A rice intelligent irrigation and drainage decision-making method based on cluster analysis

Abstract

The present invention discloses a rice intelligent irrigation and drainage decision-making method based on cluster analysis, comprising: establishing an empirical data set based on historical rice field irrigation and drainage data, collecting sample data and environmental parameters of a target sample rice field group, and standardizing all data; constructing a hierarchical cluster analysis model, classifying the target sample rice field group based on the sample data to generate multiple groups of rice fields; calculating the similarity between each group of rice fields and the rice fields in the empirical data set, and obtaining a recommended irrigation pattern for the group of rice fields; constructing an optimization function for the recommended irrigation pattern based on the environmental parameters of each group of rice fields to obtain an optimized recommended irrigation pattern; and inferring the irrigation and drainage decision for each group of rice fields based on the growth parameters of each group of rice fields in combination with the optimized recommended irrigation pattern.

Description

Technical Field

[0001] The present invention relates to the field of irrigation technology, and in particular to a rice intelligent irrigation and drainage decision-making method based on cluster analysis. Background Art

[0002] How to improve the utilization rate and water productivity of agricultural irrigation water has become a crucial issue in the agricultural water conservancy industry. Achieving efficient water-saving irrigation at low cost can effectively alleviate water resource pressure.

[0003] Agricultural irrigation is gradually shifting from extensive flooding to intelligent, precise irrigation methods. Automatic irrigation, precision irrigation, variable-rate irrigation, and irrigation methods that adapt to local conditions and make precise, real-time decisions based on crop growth will gradually develop and become more widespread. Precipitation directly impacts the efficiency of agricultural irrigation water use, and improving the effective utilization of natural precipitation has become a key decision in crop irrigation.

[0004] Most existing intelligent irrigation decisions are based on weather forecast data and real-time field moisture conditions, and rely on field moisture monitoring instruments to obtain real-time data. Installing monitoring instruments in every field in a large irrigation area with multiple rice fields is costly. Using data from a single set of monitoring equipment across multiple rice fields results in large errors, which is not conducive to improving the accuracy of irrigation and drainage decisions and meeting crop growth conditions. Summary of the Invention

[0005] In order to overcome the problem of high irrigation area monitoring costs in the existing technology when making irrigation and drainage decisions for large-scale rice fields, the present invention provides a rice intelligent irrigation and drainage decision-making method based on cluster analysis. Based on cluster analysis, the rice fields are grouped, and there is no need to install moisture monitoring equipment in each rice field. Based on historical irrigation and drainage data, the utilization rate of irrigation water is improved, and the accuracy and scientificity of irrigation and drainage decision-making are improved.

[0006] According to one aspect of the present invention, a method for intelligent rice irrigation and drainage decision-making based on cluster analysis is provided, comprising:

[0007] Establish an empirical data set based on historical rice field irrigation and drainage data, collect sample data and environmental parameters of the target sample rice field group, and standardize all data;

[0008] A hierarchical cluster analysis model was constructed to classify the target sample rice fields into multiple groups based on the sample data;

[0009] Calculate the similarity between each group of rice fields and the rice fields in the empirical dataset to obtain the recommended irrigation mode for each group of rice fields;

[0010] An optimization function for recommending irrigation patterns is constructed based on the environmental parameters of each group of rice fields to obtain the optimized recommended irrigation patterns.

[0011] Based on the environmental parameters of each group of rice fields and the optimized recommended irrigation pattern, the irrigation and drainage decisions of each group of rice fields are calculated;

[0012] As a further embodiment, the sample data includes but is not limited to measured meteorological data, transplanting time, last irrigation time, soil texture, longitude and latitude, elevation, slope, fertilization time, and rice variety;

[0013] Environmental parameters include, but are not limited to, weather forecast data, soil moisture content or water layer depth, field capacity, wilting coefficient, and the current crop growth and development stage;

[0014] The historical rice field irrigation and drainage data include but are not limited to sample data of historical rice fields, environmental parameters and the depth of controlled water layers at various growth stages.

[0015] As a further implementation plan, the working process of the hierarchical cluster analysis model is:

[0016] The sample distance is calculated based on the sample data in the target sample rice field group based on the Euclidean distance method, and the two sample rice fields with the closest sample distance are merged as a cluster group.

[0017] The inter-cluster distance between cluster groups is calculated based on the class average distance method. A grouping threshold is preset, and cluster groups with inter-cluster distances less than the grouping threshold are merged step by step from near to far to obtain multiple groups of rice fields.

[0018] As a further implementation plan, the similarity is calculated based on the sample data using the cosine similarity method, and the historical rice fields in the empirical data set with the highest similarity to the group rice fields are screened out from the similarity calculation results, and their irrigation patterns are used as the recommended irrigation patterns for the group rice fields.

