CN107991888A - Agricultural automation embedded integration system and its method of work based on machine learning - Google Patents

Agricultural automation embedded integration system and its method of work based on machine learning Download PDF

Info

Publication number
CN107991888A
CN107991888A CN201810003347.XA CN201810003347A CN107991888A CN 107991888 A CN107991888 A CN 107991888A CN 201810003347 A CN201810003347 A CN 201810003347A CN 107991888 A CN107991888 A CN 107991888A
Authority
CN
China
Prior art keywords
msub
mrow
node
layer
data
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
Application number
CN201810003347.XA
Other languages
Chinese (zh)
Inventor
刘宸
刘一宸
朱卫恩
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.)
CHANGZHOU LANXIANG ELECTRICAL APPLIANCE Co Ltd
Original Assignee
CHANGZHOU LANXIANG ELECTRICAL APPLIANCE Co Ltd
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 CHANGZHOU LANXIANG ELECTRICAL APPLIANCE Co Ltd filed Critical CHANGZHOU LANXIANG ELECTRICAL APPLIANCE Co Ltd
Priority to CN201810003347.XA priority Critical patent/CN107991888A/en
Publication of CN107991888A publication Critical patent/CN107991888A/en
Pending legal-status Critical Current

Links

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
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The present invention relates to a kind of agricultural automation embedded integration system and its method of work based on machine learning, this agricultural automation embedded integration system includes：Acquisition system and decision system, wherein acquisition system are by the crop growth environment data sending of collection to decision system；The decision system obtains the decision instruction for being adapted to crop growth by machine learning algorithm；The present invention fits the correspondence between crop growth environment data and planting decision-making using machine learning algorithm, and automation plantation is completed by the embedded system of complete set, solves the limitation of traditional agriculture, there is provided a kind of agricultural automation embedded integration system that is efficient, possessing the ability of being precisely controlled.

