An intelligent irrigation regulation method and system

Abstract

This invention discloses an intelligent irrigation regulation method and system. The method includes: collecting soil data, including humidity, light intensity, and ambient temperature; preprocessing the soil data; calculating the proximity of the current irrigation scheme to the optimal solution based on the preprocessed data using an adaptive nonlinear lemming optimization algorithm combined with dynamic weight adjustment and multi-objective optimization strategies; updating the optimal irrigation scheme and generating early warning information; and adjusting the start / stop and water volume of irrigation based on the early warning information. This invention can improve the accuracy, adaptability, and sustainability of the irrigation system, optimize water resource utilization efficiency, and enhance the ability to regulate the crop growth environment.

Description

Technical Field

[0001] This invention relates to a regulation method and system, and more particularly to an intelligent irrigation regulation method and system, belonging to the field of agricultural irrigation technology. Background Technology

[0002] Irrigation plays a vital role in agricultural production. Traditional irrigation methods rely heavily on manual experience, making it difficult to precisely control irrigation volume and timing. This can easily lead to water waste or insufficient moisture for crops, thus affecting crop growth and agricultural yield. While existing smart irrigation systems use sensors to monitor parameters such as soil moisture and ambient temperature, most systems lack efficient data analysis and decision optimization capabilities, making it difficult to meet the irrigation needs of different crops and complex environments. Furthermore, some systems fail to fully utilize intelligent algorithms to optimize irrigation strategies, resulting in low water resource utilization efficiency and impacting sustainable agricultural development. With the advancement of IoT, big data, and AI technologies, smart irrigation systems are developing towards greater precision and efficiency. Optimization control methods based on intelligent algorithms can significantly improve the accuracy of irrigation decisions and achieve rational allocation of water resources. However, existing technologies still have shortcomings in areas such as intelligent decision optimization, remote monitoring, and adaptability to complex environments. There is an urgent need for a more precise, efficient, and intelligent irrigation control method to improve agricultural production efficiency and promote the sustainable development of smart agriculture. Summary of the Invention

[0003] Purpose of the invention: The purpose of this invention is to provide an intelligent irrigation regulation method and system that can improve the accuracy of irrigation regulation.

[0004] Technical solution: The intelligent irrigation regulation method of the present invention includes:

[0005] (1) Collect soil data, including humidity, light intensity and ambient temperature;

[0006] (2) The soil data is preprocessed. Based on the preprocessed data, the degree of closeness between the current irrigation scheme and the optimal solution is calculated by using an adaptive nonlinear lemming optimization algorithm combined with dynamic weight adjustment and multi-objective optimization strategy. The optimal irrigation scheme is then updated and early warning information is generated.

[0007] (3) Adjust the start and stop of irrigation and the amount of water according to the early warning information.

[0008] Furthermore, in step (1), soil moisture data is collected by a humidity sensor, light intensity data is collected by a light sensor, and ambient temperature data is collected by a temperature sensor.

[0009] Further, step (2) includes:

[0010] (201) Normalize the collected data, linearly scaling it to the [0,1] interval while preserving the relative size relationship. The formula is as follows:

[0011]

[0012] In the formula, x′ is the normalized value, x is the original data collected; min(x) is the minimum value of all samples in the original data; max(x) is the maximum value of all samples in the original dataset;

[0013] (202) Based on the lemming optimization algorithm, the lemming population size is set to N individuals, representing the corresponding N irrigation schemes. Each individual represents a possible irrigation scheme. It is assumed that each individual corresponds to a three-dimensional vector composed of irrigation time T, irrigation frequency F, and irrigation amount W:

[0014] x i =(T i ,F i W i )

[0015] In the formula, x i This represents the irrigation scheme for the i-th individual; i = 1, 2, ..., N; T i ∈[T min ,T max ], F i ∈[F min ,F max ], W i ∈[W min W max ], where T min and T max F represents the minimum and maximum values ​​of irrigation time T. min and F max W represents the minimum and maximum values ​​of the irrigation frequency F. min and W max Let W be the minimum and maximum values ​​of the irrigation water volume.

[0016] Latin hypercube sampling is used to uniformly divide the value range of each parameter into N equal parts. A point is randomly selected from each part, and an irrigation scheme is generated for each individual based on the results of Latin hypercube sampling. The N individuals in the population are represented by a matrix X, where each row represents an individual and an irrigation scheme, as shown in the matrix below:

[0017]

[0018] In the formula, X represents the lemming population; (T N ,F N W N ) represents the Nth individual;

[0019] (203) The fitness of each individual is calculated using an objective function, which is used to simultaneously optimize economic cost, water resource utilization rate, and soil moisture deviation, as shown in the following formula:

[0020] γ=θ×C

[0021]

[0022] J = f(γ,λ,η)

[0023] In the formula, γ represents the economic cost; θ represents the economic cost weight; C represents the actual irrigation cost; λ represents the water resource utilization rate; and w i δ represents the irrigation water consumption under the current scheme; W represents the total available water resources; δ represents the water demand matching weight; ET c ET represents the theoretical water requirement of crops. α η represents the actual transpiration evaporation of the crop; α represents the soil moisture deviation; β represents the water resource utilization rate weight; m represents the soil moisture matching weight; and S represents the number of monitoring points. j S represents the soil moisture content at the j-th monitoring point; opt The optimal soil moisture content; u j Here, J represents the weight parameters; J = f(·) is the fitness function, representing the objective function value.

