CN106990769A - A kind of greenhouse comprehensive test instrument and method - Google Patents
A kind of greenhouse comprehensive test instrument and method Download PDFInfo
- Publication number
- CN106990769A CN106990769A CN201710425492.2A CN201710425492A CN106990769A CN 106990769 A CN106990769 A CN 106990769A CN 201710425492 A CN201710425492 A CN 201710425492A CN 106990769 A CN106990769 A CN 106990769A
- Authority
- CN
- China
- Prior art keywords
- greenhouse
- interval
- node
- hidden layer
- output
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012360 testing method Methods 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 55
- 230000001360 synchronised effect Effects 0.000 claims abstract description 28
- 230000003993 interaction Effects 0.000 claims abstract description 21
- 238000005070 sampling Methods 0.000 claims abstract description 19
- 238000012549 training Methods 0.000 claims description 51
- 239000011159 matrix material Substances 0.000 claims description 39
- 239000002689 soil Substances 0.000 claims description 27
- 238000010606 normalization Methods 0.000 claims description 16
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 claims description 14
- 238000011156 evaluation Methods 0.000 claims description 13
- 238000005286 illumination Methods 0.000 claims description 12
- 230000007613 environmental effect Effects 0.000 claims description 11
- 230000005540 biological transmission Effects 0.000 claims description 10
- 238000004891 communication Methods 0.000 claims description 9
- 238000012544 monitoring process Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000003860 storage Methods 0.000 claims description 8
- 229910002092 carbon dioxide Inorganic materials 0.000 claims description 7
- 239000001569 carbon dioxide Substances 0.000 claims description 7
- 230000005284 excitation Effects 0.000 claims description 4
- 238000013215 result calculation Methods 0.000 claims description 2
- 238000013528 artificial neural network Methods 0.000 abstract description 2
- 238000013480 data collection Methods 0.000 abstract description 2
- 239000000523 sample Substances 0.000 description 21
- 238000005259 measurement Methods 0.000 description 7
- 238000001514 detection method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 229910052751 metal Inorganic materials 0.000 description 3
- 239000002184 metal Substances 0.000 description 3
- 230000007935 neutral effect Effects 0.000 description 3
- 239000004033 plastic Substances 0.000 description 3
- 229920003023 plastic Polymers 0.000 description 3
- 238000013441 quality evaluation Methods 0.000 description 3
- 238000007789 sealing Methods 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 238000009833 condensation Methods 0.000 description 2
- 230000005494 condensation Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 230000017105 transposition Effects 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- 241000196324 Embryophyta Species 0.000 description 1
- 241001274961 Rubus repens Species 0.000 description 1
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009529 body temperature measurement Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000000502 dialysis Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000012528 membrane Substances 0.000 description 1
- 230000000116 mitigating effect Effects 0.000 description 1
- 210000004218 nerve net Anatomy 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- SOQBVABWOPYFQZ-UHFFFAOYSA-N oxygen(2-);titanium(4+) Chemical class [O-2].[O-2].[Ti+4] SOQBVABWOPYFQZ-UHFFFAOYSA-N 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 239000000741 silica gel Substances 0.000 description 1
- 229910002027 silica gel Inorganic materials 0.000 description 1
- GWEVSGVZZGPLCZ-UHFFFAOYSA-N titanium dioxide Inorganic materials O=[Ti]=O GWEVSGVZZGPLCZ-UHFFFAOYSA-N 0.000 description 1
- 235000010215 titanium dioxide Nutrition 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/4185—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by the network communication
- G05B19/4186—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by the network communication by protocol, e.g. MAP, TOP
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G08—SIGNALLING
- G08C—TRANSMISSION SYSTEMS FOR MEASURED VALUES, CONTROL OR SIMILAR SIGNALS
- G08C17/00—Arrangements for transmitting signals characterised by the use of a wireless electrical link
- G08C17/02—Arrangements for transmitting signals characterised by the use of a wireless electrical link using a radio link
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W84/00—Network topologies
- H04W84/18—Self-organising networks, e.g. ad-hoc networks or sensor networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Automation & Control Theory (AREA)
- Signal Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention proposes a kind of greenhouse comprehensive test instrument and method, and the comprehensive test instrument includes:Greenhouse data acquisition module, CPU and server;CPU includes:Synchronous data sampling module, d GPS locating module, human-computer interaction module, GPRS remote-transmission modules, USB device interface, DSP core processing module;The output end of greenhouse data acquisition module connects the input of synchronous data sampling module, the output end of synchronous data sampling module connects the input of DSP core processing module, d GPS locating module and GPRS remote-transmission modules connection DSP core processing module, the input of human-computer interaction module connect the output end of DSP core processing module;This method is the non-precision data assessment method based on Interval neural networks model;Synchronous data collection, data processing, position positioning and data transporting function are integrated in one by the system, and the function with remote data acquisition terminal.
Description
Technical field
The invention belongs to technical field of facility agriculture, and in particular to a kind of greenhouse comprehensive test instrument and method.
Background technology
The control of greenhouse belongs to a part for wisdom agricultural, is meeting urban consumption group demand, is mitigating planting industry wind
Danger, make full use of lamp, promote agriculture advanced technology in terms of there is important effect.In order to ensure that greenhouse meets work
The requirement of thing growth, online or offline inspection is carried out to greenhouse, and carries out greenhouse quality evaluation according to expertise,
Guiding agricultural production just has very important significance.
The following shortcoming of instrument generally existing of existing market:
(1) existing greenhouse comprehensive tester, it is impossible to while synchronizing measurement to a variety of environmental datas, more lack
Assessment according to measurement data to greenhouse, lacks directive significance for user;
(2) existing greenhouse comprehensive tester, lacks detection data distant place transmitting function, thus detection data without
Method is preserved for a long time;Simultaneously as comprehensive test instrument is mobility measurement, therefore lack the positioning function in measurement place;
(3) because the data of sensor measurement are inaccurate, if non-precision data completion greenhouse that will be based on these measurements
The quality evaluation of environment, it is necessary to the assessment technology based on non-precision data, and this technology in theory and is actually at present
Lack.
In view of the above-mentioned problems, a kind of new greenhouse integrated data measuring instrument of present invention design, is passed based on a variety of
Sensor completes the synchro measure multifactor to greenhouse;Increase GPS location function, complete positioning and measurement to measuring place
The record of time;RVFL (Random Vector Functional Link) nerve net is connected using interval random vector function
Network model, develops the greenhouse assessment models based on non-precision data, realizes the quality evaluation of greenhouse, and provide assessment
Reliability, for user refer to.The greenhouse comprehensive tester of interval RVFL neutral nets assessment models is currently based on,
It is both domestic and external research and application in there is not yet.
