CN117172385B - Sugarcane high-sugar-content harvest period prediction method and system - Google Patents
Sugarcane high-sugar-content harvest period prediction method and system Download PDFInfo
- Publication number
- CN117172385B CN117172385B CN202311190971.2A CN202311190971A CN117172385B CN 117172385 B CN117172385 B CN 117172385B CN 202311190971 A CN202311190971 A CN 202311190971A CN 117172385 B CN117172385 B CN 117172385B
- Authority
- CN
- China
- Prior art keywords
- sugar
- sugarcane
- average
- degree
- period
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 240000000111 Saccharum officinarum Species 0.000 title claims abstract description 344
- 235000007201 Saccharum officinarum Nutrition 0.000 title claims abstract description 344
- 238000003306 harvesting Methods 0.000 title claims abstract description 63
- 238000000034 method Methods 0.000 title claims abstract description 44
- CZMRCDWAGMRECN-UGDNZRGBSA-N Sucrose Chemical compound O[C@H]1[C@H](O)[C@@H](CO)O[C@@]1(CO)O[C@@H]1[C@H](O)[C@@H](O)[C@H](O)[C@@H](CO)O1 CZMRCDWAGMRECN-UGDNZRGBSA-N 0.000 claims abstract description 44
- 229930006000 Sucrose Natural products 0.000 claims abstract description 44
- 229960004793 sucrose Drugs 0.000 claims abstract description 44
- 238000000605 extraction Methods 0.000 claims abstract description 22
- 238000012216 screening Methods 0.000 claims abstract description 9
- 230000006870 function Effects 0.000 claims description 21
- 238000012360 testing method Methods 0.000 claims description 20
- 238000009825 accumulation Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 18
- 238000001556 precipitation Methods 0.000 claims description 13
- 241000609240 Ambelania acida Species 0.000 claims description 7
- 239000010905 bagasse Substances 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 238000003825 pressing Methods 0.000 claims description 5
- 238000013480 data collection Methods 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims 2
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000002354 daily effect Effects 0.000 description 14
- 238000012935 Averaging Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000003203 everyday effect Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000003066 decision tree Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 238000012353 t test Methods 0.000 description 3
- 238000000692 Student's t-test Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 241000209134 Arundinaria Species 0.000 description 1
- 238000004566 IR spectroscopy Methods 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003969 polarography Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000000551 statistical hypothesis test Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a system for predicting a high sugar harvesting period of sugarcane, and belongs to the technical field of sugarcane sugar degree prediction. Firstly, acquiring the weight of the sugarcane and the sugar extraction amount of the sugarcane, constructing a sugar conservation model based on the data, and calculating the sugar degree of different varieties and different periods of time. Then, by comparing the sugar cane sugar degree with a threshold value, whether the sugar cane sugar degree is high sugar degree is judged, and the average sugar degree in the high sugar degree period is counted. Then, weather data from the current month of sugarcane seedlings to the beginning of the period of high sugar content are collected, and average weather data are calculated. And (5) screening out meteorological variables influencing the sugar degree by calculating the correlation coefficient of the meteorological data and the sugar degree average number. Further, a prediction model is constructed according to meteorological variables and the sugar cane sugar degree, and the harvesting period of the high sugar degree of the sugar cane is predicted in real time. The invention is beneficial to improving the prediction accuracy of the high sugar harvesting period of the sugarcane, optimizing the planting management of crops, improving the yield and the quality and providing an informationized prediction tool for obtaining more profits for sugar factories.
Description
Technical Field
The invention belongs to the technical field of sugarcane sugar degree prediction, and particularly relates to a method and a system for predicting a sugarcane high sugar degree harvesting period.
Background
Sugarcane is an important commercial crop, one of its main values being determined by its sugar content. In agricultural production, accurately predicting the harvest time of high sugar content of sugarcane is important in improving sugar industry yield, optimizing crop planting and management and the like.
At present, the harvest time prediction technology of sugarcane high sugar degree is mainly divided into two types: laboratory analysis-based methods and remote sensing technology-based methods. Laboratory analysis based methods are used to determine cane sugar content by subjecting a sample of cane to chemical or physical tests such as polarography, refraction, infrared spectroscopy, and the like. Although the method has high accuracy, the method is time-consuming and labor-consuming, and large-area and high-frequency monitoring cannot be realized. The method based on the remote sensing technology is to obtain the spectral information of the sugarcane canopy by utilizing sensors such as multispectral, hyperspectral, thermal infrared and the like carried by platforms such as satellites, unmanned aerial vehicles, ground sensors and the like, and then establish a relation model between the sugarcane sugar degree and spectral characteristics by utilizing algorithms such as machine learning, deep learning and the like, so that the estimation of the sugarcane sugar degree is realized. The method has the advantages of no damage, rapidness and high flux, is influenced by interference factors such as cloud layers, atmosphere, soil and the like, also requires a large amount of sample data and calculation resources, and proper model selection and parameter adjustment, otherwise, the model is possibly over-fitted or under-fitted, and the prediction effect is influenced. Therefore, a scientific, comprehensive, time-saving and labor-saving method for predicting the high sugar harvesting period of sugarcane is needed.
Disclosure of Invention
Based on the technical problems, the invention provides a method and a system for predicting the high sugar harvesting period of sugarcane, which comprehensively consider factors such as the weight of the sugarcane, the sugar squeezing amount of the sugarcane, and meteorological data so as to improve the accuracy and the reliability of the prediction.
The invention provides a method for predicting a high sugar harvesting period of sugarcane, which comprises the following steps:
step S1: acquiring the weight of the sugarcane and the sugar squeezing amount of the sugarcane; the weight of the sugarcane comprises the weight corresponding to different varieties of sugarcane; the sugar extraction amount of the sugarcane comprises the total sugar extraction amount of different varieties of sugarcane;
step S2: constructing a sugar degree conservation model according to the weight of the sugarcane and the sugar squeezing amount of the sugarcane, and calculating the sugar degree of the sugarcane of different varieties and different time periods;
step S3: comparing the sugar cane degree with a first threshold, and if the sugar cane degree is greater than or equal to the first threshold, the sugar cane degree is high; counting starting time and ending time of a high sugar degree period of the sugarcane, and calculating a sugar degree average of the high sugar degree period of the sugarcane;
step S4: acquiring daily meteorological data from 1 day of the current month of sugarcane seedling emergence to the starting time of the high-sugar-content period of the sugarcane, and dividing the meteorological data according to different time periods to obtain average meteorological data;
step S5: calculating a correlation coefficient of the average meteorological data and the sugar degree average, and screening meteorological variables affecting the sugar degree of the sugarcane according to the correlation coefficient;
step S6: constructing a prediction model of the sugarcane high-sugar harvesting period according to the meteorological variable and the sugarcane sugar degree;
step S7: and predicting the high sugar harvesting period of the sugarcane in real time according to the prediction model.
