output functions of a hidden layer, a connection layer and the output layer, and setting the minimum value of expected errors, the maximum iteration times and the learning rate of the network;
step 2.3, randomly generating a group of random Data as input layer Data of a generation model, and then generating a group of new PM2.5 prediction Data as Fake Data in an output layer after the generation model, wherein the new PM2.5 prediction Data is recorded as D (m);
step 3, judging the authenticity of the simulation data by adopting a discrimination network, wherein the process is as follows:
3.1, creating a three-layer neural network comprising an input layer, a hidden layer and an output layer, and setting the number of nodes of the hidden layer and the output layer; the number of nodes of the hidden layer is estimated by adopting an empirical formula, wherein the empirical formula is as follows:
in the above formula, a and b are the numbers of neurons of the input layer and the output layer, respectively, and c is a constant between [0 and 10 ];
step 3.2, connecting a layer of softmax function behind the output layer, and converting the multi-classification output numerical values into relative probabilities;
wherein, ViIs the output of the pre-stage output unit of the classifier. i represents the category index, and the total number of categories is C, SiThe ratio of the index of the current element to the sum of the indexes of all elements is expressed;
3.3, randomly selecting a group of Data from the Data set as Real Data, and recording the Data as x;
step 3.4, inputting the sum D (m) and the sum D (m) as x input Data into a discrimination network, wherein the output value is a number between 0 and 1 after passing through the discrimination network, the number is used for representing the probability that the input Data is Real Data, Real is 1, and fake is 0;
and 4, predicting the PM2.5 concentration value by adopting a game neural network, wherein the process is as follows:
step 4.1, inputting the simulation data and the original data generated in the generated network into a discrimination neural network, establishing a game neural network and training;
step 4.2, calculating and distinguishing a network loss function:
LD＝-((1-y)log(1-D(G(m)))+ylogD(x)) ⑷
y is the type of input Data, and when the input Data is Real Data, y is 1, and the first half of the loss function formula is 0. D (x) is the output of the discriminant model, represents the probability that the input x is real data, and the training target is to make the output of the output D (x) of the discriminant network tend to 1;
when the input Data is the Fake Data, y is 0, the second half of the loss function formula is 0, and g (m) is the output of the generated model, and the training target at this time is to make the output of D (g (m)) tend to 0;
and 4.3, calculating and generating a network loss function:
LG＝(1-y)log(1-D(G(m))) ⑸
the training goal of the generation network is to make the data generated by G (m) have the same data distribution as the real data;
step 4.4, calculating a loss function of the game neural network:
wherein,
representing the prediction category of the discriminant model, rounding the prediction probability to 0 or 1, and changing the gradient direction;
and 4.5, performing back propagation according to the error of the loss function, and adjusting the weight of each layer of the recurrent neural network in the following manner:
the regulation rule is as follows: the discrimination of D is maximized, and the data distribution of G and real data sets is minimized;
step 4.6, judging whether the game neural network is converged, when the error is smaller than the minimum value of the expected error, the algorithm is converged, and when the maximum iteration times is reached, the algorithm is ended, so that the game neural network training is completed;
and 4.7, inputting the data to be tested into the trained game neural network, and outputting a final predicted value of the PM2.5 concentration value.
Further, the PM2.5 concentration value indicators include AQI, PM10, NO2, CO, SO2, and O3 concentrations.
In the invention, in the step 3, the discrimination network is used for discriminating the original data and the simulation data of the generation network, so as to obtain the authenticity of the simulation data of the generation network.
In the step 4, the total loss function of the game neural network is obtained by calculating the loss functions of the generating network and the discriminating network respectively, and a weight value adjustment rule of back propagation is defined according to the loss value calculated by the loss function.
