CN107644267B - Greenhouse control decision fusion method based on D-S evidence theory - Google Patents

Greenhouse control decision fusion method based on D-S evidence theory Download PDF

Info

Publication number
CN107644267B
CN107644267B CN201710812597.3A CN201710812597A CN107644267B CN 107644267 B CN107644267 B CN 107644267B CN 201710812597 A CN201710812597 A CN 201710812597A CN 107644267 B CN107644267 B CN 107644267B
Authority
CN
China
Prior art keywords
decision
value
clustering
omega
evidence theory
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
Application number
CN201710812597.3A
Other languages
Chinese (zh)
Other versions
CN107644267A (en
Inventor
孙力帆
张雅媛
郑国强
付主木
王旭栋
冀保峰
普杰信
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Henan University of Science and Technology
Original Assignee
Henan University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Henan University of Science and Technology filed Critical Henan University of Science and Technology
Priority to CN201710812597.3A priority Critical patent/CN107644267B/en
Publication of CN107644267A publication Critical patent/CN107644267A/en
Application granted granted Critical
Publication of CN107644267B publication Critical patent/CN107644267B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

A greenhouse control decision fusion method based on a D-S evidence theory comprises the following steps: step one, extracting a wild value in a measurement value of a sensor in a greenhouse by using a boxcar graph, and correcting the wild value; step two, carrying out clustering fusion on the corrected measurement values to obtain a clustering value set; and thirdly, performing decision distribution on the clustering values by using the clustering value set and the decision reference value set through a D-S evidence theory. The invention provides a greenhouse control decision fusion method based on a D-S evidence theory, which can effectively improve the fusion precision and decision reliability of index data in a greenhouse environment and simultaneously reduce decision risk.

