Unlock instant, AI-driven research and patent intelligence for your innovation.

Artificial intelligence insect situation prediction method based on multi-parameter fusion

An artificial intelligence, multi-parameter technology, applied in the direction of forecasting, neural learning methods, data processing applications, etc., can solve the problems affecting the stability and accuracy of the forecasting model, the large amount of computation of the BP artificial neural network, and the lack of convergence of the forecasting results, etc. problem, to achieve the effect of improving the level of prevention and control, improving accuracy and reducing the loss of insect pests

Pending Publication Date: 2021-11-09
NINGBO MIENERGY TECH CO LTD
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The existing agricultural forecasting models established by BP artificial neural network have shortcomings: (1) In the process of agricultural data collection, there are a large number of errors caused by man-made, equipment, and instrument accuracy limitations. are negligible, thus greatly affecting the stability and accuracy of the forecasting model
(2) When the BP artificial neural network is used for prediction, when there are few input factors, the prediction accuracy is low; when there are many input factors, the BP artificial neural network has a large amount of calculation, resulting in the prediction results not being converged
The current prediction method based on BP artificial neural network is difficult to effectively solve this contradiction.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment 1

[0042]Specific embodiment 1, aspect prediction of insect situation can construct an input layer and be 7 nodes, output layer 1 node, middle hidden layer is the BP neural network of 15 nodes. Replace P1 to P7 of the input layer with the following values ​​for their aspects:

[0043] P1: the number of detected pests;

[0044] P2: the time when the pest was detected;

[0045] P3: the spatial longitude of the pest location;

[0046] P4: spatial latitude of pest location;

[0047] P5: temperature at the location of the pest;

[0048] P6: Humidity of pest location;

[0049] P7: Atmospheric pressure where the pest is located;

[0050] while the output O 1 Then it can be replaced by the predicted number of the pest on the nth day, the activation function of the hidden layer in this model adopts the Relu function, and the activation function of the output layer adopts a linear function.

specific Embodiment 2

[0051] Specific embodiment 2 is used to predict the number trend of various pests on a certain day in the future. Assuming that there are M types of pests to be predicted, a BP neural network with an input layer of M+6 nodes, an output layer of M nodes, and an intermediate hidden layer of M+15 nodes can be constructed. Put the input layer P1 to P M Replaced by the number of input Mth pests, P M+1 to P M+6 Replaced with the following values ​​for its aspects:

[0052] P M+1 The time when the pest was detected this time,

[0053] P M+2 The spatial longitude of the detection location,

[0054] P M+3 The spatial latitude of the detection location,

[0055] P M+4 The temperature of the testing location,

[0056] P M+5 The humidity of the testing location,

[0057] P M+6 Atmospheric pressure at the location where the test was performed.

[0058] while the output O 1 to O M Then it can be replaced by the predicted number of the pest on the nth day. In this model, the...

specific Embodiment 3

[0059] Specific embodiment 3 is used to predict the number development trend of various pests on any day in the next year, and can predict the number trend of pests on any day. Assuming that there are M types of pests to be predicted, a BP neural network with an input layer of M+7 nodes, an output layer of M nodes, and an intermediate hidden layer of M+50 nodes can be constructed. Replace P1 to PM of the input layer with the number of input Mth pests, and replace PM+1 to PM+7 with the following values:

[0060] PM+1 the time when the pest was detected this time,

[0061] PM+2 the spatial longitude of the detection location,

[0062] PM+3 The spatial latitude of the detection location,

[0063] PM+4 The temperature of the location where the test was conducted,

[0064] PM+5 The humidity of the testing location,

[0065] PM+6 is the atmospheric pressure of the location where the test is performed.

[0066] PM+7 predicts the nth day in the future (n can be any integer betwee...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an artificial intelligence insect situation prediction method based on multi-parameter fusion; the method employs a multi-layer network which is formed by the interconnection of an input layer, an output layer and a plurality of hidden layer nodes, and enables a multi-layer feedforward network to be capable of building a proper linear or nonlinear relation between the input and the output; wherein the BP neural network is used for predicting the number development trend of various insects in any day in the next year, real-time monitoring and analysis of insect conditions are achieved, insect condition state data are calculated in real time, analysis data of different dimensions and early warning are provided, insect damage is prevented in advance, and insect damage loss is reduced; besides, efficient and accurate prediction is still kept when the collected data volume is large; and meanwhile, the insect situation prediction model is iteratively corrected by using external parameters collected regularly, so that the prediction accuracy is further improved, the insect control level is improved; the method serves green agriculture and organic agriculture and has a wide prospect.

Description

technical field [0001] The invention relates to the technical field of intelligent insect infestation prediction, in particular to an artificial intelligence infestation infestation method based on multi-parameter fusion. Background technique [0002] At present, in the field of agriculture, many intelligent information systems such as the insect monitoring system also use advanced technologies such as the Internet of Things and information intelligence, and the development speed is also fast. Monitor for infestation. It can also monitor the data of the insect situation in real time, and can give early warning and kill insects in time. However, most of the data collected is only the distribution of the current situation of the pest situation, such as the distribution of regions, quantities, and time, and the existing trend of the pest situation, without further analysis, calculation, and digging of more scientific and detailed algorithms to predict Future changes in the pe...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q10/04G06Q10/06G06Q50/02G06N3/04G06N3/08
CPCG06Q10/04G06Q10/0635G06Q50/02G06N3/084G06N3/045
Inventor 章晓敏章伟聪戴征武王秉旭杨吉云
Owner NINGBO MIENERGY TECH CO LTD