[0019] As a further implementation plan, the optimization function for the recommended irrigation pattern is a multi-objective optimization function established by changing the combination of controlled water layer depths during the current rice crop growth and development period, with the goals of maximizing the multi-year average rainfall utilization rate, minimizing the multi-year average irrigation frequency, and minimizing the impact on yield. The mathematical expression is as follows:

[0020]

[0021] In the above formula, Indicates the date sequence number of the current crop growth and development period; For this period Multi-year average rainfall utilization rate after normalization of the combined control water layer depth, %; For this period The multi-year average irrigation times after normalization of the combination of daily controlled water layer depth; For this period The multi-year average production index after normalization of the combination of daily controlled water layer depth; , and are the multi-year average rainfall utilization rate, the multi-year average irrigation times, and the weighted values ​​of the yield index; F is the optimization function for recommending irrigation patterns, with a step size of 1 mm and a range of ±10 mm. The optimal control water depth combination for the current crop growth and development period is selected as the optimized irrigation pattern based on the maximum value of the objective function; the water depth on the i-th day of the current crop growth and development period of rice is presented in the form of a water balance equation;

[0022] The various parameters of the water balance equation are calculated day by day, and the optimized irrigation pattern for the entire current crop growth and development period is obtained.

[0023] As a further implementation scheme, the control water layer depth combination includes a lower limit of a suitable water layer depth, an upper limit of a suitable water layer depth and an upper limit of a rainwater storage depth.

[0024] As a further implementation plan, the steps for making irrigation and drainage decisions based on the environmental parameters of the rice fields in the group and the optimized irrigation pattern include:

[0025] The predicted crop evapotranspiration of each rice field is calculated based on the environmental parameters of each rice field. The formula is as follows:

[0026]

[0027] Where, is the single crop coefficient; is the water stress coefficient, which is calculated based on the field water holding capacity and the wilting coefficient of rice; To forecast reference crop evapotranspiration, it is calculated based on weather forecast data.

[0028] The irrigation and drainage water volume is calculated based on the water balance deduction formula. The water balance deduction formula is as follows:

[0029]

[0030] and The future and Depth of water layer between fields (mm); For the future The weather forecast rainfall for the day, mm; For the future The leakage rate of the deep layer is the default value; For the future Daily forecast of crop evapotranspiration;

[0031] Calculate the irrigation and drainage water volume on a daily basis, and calculate the irrigation and drainage decisions for the entire current crop growth and development period based on the optimized irrigation pattern.

[0032] As a further implementation plan, the calculation logic of irrigation and drainage decision-making is as follows:

[0033]

[0034]

[0035] Where, Indicates the future of rice fields Daily irrigation volume; represents the drainage volume of the rice field on the jth day in the future; is the lower limit of the suitable water layer for the optimized irrigation mode, mm; is the upper limit of the suitable water layer for the optimized irrigation mode, mm; is the upper limit of rainwater storage depth of the optimized irrigation mode, mm.

[0036] According to another aspect of the present specification, a rice intelligent irrigation and drainage decision system based on cluster analysis is provided, which is used to implement a rice intelligent irrigation and drainage decision method based on cluster analysis, comprising:

[0037] The data processing module collects sample data and environmental parameters of the target rice paddy group and standardizes the data;

[0038] The rice field grouping module clusters the target rice fields into multiple groups of rice fields;

[0039] Recommendation module, which recommends irrigation patterns for grouped rice fields;

[0040] Optimization module, which optimizes the recommended irrigation pattern for group rice fields;

[0041] The irrigation and drainage decision module outputs the irrigation and drainage decisions of the rice fields in the group.

[0042] As a further implementation scheme, the data processing module is also used to establish an experience data set based on historical rice field irrigation and drainage data.

[0043] Compared with the existing technology, the beneficial effects of the present invention are: the irrigation and drainage decision-making method provided by the present invention is based on cluster analysis, which groups rice fields and does not require the installation of moisture monitoring equipment in each rice field. It also improves the utilization rate of irrigation water based on historical irrigation and drainage data, and improves the accuracy and scientificity of irrigation and drainage decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, a brief introduction will be given below to the drawings used in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 Flowchart of a method for intelligent rice irrigation and drainage decision-making based on cluster analysis in an embodiment of the present invention;

[0046] Figure 2 Schematic diagram of a rice intelligent irrigation and drainage decision-making system based on cluster analysis in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] It should be noted that:

[0048] The terms "including" and "having" and any variations thereof in the description and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions, for example, a process, method, system, product or apparatus that includes a series of steps or units is not necessarily limited to the steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0049] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices. The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily need to be executed in the order described. For example, some operations / steps may be further decomposed, while others may be combined or partially combined, so the actual execution order may vary depending on the actual situation.

