Baby cabbage irrigation strategy optimization method based on soil quality monitoring

Abstract

The invention provides a baby cabbage irrigation strategy optimization method based on soil quality monitoring, and relates to the technical field of baby cabbage irrigation strategy optimization, and the method comprises the steps: collecting historical irrigation parameters, soil quality parameters and corresponding baby cabbage quality scores; constructing a baby cabbage quality prediction model, and performing model training by taking irrigation and soil quality parameters as input and baby cabbage quality scores as labels; obtaining a value range of irrigation parameters, randomly generating an initial optimized irrigation parameter combination, inputting the initial optimized irrigation parameter combination and real-time soil quality parameters into a model, predicting a corresponding baby cabbage quality score, and constructing a fitness function; taking each group of initial optimized irrigation parameters as an individual, taking a real-time soil quality parameter as a fixed condition, calculating a fitness function value through a genetic algorithm, and obtaining an optimal irrigation parameter combination with the highest fitness; and the optimal irrigation parameter combination is corrected by utilizing the environmental data in the irrigation process. According to the method, scientific optimization of baby cabbage irrigation strategies is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of baby cabbage irrigation strategy optimization, and in particular to a baby cabbage irrigation strategy optimization method based on soil quality monitoring. Background Art

[0002] With the continued growth of the global population and the acceleration of urbanization, the demand for agricultural water is increasing, especially in arid and semi-arid regions, where water shortages are becoming increasingly prominent. In traditional agricultural irrigation practices, farmers often use empirical methods for irrigation. Especially when growing water-intensive crops such as baby cabbage, irrigation arrangements are often based on weather changes or experience. This method not only lacks scientific basis, but also easily leads to excessive or insufficient use of water resources, affecting soil quality and structure, leading to nutrient loss and soil degradation. At the same time, traditional irrigation methods make it difficult to monitor soil moisture and quality in real time, and are unable to effectively respond to environmental changes. This leads to uneven growth and reduced quality of baby cabbage, ultimately affecting agricultural output and farmers' income.

[0003] Furthermore, modern consumers have increasingly high expectations for the quality of baby cabbage, especially in the face of fierce market competition. Farmers urgently need to improve the growth quality and yield of baby cabbage. However, due to the lack of intelligent soil quality monitoring systems, farmers are often unable to make scientific decisions during irrigation based on key parameters such as soil quality and environmental parameters. This lack of information makes it difficult for farmers to effectively adjust irrigation amounts and intervals, resulting in an imbalance in soil moisture management, which in turn affects the growth environment and quality of baby cabbage. Therefore, establishing an irrigation management system based on real-time soil quality monitoring and intelligent decision-making can not only effectively improve irrigation efficiency and reduce water waste, but also ensure the scientific and stable growth of baby cabbage, becoming a key need in the transformation of agricultural modernization.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for optimizing irrigation strategies for baby cabbage based on soil quality monitoring, so as to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for optimizing irrigation strategy for baby cabbage based on soil quality monitoring, comprising the following steps:

[0008] S1, obtaining several sets of historical irrigation parameters, and simultaneously obtaining soil quality parameters and baby cabbage quality scores corresponding to each set of historical irrigation parameters, and mapping the historical irrigation parameters, soil quality parameters, and corresponding baby cabbage quality scores one by one to form a sample data set, wherein the irrigation parameters include single irrigation amount, irrigation interval, and irrigation duration, and the soil quality parameters include soil pH, soil moisture, and soil conductivity;

[0009] S2, building a baby cabbage quality prediction model based on a deep learning algorithm, taking the irrigation parameters and soil quality parameters in the sample data set as input, and the corresponding baby cabbage quality scores as labels, training the baby cabbage quality prediction model to obtain a trained baby cabbage quality prediction model;

[0010] S3, obtaining a value range for each irrigation parameter, randomly generating several groups of initial optimized irrigation parameter combinations within the value range, inputting the initial optimized irrigation parameter combinations and real-time soil quality parameters into a baby cabbage quality prediction model, obtaining a baby cabbage quality score prediction value corresponding to the combination, and constructing a fitness function based on the baby cabbage quality score prediction value, wherein the initial optimized irrigation parameter combinations include a single irrigation amount, an irrigation interval, and an irrigation duration;

[0011] S4, taking a set of initial optimized irrigation parameter combinations as an individual, the irrigation parameters in the combination as genes, and the real-time soil quality parameters as fixed environmental conditions, optimizing the irrigation parameters under their constraints, calculating the fitness function value of each individual, and obtaining the optimal irrigation parameter combination through a genetic algorithm, wherein the optimal irrigation parameter combination is the initial optimized irrigation parameter combination with the highest fitness function value;

[0012] S5, obtaining environmental data during the irrigation process, modifying the parameters in the optimal irrigation parameter combination based on the obtained environmental data, obtaining precise values of the irrigation parameters, and optimizing the baby cabbage irrigation strategy based on the precise values of the irrigation parameters, wherein the environmental parameters include light intensity and air temperature.