[0024] (204) Sort the individuals in the population in ascending order of fitness value, and retain the optimal individuals in a preset proportion; in each generation iteration, all individuals are evaluated according to the fitness function J=f(γ,λ,η); using the roulette wheel selection method, individuals with higher fitness are selected according to the probability of being selected calculated from the fitness value, as shown in the following formula:

[0025]

[0026] In the formula, f i f is the fitness value of the i-th irrigation scheme; min Where m is the minimum fitness value, m is the current irrigation scheme, and N is the total number of irrigation schemes;

[0027] (205) A nonlinear energy factor is introduced to determine whether the current algorithm is in the exploration or development stage; the formula for determining the nonlinear energy factor is as follows:

[0028]

[0029] In the formula, E (t) It is a nonlinear energy factor; t is the iteration number; T is the preset maximum iteration number; where, when E (t) When E > 1, the algorithm is in the exploration phase, performing a large-scale search; when E(t) When E ≤ 1, the algorithm is in the development stage and performs a local search. min The minimum energy threshold;

[0030] An adaptive adjustment strategy is adopted. If the optimal irrigation scheme has not been updated for several consecutive times, the exploration scope is expanded, and the adjustment formula is as follows:

[0031]

[0032] In the formula, X represents the parameter combination for the i-th irrigation scheme in generation t+1; random The solution is randomly selected from the current optimization schemes; τ is the perturbation coefficient;

[0033] (206) During the exploration phase, parameter adjustments are made, expressed by the following formula:

[0034] x i (t+1)=x i (t)+Δr·d i

[0035] In the formula, x i (t) is the parameter set of the i-th irrigation scheme in the t-th iteration; Δ is the preset adjustment range during the exploration phase; r is a random number ranging from [0,1]; d i It is an exploration direction randomly selected in this iteration;

[0036] During the development phase, a detailed search is conducted in the identified optimal areas, and the irrigation plan is optimized or adjusted to address unforeseen circumstances. The adjustment formula is as follows:

[0037] x i (t+1)=x i (t)+ν·r·d i +n i

[0038] In the formula, x i (t) is the adjustment variable for the i-th irrigation scheme in the t-th iteration; ν is the preset adjustment range during the development phase; n i It is a random noise vector;

[0039] (207) The difference between each irrigation scheme and the current optimal irrigation strategy is calculated using the distance metric method to determine their similarity and quantify the merits of each scheme. The formula is as follows:

[0040]

[0041] In the formula, D(·,·) is the Euclidean distance; x iLet x be the parameter vector for the i-th irrigation scheme; * d represents the parameter vector of the optimal irrigation strategy; d is the dimension of the optimization problem, i.e., the number of irrigation parameters; x ik Let be the value of the i-th irrigation scheme for the k-th parameter; Let be the value of the k-th parameter of the optimal irrigation strategy;

[0042] (208) Update the current optimal irrigation strategy; let the fitness value of the current optimal strategy be Fitness. best The corresponding irrigation scheme is Agent. best The fitness value for the i-th irrigation scheme is Fitness. i If Fitness best <Fitness i Then update the optimal irrigation strategy:

[0043] Fitness best =Fitness i

[0044] Agent best =Agent i

[0045] (209) Based on the results obtained by the adaptive nonlinear lemming optimization algorithm, compare them with the set objective function to determine whether the termination condition is met; if the termination condition is not met, return to step (202); if the condition is met, proceed to step (210).

[0046] (210) Output the optimal solution, determine the optimal irrigation scheme, including the optimal irrigation time, optimal irrigation frequency and optimal irrigation water volume, and generate early warning information.

[0047] Furthermore, step (3) includes: receiving early warning information and controlling the start and stop of irrigation through an intelligent timer based on the optimal irrigation plan to adjust the irrigation amount and duration.

[0048] Based on the same inventive concept, the present invention also provides an intelligent irrigation regulation system, comprising:

[0049] An environmental monitoring unit is used to collect soil data, including humidity, light intensity, and ambient temperature.

[0050] The control unit is used to preprocess soil data. Based on the preprocessed data, it calculates the proximity of the current irrigation scheme to the optimal solution using an adaptive nonlinear lemming optimization algorithm combined with dynamic weight adjustment and multi-objective optimization strategy, updates the optimal irrigation scheme, and generates early warning information.

[0051] The irrigation regulation unit is used to adjust the start / stop of irrigation and the amount of water based on early warning information.

[0052] Furthermore, the environmental monitoring unit collects soil moisture data via a humidity sensor, light intensity data via a light sensor, and ambient temperature data via a temperature sensor.

[0053] Furthermore, the method for implementing the control unit includes the following steps:

[0054] (201) Normalize the collected data, linearly scaling it to the [0,1] interval while preserving the relative size relationship. The formula is as follows:

[0055]

[0056] In the formula, x′ is the normalized value, x is the original data collected; min(x) is the minimum value of all samples in the original data; max(x) is the maximum value of all samples in the original dataset;

[0057] (202) Based on the lemming optimization algorithm, the lemming population size is set to N individuals, representing the corresponding N irrigation schemes. Each individual represents a possible irrigation scheme. It is assumed that each individual corresponds to a three-dimensional vector composed of irrigation time T, irrigation frequency F, and irrigation amount W:

[0058] x i =(T i ,F i W i )

[0059] In the formula, x i This represents the irrigation scheme for the i-th individual; i = 1, 2, ..., N; T i ∈[T min ,T max ], F i ∈[F min ,F max ], W i ∈[W min W max ], where T min and T max F represents the minimum and maximum values ​​of irrigation time T. min and F max W represents the minimum and maximum values ​​of the irrigation frequency F. min and W max Let W be the minimum and maximum values ​​of the irrigation water volume.

[0060] Latin hypercube sampling is used to uniformly divide the value range of each parameter into N equal parts. A point is randomly selected from each part, and an irrigation scheme is generated for each individual based on the results of Latin hypercube sampling. The N individuals in the population are represented by a matrix X, where each row represents an individual and an irrigation scheme, as shown in the matrix below:

[0061]

[0062] In the formula, X represents the lemming population; (T N ,F N W N ) represents the Nth individual;

[0063] (203) The fitness of each individual is calculated using an objective function, which is used to simultaneously optimize economic cost, water resource utilization rate, and soil moisture deviation, as shown in the following formula:

[0064] γ=θ×C

[0065]

[0066] J = f(γ,λ,η)

[0067] In the formula, γ represents the economic cost; θ represents the economic cost weight; C represents the actual irrigation cost; λ represents the water resource utilization rate; and w i δ represents the irrigation water consumption under the current scheme; W represents the total available water resources; δ represents the water demand matching weight; ET c ET represents the theoretical water requirement of crops. α η represents the actual transpiration evaporation of the crop; α represents the soil moisture deviation; β represents the water resource utilization rate weight; m represents the soil moisture matching weight; and S represents the number of monitoring points. j S represents the soil moisture content at the j-th monitoring point; opt The optimal soil moisture content; u j Here, J represents the weight parameters; J = f(·) is the fitness function, representing the objective function value.