[0050] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, the technical features in the various embodiments or single embodiments provided by the present invention are arbitrarily combined with each other to form a new technical solution. This combination is not restricted by the sequence of steps and / or structural composition mode, but must be based on the ability of ordinary technicians in this field to implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that this combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0051] As attached Figure 1 As shown, Figure 1 This is a flow chart of a method for intelligent rice irrigation and drainage decision-making based on cluster analysis in an embodiment of the present invention, including:

[0052] Establish an empirical data set based on historical rice field irrigation and drainage data, collect sample data and environmental parameters of the target sample rice field group, and standardize all data;

[0053] A hierarchical cluster analysis model was constructed to classify the target sample rice fields into multiple groups based on the sample data;

[0054] Calculate the similarity between each group of rice fields and the rice fields in the empirical dataset to obtain the recommended irrigation mode for each group of rice fields;

[0055] An optimization function for recommending irrigation patterns is constructed based on the environmental parameters of each group of rice fields to obtain the optimized recommended irrigation patterns.

[0056] Based on the environmental parameters of each group of rice fields and the optimized recommended irrigation pattern, the irrigation and drainage decisions of each group of rice fields were calculated.

[0057] Furthermore, the range method was used to standardize the sample data of the sample rice fields.

[0058] Furthermore, the sample data includes but is not limited to measured meteorological data, transplanting time, last irrigation time, soil texture, longitude and latitude, elevation, slope, fertilization time, and rice variety;

[0059] Environmental parameters include, but are not limited to, weather forecast data, soil moisture content or water layer depth, field capacity, wilting coefficient, and the current crop growth and development stage;

[0060] The historical rice field irrigation and drainage data include but are not limited to sample data of historical rice fields, environmental parameters and the depth of controlled water layers at various growth stages.

[0061] Furthermore, the working process of the hierarchical cluster analysis model is:

[0062] The sample distance is calculated based on the sample data in the target sample rice field group based on the Euclidean distance method, and the two sample rice fields with the closest sample distance are merged as a cluster group.

[0063] Specifically, according to the Euclidean distance method, the sample With sample The distance between The calculation formula is:

[0064]

[0065] in, , For the sample, is the vector dimension.

[0066] The inter-cluster distance between cluster groups is calculated based on the class average distance method. A grouping threshold is preset, and cluster groups with inter-cluster distances less than the grouping threshold are merged step by step from near to far to obtain multiple groups of rice fields.

[0067] Specifically, the distance between the two clusters is represented by the average distance between all data points in the two clusters. The distance used by the class average method is defined as the average square distance between the two clusters. Class and Merge classes into one back, Classes and other classes The formula for calculating the inter-class distance is:

[0068]

[0069] in, 、 、 and They are 、 、 and The number of samples in the class, ; and and Respectively represent kind, Class and Class and The inter-class distance of the class.

[0070] Furthermore, the similarity is calculated based on the sample data using the cosine similarity method. The historical rice fields in the empirical data set with the highest similarity to the group rice fields are screened out from the similarity calculation results, and their irrigation patterns are used as the recommended irrigation patterns for the group rice fields.

[0071] The specific calculation formula of the cosine similarity method is:

[0072]

[0073] in, Represents rice fields No. The standardized value of the indicator value, Represents rice fields No. The standardized value of the indicator value, Represents rice fields The average value of all indicators, Represents rice fields The average value of all indicators.

[0074] In the recommendation stage, the N historical rice fields that are most similar to the target group rice fields are found based on the similarity calculation results, and the irrigation mode preferred by the historical rice fields with the highest similarity is recommended to the target group rice fields.

[0075]

[0076] Defining a function To calculate historical rice fields Irrigation mode adopted ( The target group of rice fields under all irrigation modes ( is the set of all rice field groups), find the most interesting irrigation mode for the target rice field group. .

[0077] Furthermore, the optimization function for the recommended irrigation mode is a multi-objective optimization function established by changing the combination of controlled water layer depths during the current rice crop growth and development period, with the goals of maximizing the multi-year average rainfall utilization rate, minimizing the multi-year average irrigation times, and minimizing the impact on yield. The mathematical expression is as follows:

[0078]

[0079] In the above formula, Indicates the date sequence number of the current crop growth and development period; For this period Multi-year average rainfall utilization rate after normalization of the combined control water layer depth, %; For this period The multi-year average irrigation times after normalization of the combination of daily controlled water layer depth; For this period The multi-year average production index after normalization of the combination of daily controlled water layer depth; , and are the multi-year average rainfall utilization rate, the multi-year average irrigation times, and the weighted values ​​of the yield index; F is the optimization function for recommending irrigation patterns, with a step size of 1 mm and a range of ±10 mm. The optimal control water layer depth combination for the current crop growth and development period is selected as the optimized irrigation pattern based on the maximum value of the objective function;

[0080] The water depth on the i-th day of the current rice crop growth and development period is expressed in the form of a water balance equation, which is mathematically expressed as follows:

[0081]

[0082] Where, For the Depth of water layer in the field before irrigation or drainage, mm; For the Depth of water layer in the field at the end of the day, mm; For the Daily rainfall, mm; is the amount of irrigation on day i, mm; For the Daily evaporation and transpiration; is the leakage of rice fields; For the Daily displacement, mm.