[0013] Furthermore, the baby cabbage quality scoring standard is 1-10 points, with the quality increasing in order, and is determined by measuring the weight of the single plant, firmness, vitamin C content and combining it with expert scoring;

[0014] The specific logic for obtaining soil quality parameters is as follows: sensors are deployed in the 0-20cm tillage layer according to a 1×1m grid in the baby cabbage planting area. The data sampling frequency is 1 time / hour. The pH value, moisture and conductivity of the monitored soil are collected multiple times, and the average result is taken as the overall soil pH value, soil moisture and soil conductivity.

[0015] Furthermore, the specific logic of constructing the baby cabbage quality prediction model is as follows:

[0016] Acquire several groups of historical irrigation parameters, and simultaneously acquire the soil quality parameters and baby cabbage quality scores corresponding to each group of historical irrigation parameters, map the historical irrigation parameters, soil quality parameters and corresponding baby cabbage quality scores one by one to form a sample data set, and randomly divide the sample data set into a training set and a test set, wherein the irrigation parameters include the single irrigation amount, irrigation interval and irrigation duration, and the soil quality parameters include soil pH value, soil moisture and soil conductivity; construct a baby cabbage quality prediction model based on a deep learning algorithm, use the historical irrigation parameters and soil quality parameters in the training set as input, use the corresponding baby cabbage quality scores as labels, train the model to obtain a trained baby cabbage quality prediction model, substitute the historical irrigation parameters and soil quality parameters in the test set into the trained model, and obtain the corresponding prediction results; calculate the error between the prediction result and the actual value in the test set; determine whether the error meets a preset error threshold; if so, output the trained model, i.e., the baby cabbage quality prediction model; if not, return to continue training; the error is the mean absolute error, root mean square error and determination coefficient between the prediction result and the actual value in the test set;

[0017] The process of randomly dividing the data into training and test sets has the following logic: randomly sort the sample data set, use 80% of the sorted sample data set as the training set, and the remaining 20% as the test set.

[0018] Furthermore, the error is the mean absolute error, root mean square error, and coefficient of determination between the predicted result and the actual value in the test set, as follows:

[0019] The mean absolute error is:

[0020]

[0021] The root mean square error is:

[0022]

[0023] The coefficient of determination is:

[0024]

[0025] in, and y i Respectively represent the average value, predicted value and actual value of the baby cabbage quality score corresponding to the data of the i-th group of test samples, n is the number of test sample groups in the test sample set, and the actual value is the average value of the actual value of the baby cabbage quality score of the n groups of test samples;

[0026] When MAE≤∈ MAE , RMSE≤∈ RMSE and When , it means that the error meets the preset error threshold and the model completes training; where ∈ MAE Preset error threshold for MAE, ∈ RMSE Preset error threshold for RMSE, R 2 Preset error threshold.

[0027] Furthermore, a set of initial optimized irrigation parameter combinations was taken as an individual, the irrigation parameters within the combination were taken as genes, and the real-time soil quality parameters were taken as fixed environmental conditions. All initial irrigation parameter combinations were encoded to obtain several individuals, and an initial population was constructed based on all individuals. Each individual contained three genes, namely, the single irrigation amount, irrigation interval, and irrigation duration.

[0028] Furthermore, the specific logic for obtaining the optimal irrigation parameter combination through the genetic algorithm is as follows:

[0029] The initial population is subjected to cyclic selection, crossover, and mutation operations to generate an iterative population containing multiple new individuals. It is judged whether the maximum number of iterations has been reached. When the maximum number of iterations has been reached, the individual with the highest fitness function value in the current population is selected as the optimal irrigation parameter combination. Otherwise, the generated iterative population is used as the initial population for iterative operation until the iteration termination condition is met. The initial optimized irrigation parameter combination with the largest fitness value that appears during the iteration process is selected as the optimal irrigation parameter combination, where the iteration termination condition is the set maximum number of iterations, and the method is terminated early when the best fitness increases by less than 1% for 10 consecutive generations. The selection operation adopts a tournament selection of size 3, with a crossover probability Pc=0.8 and a mutation probability Pm=0.05.

[0030] Furthermore, the formula for calculating the fitness function is as follows:

[0031]

[0032] In the formula, SVS is the fitness function value, SCO pre The predicted value of the baby cabbage quality score output by the model θ is soil moisture, EC is soil electrical conductivity, pH is soil pH, α, β, γ and δ are preset weights, α>β>γ>δ>0, and α+β+γ+δ=1 is satisfied.