Description

Agricultural automation embedded integration system and its method of work based on machine learning
Technical field
The present invention relates to a kind of agricultural automation embedded integration system based in machine learning algorithm and its work side Method.
Background technology
China human mortality accounts for the 22% of world population, and the 7% of cultivated area Zhi Zhan worlds cultivated area, with economical at full speed Development, the continuous improvement of living standards of the people, shortage of resources, environmental degradation and the contradiction of population increase are but more and more prominent.By In operations such as the watering of traditional agriculture middle peasant China Association for Promoting Democracy row, fertilising, laxatives entirely by rule of thumb, by feel, resource utilization can be caused low, it is raw It is low to produce benefit, a series of problems, such as Planting risk is big, so how under implementation information technological guidance science accurate management, be one A not only forward position but also the scientific research task of the task of top priority.
The content of the invention
The object of the present invention is to provide a kind of agricultural automation embedded integration system, for farmer provide science, rationally, Efficient Cultivate administration decision information, and crop growth environment adjusting can also be autonomously carried out.
In order to solve the above technical problem, the present invention provides a kind of agricultural automation embedded integration system, including：Adopt Collecting system and decision system, wherein acquisition system are by the crop growth environment data sending of collection to decision system；It is described to determine Plan system obtains the decision instruction for being adapted to crop growth by machine learning algorithm.
Further, the agricultural automation embedded integration system further includes：The user terminal being connected with decision system, with And the crop growth environment regulating system that the decision instruction sent by decision system controls.
Further, the acquisition system includes：Air temperature sensor, air humidity sensor, soil temperature sensor, Soil humidity sensor, illuminance sensor；And the corresponding crop growth environment data of above-mentioned each sensor collection pass through IIC communication modes are sent to decision system.
Further, the crop growth environment data include：Air themperature data, air humidity data, the soil moisture Data, soil moisture data, illumination degrees of data；Using above-mentioned data combination date data as 6 Wei Te in machine learning algorithm Levy sample；By being trained by the Multi-layered Feedforward Networks that Back Propagation Algorithm is trained to 6 dimensional feature samples of input, and will Obtain decision instruction and pass to crop growth environment regulating system.
Further, 6 dimensional feature samples of input are trained by Multi-layered Feedforward Networks, i.e.,
Training data normalizes, and it is 1 that all training datas are normalized variance according to dimension；
Data after normalization are input to hidden layer, setting hidden layer node i and output node layer j, and two layers corresponding Weights between node are wij；The threshold values of node j is bj, the output valve of two-layer node corresponds to x respectivelyiAnd xj, and export node layer Output valve be weights and present node according to the output valve of all nodes in upper strata, present node and all nodes of last layer Threshold values and activation primitive are to realize, i.e.,Wherein f represents activation primitive；M is expression hidden layer node i Quantity；
It is as follows to the weights between hidden layer and output layer and threshold values adjustment according to gradient descent method：
η in above formula1Corresponding factor of momentum, δ between hidden layer and output layerijIt is i-th of node and defeated in hidden layer Go out the adjustment amount between j-th of node of layer；And
It is as follows for the weights between input layer and hidden layer and adjusting thresholds：
η in above formula2The factor of momentum between input layer and hidden layer, δkiAnd wkiIt is k-th of node of input layer respectively and hidden Containing the adjustment amount and weights between i-th of node of layer, xkIt is the output of input layer k nodes, biIt is the threshold value of hidden layer i-node.
Further, the crop growth environment regulating system includes：Relay；
The decision system sends decision instruction to corresponding relay, and controlled by corresponding relay light compensating lamp, take out Water valve, ventilation are turned on and off with solenoid valve, to adjust crop growth environment.
Another aspect, present invention also offers a kind of method of work of agricultural automation embedded integration system.
The method of work of shown agricultural automation embedded integration system includes the following steps：
Step S1, gathers crop growth environment data；
Step S2, the decision instruction for being adapted to crop growth is obtained by machine learning algorithm；And
Step S3, crop growth environment is adjusted by decision instruction.
