Method for predicting machine-harvested cotton yield based on temperature-light collaborative model

Abstract

The invention relates to the technical field of agricultural management, in particular to a method for predicting the yield of machine-harvested cotton based on a temperature-light collaborative model, and the method comprises the following steps: S1, multi-source data collection: collecting meteorological data, remote sensing data and ground agricultural data of a target cotton field; s2, temperature and light cooperation index calculation: calculating an effective temperature cumulant, an effective radiation cumulant and a leaf area index based on meteorological data, remote sensing data and ground agronomic data of the target cotton field, and calculating a temperature and light cooperation index based on the effective temperature cumulant and the effective radiation cumulant; s3, constructing a comprehensive prediction model: constructing the comprehensive prediction model by using a gradient lifting decision tree algorithm; and S4, yield and quality prediction: inputting the temperature-light synergy index as an input feature into the comprehensive prediction model to obtain a yield prediction value and a quality prediction value of the machine-harvested cotton. The method can quantify the synergistic effect of the temperature and illumination on the cotton through the design of the temperature-light synergistic index, and improves the prediction accuracy.

Description

Technical Field

[0001] This invention relates to the field of agricultural management technology, specifically to a method for predicting the yield of machine-harvested cotton based on a temperature-light synergy model. Background Technology

[0002] As one of the world's most important economic crops, cotton is facing increasing challenges due to accelerated agricultural modernization and rising labor costs, making mechanized harvesting an inevitable trend in its production development. The yield formation process of machine-harvested cotton is more concentrated and more sensitive to dynamic changes in environmental factors such as temperature and light under different varieties and planting methods. Therefore, achieving accurate yield prediction for machine-harvested cotton and developing a reliable and efficient technology for predicting yield and quality is of paramount practical significance for planting planning, optimizing the allocation and management of water and fertilizer resources in cotton production, ensuring the supply and demand balance in the raw cotton market, stabilizing market prices, improving the profitability of cotton farmers and the overall economic benefits of the industry, and promoting the coordinated development of the entire industrial chain.

[0003] The yield and quality of machine-harvested cotton are deeply influenced by the coupled and synergistic effects of temperature and light. The impact of temperature and light on cotton yield is a core physiological and ecological process. Cotton yield is essentially the result of the accumulation and distribution of dry matter produced by photosynthesis in different organs (especially the boll), and temperature and light are the two most critical environmental energy factors driving this process. The influence of temperature and light on cotton yield spans the entire growth period of cotton; its effects are not simply additive but a complex process of synergistic interaction between temperature and light. It also exhibits significant growth period specificity. For example, insufficient light leads to excessive vegetative growth, elongated internodes, and a loose plant structure. Low temperatures during the flowering and boll-forming stage inhibit pollen tube elongation, resulting in poor pollination and increased boll shedding. High temperatures or cloudy / rainy conditions directly lead to decreased photosynthetic enzyme activity, fertilization failure, or stunted boll development, thus significantly increasing the shedding rate. Even if bolls form, their weight and quality will decline. Suitable temperature and sufficient light are crucial conditions for ensuring high yield and quality; conversely, inadequate conditions can easily lead to reduced yield and poor fiber development. Therefore, environmental factors such as temperature and light do not act independently, but rather work synergistically and interactively to regulate physiological processes such as photosynthesis, material accumulation and distribution, thereby regulating the formation of cotton yield and quality.

[0004] Existing methods for predicting machine-harvested cotton yield based on temperature-light synergy models, such as the WOFOST model, simulate the potential growth and water-limited growth levels of crops by inputting daily meteorological data (such as solar radiation, temperature, and precipitation), soil data (such as water characteristics and nutrient status), and crop parameters (such as varietal characteristics and growth stage). This method is theoretically sound, with clearly defined physiological and physical mechanisms, thus predicting the yield and quality of machine-harvested cotton.