Optionally, constructing a sugar degree conservation model according to the weight of the sugarcane and the sugar amount of the sugarcane, and calculating sugar degrees of different varieties and different time periods, wherein the sugar degree conservation model specifically comprises the following steps:
the sugar degree conservation model is as follows:
S＝w 1 b 1 +w 2 b 2 +...+W i-1 b i-1 +w i b i
wherein the weight W= [ W ] 1 ，w 2 ，...，w i-1 ，w i ]W is a daily sugarcane weight set comprising the weight of each sugarcane, i is the number of sugarcane varieties, i is E [1,8 ]]Integer, w 1 For the first weight, w i Is known; the sugar extraction amount S of the sugarcane is the sum of the sugar extraction amounts of the sugarcane every day, and S is known; b 1 For the first degree of sugar cane, b i Unknown;
establishing an i-element once equation set according to the sugar degree conservation model, and obtaining i sugar degrees corresponding to i varieties of sugarcane every i days; the sugar squeezing period of the sugarcane of different varieties is converted into days D, and the sugar degree of the sugarcane of different varieties and in different periods is calculated, wherein the specific formula is as follows:
T＝[D/i]
wherein T is the number of time periods, D is the number of days in the sugar squeezing period of the sugarcane, i is the number of varieties of the sugarcane, and [ D/i ] is the maximum integer not exceeding D/i.
Optionally, the step of comparing the sugar cane sugar degree with a first threshold value, wherein if the sugar cane sugar degree is greater than or equal to the first threshold value, the sugar cane sugar degree is high sugar degree; counting starting time and ending time of a high sugar degree period of the sugarcane, and calculating a sugar degree average of the high sugar degree period of the sugarcane, wherein the method specifically comprises the following steps:
comparing the sugar cane degrees of different varieties and different time periods with a first threshold value, and if the sugar cane degrees are larger than or equal to the first threshold value of each variety, the sugar cane degrees are high sugar degrees, and the corresponding time periods are high sugar degree time periods; if the sugar cane sugar degree is smaller than the first threshold value of each variety, the sugar cane sugar degree is not high sugar degree, and the corresponding period is a period of non-high sugar degree;
counting starting time and ending time of a high sugar degree period of the sugarcane, and carrying out arithmetic average operation on the sugar degree of the sugarcane corresponding to the high sugar degree period of the sugarcane in the starting time and the ending time to obtain the average sugar degree of the high sugar degree period of the sugarcane.
Optionally, the acquiring daily meteorological data from 1 day of the current month of sugarcane emergence to the beginning time of the period of high sugar content of the sugarcane, dividing the meteorological data according to different time periods to obtain average meteorological data specifically includes:
acquiring day-by-day meteorological data from 1 day of the current month of sugarcane emergence to the beginning time of the high-sugar-content period of the sugarcane, wherein the meteorological data comprise the highest temperature, the average temperature, the lowest temperature, the precipitation amount, the sunshine hours, the average relative humidity and the minimum relative humidity; dividing the meteorological data by taking the current month of emergence as the beginning, and respectively dividing the meteorological data every i days, every 2i days, every 3i days and every 4i days to obtain average meteorological data; the average meteorological data comprises an average of highest temperatures, an average of average temperatures, an average of lowest temperatures, an accumulation of precipitation, an accumulation of sunshine hours, an average of average relative humidity and an average of minimum relative humidity corresponding to different dividing days.
Optionally, calculating a correlation coefficient between the average meteorological data and the average sugar degree, and screening meteorological variables affecting the sugar degree of the sugarcane according to the correlation coefficient, which specifically includes:
calculating correlation coefficients of the average highest temperature, the average temperature, the average lowest temperature, the accumulation of precipitation, the accumulation of sunshine hours, the average of average relative humidity and the average of minimum relative humidity in the average meteorological data and the average of the sugar degree to obtain the correlation coefficients of different characteristics, wherein the specific formulas are as follows:
wherein r is a correlation coefficient, n is the number of time periods for dividing data, and x α For the average meteorological data for the alpha-th time period,is the average value of the average meteorological data, y α For the degree of cane sugar corresponding to the average meteorological data in time period alpha,/->Is the average of the sugar degrees;
and judging the significance level by adopting a t hypothesis test on the correlation coefficients of the different features, and selecting meteorological variables influencing the bagasse.
Optionally, constructing a prediction model of the high sugar harvesting period of the sugarcane according to the meteorological variable and the sugar degree of the sugarcane, which specifically comprises the following steps:
constructing a data set and dividing a training set and a testing set according to the meteorological variable and the cane sugar degree; initializing a gradient lifting tree model, setting the number, depth and learning rate parameters of the tree, enabling the center of the tree to be biased to be true, fitting the gradient lifting tree model by using a training set, and calculating a loss function of the training set; randomly selecting a part of meteorological variables and dividing points when a new tree is added each time; and predicting the test set by using the new tree, updating the loss function until the number of the preset trees is reached or the loss function converges, and outputting a prediction model of the sugarcane high-sugar harvesting period.
The invention also provides a system for predicting the high sugar harvesting period of sugarcane, which comprises:
the sugarcane data acquisition module is used for acquiring the weight of sugarcane and the amount of the extracted sugarcane; the weight of the sugarcane comprises the weight corresponding to different varieties of sugarcane; the sugar extraction amount of the sugarcane comprises the total sugar extraction amount of different varieties of sugarcane;
the sugar degree conservation model construction module is used for constructing a sugar degree conservation model according to the weight of the sugarcane and the sugar squeezing amount of the sugarcane, and calculating the sugar degree of the sugarcane in different varieties and different time periods;
the sugarcane high-sugar-content time period data acquisition module is used for comparing the sugar cane sugar content with a first threshold value, and if the sugar cane sugar content is greater than or equal to the first threshold value, the sugar cane sugar content is high sugar content; counting starting time and ending time of a high sugar degree period of the sugarcane, and calculating a sugar degree average of the high sugar degree period of the sugarcane;
the meteorological data collection module is used for obtaining daily meteorological data from 1 day of the current month of sugarcane seedling emergence to the starting time of the high-sugar-content period of the sugarcane, dividing the meteorological data according to different time periods and obtaining average meteorological data;
the influence variable determining module is used for calculating a correlation coefficient of the average meteorological data and the sugar degree average, and screening meteorological variables influencing the sugar degree of the sugarcane according to the correlation coefficient;
the model definition module is used for constructing a prediction model of the high sugar harvesting period of the sugarcane according to the meteorological variable and the sugar harvesting degree of the sugarcane;
and the sugarcane harvest period prediction module is used for predicting the sugarcane high-sugar-content harvest period in real time according to the prediction model.
