A method for predicting agricultural yield under drought disaster

Abstract

The present invention discloses a method for predicting agricultural yield under drought disasters, proposes an agricultural yield prediction model under drought disaster stress, predicts the formation time of drought events under different meteorological conditions through a random forest algorithm, converts the drought formation time into a drought formation rate, and then constructs a relationship between the drought formation rate and the yield loss rate. This method can more accurately obtain the yield loss of agricultural crops under drought disasters, and provides more accurate information for early warning of agricultural drought prevention and prediction of the impact of drought on crops.

Description

Technical Field

[0001] The present invention relates to a method for predicting agricultural yield, in particular to a method for predicting agricultural yield under drought disaster. Background Art

[0002] Crop growth typically encompasses three sensitive phenological periods: the vegetative, reproductive, and mature stages. The impact of drought during these periods can significantly affect final yield. Existing drought-induced agricultural yield predictions rely on crop growth models, addressing the problem of predicting the impact of drought during different phenological periods. However, the accuracy of these predictions relies on growth models, which require specific crop models. Furthermore, driven by multiple influencing factors, the accuracy of these growth models needs to be improved. Long-term crop growth monitoring is required, and the use of this data to refine predictions. Furthermore, this approach fails to account for the impact of diverse crop production systems. Different regions employ different farming methods, primarily irrigated and rainfed, which vary in their resistance to drought stress. Improving the accuracy of agricultural yield predictions during droughts and providing a more accurate picture of crop impacts is a crucial issue in agricultural planting, risk assessment, and drought relief efforts. Summary of the Invention

[0003] Purpose of the invention: In response to the above problems, the present invention proposes a method for predicting agricultural yield under drought disasters, which can better reflect the impact of drought disasters on crops and more accurately predict crop yield losses.

[0004] Technical solution: The technical solution adopted by the present invention is a method for predicting agricultural yield under drought disasters, comprising the following steps:

[0005] Step 1: Acquire basic data, including agricultural data, meteorological data, and soil water data; continuously monitor soil water data at the location of the crop to be predicted, and predict agricultural yield when the soil water data drops to a drought event threshold or is confirmed to have entered a drought event;

[0006] Step 2: For agricultural yield predictions when soil water data drops to the drought event threshold, the drought onset time is calculated using the drought onset time prediction method based on meteorological data and soil water data. For agricultural yield predictions when a drought event is confirmed, the drought onset time is directly calculated based on the soil water data. The drought onset rate is then calculated based on the drought onset time.

[0007] Step 3: Based on the drought formation rate, the crop growth stage at the onset of the drought event, and the crop planting system used by the crop, a yield loss prediction model is used to obtain the predicted yield loss rate of the crop;

[0008] The yield loss prediction model is:

[0009] ,

[0010] Where, represents the yield loss rate, n represents the nth growth stage of the crop; represents the sensitivity of crops to drought at various growth stages; a is a constant representing the cropping system used for the crop; represents the drought formation rate;

[0011] Step 4: Calculate the crop yield prediction value based on the crop yield loss rate prediction.

[0012] The agricultural data are the historical yield data and phenological data of crops in the study area; the meteorological data include precipitation, temperature and evaporation data in the study area; and the soil water data use surface soil water data.

[0013] The growth stages of crops include: vegetative period, reproductive period and maturity period.

[0014] The method for predicting the onset time of drought includes: predicting the onset time of drought through a random forest model based on meteorological data and soil water data; establishing the random forest model includes: obtaining historical soil water data and performing drought identification to obtain the drought onset time of each drought event; obtaining historical meteorological data to obtain the average precipitation for a period of time before the onset of each drought event , average temperature Average evapotranspiration ,in i Indicates that the value is taken according to the growth stage of the crop, y is the year; the average precipitation , average temperature Average evapotranspiration As input, the drought formation time As output, a random forest model is trained and validated.

[0015] The drought formation rate is calculated based on the drought formation time. The calculation formula is:

[0016] ,

[0017] Where, represents the drought formation rate, The time when drought forms.