Description

Greenhouse control decision fusion method based on D-S evidence theory
Technical Field
The invention relates to the field of data processing and fusion of wireless sensor networks, in particular to a greenhouse control decision fusion method based on a D-S evidence theory.
Background
A greenhouse is also called a greenhouse, and refers to a room provided with cold-proof, heating, light-transmitting and other facilities for cultivating temperature-favored plants in winter. In seasons unsuitable for plant growth, the method can provide a growth period and increase yield, and is mainly used for cultivating or raising seedlings of plants such as warm vegetables, flowers and trees in low-temperature seasons. Existing greenhouses typically include a roll of film for enclosure, a fan for ventilation, and a wet curtain for increasing the humidity within the greenhouse. Most of the traditional greenhouse control methods are that a sensor is used for directly measuring environmental parameters in a temperature measuring chamber, then the environmental parameters are compared with a set threshold value, the adjustment of the environmental parameters is realized by controlling the action of a rolling film, a fan or a wet curtain, and finally the measurement value of the sensor is checked. However, since the environment of the greenhouse is very complex, the measurement results of the sensors may vary greatly at different locations, which may lead to the control system generating completely opposite control commands and thus to a chaotic environment within the greenhouse.
With the rapid development of scientific technology and the remarkable improvement of living standard, modern intelligent greenhouse environmental systems are receiving more and more attention. In order to save manpower, material resources and financial resources, the sensor network becomes the first choice of people, and the information fusion with the sensor network which cannot be separated from the sensor network is widely applied to the fields of target identification, target tracking, artificial intelligence and the like. Due to various external environmental factors and hardware influences, data measured by a sensor often has certain uncertainty, and a D-S (Dempster-Shafer) evidence theory is a well-advocated method for processing uncertainty, is already applied to many expert systems, and has a non-negligible role in decision making. However, when there is a high degree of conflict between evidences, the D-S evidence theory often repels the common rationale in the processing results, and erroneous processing results can cause erroneous control of the greenhouse environment, resulting in disruption of greenhouse function.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a greenhouse control decision fusion method based on a D-S evidence theory, which can effectively improve the fusion precision and decision reliability of index data in a greenhouse environment and reduce decision risk.
In order to achieve the purpose, the invention adopts the specific scheme that:
a greenhouse control decision fusion method based on a D-S evidence theory comprises the following steps:
step one, extracting a wild value in a measurement value of a sensor in a greenhouse by using a boxcar graph, and correcting the wild value;
step two, carrying out clustering fusion on the corrected measurement values to obtain a clustering value set;
and thirdly, performing decision distribution on the clustering values by using the clustering value set and the decision reference value set through a D-S evidence theory.
The method for correcting the outlier in the first step comprises the following steps:
step S1, calculating the wild value x and the decision reference value xmDistance between delta ═ xm-x|；
Step S2, if delta is less than or equal to 5, the corrected measurement value
Figure BDA0001404288180000021
Step S3, if Delta is more than 5 and less than or equal to 10, the corrected measurement value
Figure BDA0001404288180000022
In the formula
Figure BDA0001404288180000023
Step S4, if Delta is more than 10 and less than or equal to 15, the corrected measurement value formula
Figure BDA0001404288180000026
Step S5, if the corrected measurement value
Figure BDA0001404288180000027
If the outlier is still present, the steps S1 to S4 are repeated to perform the correction again.
And the clustering fusion method in the second step is to search the minimum distance value in the modified Mahalanobis distance matrix of the measurement values and perform clustering according to the minimum distance value.
In the third step, the corrected measurement value at a certain time is
Figure BDA0001404288180000028
Mean value of u ═ u (u)1,u2,···,un)TThen the covariance matrix is
Figure BDA0001404288180000029
And (3) solving the Mahalanobis distance according to the covariance matrix:
Figure BDA00014042881800000210
the specific clustering method comprises the steps of searching a minimum distance value in the D, aggregating two data corresponding to the minimum distance value into a class, generating a new Mahalanobis distance matrix, clustering again, and obtaining a clustering value set through clustering for a plurality of times.
Let clustering value set X be expressed as
Figure BDA00014042881800000211
The decision reference value set ω is represented as (ω)1,ω2,···,ωn) The recognition framework theta of the D-S evidence theory is expressed as (L)1，L2，......，Ln)，(L1，L2，......，Ln) Respectively, the decision assigned to the cluster value, (L)1，L2，......，Ln) And (omega)1,ω2,···,ωn) One-to-one correspondence, the specific decision distribution method is as follows:
step T1, calculating the clustering value and the mean value omegakAnd ω andk+1is a distance of
Figure BDA0001404288180000035
Step T2, compare d1And d2；
If d is1＞d2Then, then
Figure BDA0001404288180000036
Close to ωk+1，
Figure BDA0001404288180000037
For decision Lk+1More trusted, the probability distribution function for each decision is
Figure BDA0001404288180000038
In the formula (1-d)1) To represent
Figure BDA0001404288180000039
And omegak(1-d) of (A)2) To represent
Figure BDA00014042881800000310
And omegak+1The degree of similarity of (a) to (b),
Figure BDA00014042881800000311
and
Figure BDA00014042881800000312
respectively representing weighting factors, p, of the probability distribution function1And rho2The constraint factor is used for preventing the trust degree from deviating from the reality too much;
if d is1＜d2Then, then
Figure BDA00014042881800000313
Is relatively close to omegak，
Figure BDA00014042881800000314
For decision LkMore trusted, the probability distribution function for each decision is
Figure BDA00014042881800000315
In the formula (1-d)1) To represent
Figure BDA00014042881800000316
And omegak(1-d) of (A)2) To represent
Figure BDA00014042881800000317
And omegak+1The degree of similarity of (a) to (b),
Figure BDA00014042881800000318
and
Figure BDA00014042881800000319
respectively representing weighting factors, p, of the probability distribution function1And rho2The constraint factor is used for preventing the trust degree from deviating from the reality too much;
if d is1＝d2Then, then
Figure BDA00014042881800000320
And omegakAnd ωk+1The similarity of (a) to (b), the probability distribution function for each decision and (d)1＞d2Condition of (a) or (d)1＜d2In the case of (3) any allocation function is the same;
step T3, calculating m (Θ) to 1-m (L)1)-m(L2)-……-m(Ln) M (Θ) is the uncertainty for the measured data;
step T4, repeating the steps T1 to T3
Figure BDA0001404288180000041
And omegak+2～ωnComparing one by one and calculating the probability distribution function of each decision;
step T5, repeating steps T1 to T4, and pairing
Figure BDA0001404288180000042
And performing decision distribution.
The restriction factor ρ1＝5，ρ2＝3。
Has the advantages that:
1. according to the method, the data collected by the sensor are optimized through outlier processing and clustering fusion, so that the conflict degree among the measured data is effectively reduced, and a quick decision is facilitated to be made;
2. the fusion precision and decision reliability of index data in the greenhouse environment can be effectively improved, and the decision risk is reduced.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention;
FIG. 2 is a raw measurement data boxline diagram;
FIG. 3 is a boxplot of raw measurement data after correction;
fig. 4 is a result of comparing cluster values after correction/elimination of outliers at a minimum distance.
Detailed Description
Embodiments of the present invention will be specifically described below with reference to the accompanying drawings.
A greenhouse control decision fusion method based on a D-S evidence theory is shown in figure 1 and comprises the following steps:
step one, extracting a wild value in a measurement value of a sensor in a greenhouse by using a boxcar graph, and correcting the wild value;
step two, carrying out clustering fusion on the corrected measurement values to obtain a clustering value set;
and thirdly, performing probability distribution on each subset of the identification framework of the D-S evidence theory by using the clustering values and the decision reference values, establishing a decision fusion framework based on the D-S evidence theory, and performing decision on adjustment of the greenhouse through the framework. The decision reference value is generated by an expert system, is a parameter comparison standard commonly used in the field of greenhouse control, is prior art, and is not described herein again.
The method for extracting the wild value in the sensor measurement values by using the boxplot in the first step is to arrange the measured data from small to large and define the 50 th percent of the data as a median Q2Defining the 25 th% data as the lower quartile Q1，Q1Can also be located as the minimum of the measured values and Q2Median between, define 75% of the data as the upper quartile Q3，Q3Can also be located as the maximum value of the measured value and Q2The median between. The outlier is defined as greater than (Q)3+1.5IQR) Or less than (Q)1-1.5IQR) The measured value of (a). The correction method of the outlier comprises the following steps:
step S1, calculating the wild value x and the decision reference value xmDistance between delta ═ xm-x|；
Step S2, if delta is less than or equal to 5, the corrected measurement value
Figure BDA0001404288180000043
Step S3, if Delta is more than 5 and less than or equal to 10, the corrected measurement value
Figure BDA0001404288180000051
In the formula
Figure BDA0001404288180000052
Step S4, if Delta is more than 10 and less than or equal to 15, the corrected measurement value formula
Figure BDA0001404288180000055
Step S5, if the corrected measurement value
Figure BDA0001404288180000056
If the outlier is still present, the steps S1 to S4 are repeated to perform the correction again.
And the clustering fusion method in the second step is to search the minimum distance value in the modified Mahalanobis distance matrix of the measurement values and perform clustering according to the minimum distance value. The measurement value set at a certain time after correction is