The technical conception of the invention is as follows: in the historical data of PM2.5 concentration value, the historical data (AQI, PM10, NO) of indexes related to the PM2.5 concentration value2、CO、SO2、O3) Besides nonlinear correlation analysis of meteorological historical data (air temperature, relative humidity, air pressure, air speed, precipitation and the like), a generation network is introduced to mix original data with noise to output simulation data, the simulation data is sent to a discrimination network to be discriminated, and cross iteration and adjustment are carried out according to discrimination results, so that the PM2.5 concentration value prediction method based on the game neural network is provided.
The invention has the following beneficial effects: the technical scheme of the invention not only can accurately process a complete sample set with a large amount of historical data accumulation, but also can efficiently and accurately process a small sample set without a large amount of data accumulation, and a known model is not required to be defined to train data, namely, the distribution type of a learning target is not required to be preset, so that the prediction precision and the training speed of the current PM2.5 concentration value are effectively improved, the limitation of neural network training is widened, and the accurate prediction of the small sample set can be realized.
Drawings
Fig. 1 is a schematic diagram of a PM2.5 concentration value prediction method based on a game neural network.
Fig. 2 is a simulation flow chart of generating a network.
Fig. 3 is a flow chart of discrimination of the discrimination network.
FIG. 4 is a flow chart of the training of the gaming neural network
Detailed Description
The invention is further described below with reference to the accompanying drawings.
Referring to fig. 1 to 4, a method for predicting a PM2.5 concentration value based on a game neural network includes the following steps:
step 2, generating simulation data by adopting a generating network, wherein the process is as follows:
and 2.1, creating a three-layer neural network comprising an input layer, a hidden layer and an output layer, and setting the number of nodes of the hidden layer and the output layer. The number of nodes of the hidden layer is estimated by adopting an empirical formula, wherein the empirical formula is as follows:
in the above formula, a and b are the numbers of neurons of the input layer and the output layer, respectively, and c is a constant between [0 and 10 ];
2.2, respectively setting the dimensionality of data of an input layer and an output layer, training functions, connection functions and output functions of a hidden layer, a connection layer and the output layer, and setting the minimum value of expected errors, the maximum iteration times and the learning rate of the network;
step 2.3, randomly generating a group of random Data as input layer Data of a generation model, and then generating a group of new PM2.5 prediction Data as Fake Data in an output layer after the generation model, wherein the new PM2.5 prediction Data is recorded as D (m);
step 3, judging the authenticity of the simulation data by adopting a discrimination network, wherein the process is as follows:
step 3.1, creating a three-layer neural network comprising an input layer, a hidden layer and an output layer, and setting the number of nodes of the hidden layer and the output layer, wherein the number of the nodes of the hidden layer adopts an empirical formula to give an estimated value, and the empirical formula is as follows:
in the above formula, a and b are the numbers of neurons of the input layer and the output layer, respectively, and c is a constant between [0 and 10 ];
step 3.2, connecting a layer of softmax function behind the output layer, and converting the multi-classification output numerical values into relative probabilities;
wherein, ViIs the output of the pre-stage output unit of the classifier. i represents the category index, and the total number of categories is C, SiThe ratio of the index of the current element to the sum of the indexes of all elements is expressed;
3.3, randomly selecting a group of Data from the Data set as Real Data, and recording the Data as x;
step 3.4, inputting the sum D (m) and the sum D (m) as x input Data into a discrimination network, wherein the output value is a number between 0 and 1 after passing through the discrimination network, the number is used for representing the probability that the input Data is Real Data, Real is 1, and fake is 0;
and 4, predicting the PM2.5 concentration value by adopting a game neural network, wherein the process is as follows:
step 4.1, inputting the simulation data and the original data generated in the generated network into a discrimination neural network, establishing a game neural network and training;
step 4.2, calculating and distinguishing a network loss function:
LD＝-((1-y)log(1-D(G(m)))+ylogD(x)) ⑷
y is the type of input Data, when the input Data is Real Data, y is 1, the first half of the loss function formula is 0, d (x) is the output of the discriminant model, which represents the probability that the input x is Real Data (y is 1, which represents Real Data), and the training target is to make the output of the output d (x) of the discriminant network tend to 1;
when the input Data is the Fake Data, y is 0, the second half of the loss function formula is 0, and g (m) is the output of the generated model, and the training target at this time is to make the output of D (g (m)) tend to 0;
and 4.3, calculating and generating a network loss function:
LG＝(1-y)log(1-D(G(m))) ⑸
the training goal of the generation network is to make the data generated by G (m) have the same data distribution as the real data;
step 4.4, calculating a loss function of the game neural network:
wherein,
representing the prediction category of the discriminant model, rounding the prediction probability to 0 or 1 for changing the gradient direction, and setting the threshold value to 0.5;
and 4.5, performing back propagation according to the error of the loss function, and adjusting the weight of each layer of the recurrent neural network in the following manner:
the regulation rule is as follows: the discrimination of D is maximized, and the data distribution of G and real data sets is minimized;
step 4.6, judging whether the game neural network is converged, and when the error is smaller than the minimum value of the expected error, converging the algorithm; finishing the algorithm when the maximum iteration times are reached, and finishing the game neural network training;
and 4.7, inputting the data to be tested into the trained game neural network, and outputting a final predicted value of the PM2.5 concentration value.
Claims (2)
1. A PM2.5 concentration value prediction method based on a game neural network is characterized by comprising the following steps:
step 1, collecting original data, wherein the original data comprises PM2.5 concentration value historical data, PM2.5 concentration value index historical data and meteorological historical data;
step 2, generating simulation data by adopting the generated network prediction, wherein the process is as follows:
step 2.1, creating a three-layer neural network comprising an input layer, a hidden layer and an output layer, and setting the number of nodes of the hidden layer and the output layer, wherein the number of the nodes of the hidden layer adopts an empirical formula to give an estimated value, and the empirical formula is as follows:
in the above formula, a and b are the numbers of neurons of the input layer and the output layer, respectively, and c is a constant between [0 and 10 ];
2.2, respectively setting the dimensionality of data of an input layer and an output layer, training functions, connection functions and output functions of a hidden layer, a connection layer and the output layer, and setting the minimum value of expected errors, the maximum iteration times and the learning rate of the network;
step 2.3, randomly generating a group of random Data as input layer Data of a generating network, and then generating a group of new PM2.5 simulation Data as Fake Data recorded as D (m) on an output layer after the network is generated;
step 3, judging the authenticity of the simulation data by adopting a discrimination network, wherein the process is as follows:
3.1, creating a three-layer neural network comprising an input layer, a hidden layer and an output layer, and setting the number of nodes of the hidden layer and the output layer; the number of nodes of the hidden layer is estimated by adopting an empirical formula, wherein the empirical formula is as follows:
in the above formula, a and b are the numbers of neurons of the input layer and the output layer, respectively, and c is a constant between [0 and 10 ];
step 3.2, connecting a layer of softmax function behind the output layer, and converting the multi-classification output numerical values into relative probabilities;
wherein, ViIs the output of the preceding output unit of the classifier, i represents the class index, and the total number of classes is C, SiThe ratio of the index of the current element to the sum of the indexes of all elements is expressed;
3.3, randomly selecting a group of Data from the Data set as Real Data, and recording the Data as x;