[0033] Furthermore, light sensors and temperature sensors are used to obtain the light intensity and air temperature of the current baby cabbage planting area. Based on the dimensionless light intensity and air temperature, the parameters in the optimal irrigation parameter combination are corrected. The formula is as follows:

[0034]

[0035] Where Virr is the exact value of single irrigation volume, V0 is the optimal single irrigation volume within the optimal irrigation parameter combination, T int is the exact value of the irrigation interval, T0 is the optimal irrigation interval within the optimal irrigation parameter combination, D dur is the exact value of irrigation duration, D0 is the optimal irrigation duration within the optimal irrigation parameter combination, T is the air temperature, S is the light intensity, k1 and k2, m and u, p and q are three groups of weights, and k1>k2>0, m>u>0, p>q>0;

[0036] The optimization of the baby cabbage irrigation strategy is completed based on the precise values of single irrigation amount, irrigation interval and irrigation duration.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] This invention achieves intelligent optimization of irrigation strategies for baby cabbage by combining soil quality monitoring technology with deep learning algorithms. First, real-time monitoring of key soil parameters such as pH, moisture, and conductivity enables farmers to obtain accurate soil environmental data. This information helps farmers promptly identify changing trends in soil moisture, enabling them to make informed irrigation decisions. This ensures that baby cabbage receives the required water at each growth stage and avoids quality degradation caused by over- or under-irrigation.

[0039] Secondly, a deep learning algorithm, which models the relationship between irrigation parameters and baby cabbage quality, can predict the optimal irrigation plan based on historical data and real-time monitoring results. By optimizing irrigation parameter combinations through a genetic algorithm, this plan can dynamically adjust irrigation strategies to adapt to varying climate conditions and soil conditions, thereby improving irrigation flexibility and responsiveness.

[0040] By combining these technologies, this solution not only significantly improves irrigation efficiency and reduces water waste, but also ensures the growth quality of baby cabbages and enhances their market competitiveness. Furthermore, the intelligent irrigation management system provides farmers with a sustainable agricultural production method, promotes the development of smart agriculture, and provides new ideas and practical basis for the future transformation of agriculture. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 Schematic diagram of the overall method of the present invention. DETAILED DESCRIPTION

[0042] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0043] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0044] Example:

[0045] See also Figure 1 , the present invention provides a technical solution:

[0046] A method for optimizing irrigation strategy for baby cabbage based on soil quality monitoring, comprising the following steps:

[0047] S1, obtaining several sets of historical irrigation parameters, and simultaneously obtaining soil quality parameters and baby cabbage quality scores corresponding to each set of historical irrigation parameters, and mapping the historical irrigation parameters, soil quality parameters, and corresponding baby cabbage quality scores one by one to form a sample data set, wherein the irrigation parameters include single irrigation amount, irrigation interval, and irrigation duration, and the soil quality parameters include soil pH, soil moisture, and soil conductivity;

[0048] In this embodiment, the baby cabbage quality scoring standard is 1-10 points, and the quality is ranked in ascending order by measuring the weight of the single plant, firmness, vitamin C content and combining it with expert scoring.

[0049] The specific logic for obtaining soil quality parameters is as follows: sensors are deployed in the 0-20cm tillage layer according to a 1×1m grid in the baby cabbage planting area. The data sampling frequency is 1 time / hour. The pH value, moisture and conductivity of the monitored soil are collected multiple times, and the average result is taken as the overall soil pH value, soil moisture and soil conductivity.

[0050] The advantage of step S1 lies in systematically acquiring historical irrigation parameters, soil quality parameters, and baby cabbage quality scores, and mapping these data one by one to form a sample dataset. This process provides a solid foundation for subsequent data analysis and model building. Compared with existing technologies, S1 ensures data comprehensiveness and accuracy, avoiding model instability or inaccurate predictions caused by a lack of systematic data.

[0051] In this solution, step S1 not only provides a high-quality data source for model training but also, through the systematic integration of historical data, makes the irrigation strategy optimization process more scientific and reliable. This step effectively improves irrigation efficiency and the growth quality of baby cabbage, providing accurate input for subsequent deep learning models and thus improving prediction accuracy. This significantly enhances the overall solution's scientific and practicality, promoting agricultural modernization and achieving precision irrigation and sustainable development.

[0052] S2, building a baby cabbage quality prediction model based on a deep learning algorithm, taking the irrigation parameters and soil quality parameters in the sample data set as input, and the corresponding baby cabbage quality scores as labels, training the baby cabbage quality prediction model to obtain a trained baby cabbage quality prediction model;

[0053] In this embodiment, the specific logic of constructing the baby cabbage quality prediction model is as follows:

[0054] Acquire several groups of historical irrigation parameters, and simultaneously acquire the soil quality parameters and baby cabbage quality scores corresponding to each group of historical irrigation parameters, map the historical irrigation parameters, soil quality parameters and corresponding baby cabbage quality scores one by one to form a sample data set, and randomly divide the sample data set into a training set and a test set, wherein the irrigation parameters include the single irrigation amount, irrigation interval and irrigation duration, and the soil quality parameters include soil pH value, soil moisture and soil conductivity; construct a baby cabbage quality prediction model based on a deep learning algorithm, use the historical irrigation parameters and soil quality parameters in the training set as input, use the corresponding baby cabbage quality scores as labels, train the model to obtain a trained baby cabbage quality prediction model, substitute the historical irrigation parameters and soil quality parameters in the test set into the trained model, and obtain the corresponding prediction results; calculate the error between the prediction result and the actual value in the test set; determine whether the error meets a preset error threshold; if so, output the trained model, i.e., the baby cabbage quality prediction model; if not, return to continue training; the error is the mean absolute error, root mean square error and determination coefficient between the prediction result and the actual value in the test set;

[0055] The process of randomly dividing the data into training and test sets has the following logic: randomly sort the sample data set, use 80% of the sorted sample data set as the training set, and the remaining 20% as the test set.