[0083] Specifically, the daily evapotranspiration of rice is calculated using the single crop coefficient method using the following formula:

[0084]

[0085] Where, For the Daily reference crop evapotranspiration, mm, calculated using the Penman-Monteith formula; For the The single crop coefficient of rice in each growth period was obtained based on the local historical irrigation and drainage data. .

[0086] Jordi Calculated by day If the water level is lower than the lower limit of the suitable water layer for the growth period, irrigation should be carried out; if If the depth is greater than the lower limit of the suitable water layer in the growth period and less than or equal to the maximum water storage depth after rainfall in the growth period, no irrigation or drainage will be done; if If it is greater than the maximum water storage depth after rainfall during the growth period, drainage is carried out.

[0087]

[0088]

[0089] Where, is the lower limit of the suitable water layer depth, mm; is the upper limit of the suitable water layer depth, mm; is the upper limit of rainwater storage depth, mm.

[0090] Statistics on rainfall utilization rate and irrigation times in each growth stage, rainfall utilization rate The calculation formula is:

[0091]

[0092] Where, For the rainfall during the crop growth and development period; For the Drainage during crop growth and development period.

[0093] The crop water production function was used to establish the functional relationship between water deficit and yield at a specific crop growth stage. The Jensen model was used to quantify the yield index. The formula can be expressed as:

[0094]

[0095] Where, is the actual evapotranspiration; is the reference crop evapotranspiration; is the number of crop growth and development periods, i is the ordinal number of crop growth and development periods, ; For the The period water deficit sensitivity index is determined according to the rice crop growth and development period and variety.

[0096] The various parameters of the water balance equation are calculated day by day, and the optimized irrigation pattern for the entire current crop growth and development period is obtained.

[0097] Furthermore, the controlled water layer depth combination includes a lower limit of a suitable water layer depth, an upper limit of a suitable water layer depth and an upper limit of a rainwater storage depth.

[0098] Furthermore, based on the environmental parameters of the rice fields in each group and the optimized irrigation pattern, the steps for making irrigation and drainage decisions include:

[0099] The predicted crop evapotranspiration of each rice field is calculated based on the environmental parameters of each rice field. The formula is as follows:

[0100]

[0101] Where, is the single crop coefficient; is the water stress coefficient, which is calculated based on the field water holding capacity and the wilting coefficient of rice; To forecast reference crop evapotranspiration, it is calculated based on weather forecast data.

[0102] When soil water stress occurs, , when there is no soil moisture stress ; When soil water stress occurs, the soil water stress coefficient is given by the following formula:

[0103]

[0104] Where, is the amount of water consumed in the root zone (i.e., water deficit relative to field capacity), mm; is the total available water in the root zone, mm; It is the effective amount of water that crops can easily absorb from the root layer, mm.

[0105] and It is given by:

[0106]

[0107]

[0108] Where, is the field water capacity, m 3 / m 3 ; is the wilting coefficient, m 3 / m 3 ; is the varying root zone depth, m; It is the ratio of the amount of water that can be consumed in the root zone before water stress occurs to the total available water in the soil, which is taken as 0.2.

[0109] is the forecast reference crop evapotranspiration, mm / d; the forecast reference crop evapotranspiration is calculated by the locally calibrated HS model, which can be expressed as follows:

[0110]

[0111] Where, Calculated for the HS model value; and The two parameters are obtained through calibration; is the extraterrestrial radiation, MJ / m 2 / d; and are the maximum and minimum temperatures in the weather forecast, °C, respectively.

[0112] The irrigation and drainage water volume is calculated based on the water balance deduction formula. The water balance deduction formula is as follows:

[0113]

[0114] and The future and Depth of water layer between fields (mm); For the future The weather forecast rainfall for the day, mm; For the future The default value of the deep layer leakage is 2 mm / d; For the future Forecast crop evapotranspiration for the day.

[0115] Calculate the irrigation and drainage water volume on a daily basis, and calculate the irrigation and drainage decisions for the entire current crop growth and development period based on the optimized irrigation pattern.

[0116] Furthermore, the calculation logic of irrigation and drainage decision-making is:

[0117]

[0118]

[0119] Where, Indicates the future of rice fields Daily irrigation volume; Indicates the future of rice fields daily water discharge; is the lower limit of the suitable water layer for the optimized irrigation mode, mm; is the upper limit of the suitable water layer for the optimized irrigation mode, mm; is the upper limit of rainwater storage depth of the optimized irrigation mode, mm.

