Irrigation water demand measurement and calculation and dominant driving factor identification method

Abstract

The invention discloses an irrigation water demand measurement and calculation and dominant driving factor identification method, and relates to the field of agricultural irrigation, and the method specifically comprises the following steps: S1, data preprocessing; s2, calculating evapotranspiration of a reference crop; s3, correcting the crop coefficient of each stage; s4, calculating the actual water demand of the crops in the growth period; s5, calculating a crop net irrigation quota; s6, calculating the irrigation water demand; s7, constructing an enhanced regression tree model; and S8, identifying main control factors influencing the irrigation water demand. The method provides a reference basis for adjustment of a planting structure and improvement of agricultural water efficiency, provides a new view angle and decision support for agricultural management, and has important guiding significance for water resource saving and modern agricultural production development.

Description

Technical Field

[0001] The invention belongs to the technical field of agricultural irrigation, and in particular relates to a method for calculating irrigation water demand and identifying dominant driving factors based on the crop water demand mechanism. Background Art

[0002] With the increasing severity of global climate change and water shortage, accurate measurement and optimal management of agricultural irrigation water demand have become particularly important. Traditional irrigation water demand measurement methods usually rely on historical meteorological data and simple empirical formulas, which are difficult to fully reflect the growth characteristics and water demand mechanism of crops. Especially in the context of climate change and increasing uncertainty, agricultural water management requires more refined and scientific decision-making basis. Therefore, how to calculate irrigation water demand based on the actual water demand mechanism of crops has become a dominant issue in modern agricultural irrigation management.

[0003] In the prior art, the common method is to calculate the crop water requirement based on the combination of reference crop evapotranspiration (ET0) and crop coefficient (Kc), but these methods usually ignore the sensitivity of crops to meteorological conditions at different growth stages and cannot accurately reflect the actual water demand of crops. In addition, many existing methods fail to fully consider the impact of multiple factors such as effective precipitation, planting area, and irrigation efficiency on irrigation water demand, resulting in the inability to provide reliable irrigation water demand prediction results under variable climate conditions, limiting their application effect in actual agricultural irrigation management.

[0004] In addition, since irrigation water demand is affected by the combined influence of climate factors and human management factors, and the influence of each factor on irrigation water demand is usually nonlinear and complex, traditional statistical methods have certain limitations in quantitatively analyzing the driving factors of changes in irrigation water demand. Machine learning and artificial intelligence algorithms can handle high-dimensional data and nonlinear problems, thus providing a new solution to this problem. The feature importance evaluation method based on machine learning algorithms can provide new research ideas for estimating the contribution rate of factors affecting irrigation water demand and understanding the combined effects of climate and human factors. Summary of the invention

[0005] The present invention proposes a method for calculating irrigation water demand and identifying the dominant driving factors. The method combines daily meteorological data, crop water demand mechanism, effective precipitation and irrigation efficiency to accurately calculate irrigation water demand. In addition, a machine learning algorithm based on boosted regression tree (BRT) is used to quantitatively evaluate the importance and interaction relationship of human factors (irrigation area, crop planting structure, irrigation efficiency) and climate factors (precipitation, temperature, wind speed, relative humidity, sunshine hours) affecting irrigation water demand, so as to identify the dominant driving factors that cause changes in irrigation water demand. This method combines traditional agricultural irrigation theory with modern machine learning algorithms to form an innovative comprehensive measurement and analysis framework for agricultural irrigation water demand. It provides a new perspective and decision-making support for agricultural management, and has important guiding significance for promoting agricultural modernization and sustainable use of water resources.