[0056] The error is the mean absolute error, root mean square error, and coefficient of determination between the predicted result and the actual value in the test set, as follows:

[0057] The mean absolute error is:

[0058]

[0059] The root mean square error is:

[0060]

[0061] The coefficient of determination is:

[0062]

[0063] in, and y i Respectively represent the average value, predicted value and actual value of the baby cabbage quality score corresponding to the data of the i-th group of test samples, n is the number of test sample groups in the test sample set, and the actual value is the average value of the actual value of the baby cabbage quality score of the n groups of test samples;

[0064] When MAE≤∈ MAE , RMSE≤∈R MSE and When , it means that the error meets the preset error threshold and the model completes training; where ∈ MAE Preset error threshold for MAE, ∈ RMSE Preset error threshold for RMSE, R 2 Preset error threshold.

[0065] The advantage of step S2 lies in the use of a deep learning algorithm to construct a baby cabbage quality prediction model, using historical irrigation parameters and soil quality parameters as input to generate accurate quality score predictions. This process can automatically capture complex nonlinear relationships. Compared to traditional linear regression or empirical formulas, deep learning models have greater adaptability and predictive capabilities, can better handle multidimensional feature data, and improve prediction accuracy.

[0066] In this solution, the implementation of step S2 significantly enhances the scientific nature and effectiveness of the entire optimization scheme. By establishing an efficient prediction model, farmers can obtain real-time estimates of baby cabbage quality, enabling them to make more precise adjustments to their irrigation strategies. This not only improves irrigation efficiency and water resource utilization, but also ensures the quality of the vegetables, thereby promoting intelligent and sustainable agricultural production. Through the implementation of this step, the entire irrigation strategy optimization process will be more data-driven, reducing the influence of human factors on the decision-making process and comprehensively improving agricultural management.

[0067] S3, obtaining a value range for each irrigation parameter, randomly generating several groups of initial optimized irrigation parameter combinations within the value range, inputting the initial optimized irrigation parameter combinations and real-time soil quality parameters into a baby cabbage quality prediction model, obtaining a baby cabbage quality score prediction value corresponding to the combination, and constructing a fitness function based on the baby cabbage quality score prediction value, wherein the initial optimized irrigation parameter combinations include a single irrigation amount, an irrigation interval, and an irrigation duration;

[0068] The advantage of step S3 in this embodiment lies in obtaining the range of irrigation parameter values and randomly generating an initial optimized irrigation parameter combination, which is then combined with real-time soil quality parameters for comprehensive evaluation. This process effectively explores different irrigation strategies, increasing the diversity and flexibility of irrigation parameters. Compared to existing technologies, traditional irrigation methods often rely on experience or simple fixed patterns. This step, through a data-driven approach, can more scientifically adapt to different soil conditions and improve optimization accuracy.

[0069] In this solution, the implementation of step S3 significantly enhances the intelligence level of the overall optimization scheme. The randomly generated initial parameter combinations, combined with real-time soil quality data, ensure that the irrigation strategy is not only based on historical data but also adapts promptly to current environmental changes. This dynamic adjustment capability can effectively improve irrigation efficiency and the growth quality of baby cabbage, reduce water waste, and increase crop yield and quality. Overall, the implementation of this step promotes the precision and sustainable development of agricultural production, providing a more scientific decision-making basis for modern agricultural management.

[0070] S4, taking a set of initial optimized irrigation parameter combinations as an individual, the irrigation parameters in the combination as genes, and the real-time soil quality parameters as fixed environmental conditions, optimizing the irrigation parameters under their constraints, calculating the fitness function value of each individual, and obtaining the optimal irrigation parameter combination through a genetic algorithm, wherein the optimal irrigation parameter combination is the initial optimized irrigation parameter combination with the highest fitness function value;

[0071] In this embodiment, a set of initial optimized irrigation parameter combinations is taken as an individual, the irrigation parameters within the combination are taken as genes, and the real-time soil quality parameters are taken as fixed environmental conditions. All initial irrigation parameter combinations are encoded to obtain a total of several individuals, and an initial population is constructed based on all individuals. Each individual contains three genes, namely, the single irrigation amount, the irrigation interval, and the irrigation duration.

[0072] The specific logic of obtaining the optimal irrigation parameter combination through genetic algorithm is as follows:

[0073] The initial population is subjected to cyclic selection, crossover, and mutation operations to generate an iterative population containing multiple new individuals. The maximum number of iterations is determined. If the maximum number of iterations is reached, the individual with the highest fitness function value in the current population is selected as the optimal irrigation parameter combination. Otherwise, the generated iterative population is used as the initial population and iterated until the termination criterion is met. The initial optimized irrigation parameter combination with the highest fitness value is selected as the optimal irrigation parameter combination. The termination criterion is the maximum number of iterations set, and the optimal fitness is terminated early if the improvement in the optimal fitness is less than 1% for 10 consecutive generations. The selection process uses a tournament selection of size 3, with a crossover probability Pc = 0.8 and a mutation probability Pm = 0.05. Preliminary experiments have verified that Pc = 0.8 balances global search and convergence speed, while Pm = 0.05 prevents the destruction of high-quality genes.