Further, the crop growth environment data include：Air themperature data, air humidity data, the soil moisture Data, soil moisture data, illumination degrees of data；
Using above-mentioned data combination date data as 6 dimensional feature samples in machine learning algorithm；
By being trained by the Multi-layered Feedforward Networks that Back Propagation Algorithm is trained to 6 dimensional feature samples of input, and It will obtain decision instruction and pass to crop growth environment regulating system.
Further, 6 dimensional feature samples of input are trained by Multi-layered Feedforward Networks, i.e.,
Training data normalizes, and it is 1 that all training datas are normalized variance according to dimension；
Data after normalization are input to hidden layer, setting hidden layer node i and output node layer j, and two layers corresponding Weights between node are wij；The threshold values of node j is bj, the output valve of two-layer node corresponds to x respectivelyiAnd xj, and export node layer Output valve be weights and present node according to the output valve of all nodes in upper strata, present node and all nodes of last layer Threshold values and activation primitive are to realize, i.e.,Wherein f represents activation primitive；M is expression hidden layer node i Quantity；
It is as follows to the weights between hidden layer and output layer and threshold values adjustment according to gradient descent method：
η in above formula1Corresponding factor of momentum, δ between hidden layer and output layerijIt is i-th of node and defeated in hidden layer Go out the adjustment amount between j-th of node of layer；And
It is as follows for the weights between input layer and hidden layer and adjusting thresholds：
η in above formula2The factor of momentum between input layer and hidden layer, δkiAnd wkiIt is k-th of node of input layer respectively and hidden Containing the adjustment amount and weights between i-th of node of layer, xkIt is the output of input layer k nodes, biIt is the threshold value of hidden layer i-node.
Compared with prior art, the beneficial effect that is reached of the present invention is：The present invention is fitted using machine learning algorithm Correspondence between crop growth environment data and planting decision-making, and completed automatically by the embedded system of complete set Change plantation, the limitation for solving traditional agriculture, there is provided a kind of agricultural automation insertion that is efficient, possessing the ability of being precisely controlled Formula integrated system.
Brief description of the drawings
The present invention is further described with reference to the accompanying drawings and examples.
Fig. 1 is the functional block diagram of the agricultural automation embedded integration system of the present invention；
Fig. 2 is the work flow diagram of the agricultural automation embedded integration system of the present invention.
Embodiment
In conjunction with the accompanying drawings, the present invention is further explained in detail.These attached drawings are simplified schematic diagram, only with Illustration illustrates the basic structure of the present invention, therefore it only shows composition related to the present invention.
Embodiment 1
As shown in Figure 1, the present embodiment 1 provides a kind of agricultural automation embedded integration system, including：Acquisition system and Decision system, wherein acquisition system are by the crop growth environment data sending of collection to decision system；The decision system is led to Cross machine learning algorithm and obtain the decision instruction for being adapted to crop growth.
Preferably, the agricultural automation embedded integration system further includes：The user terminal being connected with decision system, with And the crop growth environment regulating system that the decision instruction sent by decision system controls.
Specifically, the acquisition system includes：Air temperature sensor, air humidity sensor, soil temperature sensor, Soil humidity sensor, illuminance sensor；And the corresponding crop growth environment data of above-mentioned each sensor collection pass through IIC communication modes are sent to decision system.
Wherein, the crop growth environment data include：Air themperature data, air humidity data, soil moisture number According to, soil moisture data, illumination degrees of data；Using above-mentioned data combination date data as 6 dimensional features in machine learning algorithm Sample；By being trained by the Multi-layered Feedforward Networks that Back Propagation Algorithm is trained to 6 dimensional feature samples of input, and will To decision instruction and pass to crop growth environment regulating system.
As a kind of preferred embodiment of sample training, the 6 dimensional feature samples by Multi-layered Feedforward Networks to input It is trained, i.e., training data normalizes, and it is 1 that all training datas are normalized variance according to dimension；By the number after normalization According to hidden layer is input to, hidden layer node i and output node layer j are set, and the weights between two layers of respective nodes are wij；Node The threshold values of j is bj, the output valve of two-layer node corresponds to x respectivelyiAnd xj, and the output valve for exporting node layer is that owned according to upper strata The threshold values and activation primitive of the output valve of node, the weights of present node and all nodes of last layer and present node to realize, I.e.Wherein f represents activation primitive；M is the quantity for representing hidden layer node i；