[0006] In order to achieve the above object, according to one aspect of the present invention, a method for calculating irrigation water demand and identifying dominant driving factors is provided, comprising the following steps:

[0007] Step 1: Basic data collection and processing: Collect the meteorological data and statistical data required for calculation, and pre-process the data;

[0008] Step 2: Calculate reference crop evapotranspiration based on meteorological data;

[0009] Step 3: Determine the standard crop coefficient of the crop based on the crop type and growth stage by referring to the FAO 56 manual, and then revise the crop coefficient of each stage according to meteorological data;

[0010] Step 4: Calculate the actual water requirement of the crop during the growth period based on the corrected crop coefficient and the reference crop evapotranspiration;

[0011] Step 5: Based on the precipitation data, calculate the effective precipitation, and use the crop water requirement and effective precipitation to calculate the crop net irrigation quota;

[0012] Step 6: Calculate irrigation water requirements based on crop planting area, irrigation efficiency and net irrigation quota;

[0013] As a further preferred embodiment of the present invention, the meteorological data in step 1 include average temperature, daily maximum temperature, daily minimum temperature, average wind speed, average relative humidity, minimum relative humidity, sunshine hours, and daily precipitation data, and the statistical data include crop planting area and the proportion of each crop, irrigation efficiency coefficient, and crop growth cycle.

[0014] As a further preferred embodiment of the present invention, in step 2, the Penman-Monteith formula is used to calculate the daily reference crop evapotranspiration.

[0015] As a further preferred embodiment of the present invention, in step 3, the entire growth period of the crop is divided into an early growth period, a rapid development period, a middle growth period and a late growth period according to the growth cycle of the crop. The basic crop coefficient is obtained by referring to the FAO56 manual, and then the crop coefficient of each stage is corrected according to the average 2m wind speed and the average minimum relative humidity of the crop at different growth stages.

[0016] As a further preferred embodiment of the present invention, in step 4, the corrected crop coefficient is linearly interpolated to obtain a crop coefficient curve during the crop growth period, and then the daily crop coefficient is multiplied by the reference crop evapotranspiration of the corresponding date to obtain the actual water requirement of the crop.

[0017] As a further preferred embodiment of the present invention, the effective precipitation in step 5 refers to the part of precipitation that can be effectively utilized by crops, excluding surface runoff and the part that seeps below the crop root zone, and is estimated by the empirical formula provided by the United States Department of Agriculture; and when the natural effective precipitation is not enough to meet the water demand of crops, irrigation supplement is required, so the net irrigation quota of crops is the difference between the water demand of crops and the effective precipitation.

[0018] As a further preferred embodiment of the present invention, the specific operation steps of step 6 are: multiplying the net irrigation quota of each crop obtained in steps 2 to 5 by the planting area of ​​the corresponding crop, and then multiplying it by the irrigation efficiency coefficient of the study area to obtain the irrigation water requirement of each crop, and adding up the irrigation water requirements of all crops to obtain the total irrigation water requirement of the study area.

[0019] In a second aspect, the present invention provides a method for identifying the dominant driving factor of irrigation water demand, comprising the following specific steps:

[0020] Step 7: Taking irrigation water demand as the dependent variable, comprehensively selecting two types of eight driving factors affecting irrigation water demand as independent variables, and constructing a boosted regression tree (BRT) model;

[0021] Step 8: Conduct multiple parameter sensitivity analyses on the BRT model and use the cross-validation method to determine the optimal parameter configuration. Based on the model operation results, use the contribution rate of each factor as an indicator to measure the importance of the explanatory variables, rank the importance of each influencing factor, and identify the main controlling factors affecting irrigation water demand.

[0022] As a further preferred embodiment of the present invention, the two categories of eight factors selected in step 7 include climate factors: temperature, precipitation, wind speed, relative humidity, and sunshine hours; and human factors: irrigation area, planting structure, and irrigation efficiency.

[0023] As a further preferred embodiment of the present invention, in step 8, a sensitivity analysis needs to be performed on the five key parameters of the BRT model to explore the model configuration that achieves the best prediction results. The model parameters include: sampling rate, learning rate, number of regression trees, regression tree complexity and loss function.

[0024] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the steps of the method for calculating irrigation water demand based on random sample consistency and identifying dominant driving factors of the present invention are implemented.