[0068] (204) Sort the individuals in the population in ascending order of fitness value, and retain the optimal individuals in a preset proportion; in each generation iteration, all individuals are evaluated according to the fitness function J=f(γ,λ,η); using the roulette wheel selection method, individuals with higher fitness are selected according to the probability of being selected calculated from the fitness value, as shown in the following formula:

[0069]

[0070] In the formula, f i f is the fitness value of the i-th irrigation scheme; minThe minimum fitness value is given by m, where m is the current irrigation scheme and N is the total number of irrigation schemes; (205) A nonlinear energy factor is introduced to determine whether the current algorithm is in the exploration or development stage; the formula for determining the nonlinear energy factor is as follows:

[0071]

[0072] In the formula, E (t) It is a nonlinear energy factor; t is the iteration number; T is the preset maximum iteration number; where, when E (t) When E > 1, the algorithm is in the exploration phase, performing a large-scale search; when E (t) When E ≤ 1, the algorithm is in the development stage and performs a local search. min The minimum energy threshold;

[0073] An adaptive adjustment strategy is adopted. If the optimal irrigation scheme has not been updated for several consecutive times, the exploration scope is expanded, and the adjustment formula is as follows:

[0074]

[0075] In the formula, X represents the parameter combination for the i-th irrigation scheme in generation t+1; random The solution is randomly selected from the current optimization schemes; τ is the perturbation coefficient;

[0076] (206) During the exploration phase, parameter adjustments are made, expressed by the following formula:

[0077] x i (t+1)=x i (t)+Δr·d i

[0078] In the formula, x i (t) is the parameter set of the i-th irrigation scheme in the t-th iteration; Δ is the preset adjustment range during the exploration phase; r is a random number ranging from [0,1]; d i It is an exploration direction randomly selected in this iteration;

[0079] During the development phase, a detailed search is conducted in the identified optimal areas, and the irrigation plan is optimized or adjusted to address unforeseen circumstances. The adjustment formula is as follows:

[0080] x i (t+1)=x i (t)+ν·r·d i +n i

[0081] In the formula, x i(t) is the adjustment variable for the i-th irrigation scheme in the t-th iteration; ν is the preset adjustment range during the development phase; n i It is a random noise vector;

[0082] (207) The difference between each irrigation scheme and the current optimal irrigation strategy is calculated using the distance metric method to determine their similarity and quantify the merits of each scheme. The formula is as follows:

[0083]

[0084] In the formula, D(·,·) is the Euclidean distance; x i Let x be the parameter vector for the i-th irrigation scheme; * d represents the parameter vector of the optimal irrigation strategy; d is the dimension of the optimization problem, i.e., the number of irrigation parameters; x ik Let be the value of the i-th irrigation scheme for the k-th parameter; Let be the value of the k-th parameter of the optimal irrigation strategy;

[0085] (208) Update the current optimal irrigation strategy; let the fitness value of the current optimal strategy be Fitness. best The corresponding irrigation scheme is Agent. best The fitness value for the i-th irrigation scheme is Fitness. i If Fitness best <Fitness i Then update the optimal irrigation strategy:

[0086] Fitness best =Fitness i

[0087] Agent best =Agent i

[0088] (209) Based on the results obtained by the adaptive nonlinear lemming optimization algorithm, compare them with the set objective function to determine whether the termination condition is met; if the termination condition is not met, return to step (202); if the condition is met, proceed to step (210).

[0089] (210) Output the optimal solution, determine the optimal irrigation scheme, including the optimal irrigation time, optimal irrigation frequency and optimal irrigation water volume, and generate early warning information.

[0090] Furthermore, the irrigation control unit includes: receiving early warning information, and controlling the start and stop of irrigation through an intelligent timer according to the optimal irrigation plan, so as to adjust the irrigation amount and duration.

[0091] Based on the same inventive concept, the present invention also provides a computing device, comprising: one or more processors, one or more memories, and one or more programs, the programs being stored in the memory and configured to be executed by the processor, the programs being loaded onto the processor to implement the steps of the intelligent irrigation regulation method according to any of the preceding claims.

[0092] Based on the same inventive concept, the present invention also provides a storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor, cause the processor to perform the steps of the intelligent irrigation regulation method according to any one of the preceding claims.

[0093] Beneficial Effects: Compared with existing technologies, this invention has the following significant advantages: 1. This invention adopts an adaptive nonlinear lemming optimization algorithm, a biomimetic intelligent optimization method, which performs optimization search by simulating the survival behavior of lemmings. This provides a novel solution for intelligent irrigation systems and has strong exploration and development capabilities; 2. By introducing a nonlinear energy factor, this invention allows the algorithm to determine whether it is in the exploration or development stage based on the number of iterations, thus enabling extensive exploration in the search space or fine optimization in a better region. This dynamic adjustment mechanism effectively avoids getting trapped in local optima and accelerates the search for the global optimum; 3. The algorithm of this invention designs a multi-objective fitness function that comprehensively considers economic cost, water resource utilization rate, and soil moisture deviation. By simultaneously optimizing multiple objectives, it ensures that the irrigation scheme can achieve optimal economic and resource utilization while guaranteeing crop growth. Attached Figure Description

[0094] Figure 1 This is a flowchart of a method according to an embodiment of the present invention;

[0095] Figure 2 This is a schematic diagram of the lemming optimization algorithm according to an embodiment of the present invention. Detailed Implementation

[0096] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0097] As attached Figure 1 As shown, the intelligent irrigation regulation method of this embodiment includes:

[0098] (1) Collect soil data, including humidity, light intensity and ambient temperature;

[0099] (2) The soil data is preprocessed. Based on the preprocessed data, the degree of closeness between the current irrigation scheme and the optimal solution is calculated by using an adaptive nonlinear lemming optimization algorithm combined with dynamic weight adjustment and multi-objective optimization strategy. The optimal irrigation scheme is then updated and early warning information is generated.

[0100] (3) Adjust the start and stop of irrigation and the amount of water according to the early warning information.

[0101] Furthermore, in step (1), soil moisture data is collected by a humidity sensor, light intensity data is collected by a light sensor, and ambient temperature data is collected by a temperature sensor.