[0005] The formation of cotton yield and quality is deeply influenced by the synergistic effects of environmental factors such as temperature and light. While existing prediction methods consider the effects of temperature and light on cotton yield and quality separately, they neglect the synergistic effect of temperature and light, failing to account for the temperature and light transmission patterns within the population microenvironment. This results in a failure to accurately reflect the quantitative relationship between temperature-light coupling parameters and key physiological indicators during the critical growth stages of machine-harvested cotton, leading to significant discrepancies between predicted and actual results. Therefore, it is necessary to construct a temperature-light synergistic prediction model for machine-harvested cotton with a clear mechanism, reliable parameters, and strong universality. This model should quantify the synergistic effect of temperature and light, closely integrate with the growth and development characteristics of machine-harvested cotton, and improve the accuracy of prediction results. The aim is to achieve more accurate, stable, and forward-looking predictions of machine-harvested cotton yield, providing core technological support for precise management and intelligent decision-making in the modern cotton industry. Summary of the Invention

[0006] To address the aforementioned issues, this invention provides a method for predicting machine-harvested cotton yield based on a temperature-light synergy model. By designing temperature-light synergy indicators, the synergistic effects of temperature and light on cotton yield and quality can be effectively quantified. Through the construction of a comprehensive prediction model, using temperature-light synergy indicators as the core, and combining historical data of the target cotton field with current data, cotton yield and quality can be predicted more comprehensively, thereby improving the accuracy of the prediction results.

[0007] To achieve the above objectives, the technical solution of the present invention is as follows: A method for predicting machine-harvested cotton yield based on a temperature-light synergy model, comprising the following steps:

[0008] S1, Multi-source data acquisition: Collect meteorological data, remote sensing data and ground agronomic data of the target cotton field.

[0009] S2, Temperature-Light Synergy Index Calculation: Based on meteorological data, remote sensing data and ground agronomic data of the target cotton field, calculate the effective temperature accumulation, effective radiation accumulation, leaf area index and water stress index, and calculate the temperature-light synergy index based on the effective temperature accumulation, effective radiation accumulation, leaf area index and water stress index.

[0010] S3, Construction of the comprehensive prediction model: The comprehensive prediction model is constructed using the gradient boosting decision tree algorithm; the comprehensive prediction model is trained using meteorological data, remote sensing data and ground agronomic data from several cotton fields to obtain the comprehensive prediction model after training.

[0011] S4, Yield and Quality Prediction: The temperature and light synergy index is used as an input feature and input into the comprehensive prediction model to obtain the yield and quality prediction values ​​of machine-harvested cotton.

[0012] Furthermore, in S1, the meteorological data includes the daily total solar radiation, the daily maximum temperature, and the daily minimum temperature; in S2, the effective temperature accumulation is calculated based on the daily maximum temperature and the daily minimum temperature, and the effective radiation accumulation is calculated based on the daily total solar radiation.

[0013] Furthermore, in S1, the remote sensing data includes multispectral images; in S2, the leaf area index is calculated based on the multispectral images.

[0014] Furthermore, in S1, ground agronomic data include planting density, cotton variety, soil moisture content, and field capacity. The water stress index is calculated based on soil moisture content and field capacity.

[0015] Furthermore, in S2, the formula for calculating the effective temperature accumulation is as follows:

[0016] CGDD=D·[(Tmax+Tmin) / 2-Tbase] (1).

[0017] Wherein, CGDD is the effective temperature accumulation, D is the number of cotton growing days, Tmax is the average daily maximum temperature over the growing days, Tmin is the average daily minimum temperature over the growing days, and Tbase is the cotton base temperature.

[0018] Furthermore, in S2, the formula for calculating the effective cumulative radiation is as follows:

[0019] CPAR = D·total solar radiation per day / 2 (2).

[0020] CPAR represents the effective cumulative radiation.

[0021] Furthermore, in S2, the formula for calculating the water stress index is as follows:

[0022] WSI = 1 - (W1 / W2) (3).

[0023] Wherein, WSI is the water stress index, W1 is the soil moisture content, and W2 is the field capacity.

[0024] Furthermore, in S2, the formula for calculating the temperature-light synergy index is as follows:

[0025] TCI=(CGDD·CPAR·LAI) / (Var+WSI) (4).

[0026] Among them, TCI is the temperature-light synergy index, Var is the canopy temperature variance, and LAI is the leaf area index.

[0027] Furthermore, in S4, the input features for calculating the cotton yield forecast include the historical average yield, historical average density, historical temperature-light synergy index, current temperature-light synergy index, and current planting density of the same cotton variety in the target cotton field.

[0028] The formula for calculating cotton yield forecasts is as follows:

[0029] Y1=(TCI1 / TCI0)×(M1 / M0)×Y0 (5).

[0030] Where Y1 is the predicted yield, TCI1 is the current temperature-light synergy index, TCI0 ​​is the historical temperature-light synergy index, M1 is the current planting density, M0 is the historical average planting density, and Y0 is the historical average yield.