It is as follows to the weights between hidden layer and output layer and threshold values adjustment according to gradient descent method：
η in above formula1Corresponding factor of momentum, δ between hidden layer and output layerijIt is i-th of node and defeated in hidden layer Go out the adjustment amount between j-th of node of layer；And
It is as follows for the weights between input layer and hidden layer and adjusting thresholds：
η in above formula2The factor of momentum between input layer and hidden layer, δkiAnd wkiIt is k-th of node of input layer respectively and hidden Containing the adjustment amount and weights between i-th of node of layer, xkIt is the output of input layer k nodes, biIt is the threshold value of hidden layer i-node.
The crop growth environment regulating system includes：Relay；The decision system sends decision instruction to phase Relay is answered, and the light compensating lamp that is controlled by corresponding relay, water pumping valve, ventilation are turned on and off with solenoid valve, to adjust farming Thing growing environment.
Wherein, the processor module that decision system uses sets chip for chip STM32F103R8T6
The timer interruption of STM32F103R8T6 so that chip every five seconds for example performs sensor data acquisition program；Utilize IIC Communications protocol is successively read air themperature Temp, air humidity Hum, soil moisture Soil_ from STM32F103R8T6 chip pins Temp, soil moisture Soil_Hum, illuminance Illuminance, and date data is also obtained by a telecommunication system, Specifically, implementing telecommunication system includes GPRS, TCP/IP communications protocol, SOCKET communications and Ethernet.
The carry out data normalization processing that will be collected, obtains Data_In；Data_In is inputted into trained completion In mapping relations, result of decision Data_out is obtained；Data_out finally is converted into corresponding level signal by corresponding to draw Foot is transferred to crop growth environment regulating system.
Embodiment 2
As shown in Fig. 2, on the basis of embodiment 1, the present embodiment 2 provides a kind of agricultural automation embedded integration system Method of work, include the following steps：
Step S1, gathers crop growth environment data；
Step S2, the decision instruction for being adapted to crop growth is obtained by machine learning algorithm；And
Step S3, crop growth environment is adjusted by decision instruction.
Wherein, the crop growth environment data include：Air themperature data, air humidity data, soil moisture number According to, soil moisture data, illumination degrees of data；Using above-mentioned data combination date data as 6 dimensional features in machine learning algorithm Sample；By being trained by the Multi-layered Feedforward Networks that Back Propagation Algorithm is trained to 6 dimensional feature samples of input, and will To decision instruction and pass to crop growth environment regulating system.
As a kind of preferred embodiment of sample training, the 6 dimensional feature samples by Multi-layered Feedforward Networks to input It is trained, i.e., training data normalizes, and it is 1 that all training datas are normalized variance according to dimension；By the number after normalization According to hidden layer is input to, hidden layer node i and output node layer j are set, and the weights between two layers of respective nodes are wij；Node The threshold values of j is bj, the output valve of two-layer node corresponds to x respectivelyiAnd xj, and the output valve for exporting node layer is that owned according to upper strata The threshold values and activation primitive of the output valve of node, the weights of present node and all nodes of last layer and present node to realize, I.e.Wherein f represents activation primitive；M is the quantity for representing hidden layer node i；
It is as follows to the weights between hidden layer and output layer and threshold values adjustment according to gradient descent method：
η in above formula1Corresponding factor of momentum, δ between hidden layer and output layerijIt is i-th of node and defeated in hidden layer Go out the adjustment amount between j-th of node of layer；And
It is as follows for the weights between input layer and hidden layer and adjusting thresholds：
η in above formula2The factor of momentum between input layer and hidden layer, δkiAnd wkiIt is k-th of node of input layer respectively and hidden Containing the adjustment amount and weights between i-th of node of layer, xkIt is the output of input layer k nodes, biIt is the threshold value of hidden layer i-node.
It is complete by above-mentioned description, relevant staff using the above-mentioned desirable embodiment according to the present invention as enlightenment Various changes and amendments can be carried out without departing from the scope of the technological thought of the present invention' entirely.The technology of this invention Property scope is not limited to the content on specification, it is necessary to determines its technical scope according to right.