[0120] A rice intelligent irrigation and drainage decision-making method based on cluster analysis includes the following steps:

[0121] The first step is to collect sample data and environmental parameters from the target sample rice fields. Based on historical irrigation and drainage data, an empirical irrigation and drainage dataset for the historical rice fields is established. Sample data includes transplanting time, last irrigation time, measured meteorological data, soil texture, longitude and latitude, elevation, slope, fertilization time, and rice variety. Environmental parameters include weather forecast data, soil moisture content or water layer depth, field holding capacity, wilting coefficient, and crop growth and development period. Historical irrigation and drainage data includes measured meteorological data from the historical rice fields, rice variety, monocrop coefficient, controlled water layer depth at each growth stage, soil texture, longitude and latitude, elevation, slope, and fertilization time. Measured meteorological data and weather forecast data include, but are not limited to, temperature, wind speed, rainfall, sunshine, and weather type.

[0122] The second step is to build a hierarchical cluster analysis model, classify the sample paddy fields according to their sample data, and generate multiple groups of paddy fields; for example, the range method is used to normalize the transplanting time, the last irrigation time, the measured meteorological data, the soil texture, the longitude and latitude, the elevation, the slope, the fertilization time, and the rice variety data to obtain dimensionless relative values. Assume that the initial data sets of each type are { | },in For the Sample No. indicator values, are the number of samples and the number of indicators respectively. The formula is as follows:

[0123]

[0124] Among them, min( )、max( ) are the minimum and maximum values ​​of the j-th index value in the sample set. The characteristic quantity of each sample is obtained -dimensional feature vector.

[0125] The distances between multiple sample rice fields were calculated using the Euclidean distance method and standardized sample data. The sample rice fields with the closest distances were merged, and then the distances between each group were calculated using the class average distance method. The groups were merged step by step from near to far to generate multiple groups of rice fields.

[0126] In the third step, the Pearson similarity method is used to calculate the similarity between each group of paddy fields and the historical paddy fields in the dataset, based on the standardized results from the second step. For example, based on the historical paddy field irrigation and drainage empirical dataset and sample data collected from the paddy fields to be irrigated or drained, the cosine similarity (Sim) between the paddy fields to be irrigated or drained and each historical paddy field is calculated. Based on this similarity, the N historical paddy fields most similar to the target paddy field group are found, and the irrigation pattern preferred by the historical paddy fields with the highest similarity is recommended to the target paddy field group.

[0127] In the fourth step, the recommended irrigation pattern is optimized based on the historical irrigation and drainage data of the historical rice fields with the highest similarity to obtain the optimized irrigation pattern.

[0128] For example, the specific method of optimizing the recommended irrigation mode is as follows:

[0129] A multi-objective optimization function was established with the goal of maximizing the average rainfall utilization rate over many years, minimizing the average irrigation times over many years, and minimizing the impact on yield. The water depth was controlled at each growth stage with a step size of 1 mm and a range of ±10 mm, including the lower limit of the suitable water layer ( ), the upper limit of suitable water layer ( ), the upper limit of rain storage ( ), the recommended irrigation mode is optimized, and the combination of controlled water layer depth that maximizes the objective function value is the optimal irrigation mode. The multi-objective optimization function can be expressed as:

[0130]

[0131] In the above formula, Indicates the date sequence number of the current crop growth and development period; For this period Multi-year average rainfall utilization rate after normalization of the combined control water layer depth, %; For this period The multi-year average irrigation times after normalization of the combination of daily controlled water layer depth; For this period The multi-year average production index after normalization of the combination of daily controlled water layer depth; , and are the multi-year average rainfall utilization rate, the multi-year average irrigation times and the weight values ​​of the yield index respectively; F is the optimization function for recommending irrigation mode, with a step size of 1mm and a range of ±10mm. The optimal control water layer depth combination for the current crop growth and development period is selected as the optimized irrigation mode according to the maximum value of the objective function; the weight value can be selected according to the actual situation. + + =1. The weight value can be selected according to the actual situation. Generally, =1 / 3. If the target rice fields focus on improving rainfall utilization efficiency, the weight of rainfall utilization efficiency can be increased, and the =2 / 3, = =1 / 6; focusing on reducing the number of irrigation times can increase the weight of irrigation times, taking =2 / 3, =1 / 6.

[0132] On day i during the rice growth period, the depth of the water layer in the field can be expressed by the water balance equation:

[0133]

[0134] Where, For the Depth of water layer in the field before irrigation or drainage, mm; For the - Depth of water layer in the field at the end of 1 day, mm; For the Daily rainfall, mm; is the amount of irrigation on day i, mm; For the Daily evaporation and transpiration; is the leakage of rice fields; For the Daily displacement, mm.

[0135] The daily evapotranspiration of rice is calculated using the single crop coefficient method using the following formula:

[0136]

[0137] Where, For the Daily reference crop evapotranspiration, mm, calculated using the Penman-Monteith formula; For the The single crop coefficient of rice in each growth period was obtained based on the local historical irrigation and drainage data. .

[0138] If the calculated value on day i is If the water level is lower than the lower limit of the suitable water layer for the growth period, irrigation should be carried out; if If the depth is greater than the lower limit of the suitable water layer in the growth period and less than or equal to the maximum water storage depth after rainfall in the growth period, no irrigation or drainage will be done; if If it is greater than the maximum water storage depth after rainfall during the growth period, drainage is carried out.

