CN108133282A - A kind of Growth of Dendrobium candidum environmental forecasting method - Google Patents
A kind of Growth of Dendrobium candidum environmental forecasting method Download PDFInfo
- Publication number
- CN108133282A CN108133282A CN201711282227.XA CN201711282227A CN108133282A CN 108133282 A CN108133282 A CN 108133282A CN 201711282227 A CN201711282227 A CN 201711282227A CN 108133282 A CN108133282 A CN 108133282A
- Authority
- CN
- China
- Prior art keywords
- error
- momentum
- states
- learning rate
- 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.)
- Granted
Links
- 241000026010 Dendrobium candidum Species 0.000 title claims abstract description 19
- 230000012010 growth Effects 0.000 title claims abstract description 19
- 230000007613 environmental effect Effects 0.000 title claims abstract description 13
- 238000013277 forecasting method Methods 0.000 title claims abstract description 4
- 238000013528 artificial neural network Methods 0.000 claims abstract description 38
- 230000008859 change Effects 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 9
- 230000007423 decrease Effects 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 11
- 238000013461 design Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 claims description 6
- 239000002689 soil Substances 0.000 claims description 6
- 238000005286 illumination Methods 0.000 claims description 4
- 229910002092 carbon dioxide Inorganic materials 0.000 claims description 3
- 239000001569 carbon dioxide Substances 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000000630 rising effect Effects 0.000 claims 2
- 230000008635 plant growth Effects 0.000 abstract description 5
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 230000007547 defect Effects 0.000 abstract description 2
- 238000000034 method Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000000259 anti-tumor effect Effects 0.000 description 1
- 239000003963 antioxidant agent Substances 0.000 description 1
- 230000003078 antioxidant effect Effects 0.000 description 1
- 235000006708 antioxidants Nutrition 0.000 description 1
- 210000001124 body fluid Anatomy 0.000 description 1
- 239000010839 body fluid Substances 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002218 hypoglycaemic effect Effects 0.000 description 1
- 210000004185 liver Anatomy 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000001151 other effect Effects 0.000 description 1
- 230000000144 pharmacologic effect Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 210000002784 stomach Anatomy 0.000 description 1
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
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Biomedical Technology (AREA)
- Mining & Mineral Resources (AREA)
- Animal Husbandry (AREA)
- Agronomy & Crop Science (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Primary Health Care (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Marine Sciences & Fisheries (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Feedback Control In General (AREA)
Abstract
The present invention relates to a kind of Growth of Dendrobium candidum environmental forecasting methods, including growing environment prediction model and the step of improved BP neural network, wherein growing environment prediction model includes making normalized to the parameter of acquisition, make neural network that there is the step of good fitness, improved BP neural network to include level-one fuzzy controller to input error signal e (n) processing and two level fuzzy controller to error originated from input change rate to sampleIt is handled.The beneficial effects of the invention are as follows：It is predicted the present invention provides a kind of improved BP neural network algorithm for plant growth environment, improved BP neural network overcomes the defects of traditional BP neural network convergence rate is slow, change step-length simultaneously, so as to make Step-varied back propagation according to growing environment using fuzzy control, convergence rate is improved.
Description
Technical field
The Forecasting Methodology of present invention design plant growth environment more particularly to a kind of Growth of Dendrobium candidum environmental forecasting side
Method.
Background technology
Dendrobium candidum is the traditional rare traditional Chinese medicine in China, pharmacological action have enhance it is immune, antitumor, anti-oxidant,
Beneficial liver stomach, hypoglycemic, nourishing Yin and promoting production of body fluid and other effects.Since Growth of Dendrobium candidum environment is more special, natural propagation rate is low, and growth is slow
The reasons such as slow, the supply of dendrobium candidum wretched insufficiency always, and be illumination, temperature an important factor for wherein influence Growth of Dendrobium candidum
Degree, humidity etc. are answered so can ensure to take in advance in the case of bad with prediction to growth environment progress monitoring in real time
To measure, unnecessary loss is avoided.
BP neural network has because with good nonlinear fitting characteristic for solving plant growth environment aspect at present
Extensive use.If it is required that predicting large area growing process, mass data is needed to carry out repeatedly effectively study, because
This learning time and space consuming are very big.