Optionally, the sugar degree conservation model construction module specifically includes:
the sugar degree conservation model is as follows:
S＝w 1 b 1 +w 2 b 2 +...+W i-1 b i-1 +w i b i
wherein the weight W= [ W ] 1 ，w 2 ，...，w i-1 ，w i ]W is a daily sugarcane weight set comprising the weight of each sugarcane, i is the number of sugarcane varieties, i is E [1,8 ]]Integer, w 1 For the first weight, w i Is known; the sugar extraction amount S of the sugarcane is the sum of the sugar extraction amounts of the sugarcane every day, and S is known; b 1 For the first degree of sugar cane, b i Unknown;
establishing an i-element once equation set according to the sugar degree conservation model, and obtaining i sugar degrees corresponding to i varieties of sugarcane every i days; the sugar squeezing period of the sugarcane of different varieties is converted into days D, and the sugar degree of the sugarcane of different varieties and in different periods is calculated, wherein the specific formula is as follows:
T＝[D/i]
wherein T is the number of time periods, D is the number of days in the sugar squeezing period of the sugarcane, i is the number of varieties of the sugarcane, and [ D/i ] is the maximum integer not exceeding D/i.
Optionally, the influence variable determining module specifically includes:
calculating correlation coefficients of the average highest temperature, the average temperature, the average lowest temperature, the accumulation of precipitation, the accumulation of sunshine hours, the average of average relative humidity and the average of minimum relative humidity in the average meteorological data and the average of the sugar degree to obtain the correlation coefficients of different characteristics, wherein the specific formulas are as follows:
wherein r is a correlation coefficient, n is the number of time periods for dividing data, and x α For the average meteorological data in the alpha-th time period, x is the average value of the average meteorological data, y α For the bagasse sugar corresponding to the average meteorological data in the alpha-th time period,is the average of the sugar degrees;
and judging the significance level by adopting a t hypothesis test on the correlation coefficients of the different features, and selecting meteorological variables influencing the bagasse.
Optionally, the sugarcane high sugar harvesting period model definition module specifically comprises:
constructing a data set and dividing a training set and a testing set according to the meteorological variable and the cane sugar degree; initializing a gradient lifting tree model, setting the number, depth and learning rate parameters of the tree, enabling the center of the tree to be biased to be true, fitting the gradient lifting tree model by using a training set, and calculating a loss function of the training set; randomly selecting a part of meteorological variables and dividing points when a new tree is added each time; and predicting the test set by using the new tree, updating the loss function until the number of the preset trees is reached or the loss function converges, and outputting a prediction model of the sugarcane high-sugar harvesting period.
Compared with the prior art, the invention has the following beneficial effects:
the invention utilizes two pieces of data which are easy to obtain and closely related to the sugar degree of the sugarcane, namely the weight of the sugarcane and the sugar amount of the sugarcane, constructs a simple and effective sugar degree conservation model, can calculate the sugar degree of the sugarcane of different varieties and different time periods, and judges whether the sugarcane reaches a high sugar degree state or not according to a set threshold value; the model capable of predicting the high sugar harvesting period of the sugarcane in real time is constructed by combining the meteorological data and the gradient lifting tree algorithm, and the proposal of the optimal harvesting time can be provided for farmers according to the current and future meteorological conditions; the high-sugar harvesting period of the sugarcane is predicted in the early growth period of the sugarcane, so that more harvesting options are provided for farmers, and loss caused by missing of the optimal harvesting period or delay of harvesting is avoided; the yield and the quality of the sugarcane are improved, and the income of peasants and the benefits of sugar-making enterprises are increased.
Drawings
FIG. 1 is a flow chart of a method for predicting the high sugar harvesting period of sugarcane according to the invention;
FIG. 2 is a block diagram of a sugarcane high sugar harvesting period prediction system according to the present invention.
Detailed Description
The invention is further described below in connection with specific embodiments and the accompanying drawings, but the invention is not limited to these embodiments.
Example 1
As shown in fig. 1, the invention discloses a method for predicting a high sugar harvesting period of sugarcane, which comprises the following steps:
step S1: acquiring the weight of the sugarcane and the sugar squeezing amount of the sugarcane; the weight of the sugarcane comprises the weight corresponding to different varieties of sugarcane; the sugar extraction amount of the sugarcane comprises the total sugar extraction amount of different types of sugarcane.
Step S2: and constructing a sugar degree conservation model according to the weight of the sugarcane and the sugar squeezing amount of the sugarcane, and calculating the sugar degree of the sugarcane of different varieties and different periods.
Step S3: comparing the sugar cane degree with a first threshold, and if the sugar cane degree is greater than or equal to the first threshold, the sugar cane degree is high; the starting time and the ending time of the high sugar degree period of the sugarcane are counted, and the average sugar degree of the high sugar degree period of the sugarcane is calculated.
Step S4: the method comprises the steps of obtaining daily meteorological data from 1 day of the current month of sugarcane seedling emergence to the starting time of a high-sugar-degree period of the sugarcane, dividing the meteorological data according to different time periods, and obtaining average meteorological data.
Step S5: and calculating a correlation coefficient of the average meteorological data and the sugar degree average, and screening meteorological variables influencing the sugar degree of the sugarcane according to the correlation coefficient.
Step S6: and constructing a prediction model of the high sugar harvesting period of the sugarcane according to the meteorological variable and the sugar degree of the sugarcane.
Step S7: and predicting the high sugar harvesting period of the sugarcane in real time according to the prediction model.
The steps are discussed in detail below:
step S1: the method for obtaining the weight of the sugarcane and the amount of the extracted sugarcane specifically comprises the following steps:
the method comprises the steps of obtaining the data of the quantity and total quantity of sugar cane of different sugar cane varieties in a sugar refinery in a sugar cane harvesting period of 3-6 months, wherein the sugar cane harvesting season is 11 months 1 day per year to 31 days 3 months in the next year, the sugar squeezing data of each variety which cannot be measured every day in actual production are obtained, the quantity of the sugar cane is in tons, the quantity of the sugar cane obtained by the sugar refinery in each day is not more than 8 varieties, the acquisition is in weight units, the unit price is determined according to the varieties, and the price of the sugar content is high.