[0102] Further, step (2) includes:

[0103] (201) Normalize the collected data, linearly scaling it to the [0,1] interval while preserving the relative size relationship. The formula is as follows:

[0104]

[0105] In the formula, x′ is the normalized value, x is the original data collected; min(x) is the minimum value of all samples in the original data; max(x) is the maximum value of all samples in the original dataset;

[0106] (202) such as Figure 2 As shown, based on the lemming optimization algorithm, the lemming population size is set to N individuals, representing the corresponding N irrigation schemes. Each individual represents a possible irrigation scheme. It is assumed that each individual corresponds to a three-dimensional vector composed of irrigation time T, irrigation frequency F, and irrigation amount W:

[0107] x i =(T i ,F i W i )

[0108] In the formula, x i This represents the irrigation scheme for the i-th individual; i = 1, 2, ..., N; T i ∈[T min ,T max ], F i ∈[F min ,F max ], W i ∈[W min W max ], where T min and T max F represents the minimum and maximum values ​​of irrigation time T. min and F max W represents the minimum and maximum values ​​of the irrigation frequency F.min and W max Let W be the minimum and maximum values ​​of the irrigation water volume.

[0109] Latin hypercube sampling is used to uniformly divide the value range of each parameter into N equal parts. A point is randomly selected from each part, and an irrigation scheme is generated for each individual based on the results of Latin hypercube sampling. The N individuals in the population are represented by a matrix X, where each row represents an individual and an irrigation scheme, as shown in the matrix below:

[0110]

[0111] In the formula, X represents the lemming population; (T N ,F N W N ) represents the Nth individual;

[0112] (203) The fitness of each individual is calculated using an objective function, which is used to simultaneously optimize economic cost, water resource utilization rate, and soil moisture deviation, as shown in the following formula:

[0113] γ=θ×C

[0114]

[0115] J = f(γ,λ,η)

[0116] In the formula, γ represents the economic cost; θ represents the economic cost weight; C represents the actual irrigation cost; λ represents the water resource utilization rate; and w i δ represents the irrigation water consumption under the current scheme; W represents the total available water resources; δ represents the water demand matching weight; ET c ET represents the theoretical water requirement of crops. α η represents the actual transpiration evaporation of the crop; α represents the soil moisture deviation; β represents the water resource utilization rate weight; m represents the soil moisture matching weight; and S represents the number of monitoring points. j S represents the soil moisture content at the j-th monitoring point; opt The optimal soil moisture content; u j Here, J represents the weight parameters; J = f(·) is the fitness function, representing the objective function value.

[0117] (204) Sort the individuals in the population in ascending order of fitness value, and retain the optimal individuals in a preset proportion; through dynamic evaluation and selection mechanism, promote the population to optimize towards a better irrigation scheme, increase the opportunities for individuals with higher fitness to mutate and mate, thereby optimizing the performance of the irrigation scheme; in each generation iteration, all individuals are evaluated according to the fitness function J=f(γ,λ,η); the roulette wheel selection method is adopted, and individuals with higher fitness are selected according to the probability of being selected calculated from the fitness value, as shown in the following formula:

[0118]

[0119] In the formula, f i f is the fitness value of the i-th irrigation scheme; min Where m is the minimum fitness value, m is the current irrigation scheme, and N is the total number of irrigation schemes;

[0120] (205) A nonlinear energy factor is introduced to determine whether the current algorithm is in the exploration or development stage; the formula for determining the nonlinear energy factor is as follows:

[0121]

[0122] In the formula, E (t) It is a nonlinear energy factor; t is the iteration number; T is the preset maximum iteration number; where, when E (t) When E > 1, the algorithm is in the exploration phase, performing a large-scale search; when E (t) When E ≤ 1, the algorithm is in the development stage and performs a local search. min This is the minimum energy threshold used to control the exploration range of the algorithm, ensuring that a certain degree of diversity is maintained during the optimization process;

[0123] An adaptive adjustment strategy is adopted. If the optimal irrigation scheme has not been updated for several consecutive times, the exploration scope is expanded, and the adjustment formula is as follows:

[0124]

[0125] In the formula, X represents the parameter combination for the i-th irrigation scheme in generation t+1; random The solution is randomly selected from the current optimization schemes; τ is the perturbation coefficient;

[0126] (206) During the exploration phase, parameter adjustments are made, expressed by the following formula:

[0127] x i (t+1)=x i (t)+Δr·d i

[0128] In the formula, x i (t) is the parameter set of the i-th irrigation scheme at the t-th iteration; Δ is the maximum adjustment range preset during the exploration phase, i.e., the maximum adjustment range for irrigation time is 7.2 hours, the maximum adjustment range for irrigation frequency is 3 times / day, and the maximum adjustment range for irrigation volume can be 25 liters / time; r is a random number in the range [0,1]; d i It is an exploration direction randomly selected in this iteration;

[0129] During the development phase, a detailed search is conducted in the identified optimal areas, and the irrigation plan is optimized or adjusted to address unforeseen circumstances. The adjustment formula is as follows:

[0130] x i (t+1)=x i (t)+v·r·d i +n i

[0131] In the formula, x i (t) is the adjustment variable for the i-th irrigation scheme in the t-th iteration; ν is the preset small adjustment range during the development phase; n i It is a random noise vector;

[0132] (207) The difference between each irrigation scheme and the current optimal irrigation strategy is calculated using the distance metric method to determine their similarity and quantify the merits of each scheme. The formula is as follows:

[0133]

[0134] In the formula, D(·,·) is the Euclidean distance; x i Let x be the parameter vector for the i-th irrigation scheme; * d represents the parameter vector of the optimal irrigation strategy; d is the dimension of the optimization problem, i.e., the number of irrigation parameters; x ik Let be the value of the i-th irrigation scheme for the k-th parameter; Let be the value of the k-th parameter of the optimal irrigation strategy;

[0135] (208) Update the current optimal irrigation strategy; let the fitness value of the current optimal strategy be Fitness. best The corresponding irrigation scheme is Agent. best The fitness value for the i-th irrigation scheme is Fitness. i If Fitness best <Fitness i Then update the optimal irrigation strategy:

[0136] Fitness best =Fitnessi

[0137] Agent best =Agent i

[0138] (209) Based on the results obtained by the adaptive nonlinear lemming optimization algorithm, compare them with the set objective function to determine whether the termination condition is met; if the termination condition is not met, return to step (202); if the condition is met, proceed to step (210).