Invention content
The purpose of the present invention is overcoming deficiency of the prior art, a kind of Growth of Dendrobium candidum ring quickly and efficiently is provided
The Forecasting Methodology in border.The purpose of the present invention is what is be achieved through the following technical solutions：
The Forecasting Methodology of this Growth of Dendrobium candidum environment, includes the following steps：
The first step, growing environment prediction model：
Normalized is made to the parameter of acquisition first, makes neural network that there is good fitness to sample：
Wherein, x ' is numerical value after normalization, and x is current value, xmaxFor sample maximum, xminFor sample minimum；
The input layer number for setting neural network is A, and A represents the principal element for influencing Growth of Dendrobium candidum environment
Number, the i.e. soil moisture, soil moisture, air themperature, air humidity, illumination, carbon dioxide and time；Output layer number of nodes is B,
B represents the principal element number for influencing Growth of Dendrobium candidum environmental forecasting, generally sets A=B；Hidden layer node C by input and
Depending on the requirement of output；
To shorten the training time of network, using serial input mode, N number of training data known to input sample, N+1
Data are trained for tutor；After, input data moves backward a data as next group of input data successively, with N+
2 data are trained for tutor, and so on are gone down；It is possible thereby to grasp the rule of neural network prediction, prediction model is：
Dn+1=F (Dn, Dn-1, Dn-2, Dn-3..., Dn-m)
F(Dn) represent that historical data passes through the function of neural network prediction mapping, DnRepresent the value in n moment environmental parameters；
The step of second step, improved BP neural network：
Data are handled first, the larger data processing of error is fallen, secondly using two level fuzzy controller to it
He is handled data error；
1) level-one fuzzy controller is to input error signal e (n) processing：Error input one caused by neural network is restrained
Grade fuzzy controller, compared with last error e (n-1), the difference (e (n)-e (n-1)) that can determine whether error be convergence or
Diverging；By the membership function of setting, be divided into error rises at a high speed PB1, error middling speed rises PM1, error low speed rises PS1,
Error stablizes 01, error low speed and declines NS1, error middling speed decline NM1 and error decline seven kinds of states of NB1 at a high speed, is obscured
Processing makes corresponding change, so as to adjust step-length according to the rule base of design for learning rate and momentum：
1. when if e (n) is PB1 states, then suitably reduce learning rate, momentum is set as 0.01, cancels this time
Iteration returns to last iteration；
2. when if e (n) is PM1 states, then suitably reduce learning rate and momentum, continue next iteration；
3. when if e (n) is PS1 states, then suitably reduce learning rate and momentum, continue next iteration；
4. when if e (n) is 01 state, then learning rate and momentum remain unchanged, and continue next iteration；
5. when if e (n) is NS1 states, then suitably increase learning rate and momentum, continue next iteration；
6. when if e (n) is NM1 states, then suitably increase learning rate and momentum, continue next iteration；
7. when if e (n) is NB1 states, then suitably increase learning rate and momentum, continue next iteration；
2) two level fuzzy controller is to error originated from input change rateIt is handled：By level-one Fuzzy Control
Treated that error is converted into error rate is input to two level fuzzy controller for device processed, by the membership function of setting, is divided into
Error rate rises at a high speed PB2, error rate middling speed rises PM2, error rate low speed rises PS2, error rate stablizes 02, error rate low speed
Decline NS2, error rate middling speed decline NM2 and error rate and decline seven kinds of states of NB2 at a high speed, phase is made to learning rate and momentum again
It should change, so as to adjust step-length：
1. when if Δ is PB2 states, then suitably reduce learning rate, momentum is set as 0.01, and cancellation this time changes
In generation, returns to last iteration；
2. when if Δ is PM2 states, then suitably reduce learning rate and momentum, continue next iteration；
3. when if Δ is PS2 states, then suitably reduce learning rate and momentum, continue next iteration；
4. when if Δ is 02 state, then learning rate and momentum remain unchanged, and continue next iteration；
5. when if Δ is NS2 states, then suitably increase learning rate and momentum, continue next iteration；
6. when if Δ is NM2 states, then suitably increase learning rate and momentum, continue next iteration；
7. when if Δ is NB2 states, then suitably increase learning rate and momentum, continue next iteration.
