CN111611547B - Method for determining nitrous oxide emission under influence of multiple factors in drainage channel of paddy field - Google Patents
Method for determining nitrous oxide emission under influence of multiple factors in drainage channel of paddy field Download PDFInfo
- Publication number
- CN111611547B CN111611547B CN202010424531.9A CN202010424531A CN111611547B CN 111611547 B CN111611547 B CN 111611547B CN 202010424531 A CN202010424531 A CN 202010424531A CN 111611547 B CN111611547 B CN 111611547B
- Authority
- CN
- China
- Prior art keywords
- nitrous oxide
- data
- var
- variable
- child
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Algebra (AREA)
- Evolutionary Biology (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Treating Waste Gases (AREA)
- Activated Sludge Processes (AREA)
Abstract
The invention discloses a method for determining nitrous oxide emission under the influence of multiple factors in a drainage ditch of a paddy field, which comprises the following steps of: (1) Collecting data of nitrous oxide emission amount of the bottom mud of the rice field drainage ditch needing to determine the nitrous oxide emission amount and corresponding environment variables; (2) preprocessing the data; (3) Constructing a Bayesian network among variables according to an appointed form, and determining a coefficient of each edge representing a regression relationship; (4) And determining the discharge amount of the ditch nitrous oxide under the given environmental condition according to the result of the Bayesian network. The method can explicitly and accurately estimate the influence of various environmental variables in the drainage ditch of the rice field on the emission amount of nitrous oxide, and provides support for China to participate in international greenhouse gas emission reduction negotiation.
Description
Technical Field
The invention relates to a method for determining nitrous oxide emission under the influence of multiple factors in a drainage ditch of a rice field, and belongs to the field of rice field environment management.
Background
Global warming caused by greenhouse gas emissions is a serious environmental problem facing human society. Nitrous oxide is one of the most important greenhouse gases, has long life cycle in the atmosphere and high heat absorption potential, causes much attention, and is an important target for internationally developing national greenhouse gas emission reduction negotiations. The paddy field system is an important source of nitrous oxide emission, and researches on nitrous oxide emission in the production process of paddy rice are many, but researches on nitrous oxide emission in drainage ditches attached to the paddy field are rare. The drainage ditch of the rice field is positioned in a transition zone of an artificial system and a natural river of the rice field, the influence of various environmental factors on nitrous oxide emission is very complex, and the nitrous oxide emission under the influence of multiple factors in the drainage ditch of the rice field needs to be determined by adopting an advanced method.
Disclosure of Invention
The invention aims to solve the problems in the prior art, and provides a method for determining nitrous oxide emission under the influence of multiple environmental factors in a drainage channel of a rice field in order to accurately estimate nitrous oxide emission under the influence of multiple environmental factors in the drainage channel of the rice field and fill the blank of the prior art.
The invention aims to realize the determination method of nitrous oxide emission under the influence of multiple factors in the drainage canal of the rice field, which comprises the following steps:
(1) Collecting n groups of nitrous oxide emission amount of the bottom mud of the paddy field drainage ditch needing to determine the nitrous oxide emission amount and data of corresponding environment variables;
(2) Preprocessing the data;
(3) Constructing a Bayesian network among variables according to an appointed form, and determining a coefficient of each edge representing a regression relationship;
(4) And determining the discharge amount of the nitrous oxide in the drainage ditch under the given environmental condition according to the result of the Bayesian network.
Preferably, the step (1) specifically comprises the following requirements:
a. the method comprises the following steps that the water power condition in a drainage ditch needs to be collected, and the water power condition is divided into three conditions of no water, water but basically static and flowing according to the field condition; environmental variable data and corresponding nitrous oxide emission under three conditions need to be collected;
b. environmental variable data collected in the drain bed mud includes: redox potential, temperature, ammoniacal nitrogen, nitrate nitrogen and sulfate radical of the bottom sludge;
c. in each hydrodynamic case, the data sets collected should be greater than 60; i.e. the total amount of data should be greater than 180 sets;
d. the sampling points are uniformly distributed in the drainage ditch, the depth of each sampling point is kept at 10 cm, and the error is not more than 1 cm.