Figure BDA0001404288180000057
Mean value of u ═ u (u)1,u2,···,un)TThen the covariance matrix is
Figure BDA0001404288180000058
And (3) solving the Mahalanobis distance according to the covariance matrix:
Figure BDA0001404288180000059
the specific clustering method comprises the steps of searching a minimum distance value in the D, aggregating two data corresponding to the minimum distance value into a class, generating a new Mahalanobis distance matrix, clustering again, and obtaining a clustering value set through clustering for a plurality of times.
In step three, let clustering value set X denote as
Figure BDA00014042881800000510
The decision reference value set ω is represented as (ω)1,ω2,···,ωn) The recognition framework theta of the D-S evidence theory is expressed as (L)1，L2，......，Ln)，(L1，L2，......，Ln) Respectively, the decision assigned to the cluster value, (L)1，L2，......，Ln) And (omega)1,ω2,···,ωn) One-to-one correspondence, the specific decision distribution method is as follows:
step T1, calculating the clustering value and the mean value omegakAnd ω andk+1is a distance of
Figure BDA00014042881800000515
Step T2, compare d1And d2；
If d is1＞d2Then, then
Figure BDA0001404288180000061
Close to ωk+1，
Figure BDA0001404288180000062
For decision Lk+1More trusted, the probability distribution function for each decision is
Figure BDA0001404288180000063
In the formula (1-d)1) To represent
Figure BDA0001404288180000064
And omegak(1-d) of (A)2) To represent
Figure BDA0001404288180000065
And omegak+1The degree of similarity of (a) to (b),
Figure BDA0001404288180000066
and
Figure BDA0001404288180000067
respectively representing weighting factors, p, of the probability distribution function1And rho2The constraint factor is used for preventing the trust degree from deviating from the reality too much;
if d is1＜d2Then, then
Figure BDA0001404288180000068
Is relatively close to omegak，
Figure BDA0001404288180000069
For decision LkMore trusted, the probability distribution function for each decision is
Figure BDA00014042881800000610
In the formula (1-d)1) To represent
Figure BDA00014042881800000611
And omegak(1-d) of (A)2) To represent
Figure BDA00014042881800000612
And omegak+1The degree of similarity of (a) to (b),
Figure BDA00014042881800000613
and
Figure BDA00014042881800000614
respectively representing weighting factors, p, of the probability distribution function1And rho2To restrict the factor, p1＝5，ρ23, the method is used for preventing the trust degree from deviating from the reality too much;
if d is1＝d2Then, then
Figure BDA00014042881800000615
And omegakAnd ωk+1The similarity of (a) to (b), the probability distribution function for each decision and (d)1＞d2Condition of (a) or (d)1＜d2In the case of (3) any allocation function is the same;
step T3, calculating m (Θ) to 1-m (L)1)-m(L2)-……-m(Ln) M (Θ) is the uncertainty for the measured data;
step T4, repeating the steps T1 to T3
Figure BDA00014042881800000616
And omegak+2～ωnComparing one by one and calculating the probability distribution function of each decision;
step T5, repeating steps T1 to T4, and pairing
Figure BDA00014042881800000617
And performing decision distribution.
In order to verify the fusion accuracy and decision reliability of the present invention, the following simulation experiments were set:
1. in order to reduce the complexity of simulation, greenhouse environment parameters are reduced, and three parameters of temperature (/ DEG C), illuminance (/ klx) and carbon dioxide volume fraction (uL/L) are reserved after reduction;
2. 2, setting the measurement time to six moments of morning, noon and evening, namely 10, 12, 14, 16, 18 and 20; 26 data are measured for each parameter at each time instant;
3. the decision results are divided into four, namely a film opening and rolling film opening and fan starting, a fan and wet curtain starting and no action, which are respectively expressed by m (1), m (2), m (3) and m (4).
The simulation results are as follows.
Taking temperature as an example for analysis, the boxplot of the measured temperature at six moments is shown in fig. 2, it can be seen that each moment has a wild value, the boxplot of the temperature after correction is shown in fig. 3, the original distribution state is not changed after correction, the midday temperature is the highest, the evening temperature is the lowest, and the whole is in normal distribution. Meanwhile, the width of the box body of the box line graph is different, and the position of the median is also different, which depends on the dispersion degree of the measured data and the accuracy of the sensor. After the measurement data is corrected, clustering is performed on the parameters at each moment, the result is shown in fig. 4, and the cluster trends of the corrected outlier and the eliminated outlier are consistent and tend to the decision reference value by comparing the corrected clustered values with the eliminated outliers.
Figure BDA0001404288180000071
Table 1 probability distribution results in simulation experiments
Probability distribution results in simulation experiments are shown in table 1, wherein t, c and i in the table are temperature, illuminance and volume fraction of carbon dioxide respectively and serve as evidences in a D-S evidence theory. As can be seen from Table 1, the uncertainty accuracy of the present invention to the measured data can reach 10-5～10-4The method has extremely high convergence rate, effectively reduces uncertainty among evidences and is beneficial to making quick decisions.
It is further noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (5)