step 3.4, inputting D (m) and x as input Data into a discrimination network, wherein the output value is a number between 0 and 1 after passing through the discrimination network, and the number is used for representing the probability that the input Data is Real Data, Real is 1, and fake is 0;
and 4, predicting the PM2.5 concentration value by adopting a game neural network, wherein the process is as follows:
step 4.1, inputting the simulation data and the original data generated in the generated network into a discrimination network, establishing a game neural network and training;
step 4.2, calculating and distinguishing a network loss function:
LD＝-((1-y)log(1-D(G(m)))+y log D(x)) ⑷
y is the type of input Data, when the input Data is Real Data, y is 1, the first half of the loss function formula is 0, d (x) is the output of the discriminant network, which represents the probability that the input x is Real Data, and the training target is to make the output of the output d (x) of the discriminant network tend to 1;
when the input Data is the Fake Data, y is 0, the second half of the loss function formula is 0, and g (m) is the output of the generated network, and the training target at this time is to make the output of D (g (m)) tend to 0;
and 4.3, calculating and generating a network loss function:
LG＝(1-y)log(1-D(G(m))) ⑸
the training goal of the generation network is to make the data generated by G (m) have the same data distribution as the real data;
step 4.4, calculating a loss function of the game neural network:
wherein,
representing the prediction category of the discrimination network, rounding the prediction probability to 0 or 1, and changing the gradient direction;
and 4.5, performing back propagation according to the error of the loss function, and adjusting the weight of each layer of the recurrent neural network in the following manner:
the regulation rule is as follows: the discrimination of D is maximized, and the data distribution of G and real data sets is minimized;
step 4.6, judging whether the game neural network is converged, when the error is smaller than the minimum value of the expected error, the algorithm is converged, and when the maximum iteration times is reached, the algorithm is ended, so that the game neural network training is completed;
and 4.7, inputting the data to be tested into the game neural network obtained after training is finished, and outputting the final predicted value of the PM2.5 concentration value.
2. The method for predicting PM2.5 concentration values based on the game neural network as claimed in claim 1, wherein the PM2.5 concentration value indexes comprise AQI, PM10, NO2, CO, SO2 and O3 concentrations.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811050495.3A CN109376903B (en) | 2018-09-10 | 2018-09-10 | PM2.5 concentration value prediction method based on game neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811050495.3A CN109376903B (en) | 2018-09-10 | 2018-09-10 | PM2.5 concentration value prediction method based on game neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109376903A CN109376903A (en) | 2019-02-22 |
| CN109376903B true CN109376903B (en) | 2021-12-17 |
Family
ID=65405373
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811050495.3A Active CN109376903B (en) | 2018-09-10 | 2018-09-10 | PM2.5 concentration value prediction method based on game neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109376903B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110766222B (en) * | 2019-10-22 | 2023-09-19 | 太原科技大学 | PM2.5 concentration prediction method based on particle swarm parameter optimization and random forest |
| CN111737429B (en) * | 2020-06-16 | 2023-11-03 | 平安科技（深圳）有限公司 | Training method, AI interview method and related equipment |
| CN112183872A (en) * | 2020-10-10 | 2021-01-05 | 东北大学 | Blast furnace gas generation amount prediction method combining generation of countermeasure network and neural network |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100612862B1 (en) * | 2004-10-05 | 2006-08-14 | 삼성전자주식회사 | Method and apparatus for summarizing sports video |
| CN105956691A (en) * | 2016-04-25 | 2016-09-21 | 北京市环境保护监测中心 | Method of calculating PM2.5 background concentration in different orientations at different observation points of prediction area |
| CN106056210B (en) * | 2016-06-07 | 2018-06-01 | 浙江工业大学 | A kind of PM2.5 concentration value Forecasting Methodologies based on hybrid neural networks |
| CN107358626B (en) * | 2017-07-17 | 2020-05-15 | 清华大学深圳研究生院 | Method for generating confrontation network calculation parallax by using conditions |
| CN107247888B (en) * | 2017-08-14 | 2020-09-15 | 吉林大学 | Method for soft measurement of total phosphorus TP (thermal transfer profile) in sewage treatment effluent based on storage pool network |