[0139]

[0140]

[0141] Where, Indicates rice fields Daily irrigation volume; Indicates rice fields daily water discharge; is the lower limit of the suitable water layer depth, mm; is the upper limit of the suitable water layer depth, mm; is the upper limit of rainwater storage depth, mm.

[0142] The rainfall utilization rate and irrigation times at each growth stage were calculated. The calculation formula for rainfall utilization rate is:

[0143]

[0144] Where, For the rainfall during the crop growth and development period; For the Drainage during crop growth and development period.

[0145] The crop water production function was used to establish the functional relationship between water deficit and yield at a specific crop growth stage. The Jensen model was used to quantify the yield index, which can be expressed as:

[0146]

[0147] Where, is the actual evapotranspiration; is the reference crop evapotranspiration; is the number of crop growth and development periods, is the ordinal number of the crop growth and development period, ; For the The water deficit sensitivity index is determined according to the rice crop growth and development period and variety. The values ​​were obtained from previous studies and are shown in Table 1.

[0148] Table 1 Water deficit sensitivity index at different growth stages of early, middle and late rice

[0149]

[0150] The actual evapotranspiration is determined by the following formula:

[0151]

[0152] Where, is the single crop coefficient; is the water stress coefficient, which is calculated based on the field water holding capacity and the wilting coefficient of rice; To forecast reference crop evapotranspiration, it is calculated based on weather forecast data.

[0153] When soil water stress occurs, , when there is no soil moisture stress ; When soil water stress occurs, the soil water stress coefficient is given by the following formula:

[0154]

[0155] Where, is the amount of water consumed in the root zone (i.e., water deficit relative to field capacity), mm; is the total available water in the root zone, mm; It is the effective amount of water that crops can easily absorb from the root layer, mm.

[0156] and It is given by:

[0157]

[0158]

[0159] Where, is the field water capacity, m 3 / m 3 ; is the wilting coefficient, m 3 / m 3 ; is the varying root zone depth, m; It is the ratio of the amount of water that can be consumed in the root zone before water stress occurs to the total available water in the soil, which is taken as 0.2.

[0160] Generate the control water layer depth dataset through NSGA-ii algorithm { }, select the optimal control water layer depth combination based on the maximum objective function value. For example, it is implemented using the Python genetic algorithm tool library Geatpy2. The specific process is as follows: 1) Define the decision variables , and ; 2) Define the sub-objective function , , ;3) Define the constraints as 1mm as the step size and ±10mm as the range change , and ; 4) Initialize the parent generation P, the population size N = 150, and calculate the fitness of the parent population P; 5) After selection, crossover, and mutation operators, generate a child population Q of size N, and calculate the fitness of the child population Q; 6) Merge the parent population P and the child population Q to form a recombinant population R of size 2N; 7) Perform fast non-dominated sorting and individual crowding calculation on the recombinant population R to generate a new parent population; 8) Perform selection, crossover, and mutation operations on the newly generated parent population to form a new child population and calculate its fitness; 9) Determine whether the termination condition is met (number of iterations ≥ 120). If the condition is met, the optimization ends and the result is output, otherwise return to 5). According to the obtained non-inferior solution set and the objective function Obtain the final irrigation pattern.

[0161] The fifth step is to collect the environmental parameters of the rice fields to be irrigated or drained in the current decision cycle, and make irrigation and drainage decisions based on the optimized irrigation mode.

[0162] Specifically, the predicted crop evapotranspiration of the area to be irrigated or drained is calculated based on the weather forecast data for the rice fields to be irrigated or drained within a preset time period in the future and the growth and development stage of the crops. The formula for crop evapotranspiration is as follows:

[0163]

[0164] Where, is the single crop coefficient; is the water stress coefficient, which is calculated based on the field water holding capacity and the wilting coefficient of rice; To forecast reference crop evapotranspiration, it is calculated based on weather forecast data.

[0165] When soil water stress occurs, , when there is no soil moisture stress ; When soil water stress occurs, the soil water stress coefficient is given by the following formula:

[0166]

[0167] Where, is the amount of water consumed in the root zone (i.e., water deficit relative to field capacity), mm; is the total available water in the root zone, mm; It is the effective amount of water that crops can easily absorb from the root layer, mm.

[0168] and It is given by:

[0169]

[0170]

[0171] Where, is the field water capacity, m 3 / m 3 ; is the wilting coefficient, m 3 / m 3 ; is the varying root zone depth, m; It is the ratio of the amount of water that can be consumed in the root zone before water stress occurs to the total available water in the soil, which is taken as 0.2.

[0172] is the forecast reference crop evapotranspiration, mm / d; the forecast reference crop evapotranspiration is calculated by the locally calibrated HS model, which can be expressed as follows:

[0173]

[0174] Where, Calculated for the HS model value; and The two parameters are obtained through calibration; is the extraterrestrial radiation, MJ / m 2 / d; and are the maximum and minimum temperatures in the weather forecast, °C, respectively.