[0074] The formula for calculating the fitness function is as follows:

[0075]

[0076] In the formula, SVS is the fitness function value, SCO pre The output of the model is the predicted value of the baby cabbage quality score θ, where soil moisture, EC, soil electrical conductivity, pH, and α, β, γ, and δ are preset weights, α = 0.5, β = 0.25, γ = 0.15, and δ = 0.1. The weights are set in this way because SCO pre It is the most critical factor in predicting the quality of baby cabbage and is directly related to the quality of the final baby cabbage product. Therefore, it is given the highest weight. Soil moisture θ is an important factor affecting plant growth. Appropriate humidity can effectively promote plant root development and water absorption. Since soil moisture has a greater impact on crop growth, its weight is set to be second only to SCO. pre Soil pH affects the effectiveness of nutrients, but its tolerance range is wide and its impact on crop growth is relatively small. In particular, changes within the appropriate range have limited impact on quality, so it is given the lowest weight.

[0077] In this formula, the dependent variable SVS is used to measure the advantages and disadvantages of a specific irrigation parameter combination, reflecting the expected impact of this combination on the quality of baby cabbage under current soil and environmental conditions. The value of the fitness function directly affects the optimal selection of irrigation strategies. The higher the value, the more advantageous the current irrigation parameter combination is for the growth of baby cabbage. Therefore, the calculation results of SVS will guide the adjustment of irrigation parameters. In terms of technical effectiveness, SVS can quantify the impact of different irrigation parameter combinations on baby cabbage growth, providing a scientific basis for decision-making, thereby improving the efficiency of water resource utilization and the growth quality of crops. For the quality prediction term α*SCO preThis item uses a linear form to directly reflect the impact of irrigation strategies on the quality of baby cabbage. The linear relationship is consistent with the positive correlation between agricultural expert scores and actual yields. Giving this item the maximum weight reflects the optimization goal of quality priority, which is significantly different from the existing irrigation strategy that only considers yield. This item uses an exponential decay form to simulate the nonlinear effect of water stress. When the soil moisture θ deviates from the optimal value by 70%, the stomatal conductance of the baby cabbage decreases exponentially. The denominator 5 controls the penalty intensity, making the function value > 0.8 in the interval [65%, 75%]. It decays rapidly after exceeding this range, which is highly consistent with the crop physiological response curve. The design is a piecewise linear function to reflect the threshold effect of salt damage. When EC = 800μS / cm, the nutrient supply and salt damage risk are balanced. When EC>1100μS / cm, the function value is negative, and the high-salt risk solution is automatically excluded. When EC<500μS / cm, the function value decreases but remains positive, reflecting that mild nutrient deficiency is acceptable. The weight γ is set to 0.15 to ensure that it does not affect the dominant optimization direction in non-saline soil. For the pH penalty item Linear constraints are used to reflect the buffering properties of pH. When pH = 6.5, trace elements such as iron and zinc are most effective. The tolerance range is set to ±1.2 to cover the natural buffering capacity of most soils. When pH < 5.5 or pH > 7.5, the function value is reset to zero, forcing the algorithm to avoid extreme acid and alkaline conditions. The weight δ is set to 0.10 to reflect its relatively minor but non-negligible role. This shows that SCO pre It was positively correlated with SVS, while |θ-70|, |EC-800|, and |pH-6.5| were negatively correlated with SVS.

[0078] The advantage of step S4 lies in treating the initial optimized irrigation parameter combination as an individual and the irrigation parameters as genes, which are then optimized using a genetic algorithm. This method simulates the process of natural selection, effectively exploring the parameter space and finding the optimal solution. Compared with traditional optimization methods, genetic algorithms have a higher global search capability and can avoid local optimal solutions, thereby improving the applicability and effectiveness of the resulting irrigation parameter combination.

[0079] In this solution, implementing step S4 significantly improves the overall optimization effect. Through iterative optimization using a genetic algorithm, irrigation parameters can be dynamically adjusted to better align with real-time soil quality and environmental conditions. This process not only increases the flexibility of irrigation strategies but also effectively improves water resource utilization efficiency and crop growth performance, ultimately achieving higher yields and better quality. Overall, the implementation of step S4 enhances the intelligence level of agricultural production, making decision-making more scientific and rational, and promoting the development of sustainable agriculture.