[0025] According to another aspect of the present invention, there is provided 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 program, the steps of the method for calculating irrigation water demand based on random sample consistency and identifying dominant driving factors of the present invention are implemented.

[0026] The beneficial effect of the present invention is that the present invention combines traditional agricultural irrigation theory with modern machine learning algorithms to form an innovative comprehensive measurement and analysis framework for agricultural irrigation water demand. Not only does it fully consider the water demand mechanism of crops to measure irrigation water demand, but it also combines machine learning algorithms to identify the leading driving factors affecting irrigation water demand, which will provide a new perspective and decision-making support for agricultural management, and has important guiding significance for water resource conservation and the development of modern agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments will be briefly introduced below. Obviously, the drawings in the following description only relate to some embodiments of the present invention, but are not intended to limit the present invention.

[0028] Figure 1 It is a flow chart of a preferred embodiment of the present invention;

[0029] Figure 2 This is a crop planting area change diagram of a preferred embodiment of the present invention;

[0030] Figure 3 This is a crop planting ratio change diagram of a preferred embodiment of the present invention;

[0031] Figure 4 This is a graph showing changes in irrigation efficiency coefficients in a preferred embodiment of the present invention;

[0032] Figure 5 This is a graph showing changes in actual water requirements and total water requirements of various crops in a preferred embodiment of the present invention;

[0033] Figure 6is a graph showing changes in total irrigation water demand in a preferred embodiment of the present invention;

[0034] Figure 7 It is a contribution diagram of driving factors for changes in irrigation water demand according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0035] To make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the embodiments of the present invention will be described clearly and completely in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them.

[0036] Unless otherwise defined, technical or scientific terms used herein shall have the same meaning as commonly understood by one of ordinary skill in the art to which the invention belongs.

[0037] The following is combined with Figure 1-7 The specific embodiments of the present invention are described in detail with examples.

[0038] like Figure 1 As shown, a method for calculating irrigation water demand and identifying dominant driving factors based on crop water demand mechanism includes the following steps:

[0039] (1) Collect historical meteorological data of the Hetao Irrigation District, including average temperature, daily maximum temperature, daily minimum temperature, average wind speed, average relative humidity, minimum relative humidity, sunshine hours, and daily precipitation data. Collect data on irrigation efficiency, crop planting types, areas and proportions, and growth cycles of various crops in the Hetao Irrigation District. Changes in crop planting areas in the Hetao Irrigation District over the past 20 years. Figure 2 As shown in Figure 2, the crop planting structure in the irrigation area is as follows: Figure 3 As shown in Figure 2, the irrigation efficiency of the irrigation area evolves as Figure 4 The growth cycle of each crop in the irrigation area is shown in Table 1.

[0040] Table 1. Growth period of main crops in Hetao irrigation area

[0041]

[0042] (2) The reference crop evaporation (ET0) is calculated based on the FAO Penman-Monteith formula:

[0043]

[0044] Where: ET0 is the reference crop evaporation (mm d -1 ), R n is the net radiation (MJ m -2 d -1 ), G is soil heat flux (MJm-2 d -1 ), e s is the saturated water vapor pressure (kPa), e a is the actual water vapor pressure (kPa), Δ is the slope of the saturated water vapor pressure curve (kPa℃ -1 ), γ is the psychrometric constant (kPa℃ -1 ), T is the average temperature (℃), u2 is the wind speed at a height of 2 meters (ms -1 ).

[0045] (3) First, the basic crop coefficients of various crops in the Hetao Irrigation Area under standard conditions were found from FAO56 (standard conditions refer to air humidity of about 45%, wind speed of about 2m / s, sufficient water supply, good management, normal growth, and large-scale high-yield crop conditions). The results are shown in Table 2.