In the embodiment, the weight of the sugarcane comprises the weight corresponding to different varieties of sugarcane; the sugar extraction amount of the sugarcane comprises the total sugar extraction amount of different types of sugarcane.
Step S2: constructing a sugar degree conservation model according to the weight of the sugarcane and the sugar squeezing amount of the sugarcane, and calculating the sugar degree of the sugarcane of different varieties and different time periods, wherein the method specifically comprises the following steps:
the sugar degree conservation model is as follows:
S＝w 1 b 1 +w 2 b 2 +...+w i-1 b i-1 +w i b i
in the formula, the weight W= [ W ] 1 ，w 2 ，...，w i-1 ，w i ]W is a daily sugarcane weight set comprising the weight of each sugarcane, i is the number of sugarcane varieties, i is E [1,8 ]]Integer, w 1 For the first weight, w i Is known; the sugar amount S of the sugarcane is daily sugarSum of amounts, S is known; b 1 For the first degree of sugar cane, b i Unknown.
Establishing an i-element once equation set according to the sugar degree conservation model, and then obtaining i sugar degrees corresponding to i varieties of sugarcane every i days; the sugar squeezing period of the sugarcane of different varieties is converted into days D, and the sugar degree of the sugarcane of different varieties and in different periods is calculated, wherein the specific formula is as follows:
T＝[D/i]
wherein T is the number of time periods, D is the number of days in the whole sugar pressing period of one year of the sugarcane, i is the number of varieties of the sugarcane, and [ D/i ] is the maximum integer not exceeding D/i.
In this example, it was assumed that 8 varieties of sugar cane were purchased daily, each having a weight of w 1 、w 2 、w 3 、w 4 、w 5 、w 6 、w 7 And w 8 The weight is known; sugar degree is b respectively 1 、b 2 、b 3 、b 4 、b 5 、b 6 、b 7 And b 8 The sugar degree is unknown; daily sugar extraction amount S of sugarcane is known; the daily conservation of sugar model is:
S＝w 1 b 1 +w 2 b 2 +...+w 7 b 7 +w 8 b 8
then a linear equation set of eight elements is listed, and 8 cane sugar degrees corresponding to 8 varieties can be obtained every 8 days, namely b 1 、b 2 、b 3 、b 4 、b 5 、b 6 、b 7 And b 8 The method comprises the steps of carrying out a first treatment on the surface of the Sugar content of 8 varieties in 18 time periods can be obtained when the sugarcane pressing time in the year is 11 months 1 day to 3 months 31 days, namely 151 days to 152 days (comprising 2 months possibly 28 days or 29 days).
Step S3: comparing the sugar cane degree with a first threshold, and if the sugar cane degree is greater than or equal to the first threshold, the sugar cane degree is high; counting starting time and ending time of a high sugar degree period of the sugarcane, and calculating sugar degree average number of the high sugar degree period of the sugarcane, wherein the method specifically comprises the following steps:
comparing the sugar cane degrees of different varieties and different time periods with a first threshold value, and if the sugar cane degrees are larger than or equal to the first threshold value of each variety, the sugar cane degrees are high sugar degrees, and the corresponding time periods are high sugar degree time periods; if the cane sugar degree is less than the first threshold value for each variety, the cane sugar degree is non-high and the corresponding time period is a non-high sugar degree time period.
Counting starting time and ending time of the high sugar degree period of the sugarcane, and carrying out arithmetic average operation on the sugar degree of the sugarcane corresponding to the high sugar degree period of the sugarcane in the starting time and the ending time to obtain the average sugar degree of the high sugar degree period of the sugarcane.
Step S4: the method comprises the steps of obtaining daily meteorological data from 1 day of a current month of sugarcane seedling emergence to the starting time of a high-sugar-degree period of the sugarcane, dividing the meteorological data according to different time periods to obtain average meteorological data, and specifically comprises the following steps:
acquiring day-by-day meteorological data from 1 day of the current month of sugarcane seedling emergence to the beginning time of a high-sugar-content period of the sugarcane, wherein the meteorological data comprise the highest temperature, the average temperature, the lowest temperature, the precipitation amount, the sunshine hours, the average relative humidity and the minimum relative humidity; dividing the meteorological data by taking the current month of seedling emergence as the beginning, if i is an even number, dividing the meteorological data by each i/2 days, each i day, each 2i days, each 3i days and each 4i days respectively, and if i is an odd number, dividing the meteorological data by each i days, each 2i days, each 3i days and each 4i days respectively to obtain average meteorological data; the average weather data includes an average of highest temperatures, an average of average temperatures, an average of lowest temperatures, an accumulation of precipitation, an accumulation of sunshine hours, an average of average relative humidity, and an average of minimum relative humidity corresponding to different dividing days.
Step S5: calculating a correlation coefficient of the average meteorological data and the average sugar degree, and screening meteorological variables affecting the sugar degree of the sugarcane according to the correlation coefficient, wherein the method specifically comprises the following steps of:
calculating the correlation coefficients of the average highest temperature, the average lowest temperature, the accumulation of precipitation, the accumulation of sunlight hours, the average of average relative humidity and the average of minimum relative humidity in the average meteorological data and the average sugar degree to obtain the correlation coefficients of different characteristics, wherein the specific formulas are as follows:
wherein r is a correlation coefficient, n is the number of time periods for dividing data, and x α As average meteorological data for the alpha-th time period,for averaging meteorological data, y α For the average meteorological data during time alpha, the corresponding bagasse is +.>Is the average number of sugar degrees.
The correlation coefficients of different features are subjected to t hypothesis test to judge the significance level, and meteorological variables influencing the sugar cane sugar degree are selected according to the significance level, and the method specifically comprises the following steps:
the assumption of t-test is: null hypothesis (H0): there is no linear relationship between cane sugar degree and the meteorological features; alternative hypothesis (H1): there is a linear relationship between cane sugar degree and this meteorological feature.
The formula for converting the correlation coefficients of different features into t values is as follows:
wherein r is a correlation coefficient, and n is the number of time periods for dividing data.
A significance level (0.05 or 0.01) is set, and according to the degree of freedom (n-2), the t value and the significance level, software such as Excel or SPSS is used for t test, and a corresponding p value is obtained. For example, in Excel, a single sample, two independent samples, or paired samples t-test can be performed using the t.test function and a two-tailed or one-tailed p-value returned; using an online calculator to input a known t value, degree of freedom and test type (single-tailed or double-tailed) and obtain a corresponding p value; and searching a probability interval corresponding to the known t value and the degree of freedom by using a t distribution table, and estimating according to the test type (single-tail or double-tail) to obtain a corresponding p value.