The content of the invention
In view of the shortcomings of the prior art, the present invention proposes a kind of greenhouse comprehensive test instrument and method.
A kind of greenhouse comprehensive test instrument, including:Greenhouse data acquisition unit, CPU and server;
The CPU, including:Synchronous data sampling module, d GPS locating module, human-computer interaction module, GPRS
Remote-transmission module, USB device interface, DSP core processing module;
The output end of the greenhouse collecting unit connects the input of synchronous data sampling module, the data syn-chronization
The output end of acquisition module connects the input of DSP core processing module, and the d GPS locating module is connected by serial communication port
DSP core processing module, the GPRS remote-transmission modules connect DSP core processing module, the USB device by serial communication port
Interface is connected with the input of DSP core processing module, the input connection DSP core processing module of the human-computer interaction module
Output end, the GPRS remote-transmission modules are connected by network with server;
The greenhouse collecting unit, for gathering greenhouse environment parameter in real time;
The synchronous data sampling module, at greenhouse environment parameter real-time Transmission to the DSP core by synchronous acquisition
Manage module;
The d GPS locating module, acquisition time, place for gathering current greenhouse environment parameter, and by this time,
Point is transmitted to DSP core processing module;
The human-computer interaction module, for carrying out information exchange with DSP core processing module, shows the greenhouse gathered in real time
Ambient parameter, selection working condition, the greenhouse environment parameter of selected collection;The working condition includes:Presence and offline
State;
The GPRS remote-transmission modules, for realizing the communication between DSP core processing module and server, will be gathered in real time
Greenhouse environment parameter transmit to server;
The USB device interface, the passage for providing USB storage device access;
The DSP core processing module, for setting up interval RVFL network models, by greenhouse environment parameter interval value and right
The interval value for the greenhouse credit rating answered trains the interval RVFL network models as training sample, it is determined that after training
Interval RVFL network models;Interval RVFL networks after training are used as according to the greenhouse environment parameter under real-time collection current environment
Input, obtain greenhouse credit rating and its confidence level under current environment.
The greenhouse environment parameter, including:Air themperature, air humidity, gas concentration lwevel, intensity of illumination, soil temperature
Degree, soil moisture.
The greenhouse collecting unit, including:Air temperature sensor, soil temperature sensor, air humidity sensing
Device, soil humidity sensor, intensity of illumination sensor, carbon dioxide sensor.
The method that the comprehensive survey of greenhouse is carried out using greenhouse comprehensive test instrument, is comprised the following steps:
Step 1:Working condition, the selected greenhouse environment parameter gathered are selected by human-computer interaction module;
Step 2:Open the collection that greenhouse data acquisition unit carries out greenhouse environment parameter;
Step 3:By synchronous data sampling module by greenhouse environment parameter real-time Transmission to the DSP core of synchronous acquisition
Manage module;
Step 4:The greenhouse environment parameter gathered in real time is shown by human-computer interaction module;
Step 5:Gather acquisition time, the place of current greenhouse data by d GPS locating module, and by this time,
Transmit to DSP core processing module in place;
Step 6:By DSP core processing module set up interval RVFL network models, by greenhouse environment parameter interval value with
The interval value of corresponding greenhouse credit rating trains the interval RVFL network models as training sample, it is determined that after training
Interval RVFL network models;Interval RVFL nets after training are used as according to the greenhouse environment parameter under real-time collection current environment
The input of network, obtains greenhouse credit rating and its confidence level under current environment;
Step 6.1:The P group greenhouse environment parameters of collection are converted into interval value;
Step 6.2:Grade classification is carried out to the greenhouse quality corresponding to monitoring point according to expertise, n are obtained
Greenhouse credit rating;
Step 6.3:The interval value of greenhouse environment parameter is normalized, the greenhouse ginseng after being normalized
Number interval value;
Step 6.4:Greenhouse credit rating is represented with interval, the interval value of greenhouse credit rating is obtained;
Step 6.5:Interval RVFL network models are set up, by the greenhouse environment parameter interval value after the normalization of P groups and correspondingly
Greenhouse credit rating interval value as training sample, train the interval RVFL network models, it is determined that training after area
Between RVFL network models;
Step 6.5.1:Interval RVFL network models are set up, input layer is set according to the number l of greenhouse environment parameter
Number, output layer node number, setting hidden layer node number m are set according to greenhouse credit rating number n;
Step 6.5.2:Input layer is set to the initial point value weights of hidden layer node, the point value threshold of hidden layer node
Value, hidden layer node are to the excitation function for exporting the initial interval right weight of node layer, hidden layer;
Step 6.5.3:It regard the greenhouse environment parameter interval value after normalization as the input area of interval RVFL network models
Between be worth, according to the initial point value weights of the input interval value and input layer of interval RVFL network models to hidden layer node, it is determined that
The interval output valve of hidden layer node;
The input interval value and input layer according to interval RVFL network models is weighed to the initial point value of hidden layer node
Value, determines that the calculation formula of the interval output valve of hidden layer node is as follows:
Wherein, UjFor the interval output valve of j-th of hidden layer node,u jThe lower limit exported for j-th of node of hidden layer,
For the upper limit of j-th of hidden layer node output, i ∈ [1, l] are input layer, and j ∈ [1, m] are hidden layer node, wI, jFor
I input layer to j-th of hidden layer node weights,x iFor the lower limit of i-th of greenhouse environment parameter after normalization,
For the higher limit of i-th of greenhouse environment parameter after normalization, θjFor the threshold value of j-th of hidden layer node;
Step 6.5.4:The lower limit of the interval output valve of hidden layer node is expressed as to the bottoming matrix of hidden layeru P×l, the upper limit of the interval output valve of hidden layer node is expressed as current limiting matrix in the output of hidden layerAnd according to hidden layer
The positive and negative output matrix to hidden layer of node to the initial weight of output node layer is reclassified, after being reclassified
Hidden layer bottoming matrix ulP×lUpper current limiting matrix uu is exported with the hidden layer after reclassifyingP×l;
Step 6.5.5:The lower limit of the interval value of greenhouse credit rating is expressed as training sample output valve lower limit square
Battle arrayd P×n, the upper limit of the interval value of greenhouse credit rating is expressed as current limiting matrix in training sample output valve
Step 6.5.6:According to the hidden layer bottoming matrix ul after reclassifyingP×l, reclassify after hidden layer it is defeated
Go out upper current limiting matrix uuP×l, current limiting matrix under training sample output valved P×n, current limiting matrix in training sample output valveUsing minimum
The hidden layer of square law training interval RVFL networks is to output layer weights, it is determined that the interval RVFL network models after training.