[0080] S5, obtaining environmental data during the irrigation process, modifying parameters within the optimal irrigation parameter combination based on the obtained environmental data to obtain precise values of the irrigation parameters, and optimizing the baby cabbage irrigation strategy based on the precise values of the irrigation parameters, wherein the environmental parameters include light intensity and air temperature;

[0081] In this embodiment, a light sensor and a temperature sensor are used to obtain the light intensity and air temperature of the current baby cabbage planting area. The parameters in the optimal irrigation parameter combination are corrected based on the dimensionless light intensity and air temperature. The formula is as follows:

[0082]

[0083] Where V irr is the exact value of single irrigation volume, V0 is the optimal single irrigation volume within the optimal irrigation parameter combination, T int is the exact value of the irrigation interval, T0 is the optimal irrigation interval within the optimal irrigation parameter combination, D dur is the exact value of irrigation duration, D0 is the optimal irrigation duration within the optimal irrigation parameter combination, T is the air temperature, S is the light intensity, k1 and k2, m and u, p and q are three groups of weights, k1 = 0.6, k2 = 0.4, m = 0.7, u = 0.3, p = 0.55, q = 0.45.

[0084] For the first formula in the above formula group, V irr It means that the actual irrigation amount adjusted according to environmental conditions can more accurately meet the water needs of crops, thereby supporting the healthy growth of crops. Its technical effect is that by dynamically adjusting the irrigation amount, it can improve the utilization efficiency of water resources, reduce water waste, and ensure that crops always have appropriate water supply under different climatic conditions, thereby improving crop yield and quality. At the same time, this method also helps to achieve smart agricultural management and optimize irrigation strategies. Temperature directly affects the transpiration of crops. At higher temperatures, the evaporation rate of crop water will increase, so the irrigation amount needs to be increased to maintain a suitable moisture state. In the formula, k1*T 2This reflects the nonlinear effect of temperature on irrigation rate. Higher temperatures generally lead to a significant increase in evaporation. Therefore, using a squared term to represent the exacerbated temperature effect more accurately reflects the impact of high temperatures on water demand. Light intensity influences crop photosynthesis and growth rate. When light intensity is high, crop growth accelerates, and water demand also increases accordingly to meet growth needs. Therefore, with greater light intensity, the corresponding irrigation rate also needs to increase. This relationship is reflected in the formula k²*S. Measured data shows that for every 5°C increase in temperature, the soil water consumption rate increases by more than 30%. Therefore, it is given a higher weight to ensure adequate water replenishment during high temperatures. Although light intensity directly affects photosynthesis, its correlation with water consumption is less than that with temperature, so it is given a lower weight.

[0085] For the second formula in the above formula group, the dependent variable T int represents the actual adjusted irrigation interval under given environmental conditions. It reflects the variation in irrigation frequency required by crops under different environmental conditions. Its technical benefit lies in the ability to dynamically adjust irrigation intervals to adapt to varying climatic conditions. Reasonable irrigation intervals improve water resource utilization efficiency, reduce water waste, and ensure that crops receive adequate water at the appropriate time, thereby improving crop growth quality and yield. The logarithmic representation is primarily used to model the nonlinear response characteristics of plant growth and water requirements. The logarithmic function effectively captures the gradual nature of the effects of temperature T and light intensity S on irrigation intervals. As temperature T and light intensity S increase, the rate of change in irrigation intervals does not increase linearly, but rather exhibits a decreasing trend. This characteristic manifests itself in actual agriculture: irrigation requirements fluctuate significantly under lower temperatures and light intensities, while variations tend to be more stable under higher conditions. Therefore, a logarithmic representation more accurately reflects this nonlinear relationship, enabling more rational and effective irrigation management. Using a logarithmic representation also improves model stability and reduces the impact of extreme values, enabling better adaptation to irrigation requirements under varying climatic conditions.

[0086] Temperature directly affects plant transpiration and growth rate. Higher temperatures lead to increased evaporation and respiration, which in turn increases the need for water. It shows that an increase in temperature will lead to a shortening of the irrigation interval, that is, more frequent irrigation is required under high temperature conditions. Light intensity S affects the photosynthesis and growth rate of plants. The higher the light intensity, the faster the plant grows and accordingly requires more water to support growth. Therefore, an increase in light intensity will lead to a decrease in T int shortened, which is expressed in the formula by High temperatures accelerate soil drying, necessitating significantly shorter irrigation intervals. Experiments have shown that when the temperature rises from 25°C to 30°C, the optimal irrigation interval decreases by 35%, so this is given a dominant weight. Light indirectly influences transpiration by regulating stomatal opening and closing, and its effects are delayed and milder than those of temperature, so it is given a lower weight.