[0046] Table 2. Basic crop coefficients of main crops in irrigation areas

[0047]

[0048] Then, according to the actual meteorological conditions of the Hetao Irrigation District, the crop coefficients at each stage are corrected. The formula is as follows:

[0049]

[0050] Where: K ctab is the basic crop coefficient obtained from the table; RH min is the average minimum relative humidity of the corresponding growth stage, %; u2 is the average wind speed at 2m in the growth stage, m / s; h is the average height of the crop in the growth stage, m.

[0051] (4) Calculate the actual water requirement ET during the crop growth period according to the following formula c :

[0052]

[0053] Where: K ci is the crop coefficient on the i-th day of the crop growth period, ET ci is the reference crop evapotranspiration on the i-th day of the growing period (mm), and n is the length of the crop growing period. The calculation results of the actual water demand and total water demand of each crop in the Hetao Irrigation District are as follows: Figure 5 As shown, in the past 20 years, the average annual total crop water requirement of major crops in the Hetao Irrigation District has shown a significant increasing trend; the average annual water requirement of wheat has shown a non-significant decreasing trend; the average annual water requirement of corn has shown a non-significant decreasing trend; the average annual water requirement of sunflower has shown a non-significant increasing trend; and the average annual water requirement of other crops has shown a non-significant decreasing trend.

[0054] (5) Estimate effective precipitation P according to the empirical formula provided by the United States Department of Agriculture e :

[0055]

[0056] Where: P is daily precipitation (mm), P e is the daily effective precipitation (mm). Then calculate the net irrigation quota I:

[0057] I=ET c -P e (5)

[0058] Among them, I is the net irrigation quota of crops (mm), ET c is the crop water requirement during the entire growth period (mm), P e is the effective precipitation (mm).

[0059] (6) Based on the crop planting area, irrigation efficiency and net irrigation quota, calculate the irrigation water requirement W:

[0060]

[0061] Where: W is the irrigation water requirement (10 8 m 3 ), j represents different crops, n is the type of crop, A j is the planting area of ​​the jth crop (10 4 hm 2 ), η is the irrigation water utilization efficiency.

[0062] The calculation results of irrigation water demand in Hetao Irrigation District in the past 20 years are as follows: Figure 6 As shown in Figure 2, the total irrigation water demand in the irrigation area showed a trend of decreasing first and then increasing, from 115.44×10 8 m 3 Reduced to 80.13×10 in 2004 8 m 3 , and then increased to 114.21×10 in 2020 8 m 3 The irrigation water demand showed a fluctuating growth trend in general. In 2004, the irrigation water demand suddenly changed, which was related to the change in crop planting structure. Specifically, the reduction in wheat and corn planting area led to a reduction in irrigation water demand in the irrigation area. After 2005, the irrigation water demand fluctuated and increased, which was caused by the expansion of corn and sunflower planting area.

[0063] (7) A boosted regression tree (BRT) model was constructed with irrigation water demand as the dependent variable and temperature, precipitation, wind speed, relative humidity, sunshine hours, irrigation area, planting structure, and irrigation efficiency as independent variables.

[0064] (8) According to the characteristics of the research object and parameter sensitivity analysis, the key parameters of the BRT model are determined: the sampling rate is 50%, the learning rate is 0.1, the tree complexity is 10, and the loss function distribution is Gaussian distribution. The contribution of each factor in the model is measured based on the SHAP value:

[0065]

[0066] Where: N is the set of factors; S is a subset of the factor set, φ i Represents the SHAP value of factor i.

[0067] The contribution rate of various factors to the change of irrigation water demand in the past 20 years is as follows: Figure 7 As shown in the figure. The contribution of human factors is greater than that of climate factors; the contribution of each factor is as follows: irrigation efficiency>irrigated area>average temperature>planting structure>relative humidity>wind speed>precipitation>sunshine hours. Therefore, the improvement of irrigation efficiency is the main driving force for the reduction of irrigation water demand. This is because in recent years, irrigation areas have vigorously promoted and implemented water-saving projects, developed channel anti-seepage technology, low-pressure pipeline water delivery technology and water-saving irrigation (drip irrigation, sprinkler irrigation) technology, which has greatly improved the irrigation efficiency coefficient of irrigation areas.