[0139] (210) Output the optimal solution, determine the optimal irrigation scheme, including the optimal irrigation time, optimal irrigation frequency and optimal irrigation water volume, and generate early warning information. The early warning information mechanism is as follows: Assuming that the current soil moisture is 30%, and the optimal irrigation scheme recommends a moisture range of 40% to 50% (the optimal irrigation scheme ensures that the soil moisture is maintained within the target range by comprehensively considering the irrigation time, frequency and amount), the control system will generate a "low humidity warning" to achieve intelligent irrigation.

[0140] Furthermore, step (3) includes: receiving early warning information and controlling the start and stop of irrigation through an intelligent timer based on the optimal irrigation plan to adjust the irrigation amount and duration.

[0141] Based on the same inventive concept, the present invention also provides an intelligent irrigation regulation system, comprising:

[0142] An environmental monitoring unit is used to collect soil data, including humidity, light intensity, and ambient temperature.

[0143] The control unit is used to preprocess soil data. Based on the preprocessed data, it calculates the proximity of the current irrigation scheme to the optimal solution using an adaptive nonlinear lemming optimization algorithm combined with dynamic weight adjustment and multi-objective optimization strategy, updates the optimal irrigation scheme, and generates early warning information.

[0144] The irrigation control unit is used to adjust the start / stop of irrigation and the amount of water based on early warning information.

[0145] Furthermore, the environmental monitoring unit collects data such as soil moisture, light intensity, and ambient temperature through humidity sensors, light sensors, and temperature sensors to accurately understand the current actual environmental conditions of the farmland, providing reliable and accurate data support for subsequent irrigation decisions.

[0146] Furthermore, the control module uses a central processing unit to process the data collected by the sensors, and employs an adaptive nonlinear lemming optimization algorithm to filter and optimize the data in order to evaluate the adaptability of the irrigation strategy, calculate the optimal result, and thus generate early warning information.

[0147] Furthermore, the steps of the adaptive nonlinear lemming optimization algorithm are as follows:

[0148] 1) Collect environmental and crop growth data through sensors, such as soil moisture, ambient temperature and light intensity; Since the collected data have different dimensions, we need to normalize all indicators, that is, use Min-Max normalization to linearly scale the data to the [0,1] interval, retain the relative size relationship, so as to facilitate subsequent processing.

[0149]

[0150] In the formula, x′ is the normalized value, x is the original data; min(x) is the minimum value of all samples in the dataset; max(x) is the maximum value of all samples in the dataset, used to adjust the upper limit;

[0151] 2) Set the population size to 50 individuals, representing 50 possible irrigation schemes. Each individual represents a possible irrigation scheme. Assume each individual corresponds to a three-dimensional vector consisting of irrigation time, frequency, and irrigation amount. These three variables are fundamental to the subsequent optimization process; they directly affect the calculation of fitness, thus determining the performance of each individual in the optimization process.

[0152] x i =(T i ,F i W i )

[0153] In the formula, x i This represents the irrigation plan for the i-th individual; i takes values ​​of 1, 2, ..., N; and the irrigation time T. i ∈[T min ,T max Irrigation frequency F i ∈[F min ,F max Irrigation volume W i ∈[W min W max ], where T min and T max F represents the minimum and maximum values ​​of irrigation time T. min and F max W represents the minimum and maximum values ​​of the irrigation frequency F. min and W max The minimum and maximum values ​​of irrigation water volume W.

[0154] Latin hypercube sampling is employed, which involves uniformly dividing the value range of each parameter into N equal parts, and then randomly selecting a point from each part to ensure sampling uniformity. Based on the results of Latin hypercube sampling, an irrigation scheme is generated for each individual. Finally, the N individuals in the population can be represented by a matrix X, where each row represents an individual, i.e., an irrigation scheme, as shown in the matrix below:

[0155]

[0156] In the formula, X represents the lemming population; (T N ,F N W N ) represents the Nth individual;

[0157] 3) The fitness of each individual is calculated by the objective function, which aims to simultaneously optimize economic cost, water resource utilization rate, and soil moisture deviation. The objective function formula is as follows:

[0158] γ=θ×C

[0159]

[0160] J = f(γ,λ,η)

[0161] In the formula, γ represents the economic cost, indicating the economic consumption level of the irrigation scheme; θ represents the economic cost weight, determining the influence of economic cost on the objective function; C represents the actual irrigation cost, including electricity, water, and equipment maintenance costs; λ represents the water resource utilization rate, measuring the rationality of water resource use; w i δ represents the irrigation water consumption under the current scheme; W represents the total available water resources; δ represents the water demand matching weight, which determines the importance of the transpiration and evaporation terms; ET represents the total irrigation water consumption under the current scheme. c ET represents the theoretical water requirement of crops. α η represents the actual transpiration and evaporation of the crop; η is the soil moisture deviation, which measures the suitability of the irrigation scheme for soil moisture; α is the water resource utilization rate weight; β is the soil moisture matching weight, which controls the influence of this term in the objective function; m is the number of monitoring points, indicating how many soil moisture monitoring points there are; S j S represents the soil moisture content at the j-th monitoring point; opt The optimal soil moisture content; u j is the weight parameter; J = f(·) is the fitness function, i.e., the objective function value to be obtained;

[0162] 4) Sort the individuals in the population in ascending order of fitness value, retaining the top 20% of the best individuals to ensure that high-quality solutions are not eliminated. Through dynamic evaluation and selection mechanisms, the population is driven towards optimizing the irrigation scheme, increasing the opportunities for individuals with higher fitness to mutate and mate, thereby optimizing the performance of the irrigation scheme. In each generation iteration, all individuals are evaluated according to the fitness function J = f(γ,λ,η); a roulette wheel selection method is used, based on each individual's fitness value, selecting individuals with higher fitness according to the probability of being selected calculated from the fitness value for further operations, ensuring that the optimization process continuously progresses towards the optimal solution, as shown in the following formula:

[0163]

[0164] In the formula, f i f is the fitness value of the i-th irrigation scheme; min Where m is the minimum fitness value, m is the current irrigation scheme, and N is the total number of irrigation schemes;

[0165] 5) During the optimization process, to balance global exploration and local development, a nonlinear energy factor is introduced to determine whether the algorithm is currently in the exploration or development phase. The exploration phase primarily involves extensively searching the search space for regions where the optimal solution might exist, while the development phase involves a refined search within the already found, relatively optimal regions. The nonlinear energy factor is dynamically adjusted based on the number of iterations to determine the optimal solution. The energy factor determination formula is shown below:

[0166]

[0167] In the formula, E (t) It is a nonlinear energy factor; t is the iteration number; T is the preset maximum iteration number; where, when E (t) When E > 1, the algorithm is in the exploration phase, performing a large-scale search; when E (t) When E ≤ 1, the algorithm is in the development stage and performs a local search. min This is the minimum energy threshold used to control the exploration range of the algorithm, ensuring that a certain degree of diversity is maintained during the optimization process;

[0168] During the optimization process, we adopt an adaptive adjustment strategy to ensure that the irrigation scheme can be continuously optimized and improve water resource utilization. If the optimal irrigation scheme is not updated for k consecutive times, it may indicate that the system is trapped in a local optimum. In this case, we need to expand the exploration scope to make the adjustment of irrigation time, frequency, and water volume more flexible. The adjustment formula is as follows:

[0169]

[0170] In the formula, X represents the parameter combination for the i-th irrigation scheme in generation t+1; random The scheme is randomly selected from the current optimization schemes to introduce new optimization directions; τ is the perturbation coefficient, usually taken as τ = 1.5, to ensure that the optimization process is neither too aggressive nor too slow;

[0171] 6) The methods for adjusting irrigation plans differ at different stages;

[0172] During the exploratory phase, extensive parameter adjustments are required to ensure that different irrigation strategies are thoroughly tested. Specific adjustments can be represented as follows:

[0173] x i (t+1)=x i (t)+Δr·d i

[0174] In the formula, x i (t) is the parameter set of the i-th irrigation scheme in the t-th iteration; Δ is the maximum adjustment range set during the exploration phase to ensure that different combinations of irrigation parameters are tried, i.e., the maximum adjustment range for irrigation time is 7.2 hours, the maximum adjustment range for irrigation frequency is 3 times / day, and the maximum adjustment range for irrigation volume can be 25 liters / time; r is a random number in the range [0,1], introducing randomness to increase the diversity of optimization; d i It is an exploration direction randomly selected in this iteration to guide the irrigation parameters to evolve towards different possible solutions;

[0175] During the development phase, a detailed search is conducted in the identified optimal areas, and the irrigation plan is optimized or adjusted to address unforeseen circumstances. The adjustment formula is as follows:

[0176] x i (t+1)=x i (t)+v·r·d i +n i

[0177] In the formula, x i (t) is the adjustment variable for the i-th irrigation scheme in the t-th iteration; v is a small adjustment range set during the development phase for more refined searching; n i This is a random noise vector introduced to avoid getting trapped in local optima. Its elements are usually randomly drawn from a normal distribution with a mean of 0 and a small standard deviation.

[0178] 7) The distance metric method is used to calculate the difference between each irrigation scheme and the current optimal irrigation strategy to determine their similarity, thereby quantifying the merits of each scheme and helping to select the optimal irrigation strategy. The formula is shown below:

[0179]

[0180] In the formula, D(·,·) is the Euclidean distance; x i Let x be the parameter vector for the i-th irrigation scheme; * d represents the parameter vector of the optimal irrigation strategy; d is the dimension of the optimization problem, i.e., the number of irrigation parameters; x ik Let be the value of the i-th irrigation scheme for the k-th parameter; Let be the value of the k-th parameter of the optimal irrigation strategy;

[0181] 8) After calculating the fitness values ​​of all irrigation schemes, the current optimal irrigation strategy needs to be updated; let the fitness value of the current optimal strategy be Fitness. best The corresponding irrigation scheme is Agent. best For the i-th irrigation scheme, its fitness value is Fitness. i If Fitness best <Fitness i Then update the optimal irrigation strategy:

[0182] Fitness best =Fitness i

[0183] Agent best =Agent i

[0184] By continuously optimizing the optimal irrigation strategy, the system can dynamically adjust the irrigation time and amount of water according to the water requirements of crops at different growth stages. The optimization algorithm will flexibly adjust the irrigation plan according to these needs to ensure that soil moisture is maintained within a suitable range, thereby promoting the crop roots to absorb water and nutrients more effectively and improving crop growth and yield.

[0185] 9) Compare the results obtained from the adaptive nonlinear lemming optimization algorithm with the set objective function to determine whether the termination condition is met; if the termination condition is not met, return to step 2); if the condition is met, proceed to step 10.

[0186] 10) Output the optimal solution, determine the optimal intelligent irrigation system control scheme, including the best irrigation time, frequency, and irrigation water volume, and generate early warning information, which is transmitted to the intelligent irrigation controller for execution. The early warning information mechanism is as follows: Assuming the current soil moisture is 30%, and the optimal irrigation scheme recommends a moisture range of 40% to 50% (the optimal irrigation scheme ensures that the soil moisture is maintained within the target range by comprehensively considering irrigation time, frequency, and volume), the control system will generate a "low humidity warning" to achieve intelligent irrigation.

[0187] Furthermore, the irrigation regulation unit relies on an intelligent irrigation controller to automatically adjust the irrigation volume and duration based on the optimal irrigation strategy calculated by the algorithm; the intelligent timer precisely controls the start and stop of irrigation to ensure that crops absorb water at the best time and effectively reduce water waste.

[0188] The environmental monitoring unit in this embodiment integrates multiple intelligent sensors for data acquisition. The control module, relying on a central processing unit, employs an adaptive nonlinear lemming optimization algorithm to filter and optimize data, assess the adaptability of irrigation strategies, calculate the optimal solution, and generate early warning information. The irrigation regulation unit then precisely controls the irrigation process. With technological advancements, this method can be deeply integrated with emerging technologies. Specifically, big data technology is used to analyze long-term accumulated irrigation and crop growth data, continuously optimizing the algorithm model and improving the accuracy and intelligence of irrigation. During its promotion and application, feedback is collected through cooperation with agricultural research institutions and large-scale growers to continuously improve the system. Continuous optimization and improvement of this method will drive smart agriculture towards high efficiency, green practices, and sustainability.

[0189] Based on the same inventive concept, this embodiment also provides a computing device, including: one or more processors, one or more memories, and one or more programs, the programs being stored in the memory and configured to be executed by the processor, the programs being loaded onto the processor to implement the steps of the intelligent irrigation regulation method according to any of the preceding claims.