[0087] For the third formula in the above formula group, the dependent variable D dur It represents the actual irrigation duration after adjustment under given environmental conditions. This value reflects the length of irrigation time required for crops based on changes in environmental factors. Its technical effect is that through this formula, the duration of irrigation can be determined more accurately to adapt to the water needs of crops under different environmental conditions. Reasonable irrigation duration can reduce the waste of water resources and ensure that crops always get enough water under different climatic conditions, thereby improving the growth effect and yield of crops. In addition, this formula can be used in intelligent irrigation systems, combined with real-time climate data, to automatically optimize irrigation plans and improve the efficiency of agricultural management. Temperature directly affects the transpiration and growth rate of crops. Higher air temperatures will lead to increased evaporation and water demand of crops. Therefore, the effect of temperature on irrigation duration is positive, and p*T in the formula represents the degree of influence of temperature on irrigation duration. Light intensity affects the photosynthesis and growth rate of plants. Under conditions of higher light intensity, the water demand of crops will also increase, so longer irrigation time is required to meet the needs of crops. q*S in the formula represents the effect of light intensity on irrigation duration. The formula uses the square root form It not only reflects the nonlinear response characteristics of plants to environmental factors, but also enhances the biological rationality and robustness of the model, making it more operational in practical applications. This design enables irrigation strategies to better adapt to different climatic conditions, thereby improving water resource utilization efficiency and crop growth. Although high temperatures require a longer single irrigation time, the irrigation volume V irr The time correction needs to be relatively conservative to avoid water leakage, so the square root function is used to suppress over-correction. Strong light requires prolonged irrigation to ensure water penetration depth, but its weight is slightly lower than the temperature term weight to prevent V irr Correction stacking leads to waste.

[0088] The optimization of the baby cabbage irrigation strategy is completed based on the precise values of single irrigation amount, irrigation interval and irrigation duration.

[0089] The advantage of step S5 is that by acquiring environmental data during the irrigation process, it dynamically adjusts the optimal irrigation parameter combination, allowing the irrigation strategy to adapt to environmental changes in real time. This method avoids rigid irrigation plans and improves irrigation flexibility and precision. Compared with existing technologies, traditional irrigation methods generally rely on preset parameters and lack a real-time response mechanism to environmental changes. Step S5, by incorporating real-time data such as light intensity and air temperature, ensures the scientific nature and effectiveness of the irrigation strategy.

[0090] In this solution, the implementation of step S5 can significantly improve the intelligence and sustainability of the overall irrigation strategy. By analyzing and applying real-time environmental data, irrigation parameters can be optimized according to different growth stages and climatic conditions, maximizing water resource utilization efficiency and crop growth quality. This dynamic adjustment capability not only reduces water waste but also effectively improves the quality and yield of baby cabbage, thereby bringing higher economic benefits to farmers. Overall, the implementation of step S5 promotes the precise management of agricultural production and provides strong support for the realization of smart agriculture.

[0091] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0092] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0093] 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, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0094] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A method for optimizing irrigation strategy for baby cabbage based on soil quality monitoring, characterized in that: The specific steps include: S1, obtaining several sets of historical irrigation parameters, and simultaneously obtaining soil quality parameters and baby cabbage quality scores corresponding to each set of historical irrigation parameters, and mapping the historical irrigation parameters, soil quality parameters, and corresponding baby cabbage quality scores one by one to form a sample data set, wherein the irrigation parameters include single irrigation amount, irrigation interval, and irrigation duration, and the soil quality parameters include soil pH, soil moisture, and soil conductivity; S2, building a baby cabbage quality prediction model based on a deep learning algorithm, taking the irrigation parameters and soil quality parameters in the sample data set as input, and the corresponding baby cabbage quality scores as labels, training the baby cabbage quality prediction model to obtain a trained baby cabbage quality prediction model; S3, obtaining a value range for each irrigation parameter, randomly generating several groups of initial optimized irrigation parameter combinations within the value range, inputting the initial optimized irrigation parameter combinations and real-time soil quality parameters into a baby cabbage quality prediction model, obtaining a baby cabbage quality score prediction value corresponding to the combination, and constructing a fitness function based on the baby cabbage quality score prediction value, wherein the initial optimized irrigation parameter combinations include a single irrigation amount, an irrigation interval, and an irrigation duration; S4, taking a set of initial optimized irrigation parameter combinations as an individual, the irrigation parameters in the combination as genes, and the real-time soil quality parameters as fixed environmental conditions, optimizing the irrigation parameters under their constraints, calculating the fitness function value of each individual, and obtaining the optimal irrigation parameter combination through a genetic algorithm, wherein the optimal irrigation parameter combination is the initial optimized irrigation parameter combination with the highest fitness function value; S5, obtaining environmental data during the irrigation process, modifying the parameters in the optimal irrigation parameter combination based on the obtained environmental data, obtaining precise values of the irrigation parameters, and optimizing the baby cabbage irrigation strategy based on the precise values of the irrigation parameters, wherein the environmental parameters include light intensity and air temperature.

2. The method for optimizing irrigation strategy for baby cabbage based on soil quality monitoring according to claim 1, characterized in that: The quality scoring standard of baby cabbage is 1-10 points, and the quality increases in order. The quality is determined by measuring the weight of the single plant, firmness, vitamin C content and combining it with expert scoring. The specific logic for obtaining soil quality parameters is as follows: sensors are deployed in the 0-20cm tillage layer according to a 1×1m grid in the baby cabbage planting area. The data sampling frequency is 1 time / hour. The pH value, moisture and conductivity of the monitored soil are collected multiple times, and the average result is taken as the overall soil pH value, soil moisture and soil conductivity.