[0175] The irrigation and drainage water volume is calculated based on the water balance deduction formula. The water balance deduction formula is as follows:

[0176]

[0177] and are the depth of the field water layer in mm on the jth and j-1th day in the future, respectively; The weather forecast rainfall for the next j day, mm; is the deep leakage on the jth day in the future, with a default value of 2 mm / d; Forecast crop evapotranspiration for the jth day in the future.

[0178] If the value calculated on the next j day is If the water level is lower than the lower limit of the suitable water layer for the growth period, irrigation should be carried out; if If the depth is greater than the lower limit of the suitable water layer in the growth period and less than or equal to the maximum water storage depth after rainfall in the growth period, no irrigation or drainage will be done; if If it is greater than the maximum water storage depth after rainfall during the growth period, drainage is carried out.

[0179]

[0180]

[0181] Where, represents the amount of irrigation water for the rice field on the jth day in the future; represents the drainage volume of the rice field on the jth day in the future; is the lower limit of the suitable water layer for the optimized irrigation mode, mm; is the upper limit of the suitable water layer for the optimized irrigation mode, mm; is the upper limit of rainwater storage depth of the optimized irrigation mode, mm.

[0182] The implementation basis of each embodiment of the present invention is to implement it through programmed processing by a device with processor function. Therefore, in engineering practice, the technical solutions and functions of each embodiment of the present invention are encapsulated into various modules. Based on this reality, on the basis of the above embodiments, Figure 2 As shown, an embodiment of the present invention provides a rice intelligent irrigation and drainage decision-making system based on cluster analysis, which is used to execute a rice intelligent irrigation and drainage decision-making method based on cluster analysis in the above method embodiment, including:

[0183] The data processing module collects sample data and environmental parameters of the target rice paddy group and standardizes the data;

[0184] The rice field grouping module clusters the target rice fields into multiple groups of rice fields;

[0185] Recommendation module, which recommends irrigation patterns for grouped rice fields;

[0186] Optimization module, which optimizes the recommended irrigation pattern for group rice fields;

[0187] The irrigation and drainage decision module outputs the irrigation and drainage decisions of the rice fields in the group.

[0188] It should be noted that the system embodiments provided by the present invention are not only used to implement the methods in the above-mentioned method embodiments, but also used to implement the methods in other method embodiments provided by the present invention. The only difference is the setting of corresponding functional modules. The principles thereof are basically the same as those of the above-mentioned system embodiments provided by the present invention. As long as those skilled in the art refer to the specific technical solutions in other method embodiments on the basis of the above-mentioned system embodiments, obtain corresponding technical means and technical solutions composed of these technical means by combining technical features, and on the premise of ensuring the practicality of the technical solutions, improve the modules in the above-mentioned system embodiments to obtain corresponding system-type embodiments, which are used to implement the methods in other method-type embodiments. For example:

[0189] Based on the content of the above system embodiment, as a preferred embodiment, the embodiment of the present invention provides a rice intelligent irrigation and drainage decision system based on cluster analysis, further comprising:

[0190] The data processing module is also used to establish an empirical data set based on historical rice field irrigation and drainage data.

[0191] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, located in one place or distributed across multiple network units. Depending on practical needs, some or all of these modules may be selected to achieve the objectives of this embodiment. Persons of ordinary skill in the art will understand and implement these embodiments without inventive effort.

[0192] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0193] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0194] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0195] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0196] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A rice intelligent irrigation and drainage decision-making method based on cluster analysis, characterized in that: include: Establish an empirical data set based on historical rice field irrigation and drainage data, collect sample data and environmental parameters of the target sample rice field group, and standardize all data; A hierarchical cluster analysis model was constructed to classify the target sample rice fields into multiple groups based on the sample data; Calculate the similarity between each group of rice fields and the rice fields in the empirical dataset to obtain the recommended irrigation mode for each group of rice fields; In the recommendation phase, the N historical rice fields that are most similar to the target group rice fields in the empirical data set are found based on the similarity calculation results, and the irrigation mode preferred by the historical rice fields with the highest similarity is recommended to the target group rice fields. An optimization function for the recommended irrigation pattern is constructed based on the environmental parameters of each group of rice fields to obtain an optimized recommended irrigation pattern. The optimization function for the recommended irrigation pattern is a multi-objective optimization function established by changing the combination of controlled water layer depths during the current rice crop growth and development period, with the goals of maximizing the multi-year average rainfall utilization rate, minimizing the multi-year average irrigation frequency, and minimizing the impact on yield. The mathematical expression is as follows: ； In the above formula, Indicates the date sequence number of the current crop growth and development period; For this period The multi-year average rainfall utilization rate after normalization of the daily control water layer depth combination; For this period The multi-year average irrigation times after normalization of the combination of daily controlled water layer depth; For this period The multi-year average production index after normalization of the combination of daily controlled water layer depth; , and are the multi-year average rainfall utilization rate, the multi-year average irrigation times, and the weighted values ​​of the yield index; F is the optimization function for recommending irrigation patterns, which sets the step size and change range to change the control water layer depth combination for the current crop growth and development period. The optimal control water layer depth combination is selected as the optimized irrigation pattern based on the maximum objective function value; the control water layer depth combination includes the lower limit of the suitable water layer depth, the upper limit of the suitable water layer depth, and the upper limit of the rainwater storage depth; The water depth on day i of the current rice crop growth and development period is presented in the form of a water balance equation. The irrigation and drainage water volume is calculated daily. The irrigation and drainage decision for the entire current crop growth and development period is calculated based on the optimized irrigation pattern. Based on the environmental parameters of each group of rice fields and the optimized recommended irrigation pattern, the irrigation and drainage decisions for each group of rice fields are calculated, and irrigation or drainage of each group of rice fields is implemented according to the irrigation and drainage decisions. The irrigation and drainage decisions are the drainage or irrigation amounts on the jth day in the future.