Under the t distribution, the p value is a probability value used to determine whether the statistical hypothesis test result has significance. It represents the probability of the observed statistic or more extreme statistic occurring with the null hypothesis (hypothesis without effects or differences) being true. The smaller the p value, the less consistent the observed statistic is with the null hypothesis, that is, the stronger the evidence of rejecting the null hypothesis. Typically, if the p-value is less than some pre-set threshold (e.g., 0.05 or 0.01), the result is considered statistically significant, that is, there is enough reason to reject the null hypothesis, accept the alternative hypothesis (the valid or differential hypothesis).
In this embodiment, taking the average of the highest temperatures and 8 varieties as examples, calculating the correlation coefficient of the average highest temperatures and the average sugar degree in the average weather data, wherein the average of the highest temperatures comprises the average of the highest temperatures of every i/2 days, every i days, every 2i days, every 3i days and every 4i days, that is, every 4 days, every 8 days, every 16 days, every 24 days and every 32 days, averaging the highest temperatures of each characteristic in the average weather data, then calculating the correlation coefficient with the average sugar degree, obtaining a sugar cane degree every 8 days for 8 varieties, averaging the sugar cane degrees corresponding to every 4 days and the sugar cane degrees corresponding to every 8 days, and similarly pushing the two sugar cane degrees corresponding to every 16 days, obtaining 5 correlation coefficients, that is, 5 t values and 5 p values, wherein the p values of each characteristic in the average weather data are firstly judged to be smaller than a set threshold value (0.05 or 0.01), and then arranging the characteristic values from large to small according to the set weather variable.
Step S6: according to meteorological variables and the sugar cane sugar degree, constructing a prediction model of the harvesting period of the high sugar degree of the sugar cane, which specifically comprises the following steps:
constructing a data set and dividing a training set and a testing set according to meteorological variables and the sugar cane sugar degree; initializing a gradient lifting tree model, setting the number, depth and learning rate parameters of the tree, enabling the center of the tree to be biased to be true, fitting the gradient lifting tree model by using a training set, and calculating a loss function of the training set; randomly selecting a part of meteorological variables and dividing points each time a new tree is added; and predicting the test set by using the new tree, updating the loss function until the number of the preset trees is reached or the loss function converges, and outputting a prediction model of the sugarcane high-sugar harvesting period.
In a conventional gradient-lifted tree, each time a new decision tree is added, a greedy algorithm is used to select the best features and partitioning points to minimize the model's loss function. However, this approach may lead to overfitting, especially if there are few training samples.
While a random gradient-lifted tree introduces randomness each time a new tree is added, similar to the randomness of an extremely random forest. The specific process is as follows:
random feature selection: each time a new decision tree is added, the best feature of all features is no longer selected, but a portion of the features is randomly selected as a candidate set. This randomly selected process helps to increase the diversity of each tree, making the model more robust.
Random division point selection: for each selected feature, a greedy algorithm is no longer used to find the best division point. Instead, a partition point is randomly selected from the candidate partition points to construct a node of the decision tree.
Residual updating: the construction of each new tree still depends on the prediction residuals of all previous trees. The prediction result of the new tree is added to the prediction of the existing model to reduce the residual.
Iterative optimization: after each new tree is added, the model is optimized according to the gradient of the loss function to adjust the weight of each tree.
By introducing randomness in the construction process of the gradient lifting tree, the random gradient lifting tree can inhibit overfitting to a certain extent, and the generalization capability of the model is improved.
Step S7: according to the prediction model, predicting the high sugar harvesting period of the sugarcane in real time specifically comprises the following steps:
environmental data to be predicted is prepared, and real-time meteorological data including current environmental factors such as temperature, precipitation, sunlight and humidity are collected. These data will be used for input into the trained predictive model.
And (3) feature processing, namely performing feature processing which is the same as that of the training set on the collected real-time environment data, wherein the feature processing comprises standardization and normalization. Ensuring that the data input to the model has the same dimensions and processing pattern.
And predicting by using the prediction model, and inputting the processed real-time environment data into the model by using the trained prediction model. The model predicts based on the input environmental data and gives an estimated sugar cane sugar content.
And (3) interpreting the prediction result, and obtaining a numerical value according to the prediction result of the model, wherein the numerical value represents the high sugar content of the sugarcane predicted by the model in the current environment. For example, if the sugar degree is higher than a certain value, the condition of high sugar degree is satisfied.
Judging the harvest time, and setting a threshold or condition according to actual conditions and business requirements to judge whether the optimal harvest time of the high sugar degree of the sugarcane is reached.
And making a decision, and making a corresponding decision according to the result of the model prediction and the threshold value judgment. If the predicted sugar level is above the threshold, indicating that the sugar cane may have entered a high sugar state, it may be considered to begin harvesting. Conversely, if the predicted sugar level is below the threshold, it may also be necessary to wait for a period of time until the sugar cane reaches a more desirable harvest sugar level.
Example 2
As shown in fig. 2, the present invention discloses a system for predicting a high sugar harvesting period of sugarcane, the system comprising:
a sugarcane data acquisition module 10 for acquiring the weight of sugarcane and the amount of the extracted sugarcane; the weight of the sugarcane comprises the weight corresponding to different varieties of sugarcane; the sugar extraction amount of the sugarcane comprises the total sugar extraction amount of different types of sugarcane.
The sugar degree conservation model construction module 20 is used for constructing a sugar degree conservation model according to the weight of the sugarcane and the sugar squeezing amount of the sugarcane, and calculating the sugar degree of the sugarcane in different varieties and different periods.
A sugar cane high sugar degree period data acquisition module 30 for comparing the sugar cane sugar degree to a first threshold value, and if the sugar cane sugar degree is greater than or equal to the first threshold value, the sugar cane sugar degree is high sugar degree; the starting time and the ending time of the high sugar degree period of the sugarcane are counted, and the average sugar degree of the high sugar degree period of the sugarcane is calculated.
The meteorological data collection module 40 is configured to obtain daily meteorological data from 1 day of the current month of sugarcane seedling to the beginning time of the high-sugar period of the sugarcane, and divide the meteorological data according to different time periods to obtain average meteorological data.
The influence variable determining module 50 is configured to calculate a correlation coefficient between the average meteorological data and the average sugar degree, and screen out meteorological variables influencing the sugar degree of the sugarcane according to the correlation coefficient.