[0031] Furthermore, in S4, the input features for calculating the predicted cotton quality value include historical quality parameters of the same cotton variety in the target cotton field, historical temperature-light synergy index, and current temperature-light synergy index; the predicted cotton quality value includes at least one of average fiber length, breaking strength, and micronaire value; historical quality parameters include historical average fiber length, historical breaking strength, and historical micronaire value; historical average fiber length, historical breaking strength, and historical micronaire value are all obtained by sampling and measuring cotton.

[0032] F1=(TCI1 / TCI0)×S0 (6).

[0033] Where F1 is the predicted quality value, TCI1 is the current temperature-light synergy index, and S0 is any one of the historical average fiber length, historical breaking strength, and historical micronaire value.

[0034] The above approach has the following beneficial effects:

[0035] 1. Existing prediction methods, while considering the impact of light and temperature on cotton yield, lack consideration of the synergistic effect of light and temperature, resulting in a large error between the predicted and actual results. This method constructs a temperature-light synergy index by correlating the temperature and light changes of cotton and combining the leaf area index of the plant and the soil water stress index, which effectively quantifies the synergistic effect and degree of influence of temperature, light, and soil on the plant.

[0036] 2. This method uses the temperature-light synergy index to predict the current cotton yield of the target cotton field based on its historical yield and changes in planting density. It also uses the temperature-light synergy index to predict the current average fiber length, breaking strength, and micronaire value of the cotton in the target cotton field based on historical quality parameters. Because the temperature-light synergy index is designed to fully consider the synergistic effects and influences of temperature, light, and soil on the plants, the accuracy of the prediction is improved.

[0037] 3. This method, through the design of a comprehensive prediction model, requires users to collect data and input it into the model. The model will then automatically output the predicted yield and quality of cotton, improving the intelligence and automation of cotton yield and quality prediction. In addition, existing prediction methods often only predict either cotton yield or quality, while this method can achieve dual prediction of both yield and quality, improving the comprehensiveness of the prediction.

[0038] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0039] Figure 1 This is a schematic diagram illustrating the steps of the method for predicting machine-harvested cotton yield based on a temperature-light synergy model according to the present invention.

[0040] Figure 2 This is a functional diagram of S1 in the method for predicting machine-harvested cotton yield based on the temperature-light synergy model of the present invention.

[0041] Figure 3 This is a functional diagram of S2 in the method for predicting machine-harvested cotton yield based on the temperature-light synergy model of the present invention.

[0042] Figure 4 This is a functional diagram of S4 in the method for predicting machine-harvested cotton yield based on the temperature-light synergy model of the present invention. Detailed Implementation

[0043] The following detailed description illustrates the specific implementation method:

[0044] Implementation, for example Figure 1 As shown, a method for predicting machine-harvested cotton yield based on a temperature-light synergy model includes the following steps:

[0045] like Figure 2 As shown, S1, multi-source data acquisition: collecting meteorological data, remote sensing data, and ground agronomic data of the target cotton field. Meteorological data includes daily total solar radiation, daily maximum temperature, and daily minimum temperature; remote sensing data includes multispectral imagery; ground agronomic data includes planting density, cotton variety, soil moisture content, and field capacity.

[0046] like Figure 3As shown in Figure S2, the temperature-light synergy index is calculated as follows: Based on meteorological data, remote sensing data, and ground agronomic data of the target cotton field, the effective temperature accumulation, effective radiation accumulation, leaf area index, and water stress index are calculated. The temperature-light synergy index is then calculated based on the effective temperature accumulation, effective radiation accumulation, leaf area index, and water stress index. Specifically, the effective temperature accumulation is calculated based on the daily maximum and minimum temperatures; the effective radiation accumulation is calculated based on the daily total solar radiation; the leaf area index is calculated based on multispectral imagery; and the water stress index is calculated based on soil moisture content and field capacity.

[0047] The formula for calculating the effective temperature accumulation is as follows:

[0048] CGDD=D·[(Tmax+Tmin) / 2-Tbase] (1).

[0049] Wherein, CGDD is the effective temperature accumulation, D is the number of cotton growing days, Tmax is the average daily maximum temperature over the growing days, Tmin is the average daily minimum temperature over the growing days, and Tbase is the cotton base temperature.

[0050] Specifically, the base temperature of cotton varies in different regions. In this embodiment, the base temperature of cotton is 12℃. Assuming that the current daily maximum temperature is 30℃ and the minimum temperature is 18℃, and the cotton growing days are 30 days, according to formula (1), the effective temperature accumulation is 1250℃. The larger the effective temperature accumulation, the greater the effective heat accumulated by the cotton, and the higher the cotton yield will be.