The beneficial effects of the invention are as follows：The present invention provides a kind of improved BP neural network algorithms to be used for plant growth ring
Border predicts that improved BP neural network overcomes the defects of traditional BP neural network convergence rate is slow, while changes step-length, so as to
Step-varied back propagation is made according to growing environment using fuzzy control, improves convergence rate.
Description of the drawings
Fig. 1 is invention neural network prediction model schematic diagram；
Fig. 2 is design of Fuzzy Controller schematic diagram；
Fig. 3 is membership function schematic diagram (level-one is identical with two level fuzzy controller degree of membership design)；
Fig. 4 is improved model schematic diagram；
Fig. 5 is fitting result chart；
Fig. 6 is prediction effect figure.
Specific embodiment
The present invention is described further with reference to embodiment.Following embodiments are served only for helping to understand the present invention.It is right
In those skilled in the art, without departing from the principle of the present invention, the present invention can also be improved,
These improvement are also fallen within the protection scope of the claims of the present invention.
In order to be predicted, plant growth environment is monitored, (period is small for 1 for the data detected using sensor
When), it enters data into improved BP neural network and is predicted, then carry out alert process.
Normalized is made to the parameter of acquisition first, makes neural network that there is good fitness to sample：
Wherein, x ' is numerical value after normalization, and x is current value, xmaxFor sample maximum, xminFor sample minimum.
The input layer number for setting neural network is A, and A represents the principal element for influencing Growth of Dendrobium candidum environment
Number, the i.e. soil moisture, soil moisture, air themperature, air humidity, illumination, carbon dioxide and time；Output layer number of nodes is B,
B represents the principal element number (generally setting A=B) for influencing Growth of Dendrobium candidum environmental forecasting；Hidden layer node C by input and
Depending on the requirement of output.
To shorten the training time of network, using serial input mode, N number of training data known to input sample, N+1
Data are trained for tutor.After, input data moves backward a data as next group of input data successively, with N+
2 data are trained for tutor, and so on are gone down.It is possible thereby to grasp the rule of neural network prediction, prediction model is：
Dn+1=F (Dn, Dn-1, Dn-2, Dn-3..., Dn-m)
F(Dn) represent that historical data passes through the function of neural network prediction mapping, DnRepresent the value in n moment environmental parameters.
The step of second step, improved BP neural network：
Data are handled first, the larger data processing of error is fallen, prevent the error that randomness is brought to entire
Prediction process affects, and secondly other data errors are handled using two level fuzzy controller.
1) level-one fuzzy controller is to input error signal e (n) processing：Error input one caused by neural network is restrained
Grade fuzzy controller, compared with last error e (n-1), the difference (e (n)-e (n-1)) that can determine whether error be convergence or
Diverging；By the membership function of setting, be divided into error rises at a high speed PB1, error middling speed rises PM1, error low speed rises PS1,
Error stablizes 01, error low speed and declines NS1, error middling speed decline NM1 and error decline seven kinds of states of NB1 at a high speed, is obscured
Processing makes corresponding change, so as to adjust step-length according to the rule base of design for learning rate and momentum：
1. when if e (n) is PB1 states, then suitably reduce learning rate, momentum is set as 0.01, cancels this time
Iteration returns to last iteration；
2. when if e (n) is PM1 states, then suitably reduce learning rate and momentum, continue next iteration；
3. when if e (n) is PS1 states, then suitably reduce learning rate and momentum, continue next iteration；
4. when if e (n) is 01 state, then learning rate and momentum remain unchanged, and continue next iteration；
5. when if e (n) is NS1 states, then suitably increase learning rate and momentum, continue next iteration；
6. when if e (n) is NM1 states, then suitably increase learning rate and momentum, continue next iteration；
7. when if e (n) is NB1 states, then suitably increase learning rate and momentum, continue next iteration.