[0190] Based on the same inventive concept, this embodiment also provides a storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor, cause the processor to perform the steps of the intelligent irrigation regulation method according to any one of the above claims.

[0191] The above embodiments are merely illustrative of the technical concept and features of the present invention, intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly, and should not be construed as limiting the scope of protection of the present invention. All equivalent transformations or modifications made according to the spirit and essence of the present invention should be covered within the scope of protection of the present invention. Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A smart irrigation regulation method, characterized in that, include: (1) Collect soil data, including humidity, light intensity and ambient temperature; (2) Soil data is preprocessed. Based on the preprocessed data, the approximation between the current irrigation scheme and the optimal solution is calculated using an adaptive nonlinear lemming optimization algorithm combined with dynamic weight adjustment and multi-objective optimization strategies. The optimal irrigation scheme is then updated, and early warning information is generated. This includes: (201) Normalize the collected data and linearly scale the data to... For intervals, retaining relative size relationships, the formula is as follows: ； In the formula, Normalized value The raw data collected; It is the minimum value of all samples in the original data; This represents the maximum value among all samples in the original dataset. (202) Based on the lemming optimization algorithm, the lemming population size is set as follows: Each individual represents the corresponding... There are several irrigation schemes, with each individual representing a possible irrigation scheme. It is assumed that each individual corresponds to a scheme determined by irrigation time. Irrigation frequency and irrigation volume The three-dimensional vector formed: ； In the formula, Indicates the first Irrigation plans for individual plants; = ; , , ,in, and For irrigation time The minimum and maximum values, and irrigation frequency The minimum and maximum values, and For irrigation water volume The minimum and maximum values; Latin hypercube sampling is used to uniformly divide the value range of each parameter into... Divide the population into equal parts, randomly select one point from each part, and generate an irrigation plan for each individual based on the results of Latin hypercube sampling; then, in the population... Each individual is represented as a matrix. Each row in the matrix represents an individual and an irrigation scheme, as shown below: ； In the formula, For lemming populations; Indicates the first Individual; (203) The fitness of each individual is calculated using an objective function, which is used to simultaneously optimize economic cost, water resource utilization rate, and soil moisture deviation, as shown in the following formula: ； ； ； ； In the formula, For economic costs; Assign economic cost weight; This refers to the actual irrigation cost; For water resource utilization rate; This represents the irrigation water consumption under the current plan; This represents the total amount of available water resources. Water demand matching weights; This represents the theoretical water requirement of the crop. This refers to the actual transpiration and evaporation of the crop. This is due to soil moisture deviation; As a weight for water resource utilization rate; Weights are assigned to soil moisture. Number of monitoring points; For the first Soil moisture content at each monitoring point; The optimal soil moisture content; For the first Weight parameters for each soil monitoring point; Let be the fitness function, and let represent the objective function value. (204) Sort the individuals in the population in ascending order of fitness value, and retain the optimal individuals in a preset proportion; in each generation iteration, all individuals are determined according to the fitness function. An evaluation is conducted; a roulette wheel selection method is used, based on the fitness value of each individual, and individuals with higher fitness are selected according to the probability of being selected calculated from the fitness value, as shown in the following formula: ； In the formula, For the first Fitness values ​​for each irrigation scheme; The minimum fitness value, For the current irrigation plan, This represents the total number of irrigation schemes; (205) A nonlinear energy factor is introduced to determine whether the current algorithm is in the exploration or development stage; the formula for determining the nonlinear energy factor is as follows: ； In the formula, It is a nonlinear energy factor; It is the number of iterations; It is the preset maximum number of iterations; where, when At that time, the algorithm is in the exploratory phase, performing a wide-ranging search; when At that time, the algorithm was in the development stage and was performing a local search. The minimum energy threshold; An adaptive adjustment strategy is adopted. If the optimal irrigation scheme has not been updated for several consecutive times, the exploration scope is expanded, and the adjustment formula is as follows: ； In the formula, For the first One irrigation scheme in The combination of parameters of the generation; The solution is randomly selected from the current optimization solutions; The disturbance coefficient; (206) When in the exploration phase, parameter adjustments are made, expressed by the following formula: ； In the formula, It is the first The irrigation scheme is in the first The parameter set at the next iteration; This is the adjustment range preset during the exploration phase; It is a range of values Random numbers between; It is an exploration direction randomly selected in this iteration; During the development phase, a detailed search is conducted in the identified optimal areas, and the irrigation plan is optimized or adjusted to address unforeseen circumstances. The adjustment formula is as follows: ； In the formula, It is the first The irrigation scheme is in the first Adjustment variables during the next iteration; This is the adjustment range preset during the development phase; It is a random noise vector; (207) The difference between each irrigation scheme and the current optimal irrigation strategy is calculated using the distance metric method to determine their similarity and quantify the merits of each scheme. The formula is as follows: ； In the formula, Euclidean distance; For the first The parameter vector of each irrigation scheme; This is the parameter vector for the optimal irrigation strategy; To optimize the number of dimensions in the problem, i.e., the number of irrigation parameters; For the first The irrigation scheme is in the first The values ​​of the parameters; When the optimal irrigation strategy is the first The values ​​of the parameters; (208) Update the current optimal irrigation strategy; let the fitness value of the current optimal strategy be _____. The corresponding irrigation scheme is For the first The fitness value of each irrigation scheme is ,if Then update the optimal irrigation strategy: ； ； (209) Based on the results obtained by the adaptive nonlinear lemming optimization algorithm, compare them with the set objective function to determine whether the set objective function value is met; if the set objective function value is not met, return to step (202); if the condition is met, proceed to step (210). (210) Output the optimal solution, determine the optimal irrigation scheme, including the optimal irrigation time, optimal irrigation frequency and optimal irrigation water volume, and generate early warning information; (3) Adjust the start and stop of irrigation and the amount of water according to the early warning information.

2. The intelligent irrigation regulation method according to claim 1, characterized in that, In step (1), soil moisture data is collected by a humidity sensor, light intensity data is collected by a light sensor, and ambient temperature data is collected by a temperature sensor.

3. The intelligent irrigation regulation method according to claim 1, characterized in that, Step (3) includes: receiving early warning information, and controlling the start and stop of irrigation through an intelligent timer according to the optimal irrigation plan to adjust the irrigation amount and duration.