[0018] In the yield loss prediction model, the parameter representing the sensitivity of crops to drought at different growth stages , calculated by particle swarm method, the objective function is to minimize the error between the predicted value and the actual value of the yield loss rate.

[0019] The development of the yield loss prediction model includes the following steps:

[0020] Step 31, obtaining basic data, including agricultural data, meteorological data, and soil water data;

[0021] Step 32, obtaining drought formation time based on soil water data and calculating drought formation rate;

[0022] Step 33, calculate the actual value of the yield loss rate, the calculation formula is:

[0023] ,

[0024] in, is the actual value of the yield loss rate, is the crop yield in the yth year when drought occurs, and Y is the theoretical yield of the crop, which is obtained by trend analysis of historical yield data;

[0025] Step 34: Based on the drought formation rate, the crop growth stage at the beginning of the drought event, and the crop planting system, a yield loss prediction model is established. The actual value of the yield loss rate obtained in step 33 is used as the dependent variable of the yield loss prediction model. The parameters of the yield loss prediction model are obtained by particle swarm analysis. The optimal solution of .

[0026] The crop yield forecast value is calculated based on the predicted crop yield loss rate. The calculation formula is:

[0027] ,

[0028] Where, is the predicted value of crop yield, and Y is the theoretical yield of the crop.

[0029] The present invention proposes a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for predicting agricultural yield under drought disasters is implemented.

[0030] The present invention provides a computer program product, comprising a computer program and / or instructions, which implement the method for predicting agricultural yield under drought disasters when executed by a processor.

[0031] Beneficial effects: Compared with the prior art, the present invention has the following advantages: The method for predicting agricultural yield under drought disasters described in the present invention does not rely on the growth model for the accuracy of the prediction results, nor does it need to be modeled for different crops, and the accuracy and timeliness are greatly improved. Driven by a variety of influencing factors, the prediction model proposed by the present invention is based on a large amount of historical data research. We found that if drought forms in a short period of time, soil water will decrease rapidly, which will have a great impact on crop growth and final yield. It is believed that the formation characteristics of drought should be the primary consideration in the yield prediction model. In addition, the method of the present invention also takes into account the impact of different crop production systems, as well as the multiple impacts of multiple small-scale droughts on crop yields within a growth cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 The present invention is a flowchart of the method for predicting agricultural yield under drought disaster. DETAILED DESCRIPTION

[0033] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0034] The method for predicting agricultural yield under drought disaster of the present invention has the following process: Figure 1 shown.

[0035] Step 1: Obtain basic data, including agricultural data, meteorological data, and soil moisture data. Continuously monitor soil moisture data for the crop locations to be predicted. Agricultural yield forecasts are performed when soil moisture data drops to the drought event threshold (i.e., the soil moisture quantile value has dropped to 20%), or when a drought event is confirmed (i.e., the soil moisture quantile value has dropped to 40%).

[0036] Among the basic data, agricultural data include historical crop yield data and phenological data in the study area; meteorological data include precipitation, temperature and evaporation data in the study area; soil water data include surface soil water.

[0037] The duration of each growth stage of the crop (usually three, vegetative period, reproductive growth period and maturity period) is obtained from the crop phenological data.

[0038] Specifically, surface soil water was used as soil water data. The drought event threshold was the 40th percentile soil water content. This quantile soil water content was obtained by preprocessing the soil water data and fitting it with an empirical distribution function.

[0039] Confirming the onset of a drought event involves preprocessing soil water data and converting them into soil water content quantile values ​​through fitting an empirical distribution function. Based on the soil water content quantile values, drought events are identified using the following rules: (1) A drought event is considered to have begun when the soil water content quantile is below 40%. (2) A drought event must have at least one moment in which the soil water content quantile is below 20%. (3) The drought must have lasted for at least two weeks.

[0040] Step 2: For agricultural yield prediction when soil water data drops to the drought event threshold, the drought onset time is obtained using the drought onset time prediction method. For agricultural yield prediction when a drought event is confirmed, the drought onset time is directly calculated based on soil water data. The drought onset rate is calculated based on the drought onset time. .