Preferably, the step (2) specifically includes the steps of:
a. respectively calculating the nitrous oxide emission, the average value and the standard deviation of oxidation-reduction potential, temperature, ammonia nitrogen, nitrate nitrogen and sulfate radical, and rejecting the sampling values of which the numerical values exceed the respective average value plus or minus 6 times of the standard deviation and the corresponding nitrous oxide and other environmental variable values; recalculating the mean and standard deviation of the environment variables after rejection, and rejecting the sample values with the mean value of + -6 times of the standard deviation and the pairs thereof for the second timeThe corresponding nitrous oxide and other environmental variable values are repeatedly executed in the same way until no data are rejected; record the final mean μ i And standard deviation σ i (i =1,2, …,6, corresponding to nitrous oxide emissions, redox potential, temperature, ammonia nitrogen, nitrate nitrogen and sulfate). At this time, the nitrous oxide emission and the edge distribution of the environmental variable data other than the water power condition are each approximated to the mathematical expectation of μ i Variance is σ i 2 A gaussian distribution of (a).
b. The number of the groups after being rejected is more than or equal to 150 groups, and the data amount under each hydrodynamic condition is more than or equal to 50 groups; if the data volume is not enough, replenishing the data again according to the requirement of the step (1), and re-executing the step (2) a until the data volume requirement is met;
c. setting double virtual variables D for the water power pulse condition in the drainage ditch according to the following rules 1 And D 2 ：
Preferably, the step (3) specifically includes the steps of:
a. a directed acyclic network, i.e., a bayesian network, is constructed as specified below. Firstly, a variable of a starting point of a directional arrow is called a parent node of an arrow pointing variable and is marked as par for convenient representation; the directed arrow points to a child node of a variable called the arrow origin variable, denoted child. Secondly, the two variables of hydrodynamic condition and temperature are the root nodes of the bayesian network, i.e. the nodes where no parent node exists. Second, the parent node of the redox potential is hydrodynamic; the father node of nitrate nitrogen is temperature, ammoniacal nitrogen and oxidation-reduction potential; the father node of the ammoniacal nitrogen is the redox potential; the father node of the sulfate radical is nitrate nitrogen, ammonia nitrogen and temperature; the father node of nitrous oxide is nitrate nitrogen, sulfate and ammonia nitrogen.
b. Regression is performed on all relationships in the bayesian network. The specific method comprises the following steps: let mth child be child m Assume that there are s par in total, and the jth par is par j . Taking a given child as a dependent variable and all par of the given child as an independent variable; estimating node groups (child) using a minimum angle regression method using corresponding processed data m ，par j ) Robust regression coefficient beta corresponding to arrow mj (m denotes the mth child, j denotes the jth par of the mth child), and the intercept λ corresponding to each regression relationship is recorded m . A child can be obtained m Given all of par j And corresponding beta mj Conditional probability f in the case, note that it must be gaussian; the method comprises the following specific steps:
preferably, the step (4) specifically includes the steps of:
a. determining the environmental variable to be estimated for the influence on the emission of nitrous oxide and recording the environmental variable as var i ；
b. Calculating the given var by applying a chain rule of probabilities according to the conditional probabilities given by the Bayesian network and the edge distribution of each variable i Conditional probability of case f (nitrous oxide | var) i ). Note that except for var i The distribution is gaussian for all conditions except hydrodynamic conditions; if var is i Is a hydrodynamic case, f (nitrous oxide | var) i ) Is a polynomial distribution; i.e. all given vars i F (nitrous oxide | var) i ) The mathematical expectation of (c) can be given explicitly.
c. With f (nitrous oxide | var) i ) As a given var i Under the condition, the equation for calculating the emission of nitrous oxide in the drainage ditch of the paddy field needs to calculate var of the condition i The specific value is substituted into the formula to obtain the value of the emission of the nitrous oxide.
The research of ecology and environmentality shows that the emission of nitrous oxide has important influence on the aggravation of global greenhouse effect, the invention provides a method for accurately and explicitly determining the emission of nitrous oxide in the paddy field drainage ditch under the influence of multiple factors by utilizing a Bayesian network model, and the method is easy to popularize in the paddy field drainage ditch systems in various regions.