3. The method for optimizing irrigation strategy for baby cabbage based on soil quality monitoring according to claim 1, characterized in that: The specific logic of constructing the baby cabbage quality prediction model is as follows: Acquire several groups of historical irrigation parameters, and simultaneously acquire the soil quality parameters and baby cabbage quality scores corresponding to each group of historical irrigation parameters, map the historical irrigation parameters, soil quality parameters and corresponding baby cabbage quality scores one by one to form a sample data set, and randomly divide the sample data set into a training set and a test set, wherein the irrigation parameters include the single irrigation amount, irrigation interval and irrigation duration, and the soil quality parameters include soil pH value, soil moisture and soil conductivity; construct a baby cabbage quality prediction model based on a deep learning algorithm, use the historical irrigation parameters and soil quality parameters in the training set as input, use the corresponding baby cabbage quality scores as labels, train the model to obtain a trained baby cabbage quality prediction model, substitute the historical irrigation parameters and soil quality parameters in the test set into the trained model, and obtain the corresponding prediction results; calculate the error between the prediction result and the actual value in the test set; determine whether the error meets a preset error threshold; if so, output the trained model, i.e., the baby cabbage quality prediction model; if not, return to continue training; the error is the mean absolute error, root mean square error and determination coefficient between the prediction result and the actual value in the test set; The process of randomly dividing the data into training and test sets has the following logic: randomly sort the sample data set, use 80% of the sorted sample data set as the training set, and the remaining 20% as the test set.

4. The method for optimizing irrigation strategy for baby cabbage based on soil quality monitoring according to claim 3, characterized in that: The error is the mean absolute error, root mean square error, and coefficient of determination between the predicted result and the actual value in the test set, as follows: The mean absolute error is: The root mean square error is: The coefficient of determination is: in, and y i Respectively represent the average value, predicted value and actual value of the baby cabbage quality score corresponding to the data of the i-th group of test samples, n is the number of test sample groups in the test sample set, and the actual value is the average value of the actual value of the baby cabbage quality score of the n groups of test samples; When MAE≤∈ MAE 、 RMSE ≤∈ RMSE and When , it means that the error meets the preset error threshold and the model completes training; where ∈ MAE Preset error threshold for MAE, ∈ RMSE Preset error threshold for RMSE, R 2 Preset error threshold.

5. The method for optimizing irrigation strategy for baby cabbage based on soil quality monitoring according to claim 1, characterized in that: A set of initial optimized irrigation parameter combinations was taken as an individual, the irrigation parameters within the combination were taken as genes, and the real-time soil quality parameters were taken as fixed environmental conditions. All initial irrigation parameter combinations were encoded to obtain several individuals. An initial population was constructed based on all individuals. Each individual contained three genes, namely, single irrigation amount, irrigation interval, and irrigation duration.

6. The method for optimizing irrigation strategy for baby cabbage based on soil quality monitoring according to claim 1, characterized in that: The specific logic of obtaining the optimal irrigation parameter combination through genetic algorithm is as follows: The initial population is subjected to cyclic selection, crossover, and mutation operations to generate an iterative population containing multiple new individuals. It is judged whether the maximum number of iterations has been reached. When the maximum number of iterations has been reached, the individual with the highest fitness function value in the current population is selected as the optimal irrigation parameter combination. Otherwise, the generated iterative population is used as the initial population for iterative operation until the iteration termination condition is met. The initial optimized irrigation parameter combination with the largest fitness value that appears during the iteration process is selected as the optimal irrigation parameter combination, where the iteration termination condition is the set maximum number of iterations, and the method is terminated early when the best fitness increases by less than 1% for 10 consecutive generations. The selection operation adopts a tournament selection of size 3, with a crossover probability Pc=0.8 and a mutation probability Pm=0.

05.

7. The method for optimizing irrigation strategy for baby cabbage based on soil quality monitoring according to claim 6, characterized in that: The formula for calculating the fitness function is as follows: In the formula, SVS is the fitness function value, SCO pre The predicted value of the baby cabbage quality score output by the model θ is soil moisture, EC is soil electrical conductivity, pH is soil pH, α, β, γ and δ are preset weights, α>β>γ>δ>0, and α+β+γ+δ=1 is satisfied.

8. The method for optimizing irrigation strategy for baby cabbage based on soil quality monitoring according to claim 1, characterized in that: Use light sensors and temperature sensors to obtain the light intensity and air temperature of the current baby cabbage planting area. Based on the dimensionless light intensity and air temperature, the parameters in the optimal irrigation parameter combination are modified. The formula is as follows: Where V irr is the exact value of single irrigation volume, V0 is the optimal single irrigation volume within the optimal irrigation parameter combination, T int is the exact value of the irrigation interval, T0 is the optimal irrigation interval within the optimal irrigation parameter combination, D dur is the exact value of irrigation duration, D0 is the optimal irrigation duration within the optimal irrigation parameter combination, T is the air temperature, S is the light intensity, k1 and k2, m and u, p and q are three groups of weights, and k1>k2>0, m>u>0, p>q>0; The optimization of the baby cabbage irrigation strategy is completed based on the precise values of single irrigation amount, irrigation interval and irrigation duration.

Images