[0068] The above is only a specific implementation of the present invention and does not limit the present invention in any way. Any simple modification, change and equivalent change made to the above embodiments according to the technical essence of the invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for calculating irrigation water demand and identifying dominant driving factors, characterized in that: The steps include: S1: Collect meteorological data and statistical data required for calculation and pre-process the data; S2: Calculate reference crop evapotranspiration based on meteorological data; S3: Determine the standard crop coefficient of crops based on crop types and growth stages, and then correct the crop coefficient of each stage according to meteorological data; S4: Calculate the actual water requirement of crops during the growing period based on the corrected crop coefficient and reference crop evapotranspiration; S5: Based on precipitation data, calculate the effective precipitation, and use the crop water requirement and effective precipitation to calculate the crop net irrigation quota; S6: Calculate irrigation water requirements based on crop planting area, irrigation efficiency and net irrigation quota; S7: Taking irrigation water demand as the dependent variable, comprehensively selecting the driving factors affecting irrigation water demand as independent variables, and constructing an enhanced regression tree model, namely the BRT model; S8: Conduct multiple parameter sensitivity analyses on the BRT model and use cross-validation methods to determine the optimal parameter configuration. Based on the model operation results, use the contribution rate of each factor as an indicator to measure the importance of the explanatory variables, rank the importance of each influencing factor, and identify the main controlling factors affecting irrigation water demand.

2. The method according to claim 1, characterized in that: The meteorological data in S1 include average temperature, daily maximum temperature, daily minimum temperature, average wind speed, average relative humidity, minimum relative humidity, sunshine hours and daily precipitation data; the statistical data include crop planting area, the proportion of each crop, irrigation efficiency coefficient and crop growth cycle.

3. The method according to claim 1, characterized in that In S2, the Penman-Monteith formula is used to calculate the daily reference crop evapotranspiration.

4. The method according to claim 1, characterized in that: In S3, the whole growth period of crops is divided into early growth period, rapid development period, middle growth period and late growth period according to the growth cycle of crops. The basic crop coefficient is obtained by referring to the FAO56 manual, and then the crop coefficient of each stage is corrected according to the average 2m wind speed and minimum relative humidity at different growth stages of crops.

5. The method according to claim 1, characterized in that: In S4, the corrected crop coefficient is linearly interpolated to obtain the crop coefficient curve during the crop growth period, and then the daily crop coefficient is multiplied by the reference crop evapotranspiration of the corresponding date to obtain the actual water requirement of the crop.

6. The method according to claim 1, characterized in that The effective precipitation in S5 refers to the part of precipitation that can be effectively used by crops; when the natural effective precipitation is not enough to meet the water needs of crops, irrigation is needed to supplement it. Therefore, the net irrigation quota for crops is the difference between the water needs of crops and the effective precipitation.

7. The method according to claim 1, characterized in that The specific operation steps of S6 are: multiply the net irrigation quota of each crop obtained by S2~S5 by the planting area of ​​the corresponding crop, and then multiply it by the irrigation efficiency coefficient of the study area to obtain the irrigation water requirement of each crop, and add up the irrigation water requirements of all crops to obtain the total irrigation water requirement of the study area.

8. The method according to claim 1, characterized in that The factors selected in S7 include climate factors and human factors; among them, climate factors include: temperature, precipitation, wind speed, relative humidity and sunshine hours; human factors include: irrigation area, planting structure and irrigation efficiency.

9. The method according to claim 1, characterized in that: In S8, it is necessary to conduct sensitivity analysis on the five key parameters of the BRT model to explore the model configuration that achieves the best prediction results. The model parameters include: sampling rate, learning rate, number of regression trees, regression tree complexity and loss function.

Images