Claims (9)

  1. A kind of 1. agricultural automation embedded integration system, it is characterised in that including：
    Acquisition system and decision system, wherein
    Acquisition system is by the crop growth environment data sending of collection to decision system；
    The decision system obtains the decision instruction for being adapted to crop growth by machine learning algorithm.
  2. 2. agricultural automation embedded integration system according to claim 1, it is characterised in that
    The agricultural automation embedded integration system further includes：The user terminal being connected with decision system, and by decision-making system The crop growth environment regulating system for the decision instruction control that system is sent.
  3. 3. agricultural automation embedded integration system according to claim 2, it is characterised in that
    The acquisition system includes：Air temperature sensor, air humidity sensor, soil temperature sensor, soil moisture pass Sensor, illuminance sensor；And
    The corresponding crop growth environment data of above-mentioned each sensor collection are sent to decision system by IIC communication modes.
  4. 4. agricultural automation embedded integration system according to claim 3, it is characterised in that
    The crop growth environment data include：Air themperature data, air humidity data, soil temperature data, soil are wet Degrees of data, illumination degrees of data；
    Using above-mentioned data combination date data as 6 dimensional feature samples in machine learning algorithm；
    By being trained by the Multi-layered Feedforward Networks that Back Propagation Algorithm is trained to 6 dimensional feature samples of input, and will To decision instruction and pass to crop growth environment regulating system.
  5. 5. agricultural automation embedded integration system according to claim 4, it is characterised in that
    6 dimensional feature samples of input are trained by Multi-layered Feedforward Networks, i.e.,
    Training data normalizes, and it is 1 that all training datas are normalized variance according to dimension；
    Data after normalization are input to hidden layer, setting hidden layer node i and output node layer j, and two layers of respective nodes Between weights be wij；The threshold values of node j is bj, the output valve of two-layer node corresponds to x respectivelyiAnd xj, and export the defeated of node layer It is the threshold values according to the output valve of all nodes in upper strata, the weights of present node and all nodes of last layer and present node to go out value And activation primitive to be to realize, i.e.,Wherein f represents activation primitive；M is the number for representing hidden layer node i Amount；
    It is as follows to the weights between hidden layer and output layer and threshold values adjustment according to gradient descent method：
    <mrow> <msub> <mi>w</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> <mo>=</mo> <msub> <mi>w</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>1</mn> </msub> <mo>*</mo> <mfrac> <mrow> <mo>&amp;part;</mo> <mrow> <mo>(</mo> <mi>w</mi> <mo>,</mo> <mi>b</mi> <mo>)</mo> </mrow> </mrow> <mrow> <mo>&amp;part;</mo> <msub> <mi>w</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> </mrow> </mfrac> <mo>=</mo> <msub> <mi>w</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>1</mn> </msub> <mo>*</mo> <msub> <mi>&amp;delta;</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> <mo>*</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>;</mo> </mrow>
    <mrow> <msub> <mi>b</mi> <mi>j</mi> </msub> <mo>=</mo> <msub> <mi>b</mi> <mi>j</mi> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>2</mn> </msub> <mo>*</mo> <mfrac> <mrow> <mo>&amp;part;</mo> <mrow> <mo>(</mo> <mi>w</mi> <mo>,</mo> <mi>b</mi> <mo>)</mo> </mrow> </mrow> <mrow> <mo>&amp;part;</mo> <msub> <mi>b</mi> <mi>j</mi> </msub> </mrow> </mfrac> <mo>=</mo> <msub> <mi>b</mi> <mi>j</mi> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>2</mn> </msub> <mo>*</mo> <msub> <mi>&amp;delta;</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> <mo>;</mo> </mrow>
    η in above formula1Corresponding factor of momentum, δ between hidden layer and output layerijIt is i-th of node and output layer in hidden layer Adjustment amount between j-th of node；And
    It is as follows for the weights between input layer and hidden layer and adjusting thresholds：
    <mrow> <msub> <mi>w</mi> <mrow> <mi>k</mi> <mi>i</mi> </mrow> </msub> <mo>=</mo> <msub> <mi>w</mi> <mrow> <mi>k</mi> <mi>i</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>1</mn> </msub> <mo>*</mo> <mfrac> <mrow> <mo>&amp;part;</mo> <mrow> <mo>(</mo> <mi>w</mi> <mo>,</mo> <mi>b</mi> <mo>)</mo> </mrow> </mrow> <mrow> <mo>&amp;part;</mo> <msub> <mi>w</mi> <mrow> <mi>k</mi> <mi>i</mi> </mrow> </msub> </mrow> </mfrac> <mo>=</mo> <msub> <mi>w</mi> <mrow> <mi>k</mi> <mi>i</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>1</mn> </msub> <mo>*</mo> <msub> <mi>&amp;delta;</mi> <mrow> <mi>k</mi> <mi>i</mi> </mrow> </msub> <mo>*</mo> <msub> <mi>x</mi> <mi>k</mi> </msub> <mo>;</mo> </mrow>
    <mrow> <msub> <mi>b</mi> <mi>i</mi> </msub> <mo>=</mo> <msub> <mi>b</mi> <mi>i</mi> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>2</mn> </msub> <mo>*</mo> <mfrac> <mrow> <mo>&amp;part;</mo> <mrow> <mo>(</mo> <mi>w</mi> <mo>,</mo> <mi>b</mi> <mo>)</mo> </mrow> </mrow> <mrow> <mo>&amp;part;</mo> <msub> <mi>b</mi> <mi>i</mi> </msub> </mrow> </mfrac> <mo>=</mo> <msub> <mi>b</mi> <mi>i</mi> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>2</mn> </msub> <mo>*</mo> <msub> <mi>&amp;delta;</mi> <mrow> <mi>k</mi> <mi>i</mi> </mrow> </msub> <mo>;</mo> </mrow>
    η in above formula2The factor of momentum between input layer and hidden layer, δkiAnd wkiIt is k-th of node of input layer and hidden layer respectively Adjustment amount and weights between i-th of node, xkIt is the output of input layer k nodes, biIt is the threshold value of hidden layer i-node.
  6. 6. agricultural automation embedded integration system according to claim 5, it is characterised in that