Step 6.6:Greenhouse environment parameter under collection current environment, and be converted to interval value in real time;
Step 6.7:The interval value of the greenhouse environment parameter gathered in real time is normalized, after being normalized
The greenhouse environment parameter interval value gathered in real time;
Step 6.8:It regard the greenhouse environment parameter interval value of the real-time collection after normalization as the interval RVFL after training
The input of network, obtains greenhouse credit rating and its confidence level under current environment;
Step 6.8.1:By the interval after the greenhouse environment parameter interval value input training of the real-time collection after normalization
RVFL networks, obtain the interval output valve of its hidden layer node;
Step 6.8.2:According to the initial power of the interval output valve and hidden layer node of hidden layer node to output node layer
Value, it is determined that the interval output valve of output node layer, i.e., the greenhouse credit rating under current environment;
The interval output valve and hidden layer node according to hidden layer node to output node layer initial weight, it is determined that
The calculation formula for exporting the interval output valve of node layer is as follows:
Wherein,YK is the interval output valve of k-th of output node layer,The upper limit exported for k-th of node of output layer,y k
For the lower limit of k-th of node output of output layer, k ∈ [1, n] are output node layer,It is j-th of hidden layer node to k-th
The upper limit of output layer node weights,β J, kFor the lower limit of j-th of hidden layer node to k-th of output layer node weights;
Step 6.8.3:Compare the size of the interval output valve of output node layer, the interval upper limit of node layer will be exported with
The maximum interval output valve of the average of limit determines that it is evaluated as evaluation result, i.e., current greenhouse credit rating class
As a result confidence level CI;
The calculation formula of the current greenhouse credit rating class is as follows:
Wherein, function max returns to the greenhouse credit rating class after assessing;
The confidence level CI of evaluation result calculation formula is as follows:
Wherein,y classThe lower limit exported for current greenhouse credit rating class correspondence output node layers,To work as
The upper limit of preceding greenhouse credit rating class correspondences output node layer output;
Step 7:When working condition is to work online state, the greenhouse collected is joined by GPRS remote-transmission modules
Number, the greenhouse credit rating under current environment and its confidence level are uploaded onto the server in database, end-of-job;
Step 8:When working condition is to work offline state, by under the greenhouse environment parameter collected, current environment
Greenhouse credit rating and its confidence level are stored in local USB storage device, remote by GPRS when network state is good
Transmission module is reached in server database, end-of-job.
Beneficial effects of the present invention:
The present invention propose be based on a kind of greenhouse comprehensive test instrument and method, the system by synchronous data collection, data processing,
Position is positioned and data transporting function is integrated in one, and is completed a variety of functions, is played assistant's function of user, and teledata
The function of acquisition terminal;The innovation of the appraisal procedure is modeled and assessed based on non-precision data, at present still not no good side
Method.The present invention is proposed to set up the assessment models based on non-precision data using Interval neural networks model, preferably solved non-
The process problem of precise information, enables developed comprehensive test instrument to provide greenhouse assessment result and the reliability assessed,
Belong to both at home and abroad pioneering.When greenhouse comprehensive test instrument is in presence, greenhouse comprehensive test instrument directly supervises greenhouse
Survey data and evaluation result is uploaded on distant server;When greenhouse comprehensive test instrument is in off-line state, greenhouse is comprehensive
Instrument is surveyed greenhouse Monitoring Data information and evaluation result, local USB is stored sequentially according to time, location sequence and stores
In equipment, service is synchronized to after the networking of greenhouse comprehensive test instrument, then greenhouse Monitoring Data information and evaluation result
In device.
Brief description of the drawings
Fig. 1 is the structured flowchart of specific embodiment of the invention medium temperature chamber environment comprehensive test instrument;
Fig. 2 is the circuit theory diagrams of specific embodiment of the invention medium temperature chamber environment comprehensive test instrument;
Fig. 3 is the flow for the method that specific embodiment of the invention medium temperature chamber environment comprehensive test instrument carries out the comprehensive survey of greenhouse
Figure;
Fig. 4 obtains the greenhouse under current environment for foundation interval RVFL network models in the specific embodiment of the invention
The flow chart of credit rating and its confidence level;
Fig. 5 be the specific embodiment of the invention in train after interval RVFL network models structural representation.
Embodiment
The specific embodiment of the invention is described in detail below in conjunction with the accompanying drawings.
A kind of greenhouse comprehensive test instrument, as shown in figure 1, including:Greenhouse data acquisition unit, CPU and
Server.
The CPU, including:Synchronous data sampling module, d GPS locating module, human-computer interaction module, GPRS
Remote-transmission module, USB device interface, DSP core processing module.
As shown in Fig. 2 the output end of the greenhouse collecting unit connects the input of synchronous data sampling module, institute
The output end for stating synchronous data sampling module connects the input of DSP core processing module, and the d GPS locating module passes through serial
Port connects DSP core processing module, and the GPRS remote-transmission modules connect DSP core processing module by serial communication port,
The USB device interface is connected with the input of DSP core processing module, the input connection DSP of the human-computer interaction module
The output end of core processing module, the GPRS remote-transmission modules are connected by network with server.
The greenhouse collecting unit, for gathering greenhouse environment parameter in real time.
The greenhouse collecting unit, including:Air temperature sensor, soil temperature sensor, air humidity sensing
Device, soil humidity sensor, intensity of illumination sensor, carbon dioxide sensor.
In present embodiment, air temperature sensor and soil temperature sensor are thermocouple temperature sensor, are aided with
Cold junction compensation circuit connects the input of synchronous data sampling module, and output analog voltage signal sends DSP core processing mould to
Block.
In present embodiment, thermocouple temperature sensor probe is with long 30mm, the cylindrical metal that basal diameter is 3m
The data wire of shell and 3m length.When measuring air themperature, directly air temperature sensor is placed among air;Measure soil
During temperature, soil temperature sensor probe is embedded at the following plant root of upper soll layer.The electric thermo-couple temperature sensing used
Device is compressed spring type temperature-sensing element, with resistance to shock is good, temperature measurement accuracy is high, high mechanical strength, good pressure-resistant performance, dependable performance it is steady
The advantages of determining.
In present embodiment, air humidity sensor model SHT10.SHT10 is believed numeral by serial type IIC interfaces
Number send DSP core processing module to.SHT10 is fabricated to probe shape, SHT10 unique package surface makes it in condensation environment
Under use.