1. A greenhouse control decision fusion method based on a D-S evidence theory is characterized in that: the method comprises the following steps:
the method comprises the following steps of firstly, extracting a wild value in a measurement value of a sensor in a greenhouse by using a boxplot, and correcting the wild value, wherein the correction method of the wild value comprises the following steps:
step S1, calculating the wild value x and the decision reference value xmDistance between delta ═ xm-x|；
Step S2, if delta is less than or equal to 5, the corrected measurement value
Figure FDA0002403975020000011
Step S3, if Delta is more than 5 and less than or equal to 10, the corrected measurement value
Figure FDA0002403975020000012
In the formula
Figure FDA0002403975020000013
Step S4, if Delta is more than 10 and less than or equal to 15, the corrected measurement value
Figure FDA0002403975020000014
In the formula
Figure FDA0002403975020000015
Step S5, if the corrected measurement value
Figure FDA0002403975020000016
Is still wild valueIf yes, repeating the steps S1-S4 to correct again;
step two, carrying out clustering fusion on the corrected measurement values to obtain a clustering value set;
and thirdly, performing decision distribution on the clustering values by using the clustering value set and the decision reference value set through a D-S evidence theory.
2. The greenhouse control decision fusion method based on the D-S evidence theory as claimed in claim 1, characterized in that: and the clustering fusion method in the second step is to search the minimum distance value in the modified Mahalanobis distance matrix of the measurement values and perform clustering according to the minimum distance value.
3. The greenhouse control decision fusion method based on the D-S evidence theory as claimed in claim 2, characterized in that: the measurement value set at a certain time after correction is
Figure FDA0002403975020000017
Mean value of u ═ u (u)1,u2,…,un)ΤThen the covariance matrix is
Figure FDA0002403975020000018
And (3) solving the Mahalanobis distance according to the covariance matrix:
Figure FDA0002403975020000019
the specific clustering method comprises the steps of searching a minimum distance value in the D, aggregating two data corresponding to the minimum distance value into a class, generating a new Mahalanobis distance matrix, clustering again, and obtaining a clustering value set through clustering for a plurality of times.
4. The greenhouse control decision fusion method based on the D-S evidence theory as claimed in claim 1, characterized in that:
in the third step, the clustering value set X is expressed as
Figure FDA00024039750200000110
The decision reference value set ω is represented as (ω)1,ω2,…,ωn) The recognition framework theta of the D-S evidence theory is expressed as (L)1，L2，......，Ln)，(L1，L2，......，Ln) Respectively, the decision assigned to the cluster value, (L)1，L2，......，Ln) And (omega)1,ω2,…,ωn) One-to-one correspondence, the specific decision distribution method is as follows:
step T1, clustering value
Figure FDA00024039750200000111
Computing
Figure FDA00024039750200000112
And mean value omegakIs a distance of
Figure FDA00024039750200000113
And
Figure FDA00024039750200000114
and ωk+1Is a distance of
Figure FDA00024039750200000115
Step T2, compare d1And d2；
If d is1＞d2Then, then
Figure FDA0002403975020000021
Close to ωk+1，
Figure FDA0002403975020000022
For decision Lk+1More trusted, the probability distribution function for each decision is
Figure FDA0002403975020000023
In the formula (1-d)1) To represent
Figure FDA0002403975020000024
And omegak(1-d) of (A)2) To represent
Figure FDA0002403975020000025
And omegak+1The degree of similarity of (a) to (b),
Figure FDA0002403975020000026
and
Figure FDA0002403975020000027
respectively representing weighting factors, p, of the probability distribution function1And rho2The constraint factor is used for preventing the trust degree from deviating from the reality too much;
if d is1＜d2Then, then
Figure FDA0002403975020000028
Is relatively close to omegak，
Figure FDA0002403975020000029
For decision LkMore trusted, the probability distribution function for each decision is
Figure FDA00024039750200000210
In the formula (1-d)1) To represent
Figure FDA00024039750200000211
And omegak(1-d) of (A)2) To represent
Figure FDA00024039750200000212
And omegak+1The degree of similarity of (a) to (b),
Figure FDA00024039750200000213
and
Figure FDA00024039750200000214
respectively representing weighting factors, p, of the probability distribution function1And rho2The constraint factor is used for preventing the trust degree from deviating from the reality too much;
if d is1＝d2Then, then
Figure FDA00024039750200000217
And omegakAnd ωk+1The similarity of (a) to (b), the probability distribution function for each decision and (d)1＞d2Condition of (a) or (d)1＜d2In the case of (3) any allocation function is the same;
step T3, calculating m (Θ) to 1-m (L)1)-m(L2)-……-m(Ln) M (Θ) is the uncertainty for the measured data;
step T4, repeating the steps T1 to T3
Figure FDA00024039750200000215
And omegak+2～ωnComparing one by one and calculating the probability distribution function of each decision;
step T5, repeating steps T1 to T4, and pairing
Figure FDA00024039750200000216
And performing decision distribution.
5. The greenhouse control decision fusion method based on the D-S evidence theory as claimed in claim 4, characterized in that: the restriction factor ρ1＝5，ρ2＝3。
CN201710812597.3A 2017-09-11 2017-09-11 Greenhouse control decision fusion method based on D-S evidence theory Active CN107644267B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710812597.3A CN107644267B (en) 2017-09-11 2017-09-11 Greenhouse control decision fusion method based on D-S evidence theory