2. The rice intelligent irrigation and drainage decision-making method based on cluster analysis according to claim 1, characterized in that: The sample data includes but is not limited to measured meteorological data, transplanting time, last irrigation time, soil texture, longitude and latitude, elevation, slope, fertilization time and rice variety; The environmental parameters include but are not limited to weather forecast data, soil moisture content or water layer depth, field water holding capacity, wilting coefficient and the current crop growth and development stage; The historical rice field irrigation and drainage data include but are not limited to sample data of historical rice fields, environmental parameters and controlled water layer depths at various growth stages.

3. The rice intelligent irrigation and drainage decision-making method based on cluster analysis according to claim 2, characterized in that: The working process of the hierarchical clustering analysis model is: The sample distance is calculated based on the sample data in the target sample rice field group based on the Euclidean distance method, and the two sample rice fields with the closest sample distance are merged into a cluster group; The inter-cluster distance between cluster groups is calculated based on the class average distance method. A grouping threshold is preset, and cluster groups with inter-cluster distances less than the grouping threshold are merged step by step from near to far to obtain multiple groups of rice fields.

4. The rice intelligent irrigation and drainage decision-making method based on cluster analysis according to claim 2, characterized in that: The similarity is calculated based on the sample data using the cosine similarity method, and the historical rice fields in the experience data set with the highest similarity to the group rice fields are screened out from the similarity calculation results, and their irrigation patterns are used as the recommended irrigation patterns for the group rice fields.

5. The rice intelligent irrigation and drainage decision-making method based on cluster analysis according to claim 2, characterized in that: Based on the environmental parameters of the rice fields in each group and the optimized irrigation pattern, the steps for making irrigation and drainage decisions include: The predicted crop evapotranspiration of each rice field is calculated based on the environmental parameters of each rice field. The formula is as follows: ； Where, is the single crop coefficient; is the water stress coefficient, which is calculated based on the field water holding capacity and the wilting coefficient of rice; To forecast reference crop evapotranspiration, it is calculated based on weather forecast data; The irrigation and drainage water volume is calculated based on the water balance deduction formula. The water balance deduction formula is as follows: ； and The future and Depth of water layer between fields; For the future The weather forecast for the day is rainfall; For the future The amount of leakage in the deep layer is the default value; For the future Daily forecast of crop evapotranspiration; Calculate the irrigation and drainage water volume on a daily basis, and calculate the irrigation and drainage decisions for the entire current crop growth and development period based on the optimized irrigation pattern.

6. The rice intelligent irrigation and drainage decision-making method based on cluster analysis according to claim 5, characterized in that: The calculation logic of the irrigation and drainage decision is: ； ； Where, Indicates the future of rice fields Daily irrigation volume; Indicates the future of rice fields daily water discharge; The lower limit of the suitable water layer for the optimized irrigation mode; The upper limit of the suitable water layer for the optimized irrigation mode; It is the upper limit of rainwater storage depth in the optimized irrigation mode.

7. A rice intelligent irrigation and drainage decision system based on cluster analysis, used to implement the rice intelligent irrigation and drainage decision method based on cluster analysis according to any one of claims 1 to 6, characterized in that: include: The data processing module collects sample data and environmental parameters of the target rice paddy group and standardizes the data; The rice field grouping module clusters the target rice fields into multiple groups of rice fields; Recommendation module, which recommends irrigation patterns for grouped rice fields; Optimization module, which optimizes the recommended irrigation pattern for group rice fields; The irrigation and drainage decision module outputs the irrigation and drainage decisions of the rice fields in the group.

8. The rice intelligent irrigation and drainage decision-making system based on cluster analysis according to claim 7, characterized in that: The data processing module is also used to establish an empirical data set based on historical rice field irrigation and drainage data.

Images