In present embodiment, soil humidity sensor is CSF11 soil moisture sensors.CSF11 soil moisture sensors
Output analog voltage signal sends DSP core processing module to by synchronous data sampling module.CSF11 soil moisture sensors
Encapsulated for cylinder, long 109mm, diameter 40mm, bottom is with 4 long 60mm, diameter 3mm metal probes.Soil moisture is passed
Sensor is inserted perpendicularly into tested soil, it is ensured that metal probe is entirely insertable the soil water in i.e. measurable amount greenhouse among soil
Point.
In present embodiment, model BH1750FVI, BH1750FVI the intensity of illumination sensor of intensity of illumination sensor
Output digit signals are sent to DSP core processing module by iic bus.BH1750FVI intensities of illumination sensor is half ball sealing
Dress, size is internal diameter 22mm, external diameter 26mm, external profile diameter 28.5mm, high 18mm.Sensor is sealed in greenhouse with insulation silica gel
Environment comprehensive test instrument surface, makes hemisphere encapsulate the intensity of illumination in exposed greenhouse i.e. measurable in greenhouse.
In present embodiment, selected by carbon dioxide sensor for S-100 carbon dioxide sensors.S-100 titanium dioxides
Carbon sensor output digit signals are sent to DSP core processing module by iic bus.Outside S-100 carbon dioxide sensors
Shape size is 32mm (length) × 12mm (width) × 38mm (height).One layer of dialysis membrane patch of attachment outside S-100 carbon dioxide sensors
The gas concentration lwevel spread in the i.e. measurable greenhouse of greenhouse comprehensive test instrument case surface.
The greenhouse environment parameter, including:Air themperature, air humidity, gas concentration lwevel, intensity of illumination, soil temperature
Degree, soil moisture.
The synchronous data sampling module, at greenhouse environment parameter real-time Transmission to the DSP core by synchronous acquisition
Manage module.
The d GPS locating module, acquisition time, place for gathering current greenhouse environment parameter, and by this time,
Point is transmitted to DSP core processing module.
In present embodiment, d GPS locating module, its major function is to provide time, place to greenhouse comprehensive test instrument
With reference to d GPS locating module carries out information exchange by serial communication port and DSP core processing module.When in presence,
Time, location sequence are obtained using d GPS locating module, record gathers the acquisition time of current greenhouse data, place, and will
This time, place regard the information as the temperature for judging this upload as a part for this greenhouse Monitoring Data information
Whether room environmental Monitoring Data and evaluation result effectively refer to;During in off-line state, the time of d GPS locating module acquisition,
Location information is stored sequentially in local USB storage device.
The human-computer interaction module, for carrying out information exchange with DSP core processing module, shows the greenhouse gathered in real time
Ambient parameter, selection working condition, the greenhouse environment parameter of selected collection;The working condition includes:Presence and offline
State.
In present embodiment, human-computer interaction module is combined the interface as man-machine interaction using LCD color screens with button,
The effect of LCD color screens is to carry out information exchange with DSP core processing module, shows the greenhouse environment parameter gathered in real time, selection
Working condition, the greenhouse environment parameter of selected collection.LCD color screens, button pass through data signal line and DSP core processing module phase
Repeatedly kick into the interaction of row information.Button more accords with the application of actual greenhouse compared to touch-screen, and touch-screen is exposed to air
In could normally use, condensation environment in, dewdrop can make touch-screen the situation of false triggering occur, and button is enclosed in function
Under panel, it is to avoid maloperation improves reliability.
The GPRS remote-transmission modules, for realizing the communication between DSP core processing module and server, will be gathered in real time
Greenhouse environment parameter transmit to server.
In present embodiment, GPRS remote-transmission modules use GPRS DTU wireless serial digital transmission modules, by serially leading to
Casually it is connected with DSP core processing module.There is provided standard RS232/485 data-interfaces for GPRS wireless modules built in this module.It is real
The reliable and effective transmission of existing data between greenhouse comprehensive test instrument and server.
The USB device interface, the passage for providing USB storage device access.
In present embodiment, need voluntarily to record the offline shape of greenhouse Monitoring Data when greenhouse comprehensive test instrument is in
During state, greenhouse comprehensive test instrument adds d GPS locating module generation time, location sequence greenhouse Monitoring Data information sequence
Row, and be stored in sequentially in time in local USB storage device, further analysis conveniently subsequently to data, processing work
Make.
The DSP core processing module, for setting up interval RVFL network models, by greenhouse environment parameter interval value and right
The interval value for the greenhouse credit rating answered trains the interval RVFL network models as training sample, it is determined that after training
Interval RVFL network models;Interval RVFL networks after training are used as according to the greenhouse environment parameter under real-time collection current environment
Input, obtain greenhouse credit rating and its confidence level under current environment.
In present embodiment, the DSP digital processing chips that DSP core processing module is used is TMS320F28335.DSP core
The main task of heart processing module coordinates cooperating between each module, and synchronous data sampling and interpersonal interactive system are gathered
Greenhouse data are sent to DSP by analog quantity interface and digital quantity interface, are given using interval RVFL neutral nets assessment models algorithm
Go out evaluation result, the greenhouse data and evaluation result of record are sent to LCD color screens, convenient use personnel understand in time to be worked as
Preceding greenhouse.
In present embodiment, the power supply plan that power unit is arranged in pairs or groups using outside 24V dc sources with 24V batteries is protected
Demonstrate,proving greenhouse comprehensive test instrument can long-term stable operation.Only a kind of in running order in the two, acquiescence uses 24V dc sources
Power supply.When being powered in outside 24V dc sources, the charging circuit of greenhouse comprehensive test instrument is according to 24V battery current states
Choose whether to charge a battery by charging circuit.Convert what output voltage 12V, 12V-5V mu balanced circuit were selected by DC-DC
Chip is LM1084 low pressure difference linear voltage regulators;The chip that 12V-3.3V mu balanced circuits are selected is steady for AMS1117 low pressure difference linearities
Depressor;The chip that 12V-2.5V mu balanced circuits are selected is LM1084 low pressure difference linear voltage regulators.
In present embodiment, shell is cuboid plastics box, and partially perforation, with handle is filled inside plastic casing
Enter after greenhouse comprehensive test instrument each several part circuit module, carry out pad pasting, sealing etc. science and engineering and make.Partially perforation is to man-machine friendship
Mutual interface, sensor lean out offer space, are carried with the handle personnel that are convenient to operation, pad pasting, sealing ensure that greenhouse is comprehensive
Survey and do not enter steam or directly water inlet inside instrument, then ensure greenhouse comprehensive test instrument each several part circuit module not because making moist, entering
Water and can not normal work.The slot of mercurial thermometer is carried outside plastic casing, is easy to operating personnel to use mercurial thermometer
Revise temperature.