| CN107766815B (en) * | 2017-10-18 | 2021-05-18 | 福州大学 | Visual auxiliary service operation method |
| CN108334977B (en) * | 2017-12-28 | 2020-06-30 | 鲁东大学 | Deep learning-based water quality prediction method and system |
| CN108268935B (en) * | 2018-01-11 | 2021-11-23 | 浙江工业大学 | PM2.5 concentration value prediction method and system based on time sequence recurrent neural network |
| CN108491497B (en) * | 2018-03-20 | 2020-06-02 | 苏州大学 | Medical text generation method based on generation type confrontation network technology |
| CN108426812B (en) * | 2018-04-08 | 2020-07-31 | 浙江工业大学 | PM2.5 concentration value prediction method based on memory neural network |
-
2018
- 2018-09-10 CN CN201811050495.3A patent/CN109376903B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109376903A (en) | 2019-02-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108426812B (en) | PM2.5 concentration value prediction method based on memory neural network | |
| CN108268935B (en) | PM2.5 concentration value prediction method and system based on time sequence recurrent neural network | |
| CN108491970B (en) | Atmospheric pollutant concentration prediction method based on RBF neural network | |
| CN106650825B (en) | Motor vehicle exhaust emission data fusion system | |
| CN110782093B (en) | PM fusing SSAE deep feature learning and LSTM2.5Hourly concentration prediction method and system | |
| CN109376903B (en) | PM2.5 concentration value prediction method based on game neural network | |
| CN103166830B (en) | A kind of Spam Filtering System of intelligent selection training sample and method | |
| CN108009674A (en) | Air PM2.5 concentration prediction methods based on CNN and LSTM fused neural networks | |
| CN110555551B (en) | Air quality big data management method and system for smart city | |
| CN109086926B (en) | Short-time rail transit passenger flow prediction method based on combined neural network structure | |
| CN109143408B (en) | Dynamic region combined short-time rainfall forecasting method based on MLP | |
| CN114781538A (en) | Air quality prediction method and system of GA-BP neural network coupling decision tree | |
| CN110610209A (en) | Air quality prediction method and system based on data mining | |
| Ning et al. | GA-BP air quality evaluation method based on fuzzy theory. | |
| Song et al. | Study on turbidity prediction method of reservoirs based on long short term memory neural network | |
| Al_Janabi et al. | Pragmatic method based on intelligent big data analytics to prediction air pollution | |
| CN115656446B (en) | Air quality detection system and method based on Internet of things | |
| Feng et al. | A dual-staged attention based conversion-gated long short term memory for multivariable time series prediction | |
| CN114973665A (en) | Short-term traffic flow prediction method combining data decomposition and deep learning | |
| CN114648095A (en) | Air quality concentration inversion method based on deep learning | |
| CN114461791A (en) | Social text sentiment analysis system based on deep quantum neural network | |
| Zhang et al. | A diverse ensemble deep learning method for short-term traffic flow prediction based on spatiotemporal correlations | |
| CN117370813A (en) | Atmospheric pollution deep learning prediction method based on K line pattern matching algorithm | |
| CN105974058A (en) | Method for rapidly detecting potassium content of tobacco leaves based on electronic nose-artificial neural network | |
| CN112991765B (en) | Method, terminal and storage medium for updating road high-emission source recognition model |
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 | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20190222 Assignee: Hangzhou Youshu Cloud Travel Information Technology Co.,Ltd. Assignor: JIANG University OF TECHNOLOGY Contract record no.: X2023980054817 Denomination of invention: A Game Neural Network Based Method for Predicting PM2.5 Concentration Values Granted publication date: 20211217 License type: Common License Record date: 20240102 Application publication date: 20190222 Assignee: Hangzhou Tianyin Computer System Engineering Co.,Ltd. Assignor: JIANG University OF TECHNOLOGY Contract record no.: X2023980054814 Denomination of invention: A Game Neural Network Based Method for Predicting PM2.5 Concentration Values Granted publication date: 20211217 License type: Common License Record date: 20240102 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20190222 Assignee: HANGZHOU YONGGUAN NETWORK TECHNOLOGY CO.,LTD. Assignor: JIANG University OF TECHNOLOGY Contract record no.: X2024980000361 Denomination of invention: A Game Neural Network Based Method for Predicting PM2.5 Concentration Values Granted publication date: 20211217 License type: Common License Record date: 20240109 |