[0051] The formula for calculating the effective cumulative radiation is as follows:

[0052] CPAR = D·total solar radiation per day / 2 (2).

[0053] CPAR represents the effective cumulative radiation.

[0054] Specifically, assuming the total daily solar radiation is 20 MJ / m², according to formula (2), the effective cumulative radiation is 300 MJ / m². The greater the effective cumulative radiation, the greater the effective radiation absorbed by the cotton, and the higher the cotton yield will be.

[0055] The formula for calculating the water stress index is as follows:

[0056] WSI = 1 - (W1 / W2) (3).

[0057] Wherein, WSI is the water stress index, W1 is the soil moisture content, and W2 is the field capacity.

[0058] Specifically, WSI=0 indicates that the soil moisture content is at field capacity, the water supply is sufficient, and the crop is not under water stress. 0<WSI≤1 indicates that water stress exists. The larger the value, the more severe the stress and the more severe the water shortage. Assuming that the soil moisture content of the target cotton field is 18% and the field capacity is 28%, according to formula (3), the water stress index is 0.36. The lower the effective soil moisture content, the more water-deficient the soil.

[0059] The formula for calculating the temperature-light synergy index is as follows:

[0060] TCI=(CGDD·CPAR·LAI) / (Var+WSI) (4).

[0061] Among them, TCI is the temperature-light synergy index, Var is the canopy temperature variance (obtained by calculating the variance of the canopy temperature in the target cotton field), and LAI is the leaf area index.

[0062] Specifically, (CGDD·CPAR·LAI) represents the positive effects of temperature, light, and leaf area on cotton yield and quality in the target cotton field under conditions of no temperature or water stress, while (Var+WSI) represents the negative effects of temperature and water stress on cotton yield and quality in the target cotton field. Assuming that the target cotton field has CGDD=1250℃, CPAR=850MJ / m², LAI=3.8, and Var=4.2℃², according to formula (4), TCI=885,000. This value is dimensionless and is used to reflect the impact of the current environment on the target cotton field.

[0063] S3, Construction of the comprehensive prediction model: The comprehensive prediction model is constructed using the gradient boosting decision tree algorithm (XGBoost model is selected in this embodiment); the comprehensive prediction model is trained using meteorological data, remote sensing data and ground agronomic data of several cotton fields to obtain the comprehensive prediction model after training.

[0064] In this embodiment, a gradient boosting decision tree algorithm (such as XGBoost) is used to construct a comprehensive prediction model.

[0065] The model's input features include:

[0066] 1. Temperature-Light Synergy Index (TCI);

[0067] 2. Cumulative effective temperature (CGDD);

[0068] 3. Cumulative Effective Radiation (CPAR);

[0069] 4. Leaf Area Index (LAI);

[0070] 5. Water Stress Index (WSI);

[0071] 6. Canopy temperature variance (Var);

[0072] 7. Planting density;

[0073] 8. Cotton varieties (coded as category variables).

[0074] The model's prediction objective is:

[0075] 1. Cotton yield per unit area (kg / hectare);

[0076] 2. Cotton quality indicators (such as average fiber length, breaking strength, micronaire value).

[0077] The model training process includes:

[0078] 1. Data preprocessing: Imput missing values, standardize numerical features, and perform one-hot encoding on categorical features.

[0079] 2. Dataset partitioning: The historical dataset is divided into training and validation sets in a 7:3 ratio.

[0080] 3. Model Training: Train the XGBoost model using the training set and adjust hyperparameters (such as learning rate, tree depth, subsampling ratio, etc.) through cross-validation to minimize prediction errors (such as mean squared error, MSE).

[0081] 4. Model Validation: Use the validation set to evaluate model performance, calculate metrics such as the coefficient of determination (R²) and root mean square error (RMSE), and ensure the model's generalization ability.

[0082] 5. Model Saving: Save the trained model as a callable file for subsequent predictions.

[0083] like Figure 4 As shown, S4, yield and quality prediction: The temperature and light synergy index is used as an input feature and input into the comprehensive prediction model to obtain the yield and quality prediction values ​​of machine-harvested cotton.

[0084] The prediction process is as follows:

[0085] First, the trained comprehensive prediction model (XGBoost model) is invoked.