The method that the comprehensive survey of greenhouse is carried out using greenhouse comprehensive test instrument, as shown in figure 3, including step once:
Step 1:Working condition, the selected greenhouse environment parameter gathered are selected by human-computer interaction module.
Step 2:Open the collection that greenhouse data acquisition unit carries out greenhouse environment parameter.
Step 3:By synchronous data sampling module by greenhouse environment parameter real-time Transmission to the DSP core of synchronous acquisition
Manage module.
Step 4:The greenhouse environment parameter gathered in real time is shown by human-computer interaction module.
In present embodiment, human-computer interaction module display mode is divided to two kinds.Mode one, the form of numeral are shown in screen
On, often gather one group of greenhouse data and refresh immediately on screen;Mode two, curve in different colors sense each
The greenhouse data display of device collection is in rectangular coordinate system, and the greenhouse data gathered according to each sensor are described
Its corresponding kind of greenhouse environment parameter variation tendency interior for a period of time.
Step 5:Gather acquisition time, the place of current greenhouse data by d GPS locating module, and by this time,
Transmit to DSP core processing module in place.
Step 6:By DSP core processing module set up interval RVFL network models, by greenhouse environment parameter interval value with
The interval value of corresponding greenhouse credit rating trains the interval RVFL network models as training sample, it is determined that after training
Interval RVFL network models;Interval RVFL nets after training are used as according to the greenhouse environment parameter under real-time collection current environment
The input of network, obtains greenhouse credit rating and its confidence level under current environment, as shown in Figure 4.
Step 6.1:The P group greenhouse environment parameters of collection are converted into interval value.
In present embodiment, greenhouse environment parameter, including:Air themperature, air humidity, gas concentration lwevel, illumination are strong
Degree, the soil moisture, soil moisture.
By the air themperature gathered, air humidity, the soil moisture, soil moisture, gas concentration lwevel and intensity of illumination
Value switchs to interval value, usesI-th of detection limit is represented, wherein,z iThe lower limit of i-th of interval parameter value is represented,Table
Show the upper limit of i-th of interval parameter value, i ∈ [1, l], l is the total number of detection limit.
Step 6.2:Grade classification is carried out to the greenhouse quality corresponding to monitoring point according to expertise, n are obtained
Greenhouse credit rating.
In present embodiment, while 4 greenhouse credit ratings are believed with vector again mark and as every group of collection
Number desired output, environmental quality grade be I when, being expressed as [1 00 0] corresponding to it;Environmental Quality Evalution grade is
During II, being expressed as [0 10 0] corresponding to it;When Environmental Quality Evalution grade is III, [0 01 are expressed as corresponding to it
0];When Environmental Quality Evalution grade is IV, being expressed as [0 00 1] corresponding to it.
In present embodiment, it is four classes that greenhouse quality is carried out into grade classification:I (excellent), II (good), III (in), IV
(poor), ranking foundation is expertise.
Step 6.3:The interval value of greenhouse environment parameter is normalized, the greenhouse ginseng after being normalized
Number interval value.
In present embodiment, the interval value of greenhouse environment parameter is normalized, method for normalizing such as formula (1) institute
Show:
Wherein, XiFor i-th of interval parameter value after normalized,x iFor i-th of interval parameter value after normalized
Lower limit,For the upper limit of i-th of interval parameter value after normalized, zminFor the minimum value of all interval parameter value lower limits,
zmaxFor the maximum of all interval parameter value upper limits, i is i-th of interval parameter value, and min represents minimum, and max represents maximum.
Step 6.4:Greenhouse credit rating is represented with interval, the interval value of greenhouse credit rating is obtained.
In present embodiment, greenhouse credit rating is represented with interval, the interval of greenhouse credit rating is obtained
Value, such as [1 00 0] are expressed as [(1,1) (0,0) (0,0) (0,0)].
Step 6.5:Interval RVFL network models are set up, by the greenhouse environment parameter interval value after the normalization of P groups and correspondingly
Greenhouse credit rating interval value as training sample, train the interval RVFL network models, it is determined that training after area
Between RVFL network models, as shown in Figure 5.
Step 6.5.1:Interval RVFL network models are set up, input layer is set according to the number l of greenhouse environment parameter
Number, output layer node number, setting hidden layer node number m are set according to greenhouse credit rating number n.
In present embodiment, because greenhouse environment parameter has 6, greenhouse credit rating is 4 grades, rule of thumb will be hidden
Number containing node layer is defined as 25, therefore determines the structure of neutral net for 6-25-4.
Step 6.5.2:Input layer is set to the initial point value weights of hidden layer node, the point value threshold of hidden layer node
Value, hidden layer node are to the excitation function for exporting the initial interval right weight of node layer, hidden layer.
In present embodiment, setting input layer arrive hidden layer node initial weight and threshold value, its for [- 1,1] it
Between random point value;Hidden layer node is set to the initial weight of output node layer, it is the random interval value between [- 1,1];
The excitation function of hidden layer is set, Sigmoid functions are set to.
Step 6.5.3:It regard the greenhouse environment parameter interval value after normalization as the input area of interval RVFL network models
Between be worth, according to the initial point value weights of the input interval value and input layer of interval RVFL network models to hidden layer node, it is determined that
The interval output valve of hidden layer node.
The input interval value and input layer according to interval RVFL network models is weighed to the initial point value of hidden layer node
Value, determines that the calculation formula such as formula (2) of the interval output valve of hidden layer node is shown:
Wherein, UjFor the interval output valve of j-th of hidden layer node,u jThe lower limit exported for j-th of node of hidden layer,
For the upper limit of j-th of hidden layer node output, i ∈ [1, l] are input layer, and j ∈ [1, m] are hidden layer node, wI, jFor
I input layer to j-th of hidden layer node weights,x iFor the lower limit of i-th of greenhouse environment parameter after normalization,
For the higher limit of i-th of greenhouse environment parameter after normalization, θjFor the threshold value of j-th of hidden layer node.