    The crop growth environment regulating system includes：Relay；
    The decision system sends decision instruction to corresponding relay, and controlled by corresponding relay light compensating lamp, water pumping valve, Ventilation is turned on and off with solenoid valve, to adjust crop growth environment.
  7. 7. a kind of method of work of agricultural automation embedded integration system, it is characterised in that include the following steps：
    Step S1, gathers crop growth environment data；
    Step S2, the decision instruction for being adapted to crop growth is obtained by machine learning algorithm；And
    Step S3, crop growth environment is adjusted by decision instruction.
  8. 8. method of work according to claim 7, it is characterised in that
    The crop growth environment data include：Air themperature data, air humidity data, soil temperature data, soil are wet Degrees of data, illumination degrees of data；
    Using above-mentioned data combination date data as 6 dimensional feature samples in machine learning algorithm；
    By being trained by the Multi-layered Feedforward Networks that Back Propagation Algorithm is trained to 6 dimensional feature samples of input, and will To decision instruction and pass to crop growth environment regulating system.
  9. 9. method of work according to claim 8, it is characterised in that
    6 dimensional feature samples of input are trained by Multi-layered Feedforward Networks, i.e.,
    Training data normalizes, and it is 1 that all training datas are normalized variance according to dimension；
    Data after normalization are input to hidden layer, setting hidden layer node i and output node layer j, and two layers of respective nodes Between weights be wij；The threshold values of node j is bj, the output valve of two-layer node corresponds to x respectivelyiAnd xj, and export the defeated of node layer It is the threshold values according to the output valve of all nodes in upper strata, the weights of present node and all nodes of last layer and present node to go out value And activation primitive to be to realize, i.e.,Wherein f represents activation primitive；M is the number for representing hidden layer node i Amount；
    It is as follows to the weights between hidden layer and output layer and threshold values adjustment according to gradient descent method：
    <mrow> <msub> <mi>w</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> <mo>=</mo> <msub> <mi>w</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>1</mn> </msub> <mo>*</mo> <mfrac> <mrow> <mo>&amp;part;</mo> <mrow> <mo>(</mo> <mi>w</mi> <mo>,</mo> <mi>b</mi> <mo>)</mo> </mrow> </mrow> <mrow> <mo>&amp;part;</mo> <msub> <mi>w</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> </mrow> </mfrac> <mo>=</mo> <msub> <mi>w</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>1</mn> </msub> <mo>*</mo> <msub> <mi>&amp;delta;</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> <mo>*</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>;</mo> </mrow>
    <mrow> <msub> <mi>b</mi> <mi>j</mi> </msub> <mo>=</mo> <msub> <mi>b</mi> <mi>j</mi> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>2</mn> </msub> <mo>*</mo> <mfrac> <mrow> <mo>&amp;part;</mo> <mrow> <mo>(</mo> <mi>w</mi> <mo>,</mo> <mi>b</mi> <mo>)</mo> </mrow> </mrow> <mrow> <mo>&amp;part;</mo> <msub> <mi>b</mi> <mi>j</mi> </msub> </mrow> </mfrac> <mo>=</mo> <msub> <mi>b</mi> <mi>j</mi> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>2</mn> </msub> <mo>*</mo> <msub> <mi>&amp;delta;</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> <mo>;</mo> </mrow>
    η in above formula1Corresponding factor of momentum, δ between hidden layer and output layerijIt is i-th of node and output layer in hidden layer Adjustment amount between j-th of node；And
    It is as follows for the weights between input layer and hidden layer and adjusting thresholds：
    <mrow> <msub> <mi>w</mi> <mrow> <mi>k</mi> <mi>i</mi> </mrow> </msub> <mo>=</mo> <msub> <mi>w</mi> <mrow> <mi>k</mi> <mi>i</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>1</mn> </msub> <mo>*</mo> <mfrac> <mrow> <mo>&amp;part;</mo> <mrow> <mo>(</mo> <mi>w</mi> <mo>,</mo> <mi>b</mi> <mo>)</mo> </mrow> </mrow> <mrow> <mo>&amp;part;</mo> <msub> <mi>w</mi> <mrow> <mi>k</mi> <mi>i</mi> </mrow> </msub> </mrow> </mfrac> <mo>=</mo> <msub> <mi>w</mi> <mrow> <mi>k</mi> <mi>i</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>1</mn> </msub> <mo>*</mo> <msub> <mi>&amp;delta;</mi> <mrow> <mi>k</mi> <mi>i</mi> </mrow> </msub> <mo>*</mo> <msub> <mi>x</mi> <mi>k</mi> </msub> <mo>;</mo> </mrow>
    <mrow> <msub> <mi>b</mi> <mi>i</mi> </msub> <mo>=</mo> <msub> <mi>b</mi> <mi>i</mi> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>2</mn> </msub> <mo>*</mo> <mfrac> <mrow> <mo>&amp;part;</mo> <mrow> <mo>(</mo> <mi>w</mi> <mo>,</mo> <mi>b</mi> <mo>)</mo> </mrow> </mrow> <mrow> <mo>&amp;part;</mo> <msub> <mi>b</mi> <mi>i</mi> </msub> </mrow> </mfrac> <mo>=</mo> <msub> <mi>b</mi> <mi>i</mi> </msub> <mo>-</mo> <msub> <mi>&amp;eta;</mi> <mn>2</mn> </msub> <mo>*</mo> <msub> <mi>&amp;delta;</mi> <mrow> <mi>k</mi> <mi>i</mi> </mrow> </msub> <mo>;</mo> </mrow>
    η in above formula2The factor of momentum between input layer and hidden layer, δkiAnd wkiIt is k-th of node of input layer and hidden layer respectively Adjustment amount and weights between i-th of node, xkIt is the output of input layer k nodes, biIt is the threshold value of hidden layer i-node.
CN201810003347.XA 2018-01-03 2018-01-03 Agricultural automation embedded integration system and its method of work based on machine learning Pending CN107991888A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810003347.XA CN107991888A (en) 2018-01-03 2018-01-03 Agricultural automation embedded integration system and its method of work based on machine learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810003347.XA CN107991888A (en) 2018-01-03 2018-01-03 Agricultural automation embedded integration system and its method of work based on machine learning

Publications (1)

Publication Number Publication Date
CN107991888A true CN107991888A (en) 2018-05-04

Family

ID=62040735

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810003347.XA Pending CN107991888A (en) 2018-01-03 2018-01-03 Agricultural automation embedded integration system and its method of work based on machine learning

Country Status (1)

Country Link
CN (1) CN107991888A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109558939A (en) * 2018-10-29 2019-04-02 广东奥博信息产业股份有限公司 A kind of crop growth model selection method neural network based and device

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101470421A (en) * 2007-12-28 2009-07-01 中国科学院沈阳应用生态研究所 Plant growth room based on artificial intelligence technology and its control system
CN102937785A (en) * 2012-11-09 2013-02-20 江苏大学 Greenhouse multi-model information fusion based methods and system for regulating and controlling greenhouse environment
CN103823371A (en) * 2014-02-12 2014-05-28 无锡中科智能农业发展有限责任公司 Neural network model-based agricultural precise fertilization system and fertilization method thereof
CN105892287A (en) * 2016-05-09 2016-08-24 河海大学常州校区 Crop irrigation strategy based on fuzzy judgment and decision making system
CN106020294A (en) * 2016-05-23 2016-10-12 太仓市新联粮食生产专业合作社 Remote-control recreational agricultural plantation system based on agricultural Internet of Things
CN106650212A (en) * 2016-10-10 2017-05-10 重庆科技学院 Intelligent plant breeding method and system based on data analysis
CN106707767A (en) * 2017-03-13 2017-05-24 山东农业大学 System and method for integrally and intelligently controlling water and fertilizer in field based on multi-source information fusion

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101470421A (en) * 2007-12-28 2009-07-01 中国科学院沈阳应用生态研究所 Plant growth room based on artificial intelligence technology and its control system
CN102937785A (en) * 2012-11-09 2013-02-20 江苏大学 Greenhouse multi-model information fusion based methods and system for regulating and controlling greenhouse environment
CN103823371A (en) * 2014-02-12 2014-05-28 无锡中科智能农业发展有限责任公司 Neural network model-based agricultural precise fertilization system and fertilization method thereof
CN105892287A (en) * 2016-05-09 2016-08-24 河海大学常州校区 Crop irrigation strategy based on fuzzy judgment and decision making system
CN106020294A (en) * 2016-05-23 2016-10-12 太仓市新联粮食生产专业合作社 Remote-control recreational agricultural plantation system based on agricultural Internet of Things
CN106650212A (en) * 2016-10-10 2017-05-10 重庆科技学院 Intelligent plant breeding method and system based on data analysis
CN106707767A (en) * 2017-03-13 2017-05-24 山东农业大学 System and method for integrally and intelligently controlling water and fertilizer in field based on multi-source information fusion

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
刘峡壁编著: "《国防工业出版社》", 31 July 2015 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109558939A (en) * 2018-10-29 2019-04-02 广东奥博信息产业股份有限公司 A kind of crop growth model selection method neural network based and device

Similar Documents

Publication Publication Date Title
Chaudhary et al. Application of wireless sensor networks for greenhouse parameter control in precision agriculture
CN104866970B (en) Intelligent Cultivate administration method and intelligent planting equipment
CN104460582B (en) A kind of Internet of Things intelligent irrigation fertilising control method and system based on fuzzy control
CN110262435A (en) Smart greenhouse control system and method based on big data analysis
CN104155925B (en) Henhouse micro climatic environment intelligent control system based on wireless sensor network
Umamaheswari et al. Integrating scheduled hydroponic system
CN106707757A (en) Dynamic regulation and control method and system for irrigation time
CN108510102A (en) A kind of water-fertilizer integral control method of irrigation using big data calculative strategy
WO2021007363A1 (en) Irrigation control with deep reinforcement learning and smart scheduling
KR20150007668A (en) Control system and method for automation of plant-curture factory
CN106613539A (en) Greenhouse water-fertilizer integrated nutrient solution automatic irrigation control method based on LabVIEW
Jiang et al. WSN-based control system of CO2 concentration in greenhouse
Wang et al. Real-time monitoring of environmental information and modeling of the photosynthetic rate of tomato plants under greenhouse conditions
CN107991888A (en) Agricultural automation embedded integration system and its method of work based on machine learning
CN107316251A (en) Plant growth intelligent management and system based on digital gradient technology
CN112099557A (en) Internet-based household plant planting method and system
CN112868455A (en) Method and device for improving planting yield of uncaria
CN117348650A (en) Intelligent greenhouse comprehensive control system for three-dimensional planting of polygonatum cyrtonema
CN104678843A (en) Indoor plant recommendation method based on human health information and air composition data
CN110716598A (en) Intelligent greenhouse control and management system based on Internet of things
CN107357935A (en) Intelligent planting equipment control method based on program cloud
CN111670733B (en) Balcony plant planting system
Michael et al. Development of an Irrigation System for Predicting Watering Time with ANFIS Method for Chili Plants
Sravani et al. Design and implementation of a smart controller in agriculture for improved productivity
CN110135403A (en) Houseplant cultural method, device, system, equipment and readable storage medium storing program for executing

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20180504

RJ01 Rejection of invention patent application after publication