[0086] Secondly, construct the input feature vector for the current cotton field, including:

[0087] 1. Current Temperature-Light Synergy Index (TCI1);

[0088] 2. Current effective temperature accumulation (CGDD);

[0089] 3. Current Cumulative Effective Radiation (CPAR);

[0090] 4. Current Leaf Area Index (LAI);

[0091] 5. Current Water Stress Index (WSI);

[0092] 6. Current canopy temperature variance (Var);

[0093] 7. Current planting density (M1);

[0094] 8. Cotton Variety Code

[0095] Then, the feature vectors are input into the model, and the model output is:

[0096] 1. Production forecast (Y1);

[0097] 2. Quality prediction value (e.g., breaking strength F1);

[0098] Finally, the results are output and visualized or reported to support production decisions.

[0099] The input features for calculating cotton yield forecasts include the historical average yield, historical average density, historical temperature-light synergy index, current temperature-light synergy index, and current planting density for the same cotton variety in the target cotton field.

[0100] The formula for calculating cotton yield forecasts is as follows:

[0101] Y1=(TCI1 / TCI0)×(M1 / M0)×Y0 (5).

[0102] Where Y1 is the predicted yield, TCI1 is the current temperature-light synergy index, TCI0 ​​is the historical temperature-light synergy index, M1 is the current planting density, M0 is the historical average planting density, and Y0 is the historical average yield.

[0103] Specifically, assuming the historical average yield of the target cotton field (Y0) = 6000 kg / ha; historical average temperature-light synergy index (TCI0) = 800,000; historical average planting density (M0) = 150,000 plants / ha; current planting density (M1) = 155,000 plants / ha; current temperature-light synergy index (TCI1) = 885,000; according to formula (5), the predicted yield is 6852 kg / ha; due to the optimization of the temperature and light environment and the reasonable increase in planting density, the current predicted yield of cotton is higher than the historical average yield.

[0104] The input features for calculating cotton quality predictions include historical quality parameters of the same cotton variety in the target cotton field, historical temperature-light synergy index, and current temperature-light synergy index; the cotton quality predictions include at least one of average fiber length, breaking strength, and micronaire value; historical quality parameters include historical average fiber length, historical breaking strength, and historical micronaire value; historical average fiber length, historical breaking strength, and historical micronaire value are all obtained through cotton sampling measurements.

[0105] F1=(TCI1 / TCI0)×S0 (6).

[0106] Where F1 is the predicted quality value, TCI1 is the current temperature-light synergy index, and S0 is any one of the historical average fiber length, historical breaking strength, and historical micronaire value.

[0107] Specifically, assuming the historical average breaking strength (S0) of the target cotton field is 30.0 cN / tex; the historical average temperature-light synergy index (TCI0) is 800,000; and the current temperature-light synergy index (TCI1) is 950,000; according to formula (6), the current average breaking strength of cotton is 35.63 cN / tex; due to the excellent light, temperature, water and heat configuration this year, the cotton fibers are well developed and the cellulose is fully deposited, so the average breaking strength in the predicted quality value of cotton is better than the historical average breaking strength.

[0108] Specific experiments are as follows:

[0109] I. Experimental Preparation

[0110] Experimental group: Input the XGBoost model using the input features from this method.

[0111] Control group: All original features (CGDD, CPAR, LAI, Var, WSI, planting density and cotton variety) were used as input features to input the XGBoost model, and the temperature-light synergy index (TCI) was not constructed.

[0112] Record the predicted yield and quality values ​​for both groups, as well as the actual yield and quality of the target cotton field (in this experiment, the breaking strength is expressed as the breaking ratio).

[0113] The target cotton field was the same experimental site in Aksu region of Xinjiang.

[0114] II. Experimental Results

[0115] Table 1 Comparison of Prediction Results

[0116]

[0117] As shown in Table 1, the relative error between the predicted and actual yield of the experimental group was approximately 1.23%, while that of the control group was approximately 3.37%. The relative error between the predicted and actual quality of the experimental group was approximately 1.07%, while that of the control group was approximately 10.16%. This discrepancy may be due to the neglect of the synergistic effect of temperature and light, thus underestimating the environmental potential. Consequently, the predicted yield and quality values ​​of the control group were both lower than those of the experimental group, and the errors were greater. In contrast, the experimental group, through the design of the Temperature-Light Synergy Index (TCI), fully considered the synergistic effect of temperature and light, and reasonably assessed the impact of the environment on cotton yield and quality, thereby making the prediction results more accurate.