The sugarcane high-sugar harvesting period model definition module 60 is used for constructing a prediction model of the sugarcane high-sugar harvesting period according to meteorological variables and the sugar cane sugar.
The sugarcane harvest time prediction module 70 is used for predicting the sugarcane high-sugar harvest time in real time according to a prediction model.
As an alternative embodiment, the sugar degree conservation model construction module 20 of the present invention specifically comprises:
the sugar degree conservation model is as follows:
S＝w 1 b 1 +w 2 b 2 +...+W i-1 b i-1 +w i b i
in the formula, the weight W= [ W ] 1 ，w 2 ，...，w i-1 ，w i ]W is a daily sugarcane weight set comprising the weight of each sugarcane, i is the number of sugarcane varieties, i is E [1,8 ]]Integer, w 1 For the first weight, w i Is known; the sugar extraction amount S of the sugarcane is the sum of the sugar extraction amounts of the sugarcane every day, and S is known; b 1 For the first degree of sugar cane, b i Unknown.
Establishing an i-element once equation set according to the sugar degree conservation model, and then obtaining i sugar degrees corresponding to i varieties of sugarcane every i days; the sugar squeezing period of the sugarcane of different varieties is converted into days D, and the sugar degree of the sugarcane of different varieties and in different periods is calculated, wherein the specific formula is as follows:
T＝[D/i]
wherein T is the number of time periods, D is the number of days in the whole sugar pressing period of one year of the sugarcane, i is the number of varieties of the sugarcane, and [ D/i ] is the maximum integer not exceeding D/i.
As an alternative embodiment, the influencing variable determination module 50 of the present invention specifically includes:
calculating the correlation coefficients of the average highest temperature, the average lowest temperature, the accumulation of precipitation, the accumulation of sunlight hours, the average of average relative humidity and the average of minimum relative humidity in the average meteorological data and the average sugar degree to obtain the correlation coefficients of different characteristics, wherein the specific formulas are as follows:
wherein r is a correlation coefficient, n is the number of time periods for dividing data, and x α As average meteorological data for the alpha-th time period,for averaging meteorological data, y α For the average meteorological data during time alpha, the corresponding bagasse is +.>Is the average number of sugar degrees.
And judging the significance level by adopting t hypothesis test on the correlation coefficients of different features, and selecting meteorological variables influencing the sugar cane sugar degree according to the significance level.
As an alternative embodiment, the sugarcane high-sugar harvest time model definition module 60 of the present invention specifically comprises:
constructing a data set and dividing a training set and a testing set according to meteorological variables and the sugar cane sugar degree; initializing a gradient lifting tree model, setting the number, depth and learning rate parameters of the tree, enabling the center of the tree to be biased to be true, fitting the gradient lifting tree model by using a training set, and calculating a loss function of the training set; randomly selecting a part of meteorological variables and dividing points each time a new tree is added; and predicting the test set by using the new tree, updating the loss function until the number of the preset trees is reached or the loss function converges, and outputting a prediction model of the sugarcane high-sugar harvesting period.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. A method for predicting a high sugar harvesting period of sugarcane, the method comprising:
step S1: acquiring the weight of the sugarcane and the sugar squeezing amount of the sugarcane; the weight of the sugarcane comprises the weight corresponding to different varieties of sugarcane; the sugar extraction amount of the sugarcane comprises the total sugar extraction amount of different varieties of sugarcane;
step S2: constructing a sugar degree conservation model according to the weight of the sugarcane and the sugar squeezing amount of the sugarcane, and calculating the sugar degree of the sugarcane of different varieties and different time periods, wherein the method specifically comprises the following steps:
the sugar degree conservation model is as follows:
；
wherein the weight of the sugar cane，/>For daily sugarcane weight collection, including the weight of each sugarcane, +.>Is the number of sugarcane varieties, and is->∈[1,8]Integer, & gt>Is the first sugarcane weight; the sugar squeezing amount of the sugarcane，/>Is the sum of the daily sugar pressing amount of the sugarcane; />Is the first degree of sugar cane;
establishing according to the sugar degree conservation modelThe first equation set is every->The day obtains->Corresponding->Degree of sugar cane; conversion of sugar squeezing period of sugarcane of different varieties into day +.>The sugar cane sugar degree of different varieties and different time periods is calculated, and the specific formula is as follows:
；
in the method, in the process of the invention,for the number of time periods->For the days of sugar squeezing of sugarcane, the formula->Is not more than->Is the largest integer of (2);
step S3: comparing the sugar cane degree with a first threshold, and if the sugar cane degree is greater than or equal to the first threshold, the sugar cane degree is high; counting starting time and ending time of a high sugar degree period of the sugarcane, and calculating a sugar degree average of the high sugar degree period of the sugarcane;
step S4: acquiring daily meteorological data from 1 day of the current month of sugarcane seedling emergence to the starting time of the high-sugar-content period of the sugarcane, and dividing the meteorological data according to different time periods to obtain average meteorological data;
step S5: calculating a correlation coefficient of the average meteorological data and the sugar degree average, and screening meteorological variables affecting the sugar degree of the sugarcane according to the correlation coefficient;
step S6: according to the meteorological variable and the sugar cane sugar degree, constructing a prediction model of the harvesting period of the sugar cane high sugar degree, which specifically comprises the following steps:
initializing a gradient lifting tree model, setting the number, depth and learning rate parameters of the tree, enabling the center of the tree to be biased to be true, fitting the gradient lifting tree model by using a training set, and calculating a loss function of the training set; randomly selecting a part of meteorological variables and dividing points when a new tree is added each time; predicting the test set by using the new tree, updating the loss function until the number of the preset tree is reached or the loss function converges, and outputting a prediction model of the sugarcane high-sugar harvesting period;
step S7: and predicting the high sugar harvesting period of the sugarcane in real time according to the prediction model.
2. A method of predicting high sugar cane harvest time according to claim 1, wherein the comparing the sugar cane sugar content to a first threshold value, the sugar cane sugar content being high sugar content if the sugar cane sugar content is greater than or equal to the first threshold value; counting starting time and ending time of a high sugar degree period of the sugarcane, and calculating a sugar degree average of the high sugar degree period of the sugarcane, wherein the method specifically comprises the following steps:
comparing the sugar cane degrees of different varieties and different time periods with a first threshold value, and if the sugar cane degrees are larger than or equal to the first threshold value of each variety, the sugar cane degrees are high sugar degrees, and the corresponding time periods are high sugar degree time periods; if the sugar cane sugar degree is smaller than the first threshold value of each variety, the sugar cane sugar degree is not high sugar degree, and the corresponding period is a period of non-high sugar degree;
counting starting time and ending time of a high sugar degree period of the sugarcane, and carrying out arithmetic average operation on the sugar degree of the sugarcane corresponding to the high sugar degree period of the sugarcane in the starting time and the ending time to obtain the average sugar degree of the high sugar degree period of the sugarcane.