2) two level fuzzy controller is to error originated from input change rateIt is handled：By level-one Fuzzy Control
Treated that error is converted into error rate is input to two level fuzzy controller for device processed, by the membership function of setting, is divided into
Error rate rises at a high speed PB2, error rate middling speed rises PM2, error rate low speed rises PS2, error rate stablizes 02, error rate low speed
Decline NS2, error rate middling speed decline NM2 and error rate and decline seven kinds of states of NB2 at a high speed, phase is made to learning rate and momentum again
It should change, so as to adjust step-length：
1. when if Δ is PB2 states, then suitably reduce learning rate, momentum is set as 0.01, and cancellation this time changes
In generation, returns to last iteration；
2. when if Δ is PM2 states, then suitably reduce learning rate and momentum, continue next iteration；
3. when if Δ is PS2 states, then suitably reduce learning rate and momentum, continue next iteration；
4. when if Δ is 02 state, then learning rate and momentum remain unchanged, and continue next iteration；
5. when if Δ is NS2 states, then suitably increase learning rate and momentum, continue next iteration；
6. when if Δ is NM2 states, then suitably increase learning rate and momentum, continue next iteration；
7. when if Δ is NB2 states, then suitably increase learning rate and momentum, continue next iteration.
Wherein to the design of the membership function of error and error rate as shown in figure 3, can be treated to be -3, -2, -1,0,1,2,
3 } seven value signals represent PB1/PB2, PM1/PM2, PS1/PS2,01/02, NS1/NS2, NM1/NM2, NB1/NB2 seven respectively
Kind state.Level-one is taken to obscure as shown in table 1 with two level fuzzy rule base design same procedure here.
Learning rate changes rule base with momentum under each state of 1 error of table/error rate
The BP neural network model structure finally improved is as shown in figure 4, first input Growth of Dendrobium candidum environmental parameter
To BP neural network, then prediction is input to substantial measurement errors in level-one fuzzy controller, the error of secondary calculating into
Row derivation is input to two level fuzzy controller, and then adaptively changing momentum and learning rate, is finally output in neural network, from
And predict the variation of growing environment.
In order to examine improved BP neural network and traditional BP neural network and the effect of other improvements BP neural network,
The same group of Growth of Dendrobium candidum environmental data detected is separately input in each network, prediction of result is compared as follows 2 institute of table
Show.
2 prediction result of table
The result shows that BP neural network of the present invention has good error convergence effect, and speed is faster.
Prediction result：The quality of a neural network is evaluated, can be seen that in terms of the fitting of neural network and prediction two.
The quality being wherein fitted is determined by training method, and the result predicted is determined by the quality being fitted.If parameter setting is not
Overfitting rationally is caused, the precision of prediction will decline.As shown in Figure 5 and Figure 6, respectively fitting and prognostic chart are in figure
The 1920 groups of data obtained for 20 days sample (1 hour period) by equally spaced, after debug data, obtain 225 groups
Data, using 125 groups of front data as training data, behind 100 groups as detection data, wherein solid line represents number of actual measurements
According to dotted line represents its predicted value, therefore this model can predict the growing environment Parameters variation of 1 hour in advance, so as to be adopted to it
Take corresponding measure.
Claims (1)
- A kind of 1. Growth of Dendrobium candidum environmental forecasting method, which is characterized in that include the following steps：The first step, growing environment prediction model：Normalized is made to the parameter of acquisition first, makes neural network that there is good fitness to sample：Wherein, x ' is numerical value after normalization, and x is current value, xmaxFor sample maximum, xminFor sample minimum；The input layer number for setting neural network is A, and A represents the principal element number for influencing Growth of Dendrobium candidum environment, i.e., The soil moisture, soil moisture, air themperature, air humidity, illumination, carbon dioxide and time；Output layer number of nodes is B, and B is represented The principal element number of Growth of Dendrobium candidum environmental forecasting is influenced, A=B is generally set；Hidden layer node C is by outputting and inputting Depending on it is required that；To shorten the training time of network, using serial input mode, N number of training data known to input sample, the N+1 data It is trained for tutor；After, input data moves