[0041] First, define when droughts develop , which means the duration of the soil moisture quantile decreasing from 40% to 20% in a certain identified drought event. Among the drought onset prediction methods, a solution that takes both accuracy and timeliness into account is to obtain historical soil moisture data, perform drought identification, and obtain the drought onset time of each drought event. Obtain historical meteorological data and obtain the average precipitation 30 days before the start of each drought event. , average temperature Average evapotranspiration . i The value is determined based on the growth stage of the crop. For example, if the crop has three growth stages, i Take 1, 2, and 3 in turn; y is a certain year. , average temperature Average evapotranspiration As an independent variable, the drought formation time As the dependent variable, the random forest model was used to model the data, and the data was divided into a training set and a validation set with a ratio of 7:3. Finally, the average precipitation for the 30 days before the start of the drought event corresponding to this agricultural yield prediction was obtained. , average temperature Average evapotranspiration ,Through the trained random forest model, the predicted drought formation events are output.

[0042] It is foreseeable that drought onset prediction methods can employ any other method capable of obtaining drought onset time. In the event of a severe drought, a more accurate drought onset prediction method is preferred, such as the drought onset prediction method using a conditional probability function disclosed in patent CN2024102894426. Other existing neural network or deep learning models can also be used to predict drought onset time.

[0043] When a drought event is confirmed, the drought formation time is directly calculated based on the soil water data. The period from the beginning of the drought event to the time when the soil moisture quantile is lower than 20% is the drought formation time.

[0044] For the prediction when the soil water data drops to the drought event threshold, the drought event starts at the time when the soil water data drops to the drought event threshold.

[0045] Step 3: Based on the drought formation rate, the crop growth stage at the beginning of the drought event, and the crop planting system used by the crop, the predicted crop yield loss rate is obtained through the yield loss prediction model.

[0046] The development of the yield loss prediction model includes the following steps:

[0047] Step 31: Obtain basic data, including agricultural data, meteorological data, and soil and water data. See Step 1 for details.

[0048] Step 32: Obtain the drought formation time and calculate the average drought formation rate. Obtain the drought formation time based on historical soil water data. The definition of the drought formation time is detailed in step 2.

[0049] After obtaining the drought formation time, calculate the drought formation rate , using the average drought formation rate formula, as follows:

[0050] ,

[0051] Step 33, calculate the yield loss rate.

[0052] Based on historical yield data and identified drought events, the actual yield loss rate can be calculated , the formula is as follows:

[0053] ,

[0054] in, is the crop yield in the yth year when drought occurs, and Y is the theoretical yield of the crop, which is obtained by trend analysis of historical yield data.

[0055] Step 34 , based on the drought formation rate, the crop growth stage at the onset of the drought event, and the crop planting system used, a yield loss prediction model is established, and the parameters in the model are solved using a particle swarm optimization method to obtain the yield loss prediction model.

[0056] Establish a yield loss prediction model to calculate the yield loss rate The predicted value is as follows:

[0057] ,

[0058] Where n represents the nth stage of the crop; Represents the sensitivity of crops to drought at various growth stages, obtained by parameter calibration: Calculate the yield loss rate Afterwards, Using random value method and particle swarm method, the predicted value of yield loss rate is continuously compared with the actual value. The comparison index can be variance or other indicators that can evaluate the error between the two, and finally the predicted value is closest to the actual value, that is, the error is minimized. The optimal value of a represents different planting systems. When rain-fed, a=0.2; when rain-fed and irrigation-mixed, a=0.15; when irrigation is used, a=0.1.

[0059] Step 4: Calculate the crop yield prediction value based on the crop yield loss rate prediction.

[0060] Calculate the predicted yield of crops using the following formula:

[0061] ,

[0062] Where, is the predicted value of crop yield, Y is the theoretical yield of the crop, The yield loss rate of crops is output by the yield loss prediction model.