3. The method for predicting the harvesting period of high sugar cane according to claim 1, wherein the step of obtaining daily meteorological data from 1 day of the current month of emergence to the beginning time of the period of high sugar cane comprises dividing the meteorological data according to different time periods to obtain average meteorological data, and specifically comprises the steps of:
acquiring day-by-day meteorological data from 1 day of the current month of sugarcane emergence to the beginning time of the high-sugar-content period of the sugarcane, wherein the meteorological data comprise the highest temperature, the average temperature, the lowest temperature, the precipitation amount, the sunshine hours, the average relative humidity and the minimum relative humidity; starting the meteorological data with the current month of emergence, and respectively starting with each month of emergenceEvery day->Every day->Tianhe every->Dividing the weather data in days to obtain average weather data; the average meteorological data comprises an average of highest temperatures, an average of average temperatures, an average of lowest temperatures, an accumulation of precipitation, an accumulation of sunshine hours, an average of average relative humidity and an average of minimum relative humidity corresponding to different dividing days.
4. The method for predicting the harvest time of high sugar cane sugar degree according to claim 1, wherein the calculating the correlation coefficient between the average meteorological data and the average sugar degree, and the screening meteorological variables affecting the sugar degree according to the correlation coefficient specifically comprises:
calculating correlation coefficients of the average highest temperature, the average temperature, the average lowest temperature, the accumulation of precipitation, the accumulation of sunshine hours, the average of average relative humidity and the average of minimum relative humidity in the average meteorological data and the average of the sugar degree to obtain the correlation coefficients of different characteristics, wherein the specific formulas are as follows:
；
in the method, in the process of the invention,for the correlation coefficient +.>For dividing the number of time periods of the data, +.>Is->Said average meteorological data, < > for each time period>For the average of the average meteorological data, +.>Is->The average meteorological data over each time period corresponds to the bagasse level, < >>Is the average of the sugar degrees;
the correlation coefficients of the different characteristics are adoptedA significance level is determined by hypothesis testing, and a meteorological variable affecting the sugar cane sugar degree is selected according to the significance level.
5. A sugarcane high sugar harvesting period prediction system, the system comprising:
the sugarcane data acquisition module is used for acquiring the weight of sugarcane and the amount of the extracted sugarcane; the weight of the sugarcane comprises the weight corresponding to different varieties of sugarcane; the sugar extraction amount of the sugarcane comprises the total sugar extraction amount of different varieties of sugarcane;
the sugar degree conservation model construction module is used for constructing a sugar degree conservation model according to the weight of the sugarcane and the sugar squeezing amount of the sugarcane, and calculating the sugar degree of the sugarcane in different varieties and different time periods, and specifically comprises the following steps:
the sugar degree conservation model is as follows:
；
wherein the weight of the sugar cane，/>For daily sugarcane weight collection, including the weight of each sugarcane, +.>Is the number of sugarcane varieties, and is->∈[1,8]Integer, & gt>Is the first sugarcane weight; the sugar squeezing amount of the sugarcane，/>Is the sum of the daily sugar pressing amount of the sugarcane; />Is the first degree of sugar cane;
establishing according to the sugar degree conservation modelThe first equation set is every->The day obtains->Corresponding->Degree of sugar cane; conversion of sugar squeezing period of sugarcane of different varieties into day +.>The sugar cane sugar degree of different varieties and different time periods is calculated, and the specific formula is as follows:
；
in the method, in the process of the invention,for the number of time periods->For the days of sugar squeezing of sugarcane, the formula->Is not more than->Is the largest integer of (2);
the sugarcane high-sugar-content time period data acquisition module is used for comparing the sugar cane sugar content with a first threshold value, and if the sugar cane sugar content is greater than or equal to the first threshold value, the sugar cane sugar content is high sugar content; counting starting time and ending time of a high sugar degree period of the sugarcane, and calculating a sugar degree average of the high sugar degree period of the sugarcane;
the meteorological data collection module is used for obtaining daily meteorological data from 1 day of the current month of sugarcane seedling emergence to the starting time of the high-sugar-content period of the sugarcane, dividing the meteorological data according to different time periods and obtaining average meteorological data;
the influence variable determining module is used for calculating a correlation coefficient of the average meteorological data and the sugar degree average, and screening meteorological variables influencing the sugar degree of the sugarcane according to the correlation coefficient;
the model definition module of the high sugar degree harvest period of the sugarcane is used for constructing a prediction model of the high sugar degree harvest period of the sugarcane according to the meteorological variable and the sugar degree of the sugarcane, and specifically comprises the following steps:
initializing a gradient lifting tree model, setting the number, depth and learning rate parameters of the tree, enabling the center of the tree to be biased to be true, fitting the gradient lifting tree model by using a training set, and calculating a loss function of the training set; randomly selecting a part of meteorological variables and dividing points when a new tree is added each time; predicting the test set by using the new tree, updating the loss function until the number of the preset tree is reached or the loss function converges, and outputting a prediction model of the sugarcane high-sugar harvesting period;
and the sugarcane harvest period prediction module is used for predicting the sugarcane high-sugar-content harvest period in real time according to the prediction model.