Step 6.5.4:The lower limit of the interval output valve of hidden layer node is expressed as to the bottoming matrix of hidden layeru P×l, the upper limit of the interval output valve of hidden layer node is expressed as current limiting matrix in the output of hidden layerAnd according to hidden layer
The positive and negative output matrix to hidden layer of node to the initial weight of output node layer is reclassified, after being reclassified
Hidden layer bottoming matrix ulP×lUpper current limiting matrix uu is exported with the hidden layer after reclassifyingP×l。
In present embodiment, the lower limit of the interval output valve of hidden layer node is expressed as to the bottoming matrix of hidden layeru P×l, as shown in formula (3):
The upper limit of the interval output valve of hidden layer node is expressed as current limiting matrix in the output of hidden layerSuch as formula (4) institute
Show:
Carried out again according to the positive and negative output matrix to hidden layer of hidden layer node to the initial weight of output node layer
Sort out, the hidden layer bottoming matrix ul after being reclassifiedP×lAs shown in formula (5):
The upper current limiting matrix uu of hidden layer output after reclassifyingP×lAs shown in formula (6):
Wherein, ulP, jThe lower bound exported for the hidden layer after reclassifying, as shown in formula (7):
uuP, jThe upper bound exported for the hidden layer after reclassifying, as shown in formula (8):
Wherein, β_jHidden layer is represented to the lower bound of output layer weights,Hidden layer is represented to the upper bound of output layer weights,u P, jThe lower limit of hidden layer output valve is represented,Represent the upper limit of hidden layer output valve.
Step 6.5.5:The lower limit of the interval value of greenhouse credit rating is expressed as training sample output valve lower limit square
Battle arrayd P×n, the upper limit of the interval value of greenhouse credit rating is expressed as current limiting matrix in training sample output valve
In present embodiment, the lower limit of the interval value of greenhouse credit rating is expressed as training sample output valve lower limit
Matrixd P×n, as shown in formula (9):
The upper limit of the interval value of greenhouse credit rating is expressed as current limiting matrix in training sample output valveSuch as formula
(10) shown in:
Step 6.5.6:According to the hidden layer bottoming matrix ul after reclassifyingP×l, reclassify after hidden layer it is defeated
Go out upper current limiting matrix uuP×l, current limiting matrix under training sample output valved P×n, current limiting matrix in training sample output valveUsing minimum
The hidden layer of square law training interval RVFL networks is to output layer weights, it is determined that the interval RVFL network models after training.
In present embodiment, using the hidden layer of least square in training interval RVFL networks to output layer weights such as formula
(11) and shown in formula (12):
β=(ulT·ul)-1·(ulT·d) (11)
Wherein, ulTFor ul transposition, uuTFor uu transposition, ()-1Represent finding the inverse matrix.
Step 6.6:Greenhouse environment parameter under collection current environment, and be converted to interval value in real time.
Step 6.7:The interval value of the greenhouse environment parameter gathered in real time is normalized, after being normalized
The greenhouse environment parameter interval value gathered in real time.
Step 6.8:It regard the greenhouse environment parameter interval value of the real-time collection after normalization as the interval RVFL after training
The input of network, obtains greenhouse credit rating and its confidence level under current environment.
Step 6.8.1:By the interval after the greenhouse environment parameter interval value input training of the real-time collection after normalization
RVFL networks, obtain the interval output valve of its hidden layer node.
Step 6.8.2:According to the initial power of the interval output valve and hidden layer node of hidden layer node to output node layer
Value, it is determined that the interval output valve of output node layer, i.e., the greenhouse credit rating under current environment.
The interval output valve and hidden layer node according to hidden layer node to output node layer initial weight, it is determined that
Shown in the calculation formula such as formula (13) for the interval output valve for exporting node layer:
Wherein, YkFor k-th output node layer interval output valve,The upper limit exported for k-th of node of output layer,y k
For the lower limit of k-th of node output of output layer, k ∈ [1, n] are output node layer,It is j-th of hidden layer node to k-th
The upper limit of output layer node weights,β J, kFor the lower limit of j-th of hidden layer node to k-th of output layer node weights.
Step 6.8.3:Compare the size of the interval output valve of output node layer, the interval upper limit of node layer will be exported with
The maximum interval output valve of the average of limit is as evaluation result, i.e., shown in current greenhouse credit rating class such as formulas (14),
And determine shown in the confidence level CI such as formulas (15) of its evaluation result:
Wherein, function max returns to the greenhouse credit rating class after assessing.
Wherein,y classThe lower limit exported for current greenhouse credit rating class correspondence output node layers,To work as
The upper limit of preceding greenhouse credit rating class correspondences output node layer output.
Step 7:When working condition is to work online state, the greenhouse collected is joined by GPRS remote-transmission modules
Number, the greenhouse credit rating under current environment and its confidence level are uploaded onto the server in database, end-of-job;
Step 8:When working condition is to work offline state, by under the greenhouse environment parameter collected, current environment
Greenhouse credit rating and its confidence level are stored in local USB storage device, remote by GPRS when network state is good
Transmission module is reached in server database, end-of-job.
Claims (7)
1. a kind of greenhouse comprehensive test instrument, it is characterised in that including:Greenhouse data acquisition unit, CPU and
Server;
The CPU, including:Synchronous data sampling module, d GPS locating module, human-computer interaction module, GPRS teletransmissions
Module, USB device interface, DSP core processing module;
The output end of the greenhouse data acquisition unit connects the input of synchronous data sampling module, the data syn-chronization
The output end of acquisition module connects the input of DSP core processing module, and the d GPS locating module is connected by serial communication port
DSP core processing module, the GPRS remote-transmission modules connect DSP core processing module, the USB device by serial communication port
Interface is connected with the input of DSP core processing module, the input connection DSP core processing module of the human-computer interaction module
Output end, the GPRS remote-transmission modules are connected by network with server;
The greenhouse data acquisition unit, for gathering greenhouse environment parameter in real time;
The synchronous data sampling module, for the greenhouse environment parameter real-time Transmission of synchronous acquisition to DSP core to be handled into mould
Block;
The d GPS locating module, acquisition time, place for gathering current greenhouse environment parameter, and this time, place are passed
Transport to DSP core processing module;
The human-computer interaction module, for carrying out information exchange with DSP core processing module, shows the greenhouse gathered in real time
Parameter, selection working condition, the greenhouse environment parameter of selected collection;The working condition includes:Presence and off-line state;
The GPRS remote-transmission modules, for realizing the communication between DSP core processing module and server, by the temperature gathered in real time
Room environmental parameter is transmitted to server;
The USB device interface, the passage for providing USB storage device access;
The DSP core processing module, for setting up interval RVFL network models, by greenhouse environment parameter interval value and corresponding
The interval value of greenhouse credit rating trains the interval RVFL network models as training sample, it is determined that the interval after training
RVFL network models;The defeated of interval RVFL networks after training is used as according to the greenhouse environment parameter under real-time collection current environment
Enter, obtain greenhouse credit rating and its confidence level under current environment.