Claims

1. A method for predicting agricultural yield under drought disaster, characterized in that: The following steps are involved: Step 1: Acquire basic data, including agricultural data, meteorological data, and soil water data; continuously monitor soil water data at the location of the crop to be predicted, and predict agricultural yield when the soil water data drops to a drought event threshold or is confirmed to have entered a drought event; Step 2: For agricultural yield predictions when soil water data drops to the drought event threshold, the drought onset time is calculated using the drought onset time prediction method based on meteorological data and soil water data. For agricultural yield predictions when a drought event is confirmed, the drought onset time is directly calculated based on the soil water data. The drought onset rate is then calculated based on the drought onset time. Step 3: Based on the drought formation rate, the crop growth stage at the onset of the drought event, and the crop planting system used by the crop, a yield loss prediction model is used to obtain the predicted yield loss rate of the crop; The yield loss prediction model is: ， Where, represents the yield loss rate, n represents the nth growth stage of the crop; represents the sensitivity of crops to drought at various growth stages; a is a constant representing the cropping system used for the crop; represents the drought formation rate; Step 4: Calculate the crop yield prediction value based on the crop yield loss rate prediction.

2. The method for predicting agricultural yield under drought disaster according to claim 1, characterized in that: The agricultural data are the historical yield data and phenological data of crops in the study area; the meteorological data include precipitation, temperature and evaporation data in the study area; and the soil water data use surface soil water data.

3. The method for predicting agricultural yield under drought disaster according to claim 1, characterized in that: The growth stages of crops include: vegetative period, reproductive period and maturity period.

4. The method for predicting agricultural yield under drought disaster according to claim 1, characterized in that: The method for predicting the onset time of drought includes: predicting the onset time of drought through a random forest model based on meteorological data and soil water data; establishing the random forest model includes: obtaining historical soil water data and performing drought identification to obtain the drought onset time of each drought event; obtaining historical meteorological data to obtain the average precipitation for a period of time before the onset of each drought event , average temperature Average evapotranspiration ,in i Indicates that the value is taken according to the growth stage of the crop, y is the year; the average precipitation , average temperature Average evapotranspiration As input, the drought onset time As output, a random forest model is trained and validated.

5. The method for predicting agricultural yield under drought disaster according to claim 1, characterized in that: The drought formation rate is calculated based on the drought formation time. The calculation formula is: ， Where, represents the drought formation rate, The time when drought forms.

6. The method for predicting agricultural yield under drought disaster according to claim 1, characterized in that: In the yield loss prediction model, the parameter representing the sensitivity of crops to drought at different growth stages , calculated by particle swarm method, the objective function is to minimize the error between the predicted value and the actual value of the yield loss rate.

7. The method for predicting agricultural yield under drought disaster according to claim 1, characterized in that: The development of the yield loss prediction model includes the following steps: Step 31, obtaining basic data, including agricultural data, meteorological data, and soil water data; Step 32, obtaining drought formation time based on soil water data and calculating drought formation rate; Step 33, calculate the actual value of the yield loss rate, the calculation formula is: ， in, is the actual value of the yield loss rate, is the crop yield in the yth year when drought occurs, and Y is the theoretical yield of the crop; Step 34: Based on the drought formation rate, the crop growth stage at the beginning of the drought event, and the crop planting system, a yield loss prediction model is established. The actual value of the yield loss rate obtained in step 33 is used as the dependent variable of the yield loss prediction model. The parameters of the yield loss prediction model are obtained by particle swarm analysis. The optimal solution of .

8. The method for predicting agricultural yield under drought disaster according to claim 1, characterized in that: The crop yield forecast value is calculated based on the predicted crop yield loss rate. The calculation formula is: ， Where, is the predicted value of crop yield, and Y is the theoretical yield of the crop.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for predicting agricultural yield under drought disaster according to any one of claims 1 to 8 is implemented.

10. A computer program product comprising a computer program and / or instructions, characterized in that When the computer program and / or instructions are executed by a processor, the method for predicting agricultural yield under drought disaster according to any one of claims 1 to 8 is implemented.

Images