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710812597.3A CN107644267B (en) 2017-09-11 2017-09-11 Greenhouse control decision fusion method based on D-S evidence theory

Publications (2)

Publication Number Publication Date
CN107644267A CN107644267A (en) 2018-01-30
CN107644267B true CN107644267B (en) 2020-04-24

Family

ID=61111086

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710812597.3A Active CN107644267B (en) 2017-09-11 2017-09-11 Greenhouse control decision fusion method based on D-S evidence theory

Country Status (1)

Country Link
CN (1) CN107644267B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110097296B (en) * 2019-05-16 2022-07-05 西南交通大学 Optimal emergency treatment technology screening method and emergency decision-making system for basin sudden water pollution accident
CN111193742A (en) * 2019-12-31 2020-05-22 广东电网有限责任公司 D-S evidence theory-based power communication network anomaly detection method
CN112785004A (en) * 2021-02-03 2021-05-11 中国计量大学 Greenhouse intelligent decision-making method based on rough set theory and D-S evidence theory

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101282243A (en) * 2008-03-05 2008-10-08 中科院嘉兴中心微系统所分中心 Method for recognizing distributed amalgamation of wireless sensor network
CN101996157A (en) * 2010-10-23 2011-03-30 山东科技大学 Multisource information fusion method in evidence high-conflict environment
CN106154182A (en) * 2016-08-26 2016-11-23 上海电力学院 A kind of based on the lithium battery method for diagnosing faults improving D S evidence theory

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9699205B2 (en) * 2015-08-31 2017-07-04 Splunk Inc. Network security system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101282243A (en) * 2008-03-05 2008-10-08 中科院嘉兴中心微系统所分中心 Method for recognizing distributed amalgamation of wireless sensor network
CN101996157A (en) * 2010-10-23 2011-03-30 山东科技大学 Multisource information fusion method in evidence high-conflict environment
CN106154182A (en) * 2016-08-26 2016-11-23 上海电力学院 A kind of based on the lithium battery method for diagnosing faults improving D S evidence theory

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
《融合粗糙集和证据理论的温室环境控制推理决策方法》;王俊 等;《农业工程学报》;20120930;第28卷(第17期);第172-176页 *

Also Published As

Publication number Publication date
CN107644267A (en) 2018-01-30

Similar Documents

Publication Publication Date Title
CN107644267B (en) Greenhouse control decision fusion method based on D-S evidence theory
US10627785B2 (en) Plant production feedback control loop optimization
AU2018328254A1 (en) System and method for controlling a growth environment of a crop
CN106371318B (en) Facilities environment multiobjective optimization control method based on cooperative game
Liu et al. A WSN-based prediction model of microclimate in a greenhouse using extreme learning approaches
Hu et al. Strong neutral spatial effects shape tree species distributions across life stages at multiple scales
CN115018105A (en) Winter wheat meteorological yield prediction method and system
Zhu et al. Remote monitoring and management system of intelligent agriculture under the internet of things and deep learning
CN115013949A (en) Indoor temperature control system and method based on image recognition
CN115292334A (en) Intelligent planting method and system based on vision, electronic equipment and storage medium
CN117789893B (en) Breeding data prediction method based on correlation analysis
CN117311417B (en) Intelligent agricultural information comprehensive management method and system based on Internet of things
CN108388910A (en) A kind of high conflicting evidence fusion method based on distance function
CN107656564B (en) Agricultural greenhouse internal environment control device
Zhang et al. Modeling and simulation of temperature control system in plant factory using energy balance
CN117475429A (en) Control method and device based on key environmental parameters of straw mushroom biological characteristics
CN109117996B (en) Method for constructing greenhouse winter temperature prediction model
Wang et al. Genetic groups in the common plant species Castanopsis chinensis and their associations with topographic habitats
CN111625030A (en) Greenhouse environment control method, device, equipment, system and storage medium
KR102471742B1 (en) Method for building an optimal linear model of production and environment of smart farm
CN110073857A (en) A kind of greenhouse facade ventilating and thermal insulating global anti-wind system and control method
Pagé-Fortin Class-Incremental Learning of Plant and Disease Detection: Growing Branches with Knowledge Distillation
US20210272137A1 (en) Apparatus for Fast Clustering of Massive Data Based on Variate-Specific Population Strata
CN108595843B (en) Dynamically self-adaptive crowd-sourced design scheme data optimization method
Shantkumari et al. Adaptive machine learning approach for grape leaf segmentation

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