backward a data as next group of input data successively, with N+2 Data are trained for tutor, and so on are gone down；It is possible thereby to grasp the rule of neural network prediction, prediction model is：Dn+1=F (Dn, Dn-1, Dn-2, Dn-3..., Dn-m)F(Dn) represent that historical data passes through the function of neural network prediction mapping, DnRepresent the value in n moment environmental parameters；The step of second step, improved BP neural network：Data are handled first, the larger data processing of error is fallen, secondly other are counted using two level fuzzy controller It is handled according to error；1) level-one fuzzy controller is to input error signal e (n) processing：Error caused by neural network is restrained inputs level-one mould Fuzzy controllers are compared with last error e (n-1), and the difference (e (n)-e (n-1)) that can determine whether error is convergence or hair It dissipates；By the membership function of setting, it is divided into error and rises PB1 at a high speed, error middling speed rising PM1, error low speed rising PS1, misses Difference stablizes 01, error low speed and declines NS1, error middling speed decline NM1 and error decline seven kinds of states of NB1 at a high speed, carries out fuzzy place Reason makes corresponding change, so as to adjust step-length according to the rule base of design for learning rate and momentum：1. when if e (n) is PB1 states, then learning rate is suitably reduced, momentum is set as 0.01, cancels this time iteration, Return to last iteration；2. when if e (n) is PM1 states, then suitably reduce learning rate and momentum, continue next iteration；3. when if e (n) is PS1 states, then suitably reduce learning rate and momentum, continue next iteration；4. when if e (n) is 01 state, then learning rate and momentum remain unchanged, and continue next iteration；5. when if e (n) is NS1 states, then suitably increase learning rate and momentum, continue next iteration；6. when if e (n) is NM1 states, then suitably increase learning rate and momentum, continue next iteration；7. when if e (n) is NB1 states, then suitably increase learning rate and momentum, continue next iteration；2) two level fuzzy controller is to error originated from input change rateIt is handled：By level-one fuzzy controller Error that treated is converted into error rate and is input to two level fuzzy controller, by the membership function of setting, is divided into error Rate rises at a high speed PB2, error rate middling speed rises PM2, error rate low speed rises PS2, error rate stablizes the decline of 02, error rate low speed NS2, error rate middling speed decline NM2 and error rate declines at a high speed seven kinds of states of NB2, learning rate and momentum are made accordingly changing again Become, so as to adjust step-length：1. when if Δ is PB2 states, then suitably reduce learning rate, momentum is set as 0.01, cancels this time iteration, returns To last iteration；2. when if Δ is PM2 states, then suitably reduce learning rate and momentum, continue next iteration；3. when if Δ is PS2 states, then suitably reduce learning rate and momentum, continue next iteration；4. when if Δ is 02 state, then learning rate and momentum remain unchanged, and continue next iteration；5. when if Δ is NS2 states, then suitably increase learning rate and momentum, continue next iteration；6. when if Δ is NM2 states, then suitably increase learning rate and momentum, continue next iteration；7. when if Δ is NB2 states, then suitably increase learning rate and momentum, continue next iteration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711282227.XA CN108133282B (en) | 2017-12-07 | 2017-12-07 | Dendrobium officinale growth environment prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711282227.XA CN108133282B (en) | 2017-12-07 | 2017-12-07 | Dendrobium officinale growth environment prediction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108133282A true CN108133282A (en) | 2018-06-08 |
| CN108133282B CN108133282B (en) | 2021-03-30 |
Family
ID=62390151
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711282227.XA Active CN108133282B (en) | 2017-12-07 | 2017-12-07 | Dendrobium officinale growth environment prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108133282B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109324503A (en) * | 2018-08-28 | 2019-02-12 | 南京理工大学 | Multilayer neural network electric system control method based on robust integral |
| CN112666246A (en) * | 2020-12-17 | 2021-04-16 | 贵州中医药大学 | Method for screening dendrobium officinale cultivated by imitating wild rock fissure epiphytic growth |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6128609A (en) * | 1997-10-14 | 2000-10-03 | Ralph E. Rose | Training a neural network using differential input |
| US20150106310A1 (en) * | 2013-10-16 | 2015-04-16 | University Of Tennessee Research Foundation | Method and apparatus for constructing a neuroscience-inspired artificial neural network |