2. greenhouse comprehensive test instrument according to claim 1, it is characterised in that the greenhouse environment parameter, including:Air
Temperature, air humidity, gas concentration lwevel, intensity of illumination, the soil moisture, soil moisture.
3. greenhouse comprehensive test instrument according to claim 1, it is characterised in that the greenhouse data acquisition unit,
Including:Air temperature sensor, soil temperature sensor, air humidity sensor, soil humidity sensor, intensity of illumination sensing
Device, carbon dioxide sensor.
4. the method for the comprehensive survey of greenhouse is carried out using the greenhouse comprehensive test instrument described in claim 1, it is characterised in that including
Following steps:
Step 1:Working condition, the selected greenhouse environment parameter gathered are selected by human-computer interaction module;
Step 2:Open the collection that greenhouse data acquisition unit carries out greenhouse environment parameter;
Step 3:The greenhouse environment parameter real-time Transmission of synchronous acquisition to DSP core is handled by mould by synchronous data sampling module
Block;
Step 4:The greenhouse environment parameter gathered in real time is shown by human-computer interaction module;
Step 5:Gather acquisition time, the place of current greenhouse data by d GPS locating module, and by this time, place
Transmit to DSP core processing module;
Step 6:Interval RVFL network models are set up by DSP core processing module, by greenhouse environment parameter interval value and correspondingly
Greenhouse credit rating interval value as training sample, train the interval RVFL network models, it is determined that training after area
Between RVFL network models;Interval RVFL networks after training are used as according to the greenhouse environment parameter under real-time collection current environment
Input, obtains greenhouse credit rating and its confidence level under current environment;
Step 7:When working condition for work online state when, by GPRS remote-transmission modules by the greenhouse environment parameter collected,
Greenhouse credit rating and its confidence level under current environment are uploaded onto the server in database, end-of-job;
Step 8:When working condition is to work offline state, by the greenhouse under the greenhouse environment parameter collected, current environment
Environmental quality grade and its confidence level are stored in local USB storage device, pass through GPRS teletransmission moulds when network state is good
Block is reached in server database, end-of-job.
5. the method according to claim 4, it is characterised in that the step 6 comprises the following steps:
Step 6.1:The P group greenhouse environment parameters of collection are converted into interval value;
Step 6.2:Grade classification is carried out to the greenhouse quality corresponding to monitoring point according to expertise, n greenhouse is obtained
Environmental quality grade;
Step 6.3:The interval value of greenhouse environment parameter is normalized, the greenhouse environment parameter area after being normalized
Between be worth;
Step 6.4:Greenhouse credit rating is represented with interval, the interval value of greenhouse credit rating is obtained;
Step 6.5:Interval RVFL network models are set up, greenhouse environment parameter interval value and corresponding temperature after P groups are normalized
The interval value of room environmental credit rating trains the interval RVFL network models as training sample, it is determined that the interval after training
RVFL network models;
Step 6.6:Greenhouse environment parameter under collection current environment, and be converted to interval value in real time;
Step 6.7:The interval value of the greenhouse environment parameter gathered in real time is normalized, it is real-time after being normalized
The greenhouse environment parameter interval value of collection;
Step 6.8:It regard the greenhouse environment parameter interval value of the real-time collection after normalization as the interval RVFL networks after training
Input, obtain greenhouse credit rating and its confidence level under current environment.
6. the method according to claim 5, it is characterised in that the step 6.5 comprises the following steps:
Step 6.5.1:Interval RVFL network models are set up, input layer number is set according to the number l of greenhouse environment parameter,
Output layer node number, setting hidden layer node number m are set according to greenhouse credit rating number n;
Step 6.5.2:Set input layer to the initial point value weights of hidden layer node, the point value threshold value of hidden layer node,
Hidden layer node is to the output initial interval right weight of node layer, the excitation function of hidden layer;
Step 6.5.3:Using the greenhouse environment parameter interval value after normalization as interval RVFL network models input interval value,
According to the initial point value weights of the input interval value and input layer of interval RVFL network models to hidden layer node, hidden layer is determined
The interval output valve of node;
The input interval value and input layer according to interval RVFL network models to hidden layer node initial point value weights, really
The calculation formula for determining the interval output valve of hidden layer node is as follows:
Wherein, UjFor the interval output valve of j-th of hidden layer node,u jThe lower limit exported for j-th of node of hidden layer,For jth
The upper limit of individual hidden layer node output, i ∈ [1, l] are input layer, and j ∈ [1, m] are hidden layer node, wI, jIt is defeated for i-th
Enter node layer to the weights of j-th of hidden layer node,x iFor the lower limit of i-th of greenhouse environment parameter after normalization,For normalizing
The higher limit of i-th of greenhouse environment parameter, θ after changejFor the threshold value of j-th of hidden layer node;
Step 6.5.4:The lower limit of the interval output valve of hidden layer node is expressed as to the bottoming matrix of hidden layeru P×l, will
The upper limit of the interval output valve of hidden layer node is expressed as current limiting matrix in the output of hidden layerAnd arrived according to hidden layer node
The positive and negative output matrix to hidden layer for exporting the initial weight of node layer is reclassified, implicit after being reclassified
Layer bottoming matrix ulP×lUpper current limiting matrix uu is exported with the hidden layer after reclassifyingP×l;
Step 6.5.5:The lower limit of the interval value of greenhouse credit rating is expressed as current limiting matrix under training sample output valved P×n, the upper limit of the interval value of greenhouse credit rating is expressed as current limiting matrix in training sample output valve
Step 6.5.6:According to the hidden layer bottoming matrix ul after reclassifyingP×l, reclassify after hidden layer output on
Current limiting matrix uuP×l, current limiting matrix under training sample output valved P×n, current limiting matrix in training sample output valveUsing least square
The hidden layer of method training interval RVFL networks is to output layer weights, it is determined that the interval RVFL network models after training.
7. the method according to claim 5, it is characterised in that the step 6.8 comprises the following steps:
Step 6.8.1:By the interval RVFL nets after the greenhouse environment parameter interval value input training of the real-time collection after normalization
Network, obtains the interval output valve of its hidden layer node;
Step 6.8.2:According to the initial point value power of the interval output valve and hidden layer node of hidden layer node to output node layer
Value, it is determined that the interval output valve of output node layer, i.e., the greenhouse credit rating under current environment;
The interval output valve and hidden layer node according to hidden layer node to output node layer initial point value weights, it is determined that
The calculation formula for exporting the interval output valve of node layer is as follows:
Wherein, YkFor k-th output node layer interval output valve,The upper limit exported for k-th of node of output layer,y kTo be defeated
Go out the lower limit of k-th of node output of layer, k ∈ [1, n] are output node layer,For j-th of hidden layer node to k-th of output
The upper limit of node layer weights,β J, kFor the lower limit of j-th of hidden layer node to k-th of output layer node weights;
Step 6.8.3:Compare the size of the interval output valve of output node layer, the interval upper limit and lower limit of node layer will be exported
The maximum interval output valve of average determines its evaluation result as evaluation result, i.e., current greenhouse credit rating class
Confidence level CI;
The calculation formula of the current greenhouse credit rating class is as follows:
Wherein, function max returns to the greenhouse credit rating class after assessing;
The confidence level CI of evaluation result calculation formula is as follows:
Wherein,y classThe lower limit exported for current greenhouse credit rating class correspondence output node layers,For current temperature
The upper limit of room environmental credit rating class correspondence output node layer outputs.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710425492.2A CN106990769A (en) | 2017-06-08 | 2017-06-08 | A kind of greenhouse comprehensive test instrument and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710425492.2A CN106990769A (en) | 2017-06-08 | 2017-06-08 | A kind of greenhouse comprehensive test instrument and method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106990769A true CN106990769A (en) | 2017-07-28 |
Family
ID=59421896
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710425492.2A Pending CN106990769A (en) | 2017-06-08 | 2017-06-08 | A kind of greenhouse comprehensive test instrument and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106990769A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108133085A (en) * | 2017-12-08 | 2018-06-08 | 北方工业大学 | Method and system for predicting equipment temperature in electronic equipment cabin |
CN109597687A (en) * | 2018-10-31 | 2019-04-09 | 东软集团股份有限公司 | Data synchronous resource allocation methods, device, storage medium and electronic equipment |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6490573B1 (en) * | 2000-04-11 | 2002-12-03 | Philip Chidi Njemanze | Neural network for modeling ecological and biological systems |
CN102903219A (en) * | 2012-10-23 | 2013-01-30 | 中国矿业大学 | Remote wireless real-time greenhouse environment monitoring system |
CN104729965A (en) * | 2015-01-28 | 2015-06-24 | 东北大学 | PM2.5 concentration detection method based on interzone radial basis function nerve network |
CN104869126A (en) * | 2015-06-19 | 2015-08-26 | 中国人民解放军61599部队计算所 | Network intrusion anomaly detection method |
CN105352907A (en) * | 2015-11-27 | 2016-02-24 | 南京信息工程大学 | Infrared gas sensor based on radial basis network temperature compensation and detection method |
CN106780468A (en) * | 2016-12-22 | 2017-05-31 | 中国计量大学 | View-based access control model perceives the conspicuousness detection method of positive feedback |
-
2017
- 2017-06-08 CN CN201710425492.2A patent/CN106990769A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6490573B1 (en) * | 2000-04-11 | 2002-12-03 | Philip Chidi Njemanze | Neural network for modeling ecological and biological systems |
CN102903219A (en) * | 2012-10-23 | 2013-01-30 | 中国矿业大学 | Remote wireless real-time greenhouse environment monitoring system |
CN104729965A (en) * | 2015-01-28 | 2015-06-24 | 东北大学 | PM2.5 concentration detection method based on interzone radial basis function nerve network |
CN104869126A (en) * | 2015-06-19 | 2015-08-26 | 中国人民解放军61599部队计算所 | Network intrusion anomaly detection method |
CN105352907A (en) * | 2015-11-27 | 2016-02-24 | 南京信息工程大学 | Infrared gas sensor based on radial basis network temperature compensation and detection method |
CN106780468A (en) * | 2016-12-22 | 2017-05-31 | 中国计量大学 | View-based access control model perceives the conspicuousness detection method of positive feedback |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108133085A (en) * | 2017-12-08 | 2018-06-08 | 北方工业大学 | Method and system for predicting equipment temperature in electronic equipment cabin |
CN108133085B (en) * | 2017-12-08 | 2021-12-07 | 北方工业大学 | Method and system for predicting equipment temperature in electronic equipment cabin |
CN109597687A (en) * | 2018-10-31 | 2019-04-09 | 东软集团股份有限公司 | Data synchronous resource allocation methods, device, storage medium and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104123444B (en) | Real-time irrigation forecast system and method based on region Soil Moisture Monitoring and remotely-sensed data | |
CN101965807B (en) | System for individually recording production performance of layers and automatically monitoring environment of layer house | |
CN202013348U (en) | Wireless network-based content measuring system for CO, CO2 and SO2 in air | |
CN206592248U (en) | The Portable wind-driven generator group field test of off-network type and analysis system | |
CN109387452A (en) | Forest land surface dead combustible fuel hydrated comples ion system based on artificial neural network | |
CN110046834A (en) | Workplace pernicious gas health risk quantitative evaluating system and its appraisal procedure | |
CN201860640U (en) | System for individually recording performance of layer chicken and automatically monitoring environment of henhouse | |
CN202676649U (en) | Capacitive sensor for synchronous real-time self-compensation measurement of soil moisture content | |
CN106990769A (en) | A kind of greenhouse comprehensive test instrument and method | |
CN206833234U (en) | A kind of greenhouse comprehensive test instrument | |
CN107389127A (en) | Ensilage nondestructive detection system and method based on ZigBee | |
CN107015590A (en) | A kind of hen house cultivation surroundings monitoring system and method based on Arduino | |
CN108006894A (en) | A kind of Portable type central air-conditioning energy efficiency monitoring method for early warning | |
CN104089656A (en) | Storage yard coal spontaneous combustion detection method and device | |
CN107714000B (en) | Milk cow health condition detection method and device | |
CN105957325A (en) | Monitoring integrated device of measurement laboratory | |
CN107654340B (en) | The digitlization vertical axis wind power generation test device communicated using more Wi-Fis | |
CN105185076B (en) | Auxiliary programming equipment is afforested in Portable city | |
Zhang et al. | Research of temperature and humidity monitoring system based on WSN and fuzzy control | |
CN106871962A (en) | A kind of integrated form is detecting device integrated and detection method | |
CN206223784U (en) | A kind of soil multi-point multi-layer humiture on-Line Monitor Device based on cloud platform | |
CN107654341B (en) | The vertical axis wind power generation monitoring device of survey is sentenced based on power observation and data exception | |
CN206097446U (en) | Plant nutrient solution detecting system based on zigBee wireless technology | |
CN107248269A (en) | Soil moisture content data uploading device and method | |
CN204066379U (en) | Power carrier collector |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170728 |
|
WD01 | Invention patent application deemed withdrawn after publication |