4. An intelligent irrigation regulation system, characterized in that, include: An environmental monitoring unit is used to collect soil data, including humidity, light intensity, and ambient temperature. The control unit is used to preprocess soil data. Based on the preprocessed data, it calculates the proximity of the current irrigation scheme to the optimal solution using an adaptive nonlinear lemming optimization algorithm combined with dynamic weight adjustment and a multi-objective optimization strategy. It then updates the optimal irrigation scheme and generates early warning information. The process includes the following steps: (201) Normalize the collected data and linearly scale the data to... For intervals, retaining relative size relationships, the formula is as follows: ； In the formula, Normalized value The raw data collected; It is the minimum value of all samples in the original data; This represents the maximum value among all samples in the original dataset. (202) Based on the lemming optimization algorithm, the lemming population size is set as follows: Each individual represents the corresponding... There are several irrigation schemes, with each individual representing a possible irrigation scheme. It is assumed that each individual corresponds to a scheme determined by irrigation time. Irrigation frequency and irrigation volume The three-dimensional vector formed: ； In the formula, Indicates the first Irrigation plans for individual plants; = ; , , ,in, and For irrigation time The minimum and maximum values, and irrigation frequency The minimum and maximum values, and For irrigation water volume The minimum and maximum values; Latin hypercube sampling is used to uniformly divide the value range of each parameter into... Divide the population into equal parts, randomly select one point from each part, and generate an irrigation plan for each individual based on the results of Latin hypercube sampling; then, in the population... Each individual is represented as a matrix. Each row in the matrix represents an individual and an irrigation scheme, as shown below: ； In the formula, For lemming populations; Indicates the first Individual; (203) The fitness of each individual is calculated using an objective function, which is used to simultaneously optimize economic cost, water resource utilization rate, and soil moisture deviation, as shown in the following formula: ； ； ； ； In the formula, For economic costs; Assign economic cost weight; This refers to the actual irrigation cost; For water resource utilization rate; This represents the irrigation water consumption under the current plan; This represents the total amount of available water resources. Water demand matching weights; This represents the theoretical water requirement of the crop. This refers to the actual transpiration and evaporation of the crop. This is due to soil moisture deviation; As a weight for water resource utilization rate; Weights are assigned to soil moisture. Number of monitoring points; For the first Soil moisture content at each monitoring point; The optimal soil moisture content; These are weight parameters; Let be the fitness function, and let represent the objective function value. (204) Sort the individuals in the population in ascending order of fitness value, and retain the optimal individuals in a preset proportion; in each generation iteration, all individuals are determined according to the fitness function. An evaluation is conducted; a roulette wheel selection method is used, based on the fitness value of each individual, and individuals with higher fitness are selected according to the probability of being selected calculated from the fitness value, as shown in the following formula: ； In the formula, For the first Fitness values ​​for each irrigation scheme; The minimum fitness value, For the current irrigation plan, This represents the total number of irrigation schemes; (205) A nonlinear energy factor is introduced to determine whether the current algorithm is in the exploration or development stage; the formula for determining the nonlinear energy factor is as follows: ； In the formula, It is a nonlinear energy factor; It is the number of iterations; It is the preset maximum number of iterations; where, when At that time, the algorithm is in the exploratory phase, performing a wide-ranging search; when At that time, the algorithm was in the development stage and was performing a local search. The minimum energy threshold; An adaptive adjustment strategy is adopted. If the optimal irrigation scheme has not been updated for several consecutive times, the exploration scope is expanded, and the adjustment formula is as follows: ； In the formula, For the first One irrigation scheme in The combination of parameters of the generation; The solution is randomly selected from the current optimization solutions; The disturbance coefficient; (206) When in the exploration phase, parameter adjustments are made, expressed by the following formula: ； In the formula, It is the first The irrigation scheme is in the first The parameter set at the next iteration; This is the adjustment range preset during the exploration phase; It is a range of values Random numbers between; It is an exploration direction randomly selected in this iteration; During the development phase, a detailed search is conducted in the identified optimal areas, and the irrigation plan is optimized or adjusted to address unforeseen circumstances. The adjustment formula is as follows: ； In the formula, It is the first The irrigation scheme is in the first Adjustment variables during the next iteration; This is the adjustment range preset during the development phase; It is a random noise vector; (207) The difference between each irrigation scheme and the current optimal irrigation strategy is calculated using the distance metric method to determine their similarity and quantify the merits of each scheme. The formula is as follows: ； In the formula, Euclidean distance; For the first The parameter vector of each irrigation scheme; This is the parameter vector for the optimal irrigation strategy; To optimize the number of dimensions in the problem, i.e., the number of irrigation parameters; For the first The irrigation scheme is in the first The values ​​of the parameters; When the optimal irrigation strategy is the first The values ​​of the parameters; (208) Update the current optimal irrigation strategy; let the fitness value of the current optimal strategy be _____. The corresponding irrigation scheme is For the first The fitness value of each irrigation scheme is ,if Then update the optimal irrigation strategy: ； ； (209) Based on the results obtained by the adaptive nonlinear lemming optimization algorithm, compare them with the set objective function to determine whether the termination condition is met; if the termination condition is not met, return to step (202); if the condition is met, proceed to step (210). (210) Output the optimal solution, determine the optimal irrigation scheme, including the optimal irrigation time, optimal irrigation frequency and optimal irrigation water volume, and generate early warning information; The irrigation control unit is used to adjust the start / stop of irrigation and the amount of water based on early warning information.

5. The intelligent irrigation regulation system according to claim 4, characterized in that, The environmental monitoring unit collects soil moisture data through a humidity sensor, light intensity data through a light sensor, and ambient temperature data through a temperature sensor.

6. The intelligent irrigation regulation system according to claim 4, characterized in that, The irrigation control unit includes: receiving early warning information, and controlling the start and stop of irrigation through an intelligent timer based on the optimal irrigation plan, so as to adjust the irrigation amount and duration.

7. A computing device, characterized in that, include: One or more processors, one or more memories, and one or more programs, said programs being stored in the memory and configured to be executed by the processor, said programs, when loaded onto the processor, implementing the steps of the intelligent irrigation regulation method according to any one of claims 1 to 3.

8. A storage medium, characterized in that, The storage medium stores a computer program, which includes program instructions that, when executed by a processor, cause the processor to perform the steps of the intelligent irrigation regulation method according to any one of claims 1 to 3.

Images