| CN104835103A (en) * | 2015-05-11 | 2015-08-12 | 大连理工大学 | Mobile network health evaluation method based on neural network and fuzzy comprehensive evaluation |
| CN105701280A (en) * | 2016-01-05 | 2016-06-22 | 浙江大学城市学院 | Southern America white-leg shrimp pond culture water quality prediction method |
| CN107255920A (en) * | 2017-06-21 | 2017-10-17 | 武汉理工大学 | PID control method and apparatus and system based on network optimization algorithm |
-
2017
- 2017-12-07 CN CN201711282227.XA patent/CN108133282B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6128609A (en) * | 1997-10-14 | 2000-10-03 | Ralph E. Rose | Training a neural network using differential input |
| US20150106310A1 (en) * | 2013-10-16 | 2015-04-16 | University Of Tennessee Research Foundation | Method and apparatus for constructing a neuroscience-inspired artificial neural network |
| CN104835103A (en) * | 2015-05-11 | 2015-08-12 | 大连理工大学 | Mobile network health evaluation method based on neural network and fuzzy comprehensive evaluation |
| CN105701280A (en) * | 2016-01-05 | 2016-06-22 | 浙江大学城市学院 | Southern America white-leg shrimp pond culture water quality prediction method |
| CN107255920A (en) * | 2017-06-21 | 2017-10-17 | 武汉理工大学 | PID control method and apparatus and system based on network optimization algorithm |
Non-Patent Citations (2)
| Title |
|---|
| ABDELLA M: "Treatment of missing data using neural networks and genetic algorithms", 《PROCEEDINGS OF THE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS （IJCNN）》 * |
| 丁金婷: "模糊方法改进的反向传输神经网络预测南美白对虾养殖的水质", 《浙江大学学报（农业与生命科学版）》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109324503A (en) * | 2018-08-28 | 2019-02-12 | 南京理工大学 | Multilayer neural network electric system control method based on robust integral |
| CN112666246A (en) * | 2020-12-17 | 2021-04-16 | 贵州中医药大学 | Method for screening dendrobium officinale cultivated by imitating wild rock fissure epiphytic growth |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108133282B (en) | 2021-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107168402B (en) | Environment of chicken house temperature intelligent monitoring system based on CAN fieldbus | |
| CN110580021B (en) | Granary environmental safety intelligent monitoring system based on field bus | |
| CN108920812B (en) | Machining surface roughness prediction method | |
| CN105701280B (en) | Penaeus Vannmei encloses pool cultivation water prediction technique | |
| CN109635318A (en) | A kind of aero-engine sensor intelligent analytic redundancy design method based on KEOS-ELM algorithm | |
| CN112016736A (en) | Photovoltaic power generation power control method based on gate control convolution and attention mechanism | |
| Kadir et al. | Wheat yield prediction: Artificial neural network based approach | |
| CN108133282A (en) | A kind of Growth of Dendrobium candidum environmental forecasting method | |
| CN112713881B (en) | Synchronous clock maintaining system and method based on edge calculation | |
| Li et al. | Application of ARIMA and LSTM in relative humidity prediction | |
| CN107045659A (en) | Predict the method and device of photovoltaic generation power | |
| CN105574619A (en) | Solar photovoltaic power generation output prediction system, and prediction method thereof | |
| CN117406593A (en) | Intelligent data processing method and system applied to fishery management | |
| Li et al. | The research of temperature compensation for thermopile sensor based on improved PSO-BP algorithm | |
| CN103675012B (en) | The industrial melt index soft measurement instrument of BP particle group optimizing and method | |
| Huang et al. | The application of neural network in wind speed forecasting | |
| Danilin et al. | The research of operation accuracy of a memristor-based artificial neural network with an input signal containing noise and pulse interference | |
| Jun-Ma et al. | Photovoltaic power generation prediction based on MEA-BP neural network | |
| Ansari et al. | Wind power forecasting using artificial neural network | |
| Ge et al. | Research on Temperature and Humidity Prediction Model of Granary Based on RNN-LSTM | |
| CN112816211A (en) | Belt conveyor fault diagnosis method based on PSO-BP algorithm | |
| Kamarzaman et al. | Adaptive mechanism for enhanced performance of shark smell optimization | |
| Yin et al. | Forecasting the Intrinsic Viscosity of Polyester Based on Improved Extreme Learning Machine | |
| Chen | Research on multi-sensor data fusion technology based on PSO-RBF neural network | |
| CN108334975B (en) | Oxygen content dynamic prediction method and device for anhydrous storage environment |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220707 Address after: 310015 No. 51, Huzhou street, Hangzhou, Zhejiang Patentee after: Zhejiang University City College Address before: 310015 No. 50 Huzhou Street, Hangzhou City, Zhejiang Province Patentee before: Zhejiang University City College |