Has the beneficial effects that: the invention provides a method for accurately and explicitly determining the nitrous oxide emission in the drainage ditch of the rice field by using the Bayesian network theory, and can provide a basis for the country to participate in international greenhouse gas emission reduction negotiation and control and reduce the nitrous oxide emission of the rice field system.
Drawings
FIG. 1 is a flow chart of the present invention
FIG. 2 is a Bayesian network as specified by the present invention.
Detailed Description
The method for determining the nitrous oxide emission under the influence of multiple factors in the drainage ditch of the paddy field comprises the following steps:
(1) Collecting n groups of nitrous oxide emission amount of the bottom mud of the paddy field drainage ditch needing to determine the nitrous oxide emission amount and data of corresponding environment variables; the step (1) specifically comprises the following requirements:
a. the method comprises the following steps that the water power condition in a drainage ditch needs to be collected, and the water power condition is divided into three conditions of no water, water but basically static and flowing according to the field condition; environmental variable data and corresponding nitrous oxide emission under three conditions need to be collected;
b. environmental variable data collected in the drain bed mud includes: redox potential, temperature, ammoniacal nitrogen, nitrate nitrogen and sulfate radical of the bottom sludge;
c. in each hydrodynamic case, the data sets collected should be greater than 60; i.e. the total data amount n should be greater than 180 sets;
d. the positions of sampling points are uniformly distributed in the drainage ditch, the depth of each sampling point is kept at 10 cm, and the error does not exceed 1 cm;
(2) Preprocessing the data; the step (2) specifically comprises the following steps:
a. respectively calculating the emission of nitrous oxide, oxidation-reduction potential, temperature and ammoniaThe average values and standard deviations of nitrogen, nitrate nitrogen and sulfate radicals, and the sampling values of which the numerical values exceed the respective average values by +/-6 times of standard deviations and the corresponding values of nitrous oxide and other environmental variables are removed; after the elimination, recalculating the average value and the standard deviation of the environment variables, and eliminating the sampling value of the average value plus or minus 6 times of the standard deviation and the corresponding nitrous oxide and other environment variables for the second time, and repeating the operation until no data are eliminated; record the final mean μ i And standard deviation σ i (i =1,2, …,6, corresponding to nitrous oxide emissions, redox potential, temperature, ammonia nitrogen, nitrate nitrogen and sulfate); at this time, the nitrous oxide emission and the edge distribution of the environmental variable data other than the water power condition are each approximated to the mathematical expectation of μ i Variance is σ i 2 (ii) a gaussian distribution of;
b. the number of the groups after being rejected is more than or equal to 150 groups, and the data amount under each hydrodynamic condition is more than or equal to 50 groups; if the data volume is not enough, replenishing the data again according to the requirement of the step (1), and performing the step (2) a again until the requirement of the data volume is met;
c. setting double virtual variables D for the water power pulse condition in the drainage ditch according to the following rules 1 And D 2 ：
The step (3) specifically comprises the following steps:
a. constructing a directed acyclic network, namely a Bayesian network, according to the following specified requirements; firstly, a directional arrow starting point variable is called as a parent node of an arrow pointing variable and is marked as par for convenient representation; the directional variable of the directional arrow is called as a child node of the variable of the starting point of the arrow and is marked as child; secondly, the two variables of the hydrodynamic condition and the temperature are root nodes of the Bayesian network, namely nodes without father nodes; second, the parent node of the redox potential is hydrodynamic; the father node of nitrate nitrogen is temperature, ammoniacal nitrogen and oxidation-reduction potential; the father node of the ammoniacal nitrogen is the oxidation-reduction potential; the father node of the sulfate radical is nitrate nitrogen, ammonia nitrogen and temperature; the father node of nitrous oxide is nitrate nitrogen, sulfate radical and ammonia nitrogen;
b. performing regression on all relations in the Bayesian network; the specific method comprises the following steps: let mth child be child m Assume that there are s par in total, and the jth par is par j (ii) a Taking a given child as a dependent variable and all par of the given child as an independent variable; estimating node groups (child) using a minimum angle regression method using corresponding processed data m ，par j ) Robust regression coefficient beta corresponding to arrow mj (m denotes the mth child, j denotes the jth par of the mth child), and the intercept λ corresponding to each regression relationship is recorded m (ii) a A child can be obtained m Given all of par j And corresponding beta mj Conditional probability f in the case, note that it must be gaussian; the method specifically comprises the following steps:
(3) Constructing a Bayesian network among variables according to an appointed form, and determining a coefficient of each edge representing a regression relationship;
(4) Determining the amount of exhaust ditch nitrous oxide emissions under given environmental conditions based on the results of the bayesian network, said step (4) comprising the steps of:
a. determining the environmental variables whose influence on nitrous oxide emissions needs to be estimated, and recording as var i ；
b. According to the conditional probability given by the Bayes network and the edge distribution of each variable, a given var is calculated by applying the chain rule of the probability i Conditional probability of case f (nitrous oxide | var) i ). Note that except for var i The distribution is gaussian for all conditions other than hydrodynamic conditions; if var i Is a hydrodynamic condition, f (nitrous oxide | var) i ) Is a polynomial distribution; i.e. all given vars i F (nitrous oxide | var) i ) The mathematical expectations of (c) can all be given explicitly;
c. with f (nitrous oxide | var) i ) As a given var i Under the condition, the equation for calculating the discharge amount of nitrous oxide in the drainage ditch of the paddy field needs to calculate the var of the condition i The specific value is substituted into the formula to obtain the value of the emission of the nitrous oxide.
The invention is further explained by combining the attached drawings and the measured data of the water drainage ditch of a certain provincial paddy field in the east of China:
(1) According to the flow chart, 192 groups of bottom mud nitrous oxide and environmental data are collected at different periods of the selected paddy field drainage ditch, and the data of each hydrodynamic condition are 64 groups;
(2) After pretreatment, 160 groups of data meeting the requirements are screened out, and 50 groups of data are screened out under each hydrodynamic condition; using these data to calculate the average and standard deviation of the redox potential, temperature, ammoniacal nitrogen, nitrate nitrogen and sulphate respectively;
(3) According to the bayesian network shown in fig. 2, each child node variable is used as a dependent variable, the corresponding parent node variable is used as an independent variable, a regression coefficient of an arrow is estimated by using minimum angle regression, and a corresponding intercept is recorded;
(4) The influence of nitrate nitrogen on the emission of nitrous oxide needs to be calculated, and from the properties of a bayesian network and a gaussian distribution, the following can be calculated:
f (nitrous oxide | nitrate nitrogen) -N (20 + 0.33X nitrate nitrogen, 25.4)
Therefore, when the nitrate nitrogen content in the substrate sludge is 1.2mg/L, the regional nitrous oxide emission is 20.4 mg/(hectare-day).
Claims (1)
1. The method for determining the nitrous oxide emission under the influence of multiple factors in the drainage channel of the paddy field is characterized by comprising the following steps of:
(1) Collecting n groups of nitrous oxide emission amount of the bottom mud of the paddy field drainage ditch needing to determine the nitrous oxide emission amount and data of corresponding environment variables;
(2) Preprocessing the data;
(3) Constructing a Bayesian network among variables according to an appointed form, and determining a coefficient of each edge representing a regression relationship;
(4) Determining the discharge amount of nitrous oxide in a drainage ditch under given environmental conditions according to the result of the Bayesian network;
the step (1) specifically comprises the following requirements:
a. the method comprises the following steps that the water power condition in a drainage ditch needs to be collected, and the water power condition is divided into three conditions of no water, water but basically static and flowing according to the field condition; environmental variable data and corresponding nitrous oxide emission under three conditions need to be collected;
b. the environmental variable data collected in the drain bottom mud includes: redox potential, temperature, ammoniacal nitrogen, nitrate nitrogen and sulfate radical of the bottom sludge;
c. in each hydrodynamic case, the data sets collected should be greater than 60; i.e. the total data amount n should be greater than 180 sets;
d. the positions of sampling points are uniformly distributed in the drainage ditch, the depth of each sampling point is kept at 10 cm, and the error is not more than 1 cm;
the step (2) specifically comprises the following steps:
a. respectively calculating the nitrous oxide emission, the average value and standard deviation of oxidation-reduction potential, temperature, ammonia nitrogen, nitrate nitrogen and sulfate radical, and removing sampling values with numerical values exceeding the respective average value +/-6 times of standard deviation and corresponding nitrous oxide and other environmental variable values; after the elimination, recalculating the average value and the standard deviation of the environment variables, and eliminating the sampling value of the average value plus or minus 6 times of the standard deviation and the corresponding nitrous oxide and other environment variables for the second time, and repeating the operation until no data are eliminated; record the final average value μ i And standard deviation σ i (i =1,2, …,6, corresponding to nitrous oxide emissions, redox potential, temperature, ammonia nitrogen, nitrate nitrogen and sulfate); at this time, the nitrous oxide emission amount and the edge distribution of the environmental variable data other than the water power condition are each approximated to the mathematical expectation μ i Variance of
(ii) a gaussian distribution of;
b. the number of the groups after being rejected is more than or equal to 150 groups, and the data amount under each hydrodynamic condition is more than or equal to 50 groups; if the data volume is not enough, replenishing the data again according to the requirement of the step (1), and re-executing the step (2) a until the data volume requirement is met;
c. setting double virtual variables D for the water power pulse condition in the drainage ditch according to the following rules 1 And D 2 ：
The step (3) specifically comprises the following steps:
a. constructing a directed acyclic network, namely a Bayesian network, according to the following specified requirements; firstly, a variable at the starting point of the directional arrow is called as a parent node of an arrow pointing variable and is marked as par; the directional variable of the directional arrow is called as a child node of the variable of the starting point of the arrow and is marked as child; secondly, the hydrodynamic condition and the temperature are root nodes of the Bayesian network, namely nodes without father nodes; second, the parent node of the redox potential is hydrodynamic; the father node of nitrate nitrogen is temperature, ammoniacal nitrogen and oxidation-reduction potential; the father node of the ammoniacal nitrogen is the redox potential; the father node of the sulfate radical is nitrate nitrogen, ammonia nitrogen and temperature; the father node of nitrous oxide is nitrate nitrogen, sulfate radical and ammonia nitrogen;
b. performing regression on all relations in the Bayesian network; the specific method comprises the following steps: let mth child be child m Assume that there are s par in total, and the jth par is par j (ii) a Taking a given child as a dependent variable and all par of the given child as an independent variable; estimating by using the corresponding processed data and the minimum angle regression methodNode group (child) m ，par j ) Robust regression coefficient beta corresponding to arrow mj (m denotes the mth child, j denotes the jth par of the mth child), and records the intercept lambda corresponding to each regression relationship m (ii) a A child can be obtained m Given all of par j And corresponding beta mj The conditional probability f of the case, which must be a gaussian distribution; the method specifically comprises the following steps:
the step (4) comprises the following steps:
a. determining the environmental variable to be estimated for the influence on the emission of nitrous oxide and recording the environmental variable as var i ；
b. Calculating the given var by applying a chain rule of probabilities according to the conditional probabilities given by the Bayesian network and the edge distribution of each variable i Conditional probability f (nitrous oxide | var) of case i ) Except for var i The distribution is gaussian for all conditions except hydrodynamic conditions; if var is i Is a hydrodynamic case, f (nitrous oxide | var) i ) Is a polynomial distribution; i.e. all given vars i F (nitrous oxide | var) i ) The mathematical expectations of (a) can all be given explicitly;
c. with f (nitrous oxide | var) i ) As a given var i Under the condition, the equation for calculating the discharge amount of nitrous oxide in the drainage ditch of the paddy field needs to calculate the var of the condition i The specific value is substituted into the formula to obtain the value of the emission of the nitrous oxide.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010424531.9A CN111611547B (en) | 2020-05-19 | 2020-05-19 | Method for determining nitrous oxide emission under influence of multiple factors in drainage channel of paddy field |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010424531.9A CN111611547B (en) | 2020-05-19 | 2020-05-19 | Method for determining nitrous oxide emission under influence of multiple factors in drainage channel of paddy field |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111611547A CN111611547A (en) | 2020-09-01 |
| CN111611547B true CN111611547B (en) | 2023-04-14 |
Family
ID=72204841
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010424531.9A Active CN111611547B (en) | 2020-05-19 | 2020-05-19 | Method for determining nitrous oxide emission under influence of multiple factors in drainage channel of paddy field |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111611547B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114936721B (en) * | 2022-07-15 | 2022-10-28 | 北京师范大学 | Method and device for determining discharge amount of nitrous oxide in river and electronic equipment |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10238323B2 (en) * | 2014-06-06 | 2019-03-26 | Dexcom, Inc. | Fault discrimination and responsive processing based on data and context |
| CN109190865B (en) * | 2018-06-25 | 2021-09-28 | 西南交通大学 | Water quality index evaluation method, prediction method and water area water quality grade evaluation method |
-
2020
- 2020-05-19 CN CN202010424531.9A patent/CN111611547B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN111611547A (en) | 2020-09-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Karul et al. | Case studies on the use of neural networks in eutrophication modeling | |
| CN110728035A (en) | Pollutant total amount control method based on control of section water quality reaching standard | |
| CN109711617B (en) | Medium-and-long-term runoff prediction method based on BLSTM deep learning | |
| CN105784556B (en) | A kind of air fine particles PM based on Self-organized Fuzzy Neural Network2.5Flexible measurement method | |
| CN104376380A (en) | Ammonia concentration predicting method based on recursion self-organization neural network | |
| CN111080157B (en) | Method and system for scheduling phosphorus discharge amount of cascade hydropower station | |
| CN109308544B (en) | Blue algae bloom prediction method based on contrast divergence-long and short term memory network | |
| CN111611547B (en) | Method for determining nitrous oxide emission under influence of multiple factors in drainage channel of paddy field | |
| CN115169232B (en) | Daily peak load prediction method, computer equipment and readable storage medium | |
| CN112036595B (en) | Short-term wind power prediction method and system based on multi-position numerical weather forecast | |
| CN114912079B (en) | Method, device and equipment for calculating carbon emission of sewage plant and storage medium | |
| Cui et al. | Modeling the response of phytoplankton to reduced external nutrient load in a subtropical Chinese reservoir using DYRESM-CAEDYM | |
| CN110956312A (en) | Photovoltaic power distribution network voltage prediction method based on EMD-CNN deep neural network | |
| CN112330065A (en) | Runoff forecasting method based on basic flow segmentation and artificial neural network model | |
| Sharadqah et al. | Nonlinear Rainfall Yearly Prediction based on Autoregressive Artificial Neural Networks Model in Central Jordan using Data Records: 1938-2018 | |
| CN114219131A (en) | Watershed runoff prediction method based on LSTM | |
| Hao et al. | Prediction of nitrogen oxide emission concentration in cement production process: a method of deep belief network with clustering and time series | |
| CN106909798A (en) | A kind of Daily rainfall multi-mode collection approach based on cumulative probability curve | |
| CN110555566B (en) | B-spline quantile regression-based photoelectric probability density prediction method | |
| CN116843012A (en) | Time sequence prediction method integrating personalized context and time domain dynamic characteristics | |
| CN115130736A (en) | Method and system for determining hydrological ecological response relation and ecological flow based on Bayesian hierarchical model | |
| CN116307068A (en) | Multi-city multi-atmosphere pollutant prediction method based on four-dimensional directed GCN-LSTM model | |
| CN114978931A (en) | Network traffic prediction method and device based on manifold learning and storage medium | |
| Gupta et al. | Single Step-Ahead Solar Irradiation Forecasting Based on Empirical Mode Decomposition with Back Propagation Neural Network | |
| CN114154404A (en) | Method for deducing running state and parameters of adjacent hydropower stations by using observation data |
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 |