[0118] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for predicting machine-harvested cotton yield based on a temperature-light synergy model, characterized in that, Includes the following steps: S1, Multi-source data acquisition: Collect meteorological data, remote sensing data and ground agronomic data of the target cotton field; S2, Calculation of temperature-light synergy index: Based on meteorological data, remote sensing data and ground agronomic data of the target cotton field, calculate the effective temperature accumulation, effective radiation accumulation, leaf area index and water stress index, and calculate the temperature-light synergy index based on the effective temperature accumulation, effective radiation accumulation, leaf area index and water stress index. S3, Construction of the comprehensive prediction model: The comprehensive prediction model is constructed using the gradient boosting decision tree algorithm; the comprehensive prediction model is trained using meteorological data, remote sensing data and ground agronomic data from several cotton fields, and the trained comprehensive prediction model is obtained. S4, Yield and Quality Prediction: The temperature and light synergy index is used as an input feature and input into the comprehensive prediction model to obtain the yield and quality prediction values ​​of machine-harvested cotton.

2. The method for predicting machine-harvested cotton yield based on a temperature-light synergy model according to claim 1, characterized in that, In S1, meteorological data include daily total solar radiation, daily maximum temperature, and daily minimum temperature; In S2, the effective temperature accumulation is calculated based on the daily maximum and minimum temperatures, and the effective radiation accumulation is calculated based on the daily total solar radiation.

3. The method for predicting machine-harvested cotton yield based on a temperature-light synergy model according to claim 1, characterized in that, In S1, the remote sensing data includes multispectral imagery; In S2, the leaf area index was calculated based on multispectral images.

4. The method for predicting machine-harvested cotton yield based on a temperature-light synergy model according to claim 1, characterized in that, In S1, ground agronomic data include planting density, cotton variety, soil moisture content, and field water holding capacity; The water stress index is calculated based on soil moisture content and field capacity.

5. The method for predicting machine-harvested cotton yield based on a temperature-light synergy model according to claim 1, characterized in that, In S2, the formula for calculating the effective temperature accumulation is as follows: CGDD=D·[(Tmax+Tmin) / 2-Tbase] (1); Wherein, CGDD is the effective temperature accumulation, D is the number of cotton growing days, Tmax is the average daily maximum temperature over the growing days, Tmin is the average daily minimum temperature over the growing days, and Tbase is the cotton base temperature.

6. The method for predicting machine-harvested cotton yield based on a temperature-light synergy model according to claim 5, characterized in that, In S2, the formula for calculating the effective cumulative radiation is as follows: CPAR = D·total solar radiation per day / 2 (2); CPAR represents the effective cumulative radiation.

7. The method for predicting machine-harvested cotton yield based on a temperature-light synergy model according to claim 6, characterized in that, In S2, the formula for calculating the water stress index is as follows: WSI = 1 - (W1 / W2) (3); Wherein, WSI is the water stress index, W1 is the soil moisture content, and W2 is the field capacity.

8. The method for predicting machine-harvested cotton yield based on a temperature-light synergy model according to claim 7, characterized in that, In S2, the formula for calculating the temperature-light synergy index is as follows: TCI=(CGDD·CPAR·LAI) / (Var+WSI) (4); Among them, TCI is the temperature-light synergy index, Var is the canopy temperature variance, and LAI is the leaf area index.

9. The method for predicting machine-harvested cotton yield based on a temperature-light synergy model according to claim 8, characterized in that, In S4, the input features for calculating the cotton yield forecast include the historical average yield, historical average density, historical temperature-light synergy index, current temperature-light synergy index, and current planting density of the same cotton variety in the target cotton field. The formula for calculating cotton yield forecasts is as follows: Y1=(TCI1 / TCI0)×(M1 / M0)×Y0 (5); Where Y1 is the predicted yield, TCI1 is the current temperature-light synergy index, TCI0 ​​is the historical temperature-light synergy index, M1 is the current planting density, M0 is the historical average planting density, and Y0 is the historical average yield.

10. The method for predicting machine-harvested cotton yield based on a temperature-light synergy model according to claim 9, in S4, the input features for calculating the predicted cotton quality value include historical quality parameters of the same cotton variety in the target cotton field, historical temperature-light synergy index, and current temperature-light synergy index; the predicted cotton quality value includes at least one of average fiber length, breaking strength, and micronaire value; the historical quality parameters include historical average fiber length, historical breaking strength, and historical micronaire value; the historical average fiber length, historical breaking strength, and historical micronaire value are all obtained by sampling and measuring cotton. F1=(TCI1 / TCI0)×S0 (6); in, F1 is the predicted quality value, TCI1 is the current temperature-light synergy index, and S0 is any one of the historical average fiber length, historical breaking strength, and historical micronaire value.

Images