6. The sugarcane high-sugar harvest time prediction system according to claim 5, wherein the influencing variable determination module specifically comprises:
calculating correlation coefficients of the average highest temperature, the average temperature, the average lowest temperature, the accumulation of precipitation, the accumulation of sunshine hours, the average of average relative humidity and the average of minimum relative humidity in the average meteorological data and the average of the sugar degree to obtain the correlation coefficients of different characteristics, wherein the specific formulas are as follows:
；
in the method, in the process of the invention,for the correlation coefficient +.>For dividing the number of time periods of the data, +.>Is->Said average meteorological data, < > for each time period>For the average of the average meteorological data, +.>Is->The average meteorological data over each time period corresponds to the bagasse level, < >>Is the average of the sugar degrees;
the correlation coefficients of the different characteristics are adoptedA significance level is determined by hypothesis testing, and a meteorological variable affecting the sugar cane sugar degree is selected according to the significance level.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311190971.2A CN117172385B (en) | 2023-09-15 | 2023-09-15 | Sugarcane high-sugar-content harvest period prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311190971.2A CN117172385B (en) | 2023-09-15 | 2023-09-15 | Sugarcane high-sugar-content harvest period prediction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117172385A CN117172385A (en) | 2023-12-05 |
| CN117172385B true CN117172385B (en) | 2024-03-19 |
Family
ID=88931636
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311190971.2A Active CN117172385B (en) | 2023-09-15 | 2023-09-15 | Sugarcane high-sugar-content harvest period prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117172385B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015176339A (en) * | 2014-03-14 | 2015-10-05 | 株式会社Ｊｓｏｌ | Device, method and program for prediction of crop harvest time |
| CN113505920A (en) * | 2021-06-25 | 2021-10-15 | 国家粮食和物资储备局科学研究院 | Method and system for predicting content of vomitoxin in wheat harvest period |
| CN113592193A (en) * | 2021-08-19 | 2021-11-02 | 中化现代农业有限公司 | Crop harvest time prediction method and device and storage medium |
| CN114330931A (en) * | 2022-01-24 | 2022-04-12 | 南宁师范大学 | Construction method of Guiping litchi yield major and minor annual grade region prediction model based on meteorological conditions |
| JP2022136058A (en) * | 2021-03-05 | 2022-09-15 | 国立研究開発法人農業・食品産業技術総合研究機構 | Method of generating prediction model for predicting crop production performance, generation apparatus, and generation program |
| CN115600760A (en) * | 2022-11-09 | 2023-01-13 | 南宁师范大学（Cn） | Sugarcane region yield per unit prediction method and system |
| CN115688997A (en) * | 2022-10-21 | 2023-02-03 | 浙江领见数智科技有限公司 | Accumulated temperature-based tea leaf picking period prediction method and system |
-
2023
- 2023-09-15 CN CN202311190971.2A patent/CN117172385B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015176339A (en) * | 2014-03-14 | 2015-10-05 | 株式会社Ｊｓｏｌ | Device, method and program for prediction of crop harvest time |
| JP2022136058A (en) * | 2021-03-05 | 2022-09-15 | 国立研究開発法人農業・食品産業技術総合研究機構 | Method of generating prediction model for predicting crop production performance, generation apparatus, and generation program |
| CN113505920A (en) * | 2021-06-25 | 2021-10-15 | 国家粮食和物资储备局科学研究院 | Method and system for predicting content of vomitoxin in wheat harvest period |
| CN113592193A (en) * | 2021-08-19 | 2021-11-02 | 中化现代农业有限公司 | Crop harvest time prediction method and device and storage medium |
| CN114330931A (en) * | 2022-01-24 | 2022-04-12 | 南宁师范大学 | Construction method of Guiping litchi yield major and minor annual grade region prediction model based on meteorological conditions |
| CN115688997A (en) * | 2022-10-21 | 2023-02-03 | 浙江领见数智科技有限公司 | Accumulated temperature-based tea leaf picking period prediction method and system |
| CN115600760A (en) * | 2022-11-09 | 2023-01-13 | 南宁师范大学（Cn） | Sugarcane region yield per unit prediction method and system |
Non-Patent Citations (4)
| Title |
|---|
| "Estimating Crop Yields With Remote Sensing And Deep Learning";Renato Luiz de Freitas Cunha等;《2020 IEEE Latin American GRSS & ISPRS Remote Sensing Conference (LAGIRS)》;20200326;273-278 * |
| "影响广西蔗区甘蔗糖分的气象因素分析及糖分预测模型的建立";毛乐等;《分子植物育种》;20221107;1-17 * |
| 基于卫星遥感预测作物成熟期的可行性分析;蒙继华;吴炳方;;遥感技术与应用;20130415(第02期);7-15 * |
| 基于随机森林算法和气象因子的砀山酥梨始花期预报;李德;陈文涛;乐章燕;范孝玲;孙义;孟雅婷;杨健;;农业工程学报;20200623(第12期);151-159 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117172385A (en) | 2023-12-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109711102B (en) | Method for rapidly evaluating crop disaster loss | |
| CN111126662A (en) | Irrigation decision making method, device, server and medium based on big data | |
| CN117455062A (en) | Crop yield prediction algorithm based on multi-source heterogeneous agricultural data | |
| CN116227758B (en) | Agricultural product maturity prediction method and system based on remote sensing technology and deep learning | |
| CN115018105A (en) | Winter wheat meteorological yield prediction method and system | |
| CN118364975B (en) | Wheat yield prediction method of multi-source data-driven hybrid mechanism learning model | |
| Doraiswamy et al. | Operational prediction of crop yields using MODIS data and products | |
| Song et al. | Estimating reed loss caused by Locusta migratoria manilensis using UAV-based hyperspectral data | |
| CN112819227A (en) | County scale winter wheat yield per unit prediction method and system | |
| CN115577866A (en) | Method and device for predicting waiting period, electronic equipment and storage medium | |
| CN116485757A (en) | Winter wheat total nitrogen content prediction method | |
| CN117077868A (en) | Model-based grain yield prediction system and method | |
| Tang et al. | Estimating wheat grain yield by assimilating phenology and LAI with the WheatGrow model based on theoretical uncertainty of remotely sensed observation | |
| CN117694070A (en) | Nutrient element inversion evaluation and intelligent variable accurate fertilization decision system | |
| CN117172385B (en) | Sugarcane high-sugar-content harvest period prediction method and system | |
| Wang et al. | Monitoring model for predicting maize grain moisture at the filling stage using NIRS and a small sample size | |
| Rahimi Jamnani et al. | Optimization of sugarcane harvest using remote sensing | |
| CN116735538B (en) | Crop nitrogen accumulation remote sensing prediction method based on nitrogen distribution model | |
| CN112749837B (en) | Method for estimating cereal crop yield based on post-flower NDVI accumulation value | |
| Vinayaka et al. | AI-Enhanced Remote Sensing Applications in Indian Sugarcane Research: A Comprehensive Review | |
| Data | Crop yield estimation using decision trees and random forest machine learning algorithms on data from terra | |
| CN117391221B (en) | NDVI prediction integrated optimization method and system based on machine learning | |
| Dodge | Digital Phenotyping in Cotton Breeding Using Growth Rate Modeling Based on Visible Light Data Collected With Unmanned Aerial Systems | |
| CN118262234A (en) | Remote sensing estimation method for sugar cane yield of land parcel scale | |
| Patane et al. | Chlorophyll and nitrogen estimation of grape using colour image processing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |