Plant greenhouse intelligent monitoring and environment optimization management system and method

Abstract

The invention discloses a plant greenhouse intelligent monitoring and environment optimization management system and method. The method comprises the steps that a detection module obtains environment data, resource data and equipment feedback data of a vegetable greenhouse in real time; the acquisition module acquires vegetable types, growth days and growth parameters; the health prediction module calculates the current growth index according to the growth parameters and predicts the index of the next growth stage; the establishing module calculates a difference value between the predicted growth index and the standard growth index through a target function; the parameter optimization module adopts an improved dung beetle optimization algorithm, adjusts an optimization process according to equipment feedback and outputs optimal cultivation parameters; the condition generation module generates a resource scarcity state according to the resource data and provides adjustment limitation; the processing module adjusts a theoretical range based on the optimal parameter and the resource constraint; and the in-greenhouse adjustment module dynamically adjusts the greenhouse environment according to the optimization result. Through integration of sensor data and an optimization algorithm, dynamic adjustment of the resource ratio in the vegetable greenhouse is realized, and the vegetable growth efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vegetable cultivation, and particularly to an intelligent monitoring and environmental optimization management system and method for a plant greenhouse. Background Art

[0002] With the construction of polar regions and the in-depth development of scientific research, the demand for self-sufficiency in polar regions is gradually increasing. Especially in terms of food supply, vegetables, as one of the daily essential foods, their cultivation management has become particularly important. However, the special climatic conditions in polar regions, such as extremely cold weather, long periods of darkness, polar day and night phenomena, and severe wind and snow, make vegetable cultivation face severe challenges. Traditional agricultural cultivation methods cannot adapt to these extreme conditions. Therefore, the development of an efficient vegetable cultivation management system suitable for polar climatic conditions has become an inevitable demand.

[0003] During the process of vegetable cultivation in polar regions, how to ensure the growth of vegetables and increase yields in extreme environments is a task that combines technicality and challenge. First of all, factors such as low temperature, insufficient light, and large humidity changes in the polar environment greatly affect the photosynthesis and growth efficiency of plants. Secondly, the scarcity of resources limits the application of traditional agricultural technologies, making how to efficiently utilize existing resources a key issue.

[0004] Existing vegetable cultivation management systems can control environmental factors such as temperature, humidity, and light to a certain extent, but most of them only focus on a single factor and often cannot respond in real time to the actual needs of vegetable growth. The limitations of such systems lead to many uncertainties in the growth and cultivation of vegetables in polar research stations. In addition, traditional environmental control and resource optimization methods are relatively extensive and do not fully consider the finiteness and dynamic changes of resources in the polar environment.

[0005] With the development of artificial intelligence and optimization algorithms, more and more intelligent management technologies have begun to be applied in the agricultural field. By combining sensor technology, data acquisition and processing technology, machine learning, and optimization algorithms, it is possible to more precisely adjust environmental conditions and dynamically optimize the growth environment of vegetables according to real-time data, thereby improving production efficiency. In particular, by using the vegetable growth prediction model and intelligent optimization algorithm in machine learning, it is possible to scientifically predict and precisely control the growth process of vegetables in the polar environment.

[0006] However, the existing technologies still face the following deficiencies: First, it is difficult to monitor and analyze the growth conditions of various vegetables in real time and accurately, especially in the context of scarce polar resources; second, there is a lack of intelligent optimization algorithms that can adapt to the dynamic changes in the polar environment, and it is impossible to effectively improve the yield and quality of vegetables under limited resource conditions; third, there is still a lack of effective technical means on how to flexibly adjust the ratio of the environment and resources in extreme environments to ensure the stability and health of vegetable growth.

[0007] Therefore, developing a vegetable planting management system that can adapt to the special environment of polar research stations to monitor and predict the growth status of vegetables in real time, and at the same time optimize the use of resources and adjust environmental conditions through intelligent algorithms has become the key to the sustainable development and self-sufficiency of polar research stations. Summary of the Invention

[0008] Aiming at the deficiencies of the existing technologies, the purpose of the present invention is to provide a plant greenhouse intelligent monitoring and environment optimization management system and method, which are used to dynamically adjust the resource ratio in the vegetable greenhouse and improve the vegetable growth efficiency.

[0009] To achieve the above purpose, the present invention provides the following technical solutions: A plant greenhouse intelligent monitoring and environment optimization management system, including:

[0010] A detection module, which is used to detect in real time the in-greenhouse environment data inside the vegetable greenhouse of the polar research station, the remaining resource data of the polar research station, and the equipment feedback data inside the vegetable greenhouse;

[0011] An acquisition module, which is used to acquire the vegetable types, vegetable growth days, and vegetable growth parameters of various vegetables in the vegetable greenhouse;

[0012] A health prediction module, connected to the detection module and the acquisition module, which is used to process the current growth index according to the vegetable growth parameters, organize the in-greenhouse environment data, the vegetable types, the vegetable growth days, and the current growth index into vegetable cultivation parameters and input them into a pre-trained vegetable growth prediction model to predict the predicted growth index of the vegetables in the next growth period;

[0013] A establishment module, connected to the health prediction module, which is used to establish an objective function according to the predicted growth index and a preset standard growth index, and the objective function is used to output an objective function value;

[0014] The parameter optimization module, connected to the detection module and the establishment module, is used to introduce the dung beetle optimization algorithm, improve the dung beetle optimization algorithm according to the device feedback data, take each group of the vegetable cultivation parameters as a dung beetle individual, use the improved dung beetle optimization algorithm to iteratively update the position of the dung beetle individual, calculate the objective function value corresponding to the dung beetle individual, and output the position of the corresponding dung beetle individual as the global optimal solution when the objective function value reaches the minimum;

[0015] The condition generation module, connected to the detection module, is used to process the remaining resource data to obtain the resource shortage state, and process the adjustment limit conditions based on the resource shortage state;

[0016] The processing module, connected to the parameter optimization module and the condition generation module, is used to process the vegetable cultivation parameters in the global optimal solution to obtain the theoretical adjustment range, and limit the theoretical adjustment range according to the adjustment limit conditions to obtain the resource optimization adjustment range;

[0017] The in-shed adjustment module, connected to the processing module, is used to dynamically adjust the air environment and the culture solution environment in the vegetable greenhouse according to the resource optimization adjustment range.

[0018] Further, the in-shed environment data includes air environment parameters and culture solution environment parameters. The air environment parameters include air temperature, air humidity, carbon dioxide concentration, and light intensity; the culture solution environment parameters include the nutrient element content of the culture solution, the pH value of the culture solution, the conductivity of the culture solution, and the oxygen content of the culture solution.

[0019] Further, the health prediction module includes:

[0020] The analysis unit is used to perform a correlation analysis on each of the air environment parameters, the culture solution environment parameters, and the current growth index to obtain a correlation analysis result;

[0021] The screening unit, connected to the analysis unit, is used to screen each of the air environment parameters and the culture solution environment parameters according to the correlation analysis result to obtain key environment parameters, and the key environment parameters include air temperature, light intensity, air humidity, carbon dioxide concentration, and the pH value of the culture solution.

[0022] Further, the health prediction module further includes:

[0023] The storage unit is used to save the historical key environment parameters, historical vegetable types, historical growth days, and historical growth indexes at multiple historical moments;

[0024] A training unit, connected to the storage unit, is configured to introduce an initial model, and use the historical key environmental parameters, the historical vegetable types, the historical growth days, and the historical growth indices at multiple historical moments as inputs, and use the historical growth indices after the next growth time period as outputs to retrain the initial model to obtain the vegetable growth prediction model.

[0025] Further, the formula of the vegetable growth prediction model is configured as:

[0026]

[0027] f4(C) = k C ·C

[0028]

[0029] wherein, T is used to represent the air temperature, L is used to represent the light intensity, H is used to represent the air humidity, C is used to represent the carbon dioxide concentration, pH is used to represent the pH value of the culture solution, S is used to represent the vegetable type, D is used to represent the vegetable growth days, GI is used to represent the vegetable growth index, GI next is used to represent the predicted growth index, f1(T) is used to represent the influence function of temperature on vegetable growth, T0 is used to represent the preset suitable temperature value, σ T is used to represent the preset standard deviation of temperature fluctuation, f2(L) is used to represent the influence of light intensity on vegetable growth, A L is used to represent the preset light intensity coefficient, L max is used to represent the preset maximum light intensity, f3(H) is used to represent the influence function of air humidity on vegetable growth, H0 is used to represent the preset humidity threshold, σ H is used to represent the preset humidity change sensitivity value, f4(C) is used to represent the influence function of carbon dioxide concentration on vegetable growth, k C is used to represent the preset carbon dioxide concentration coefficient, f5(pH) is used to represent the influence function of the pH value of the culture solution on vegetable growth, pH0 is used to represent the preset most suitable pH value, ΔpH is used to represent the preset pH value adaptation range, f6(S, D) is used to represent the influence function of vegetable type and growth days on the growth index, k S is used to represent the preset vegetable type coefficient, γ D is used to represent the preset growth rate coefficient, t is used to represent time, and λ and μ are respectively used to represent the preset first time decay coefficient and the second time decay coefficient.

[0030] Further, the device feedback data includes multiple irrigation device operation parameters, temperature and humidity adjustment device operation parameters, and lighting device operation parameters.

[0031] Further, the parameter optimization module includes:

[0032] A first calculation unit, configured to calculate the damage degree of the irrigation equipment, the damage degree of the temperature and humidity adjustment equipment, and the damage degree of the lighting equipment according to the operation parameters of each irrigation equipment, the operation parameters of each temperature and humidity adjustment equipment, and the operation parameters of each lighting equipment;

[0033] A second calculation unit, connected to the first calculation unit, configured to input the damage degree of the irrigation equipment, the damage degree of the temperature and humidity equipment, and the damage degree of the lighting equipment into a preset first improvement formula and a second improvement formula respectively, and improve the stealing behavior probability and the environmental disturbance intensity in the dung beetle optimization algorithm to obtain an optimized stealing behavior probability and an optimized environmental disturbance intensity;

[0034] An algorithm improvement unit, connected to the second calculation unit, configured to improve the dung beetle optimization algorithm according to the optimized stealing behavior probability and the optimized environmental disturbance intensity to obtain the improved dung beetle optimization algorithm.

[0035] Further, the first improvement formula is configured as:

[0036]

[0037] where, P opt is used to represent the optimized stealing behavior probability, P c is used to represent the initial stealing behavior probability, κ i is used to represent the irrigation equipment damage attenuation coefficient, D i is used to represent the damage degree of the irrigation equipment, D t is used to represent the damage degree of the temperature and humidity equipment, D t0 is used to represent a preset damage threshold of the temperature and humidity equipment, λ t is used to represent a sensitivity coefficient of the influence of the preset damage of the temperature and humidity equipment, D l is used to represent the damage degree of the lighting equipment, v l is used to represent the influence intensity of the lighting equipment damage, γ is used to represent the adjustment coefficient of the stealing behavior probability, α is used to represent a preset third time attenuation coefficient, and t is used to represent time.

[0038] Further, the second improvement formula is configured as:

[0039]

[0040] where, I opt is used to represent the optimized environmental disturbance intensity, Γ(α1, D i ·θ i) is used to represent the Gamma function, α1 is used to represent the shape parameter of the Gamma function, θ i is used to represent the preset irrigation equipment scale parameter, ζ(β, D t ) is used to represent the Riemann Zeta function, β is used to represent the shape parameter of the Riemann Zeta function, Erf(γ1D l ) is used to represent the error function, γ1 is used to represent the sensitivity parameter of the error function, P e is used to represent the environmental perturbation intensity in the dung beetle optimization algorithm, δ is used to represent the adjustment factor of the logarithmic function, and λ1 is used to represent the preset time decay factor.

[0041] A method for intelligent monitoring and environmental optimization management of a plant greenhouse, which is applied to the above-mentioned intelligent monitoring and environmental optimization management system of the plant greenhouse, and includes:

[0042] Step S1, the detection module detects the internal greenhouse environment data of the vegetable greenhouse in the polar research station, the remaining resource data of the polar research station, and the equipment feedback data in the vegetable greenhouse in real time; the acquisition module acquires the vegetable types, vegetable growth days, and vegetable growth parameters of various vegetables in the vegetable greenhouse;

[0043] Step S2, the health prediction module processes the current growth index according to the vegetable growth parameters, and collates the greenhouse environment data, the vegetable types, the vegetable growth days, and the current growth index to form vegetable cultivation parameters and inputs them into a pre-trained vegetable growth prediction model to predict the predicted growth index of the vegetables in the next growth period;

[0044] Step S3, the parameter optimization module introduces the dung beetle optimization algorithm, improves the dung beetle optimization algorithm according to the equipment feedback data, takes each group of the vegetable cultivation parameters as a dung beetle individual, uses the improved dung beetle optimization algorithm to iteratively update the position of the dung beetle individual, and calculates the objective function value corresponding to the dung beetle individual. When the minimum value of the objective function value is obtained, the position of the corresponding dung beetle individual is output as the global optimal solution;

[0045] Step S4, the condition generation module processes the resource shortage state according to the remaining resource data, and processes the adjustment restriction conditions based on the resource shortage state; the processing module processes the theoretical adjustment range according to the vegetable cultivation parameters in the global optimal solution, and restricts the theoretical adjustment range according to the adjustment restriction conditions to obtain the resource optimization adjustment range;

[0046] Step S5, the in-greenhouse adjustment module dynamically adjusts the air environment and the culture solution environment in the vegetable greenhouse according to the resource optimization adjustment range.

[0047] Advantages of the present invention:

[0048] In the present invention, by real-time monitoring of environmental data, resource data, and equipment feedback data in the vegetable greenhouse, combined with the health prediction module and the improved dung beetle optimization algorithm, the growth situation of vegetables can be accurately predicted, and the environmental conditions and resource allocation can be dynamically adjusted according to the prediction results to ensure that the vegetables obtain the best growth conditions in the polar environment. Through refined adjustment and optimization, the growth efficiency and yield of vegetables can be significantly improved;

[0049] The present invention adjusts the resource usage strategy according to the real-time resource shortage state, effectively avoids resource waste, ensures the optimal allocation of limited water, fertilizers, and energy, and improves the utilization rate of resources;

[0050] The present invention also realizes the full-automatic management of the vegetable planting process. From data collection, health prediction to resource optimization and environmental regulation, it reduces manual intervention and improves management efficiency and accuracy. Brief Description of the Drawings

[0051] Figure 1 is a schematic structural diagram of the intelligent monitoring and environmental optimization management system for the plant greenhouse in the present invention;

[0052] Figure 2 is a step flow chart of the intelligent monitoring and environmental optimization management method for the plant greenhouse in the present invention.

[0053] Reference Signs: 1, detection module; 2, acquisition module; 3, health prediction module; 31, analysis unit; 32, screening unit; 33, storage unit; 34, training unit; 4, establishment module; 5, parameter optimization module; 51, first calculation unit; 52, second calculation unit; 53, algorithm improvement unit; 6, condition generation module; 7, processing module; 8, in-shed adjustment module. Detailed Embodiments

[0054] The present invention will be further described in detail below with reference to the drawings and embodiments. The same components are denoted by the same reference signs. It should be noted that the terms "front", "rear", "left", "right", "upper", and "lower" used in the following description refer to the directions in the drawings, and the terms "bottom surface" and "top surface", "inner" and "outer" refer to the directions towards or away from the geometric center of a specific component, respectively.

[0055] Example 1, referring to Figure 1 , which is the first embodiment of the present invention. This embodiment provides an intelligent monitoring and environmental optimization management system for a plant greenhouse, which can dynamically adjust the resource ratio in the vegetable greenhouse and improve the vegetable growth efficiency, including:

[0056] The detection module 1 is used to detect in real time the internal greenhouse environment data of the vegetable greenhouse in the polar research station, the remaining resource data of the polar research station, and the equipment feedback data in the vegetable greenhouse;

[0057] The acquisition module 2 is used to acquire the vegetable species, the vegetable growth days, and the vegetable growth parameters of various vegetables in the vegetable greenhouse;

[0058] The health prediction module 3 is connected to the detection module 1 and the acquisition module 2. It is used to process the current growth index according to the vegetable growth parameters, organize the greenhouse environment data, vegetable species, vegetable growth days, and the current growth index into vegetable cultivation parameters, and input them into the pre-trained vegetable growth prediction model to predict the predicted growth index of the vegetables in the next growth period;

[0059] The establishment module 4 is connected to the health prediction module 3. It is used to establish an objective function according to the predicted growth index and the preset standard growth index, and the objective function is used to output the objective function value;

[0060] The parameter optimization module 5 is connected to the detection module 1 and the establishment module 4. It is used to introduce the dung beetle optimization algorithm, improve the dung beetle optimization algorithm according to the equipment feedback data, regard each group of vegetable cultivation parameters as a dung beetle individual, use the improved dung beetle optimization algorithm to iteratively update the position of the dung beetle individual, calculate the objective function value corresponding to the dung beetle individual, and output the position of the corresponding dung beetle individual as the global optimal solution when the objective function value reaches the minimum;

[0061] The condition generation module 6 is connected to the detection module 1. It is used to process the resource shortage state according to the remaining resource data and obtain the adjustment limit conditions based on the resource shortage state;

[0062] The processing module 7 is connected to the parameter optimization module 5 and the condition generation module 6. It is used to process the theoretical adjustment range according to the vegetable cultivation parameters in the global optimal solution, and limit the theoretical adjustment range according to the adjustment limit conditions to obtain the resource optimization adjustment range;

[0063] The in-greenhouse adjustment module 8 is connected to the processing module 7. It is used to dynamically adjust the air environment and the culture solution environment in the vegetable greenhouse according to the resource optimization adjustment range.

[0064] Preferably, the greenhouse environment data includes air environment parameters and culture solution environment parameters. The air environment parameters include air temperature, air humidity, carbon dioxide concentration, and light intensity; the culture solution environment parameters include the content of nutrient elements in the culture solution, the pH value of the culture solution, the conductivity of the culture solution, and the oxygen content of the culture solution.

[0065] The working principle of Embodiment 1:

[0066] Detection Module 1: Multiple sensors are installed in the greenhouse to monitor the environmental data, resource data, and equipment status in the greenhouse in real time. Specifically, the air environment parameters include air temperature, humidity, carbon dioxide concentration, and light intensity; the culture solution environment parameters include nutrient element content, pH value, conductivity, and oxygen content. The equipment feedback data includes the operation status data of the temperature and humidity regulator, lighting equipment, and water and fertilizer system. These data are sent to the central control system in real time through wireless transmission.

[0067] Acquisition Module 2: Regularly collect the vegetable variety, planting days, and growth parameters of each vegetable. Vegetable growth parameters refer to the key physical and physiological characteristic parameters during the growth process of vegetables, such as the height of vegetables, growth rate, leaf area, root development, etc. These parameters can reflect the growth status of vegetables and predict their growth trends. In the vegetable planting system of polar research stations, Acquisition Module 2 will collect the following vegetable growth parameters from the data of sensors or manual input:

[0068] Vegetable height: The height of vegetables is monitored in real time through laser ranging sensors or optical sensors installed in the greenhouse.

[0069] Growth rate: The growth rate of vegetables is calculated by regularly measuring the change rate of the height of vegetables and other physiological characteristics (such as leaf growth).

[0070] Leaf area: Image recognition technology or leaf area sensors are used to monitor the growth area of leaves to estimate their photosynthetic capacity.

[0071] Health Prediction Module 3: This module calculates the current growth index based on the collected vegetable growth parameters. The calculation process of the current growth index includes:

[0072] First, configure weights for each vegetable growth parameter: The influence of each growth parameter on the growth index is different. According to agricultural research or historical data, set the weights of each growth parameter: the weight of vegetable height is 0.4, the weight of growth rate is 0.3, and the weight of leaf area is 0.3.

[0073] Then, input the above weights into the preset growth index formula to calculate the growth index. The growth index formula is configured as: GI = 0.4H’ + 0.3V + 0.3A

[0074] Among them, GI is used to represent the current vegetable growth index, H’ is used to represent the vegetable height, V is used to represent the growth rate, and A is used to represent the leaf area.

[0075] The health prediction module 3 also combines the real-time monitored environmental data and information such as vegetable types and growth days to organize them into vegetable cultivation parameters. Subsequently, these parameters are input into a pre-trained vegetable growth prediction model to predict the predicted growth index of the vegetables in the next growth cycle.

[0076] Establishment module 4: Based on the output of the health prediction module 3 (i.e., the predicted growth index), compare it with the preset standard growth index to establish an objective function. The objective function calculates the difference between the two and outputs the objective function value for subsequent optimization processing.

[0077] Parameter optimization module 5: In the parameter optimization module 5, the dung beetle optimization algorithm is introduced, and the optimization process of the dung beetle optimization algorithm is adjusted according to the device feedback data. Each set of vegetable cultivation parameters is regarded as a dung beetle individual, and the optimization algorithm iteratively updates the positions of the dung beetle individuals and calculates the objective function values corresponding to each dung beetle individual. When the objective function value reaches the minimum, the system outputs the position of this dung beetle individual as the global optimal solution to provide the optimal vegetable cultivation parameters.

[0078] Condition generation module 6: The main function of the condition generation module 6 is to judge whether there is a state of resource shortage according to the real-time monitored resource data (such as the remaining resource data of water source, fertilizer, electricity, etc.), and generate corresponding adjustment limit conditions. It will adjust the resource allocation of the vegetable growth environment according to the actual resource situation to ensure the efficient use of resources.

[0079] The functions of the condition generation module 6 include:

[0080] Resource data monitoring: This module receives the data from the detection module 1 in real time, including information such as water, fertilizer, carbon dioxide, air temperature and humidity, etc., to monitor the resource consumption situation of the polar research station.

[0081] Resource shortage judgment: According to the set warning threshold (for example, the remaining amount of water source is lower than a certain percentage), automatically judge whether to enter the resource shortage state. If a certain resource is lower than the set threshold, the system will start the resource adjustment strategy.

[0082] Adjustment limit condition generation: When the resource shortage state is confirmed, the condition generation module 6 will generate specific adjustment limit conditions. For example, the system may limit the irrigation water volume, reduce the carbon dioxide concentration in the greenhouse, or limit the nutrient element concentration in the culture solution.

[0083] Specific embodiments of the condition generation module 6:

[0084] Water resource monitoring and adjustment: Water resource sensors are equipped in the system to monitor the remaining amount of the water source in real time.

[0085] Set that when the remaining amount of water source is lower than 20%, the system automatically enters the "water resource shortage" state. After receiving this information, the condition generation module 6 generates adjustment limit conditions: such as reducing the amount of water per irrigation, or adjusting the irrigation frequency to extend the time interval between each irrigation.

[0086] Fertilizer resource shortage: The fertilizer concentration is monitored in real time through sensors. When the fertilizer concentration is lower than the preset threshold, the system will enter the "fertilizer resource shortage" state. Generate adjustment limit conditions, such as reducing the amount of fertilizer applied, adjusting the fertilizer ratio, or using alternative resources (such as compost) to maintain the nutritional requirements of plants.

[0087] Processing module 7: According to the vegetable cultivation parameters in the global optimal solution, the processing module 7 calculates the theoretical adjustment range, and combines the resource adjustment limits provided by the condition generation module 6 to determine the resource optimization adjustment range. This range is used to further guide the precise adjustment of the environment to ensure the maximum improvement of vegetable growth efficiency under limited resources.

[0088] The functions of the processing module 7 include:

[0089] Calculation of the theoretical adjustment range: According to the global optimal solution, the processing module 7 calculates the theoretical adjustment range, which represents the best interval of various environmental parameters required for vegetable growth under ideal conditions.

[0090] Generation of the resource optimization adjustment range: Combining the resource adjustment limits of the condition generation module 6, the processing module 7 restricts the theoretical adjustment range according to the state of resource shortage to obtain the final resource optimization adjustment range. This range ensures the best adjustment of the environmental parameters required for vegetable growth under limited resources.

[0091] Specific embodiments of the processing module 7:

[0092] Temperature and humidity adjustment: Through an optimization algorithm, the system calculates that the ideal temperature is 22 °C and the humidity is 70%.

[0093] If the condition generation module 6 finds that the current water resources are insufficient and the system needs to reduce water consumption, it may lead to a decrease in humidity. After receiving this restriction, the processing module 7 will adjust the ideal humidity range from 70% to the range of 65% - 70%. In this way, the humidity will not be too low and the effective utilization of water resources can be ensured.

[0094] Light adjustment: Assume that the recommended light intensity in the global optimal solution is 300 μmol / m 2 / s to support the optimal photosynthesis of vegetables. If the condition generation module 6 detects a shortage of fertilizer resources, the system decides to reduce the light intensity to lower the nutrient demand. Then, the processing module 7 outputs an adjusted light intensity range, such as 250 μmol / m2 / s to 300 μmol / m2 / s, according to this condition. This adjusted range can ensure photosynthesis while reducing fertilizer consumption.

[0095] Adjustment of culture solution components: The recommended pH value of the culture solution by the global optimal solution is 6.5, and the conductivity is 2.0 mS / cm. When the condition generation module 6 finds that the nitrogen source in the culture solution is exhausted, resulting in large fluctuations in the pH value, the system will adjust the pH value range according to the shortage of resources. The processing module 7 outputs a new pH value range (6.0 to 6.5) according to the adjustment limit conditions to adapt to the current shortage of fertilizer resources and ensure the healthy growth of vegetables.

[0096] In-shed adjustment module 8: This module automatically adjusts the environmental conditions in the vegetable greenhouse according to the resource optimization adjustment range output by the processing module 7, including air temperature, light intensity, air humidity, carbon dioxide concentration, and pH value of the culture solution. By responding to changes in the external environment in real time and through fine control, it ensures that vegetables are in the best growth conditions at each growth stage.

[0097] Preferably, the health prediction module 3 includes:

[0098] An analysis unit 31 for performing a correlation analysis on each air environment parameter, culture solution environment parameter, and the current growth index to obtain a correlation analysis result;

[0099] A screening unit 32 connected to the analysis unit 31 for screening each air environment parameter and culture solution environment parameter according to the correlation analysis result to obtain key environmental parameters, which include air temperature, light intensity, air humidity, carbon dioxide concentration, and pH value of the culture solution.

[0100] Specifically, in this embodiment, the core task of the analysis unit 31 is to perform a statistical analysis on the relationship between each air environment parameter, culture solution environment parameter, and the current vegetable growth index. Through correlation analysis, it identifies which parameters have the greatest impact on vegetable growth.

[0101] The analysis unit 31 obtains each environmental parameter and the current growth index from sensors or data acquisition systems in real time, and then performs a correlation analysis: using statistical methods such as Pearson Correlation Coefficient and Mutual Information to calculate the correlation between each air and culture solution environment parameter and the current vegetable growth index.

[0102] The calculation process is as follows: Calculate the correlation coefficient between the air temperature and the growth index; calculate the correlation between the light intensity and the growth index; compare the correlations between other parameters such as carbon dioxide concentration, humidity, and pH value and the growth index.

[0103] The correlation analysis results will return a correlation matrix of a parameter and the growth index:

[0104] The correlation between the air temperature and the growth index is 0.85 (highly correlated);

[0105] The correlation between the light intensity and the growth index is 0.75 (moderately correlated);

[0106] The correlation between the carbon dioxide concentration and the growth index is 0.65 (moderately correlated);

[0107] The correlation between the pH value of the culture solution and the growth index is 0.9 (highly correlated).

[0108] The screening unit 32 receives the correlation analysis results from the analysis unit 31, such as the above-mentioned correlation matrix. Then it performs parameter screening: According to the set threshold (for example, a parameter with a correlation coefficient greater than 0.7 is considered to have a greater impact on the growth index), it screens out the environmental parameters that have a significant correlation with the growth index.

[0109] In this embodiment, parameters such as air temperature (0.85), light intensity (0.75), carbon dioxide concentration (0.65), and pH value of the culture solution (0.9) are screened as key environmental parameters.

[0110] Preferably, the health prediction module 3 further includes:

[0111] A storage unit 33 for saving historical key environmental parameters, historical vegetable types, historical growth days, and historical growth indices at multiple historical moments;

[0112] A training unit 34, connected to the storage unit 33, for introducing an initial model, and using the historical key environmental parameters, historical vegetable types, historical growth days, and historical growth indices at multiple historical moments as inputs, and the historical growth index after the next growth period as the output, to retrain the initial model to obtain a vegetable growth prediction model.

[0113] Specifically, in this embodiment, the main task of the storage unit 33 is to save the key environmental parameters, growth parameters, and growth indices at multiple historical moments. These historical data are used to train and optimize the vegetable growth prediction model to ensure that the model can make effective predictions based on historical data.

[0114] Among them, the historical key environmental parameters include environmental data such as air temperature, humidity, light intensity, carbon dioxide concentration, and pH value of the culture solution collected in real time.

[0115] Historical vegetable types: Record the growth information of different types of vegetables.

[0116] Historical growth days: Record the growth days of vegetables, including the growth time of each planting cycle.

[0117] Historical growth index: The growth index calculated by measuring the height, leaf area, root development, etc. of vegetables.

[0118] All historical data (such as environmental parameters, growth index, etc.) will be stored in the form of a time series. The data at each time point includes environmental parameters, vegetable types, growth days, and growth index. The storage method can be a database or a local data file to ensure data scalability and access efficiency.

[0119] Specific embodiment of storage unit 33: Assume that multiple sensors are installed in a vegetable greenhouse to monitor environmental data in real time, such as temperature (22°C), humidity (70%), carbon dioxide concentration (800 ppm), light intensity (350 μmol / m 2 / s), etc. The data collected each time and the corresponding vegetable types (such as "tomato", "lettuce") and growth days (such as 15 days) are recorded and stored. The corresponding growth index (such as 0.75) will also be stored as part of the historical data. Every once in a while (for example, 1 hour), new data will be recorded and added to the database to form a time series of historical data.

[0120] The task of training unit 34 is to use the historical data in storage unit 33 to train a vegetable growth prediction model. By introducing an initial model and training it with historical data, training unit 34 can continuously optimize the model and improve the prediction accuracy after each new data input.

[0121] The initial model can be a machine learning model (such as linear regression, support vector machine, neural network, etc.), which has been preliminarily trained and has basic prediction capabilities.

[0122] The input of this initial model is historical key environmental parameters, vegetable types, growth days, and historical growth index, and the output is the predicted growth index for the next growth cycle.

[0123] Training unit 34 takes the data at multiple historical moments as input, including the historical key environmental parameters, vegetable types, historical growth days, and historical growth index at each time point.

[0124] Training unit 34 calculates the growth index for the next growth cycle based on the historical data and uses these data as the output of the model to continuously adjust and optimize the initial model.

[0125] Model Update: Use optimization algorithms such as gradient descent or random forest to continuously adjust the parameters of the model by minimizing the error until the prediction ability of the model reaches the optimal state.

[0126] Specific Example of Training Unit 34: Initial Model: Use a neural network based on a multi-layer perceptron (MLP) as the initial model. The input of this neural network includes historical key environmental parameters, historical vegetable types, historical growth days, and historical growth indices, and the output is the vegetable growth index.

[0127] Training Process of Training Unit 34:

[0128] Assume the historical data is as follows: Time point T1: Air temperature 20°C, humidity 65%, light intensity 300 μmol / m2 / s, carbon dioxide concentration 800 ppm, vegetable type "tomato", growth days 15 days, historical growth index 0.75. Time point T2: Air temperature 22°C, humidity 70%, light intensity 350 μmol / m 2 / s, carbon dioxide concentration 850 ppm, vegetable type "tomato", growth days 16 days, historical growth index 0.78.

[0129] Training Unit 34 inputs the data of T1 and T2 into the initial model, and the goal is to predict the growth index at T3 through the model.

[0130] Training Unit 34 uses the data of T1 and T2 for backpropagation, calculates the error predicted by the current model, and then updates the weights of the model through the gradient descent algorithm.

[0131] Every time new data (such as the data at time point T3) arrives, Training Unit 34 inputs the new historical data into the model, updates the model, and gradually improves the prediction accuracy. After multiple rounds of training, the model can predict the growth index at the next time point based on historical environmental data and vegetable growth parameters. For example, the model may predict that at the next time point (T4), the growth index of tomatoes is 0.80. This predicted value will be compared with the actual growth index and used to further adjust the model.

[0132] Real-time Update of Training Unit 34: Every time new historical data is collected, Training Unit 34 will readjust the model to cope with the impact of environmental changes on vegetable growth. Over time, Training Unit 34 gradually improves the model and enhances its prediction accuracy for future growth cycles. Assume that the system collects environmental data in the vegetable greenhouse every day and updates the model once. With the accumulation of more historical data, the prediction ability of the model will gradually increase, and it can provide more accurate growth index predictions in each growth cycle.

[0133] Preferably, the formula of the vegetable growth prediction model is configured as:

[0134]

[0135] f4(C) = k C ·C

[0136]

[0137] wherein, T is used to represent the air temperature, L is used to represent the light intensity, H is used to represent the air humidity, C is used to represent the carbon dioxide concentration, pH is used to represent the pH value of the culture solution, S is used to represent the vegetable variety, D is used to represent the vegetable growth days, GI is used to represent the vegetable growth index, GI next is used to represent the predicted growth index, f1(T) is used to represent the influence function of temperature on vegetable growth, T0 is used to represent the preset suitable temperature value, σ T is used to represent the preset standard deviation of temperature fluctuation, f2(L) is used to represent the influence of light intensity on vegetable growth, A L is used to represent the preset light intensity coefficient, L max is used to represent the preset maximum light intensity, f3(H) is used to represent the influence function of air humidity on vegetable growth, H0 is used to represent the preset humidity threshold, σ H is used to represent the preset humidity change sensitivity value, f4(C) is used to represent the influence function of carbon dioxide concentration on vegetable growth, k C is used to represent the preset carbon dioxide concentration coefficient, f5(pH) is used to represent the influence function of the pH value of the culture solution on vegetable growth, pH0 is used to represent the preset most suitable pH value, ΔpH is used to represent the preset pH value adaptation range, f6(S, D) is used to represent the influence function of vegetable variety and growth days on the growth index, k S is used to represent the preset vegetable variety coefficient, γ D is used to represent the preset growth rate coefficient, t is used to represent time, and λ and μ are respectively used to represent the preset first time decay coefficient and second time decay coefficient.

[0138] Specifically, in this embodiment, the value range of GI is between 0 and 1, representing the growth state of the vegetable, 0 indicating completely un-grown, and 1 indicating completely grown.

[0139] The air temperature is monitored by a temperature sensor in degrees Celsius (°C). The sensor should be arranged at multiple positions in the vegetable growth area to measure and record the temperature changes in real time; the light intensity is measured using a light sensor (such as a photoelectric sensor) in lux, and data is obtained according to the light changes in different areas; the air humidity is monitored by an air humidity sensor arranged in the vegetable greenhouse in percentage (%), and this data is used to evaluate the impact of humidity changes in the greenhouse; the carbon dioxide concentration (C) is measured by installing a carbon dioxide concentration sensor in ppm (parts per million); the cultivation pH value is measured using a pH sensor, and the ideal pH value range varies depending on the vegetable variety; the vegetable variety is obtained through planting records; the number of growth days is obtained by real-time tracking and recording based on the planting date; the current growth index ranges from 0 to 1.

[0140] After collecting the above data, use this formula to predict the vegetable growth index for the next growth cycle. By comparing the actual data with the predicted growth index, continuously adjust the parameters of each function (such as temperature standard deviation, light intensity coefficient, etc.) to improve the prediction accuracy. In addition, according to the prediction results of the growth index, automatically adjust the environmental parameters, provide warnings or suggestions to the greenhouse management personnel to ensure that the vegetable growth conditions are in the optimal state.

[0141] Preferably, the device feedback data includes the operating parameters of multiple irrigation devices, the operating parameters of temperature and humidity adjustment devices, and the operating parameters of lighting devices.

[0142] Preferably, the parameter optimization module 5 includes:

[0143] The first calculation unit 51 is used to calculate the damage degree of the irrigation device, the damage degree of the temperature and humidity device, and the damage degree of the lighting device according to the operating parameters of each irrigation device, the operating parameters of each temperature and humidity adjustment device, and the operating parameters of each lighting device;

[0144] The second calculation unit 52 is connected to the first calculation unit 51 and is used to input the damage degree of the irrigation device, the damage degree of the temperature and humidity device, and the damage degree of the lighting device into the preset first improvement formula and second improvement formula respectively to improve the theft behavior probability and environmental disturbance intensity in the dung beetle optimization algorithm to obtain the optimized theft behavior probability and optimized environmental disturbance intensity;

[0145] The algorithm improvement unit 53 is connected to the second calculation unit 52 and is used to improve the dung beetle optimization algorithm according to the optimized theft behavior probability and optimized environmental disturbance intensity to obtain the improved dung beetle optimization algorithm.

[0146] Specifically, in this embodiment, the operating parameters of the device are an important basis for judging the performance and life of the device and can reflect the damage situation of the device. Specifically, the device feedback data includes the following content:

[0147] Operating parameters of irrigation equipment: such as irrigation water flow rate, irrigation frequency, water pressure, etc.

[0148] Operating parameters of temperature and humidity regulation equipment: such as temperature set value, actual temperature, humidity set value and actual humidity of temperature regulation equipment.

[0149] Operating parameters of lighting equipment: such as light intensity, switching time, current consumption of lighting equipment, etc.

[0150] The feedback data of these devices are monitored in real time through sensors and transmitted to the system for subsequent calculation of equipment damage degree and optimization.

[0151] The main function of the first calculation unit 51 is to calculate the damage degree of each device according to the operating parameters of the device collected in real time. The damage degree of the device reflects the possible aging or damage of the device during use in the polar environment, and this information will affect the operating efficiency of the device and the impact on the vegetable growth environment.

[0152] Calculation of the damage degree of irrigation equipment: By monitoring the changes in water flow rate and water pressure of the irrigation system, the performance degradation of the irrigation system is calculated. For example, a decrease in water flow rate or unstable water pressure may indicate a malfunction or damage to the irrigation equipment.

[0153] Calculation of the damage degree of temperature and humidity regulation equipment: By monitoring the deviation between the actual temperature and the set temperature, as well as the humidity change, the damage situation of the temperature and humidity regulation equipment is calculated. If the temperature control system cannot maintain the set temperature, or the humidity control fails, it may indicate the aging or malfunction of the equipment.

[0154] Calculation of the damage degree of lighting equipment: By monitoring the gap between the light intensity and the set value, the performance change of the lighting equipment is calculated. If the light intensity of the lighting equipment is lower than the set standard, it indicates that the equipment may have a malfunction or reduced efficiency.

[0155] In a specific embodiment:

[0156] Calculation of the damage degree of irrigation equipment: The water flow rate of the irrigation equipment is set to 10 L / min, and the actual flow rate is 8 L / min. The system calculates that the damage degree of the irrigation equipment is 20%.

[0157] Calculation of the damage degree of temperature and humidity regulation equipment: The set temperature is 22 °C, the actual temperature is 20 °C, and the temperature deviation is 2 °C. The system calculates that the damage degree of the temperature control equipment is 15%.

[0158] Calculation of the damage degree of lighting equipment: The set light intensity is 300 μmol / m2 / s, and the actual light intensity is 270 μmol / m2 / s. The system calculates that the damage degree of the lighting equipment is 10%.

[0159] These degrees of damage will be transmitted to the second computing unit 52 for improving the relevant parameters of the dung beetle optimization algorithm.

[0160] The task of the second computing unit 52 is to improve two key parameters in the dung beetle optimization algorithm based on the degree of damage to the device: the probability of stealing behavior and the intensity of environmental disturbance. Through these optimizations, the dung beetle optimization algorithm can better adapt to the fluctuations in device performance and find the global optimal solution.

[0161] Optimization of the probability of stealing behavior: The probability of stealing behavior in the dung beetle optimization algorithm determines the way in which dung beetle individuals obtain information during the optimization process. The aging and performance degradation of the device will affect the information transmission efficiency, so it is necessary to adjust the probability of stealing behavior according to the degree of damage to the device.

[0162] Optimization of the intensity of environmental disturbance: The intensity of environmental disturbance in the dung beetle optimization algorithm affects the exploration range of dung beetle individuals in the search space. The degree of damage to the device will cause fluctuations in the greenhouse environment, so it is necessary to adjust the disturbance intensity to improve the optimization efficiency.

[0163] The algorithm improvement unit 53 applies the optimized probability of stealing behavior and disturbance intensity output by the second computing unit 52 to the dung beetle optimization algorithm to adjust the search strategy of dung beetle individuals. Specifically, the optimized dung beetle optimization algorithm can stably find the global optimal solution by dynamically adjusting the information acquisition strategy and disturbance intensity during the search process when facing device performance fluctuations and environmental disturbances.

[0164] Adjust the stealing behavior: According to the optimized probability of stealing behavior, adjust the way in which dung beetle individuals obtain information. When the device is severely damaged, reduce the probability of stealing behavior and enhance the individual's autonomous search ability.

[0165] Adjust the disturbance intensity: According to the optimized intensity of environmental disturbance, adjust the exploration amplitude of dung beetle individuals in the search space. When the device is severely damaged, reduce the disturbance intensity to reduce the impact of environmental disturbance on the optimization process.

[0166] In the case of device damage (for example, the irrigation device is damaged by 20%, the temperature and humidity regulation device is damaged by 15%, and the lighting device is damaged by 10%), the algorithm improvement unit 53 transmits this information to the dung beetle optimization algorithm to adjust its probability of stealing behavior and disturbance intensity. After improvement, the dung beetle optimization algorithm will search more stably. Even if the device has a performance decline, it can effectively avoid the early convergence of the algorithm and improve the efficiency of finding the global optimal solution.

[0167] Preferably, the first improvement formula is configured as:

[0168]

[0169] Where P optUsed to represent the probability of optimizing theft behavior, P c Used to represent the initial probability of theft behavior, k i Used to represent the attenuation coefficient of the damaged irrigation equipment, D i Used to represent the degree of damage of the irrigation equipment, D t Used to represent the degree of damage of the temperature and humidity equipment, D t0 Used to represent the preset damage threshold of the temperature and humidity equipment, λ t Used to represent the sensitivity coefficient of the influence of the preset damage of the temperature and humidity equipment, D l Used to represent the degree of damage of the lighting equipment, vl is used to represent the influence intensity of the damage of the lighting equipment, γ is used to represent the adjustment coefficient of the probability of theft behavior, α is used to represent the preset third time attenuation coefficient, and t is used to represent time.

[0170] Specifically, in this embodiment, the degree of damage D of the irrigation equipment i is 0.3, the degree of damage D of the temperature and humidity equipment t is 0.5, the degree of damage D of the lighting equipment l is 0.2, the current probability of theft behavior P c is 0.4, and the following constants are set simultaneously: D t0 = 0.4, κ i = 2.0, λ t = 1.5, ν l = 1.0, γ = 0.5, α = 0.05;

[0171] Then

[0172]

[0173] According to the first improvement formula, all influence functions can be combined to calculate the final optimized probability of theft behavior. Assuming that the influence of the third time attenuation coefficient in a short period of time is 1, it can be simplified as:

[0174]

[0175] Since the influence of the time decay factor is small at this time, the above formula is further simplified as:

[0176] P opt ≈f1(D i )·f2(D t )·f3(D l )·f4(P c ) = 0.5488·0.1393·0.1813·0.3275

[0177] Finally, P opt≈0.0137. This result indicates that in the improved dung beetle optimization algorithm, the likelihood of theft behavior occurring is greatly reduced (close to zero). This result can effectively improve the stability and efficiency of the algorithm, reduce the search for invalid solutions, and optimize the control strategy of the device.

[0178] Preferably, the second improvement formula is configured as:

[0179]

[0180] Where I opt is used to represent the optimized environmental perturbation intensity, Γ(α1, D i ·θ i ) is used to represent the Gamma function, α1 is used to represent the shape parameter of the Gamma function, and θ i is used to represent the preset irrigation equipment scale parameter, ζ(β, D t ) is used to represent the Riemann Zeta function, β is used to represent the shape parameter of the Riemann Zeta function, Erf(γ1D l ) is used to represent the error function, γ1 is used to represent the sensitivity parameter of the error function, P e is used to represent the environmental perturbation intensity in the dung beetle optimization algorithm, δ is used to represent the adjustment factor of the logarithmic function, and λ1 is used to represent the preset time decay factor.

[0181] Specifically, in this embodiment, the irrigation equipment scale parameter represents the scale of equipment damage. The shape parameter of the Riemann Zeta function is used to control the influence of temperature and humidity equipment damage on the perturbation intensity. The sensitivity parameter of the error function represents the influence of lighting equipment damage on the perturbation intensity. The adjustment factor of the logarithmic function represents the degree of influence of the perturbation intensity on optimization. Through the second improvement formula, the environmental perturbation intensity in the dung beetle optimization algorithm can be effectively optimized, thereby improving the global search ability and stability of the algorithm. By adjusting factors such as equipment damage degree and environmental perturbation intensity, this system can adjust the optimization process in real time to ensure the efficiency and accuracy of the algorithm.

[0182] A method for intelligent monitoring and environmental optimization management of a plant greenhouse, which is applied to the above-mentioned plant greenhouse intelligent monitoring and environmental optimization management system, as Figure 2 shown, includes:

[0183] Step S1, the detection module 1 detects the internal greenhouse environment data of the vegetable greenhouse in the polar research station, the remaining resource data of the polar research station, and the equipment feedback data in the vegetable greenhouse in real time; the acquisition module 2 acquires the vegetable types, vegetable growth days, and vegetable growth parameters of various vegetables in the vegetable greenhouse;

[0184] Step S2: The health prediction module 3 processes the vegetable growth parameters to obtain the current growth index, collates the greenhouse environment data, vegetable variety, vegetable growth days, and current growth index to form vegetable cultivation parameters, and inputs them into the pre-trained vegetable growth prediction model to predict the predicted growth index of the vegetables in the next growth period;

[0185] Step S3: The parameter optimization module 5 introduces the dung beetle optimization algorithm, improves the dung beetle optimization algorithm according to the device feedback data, takes each group of vegetable cultivation parameters as a dung beetle individual, uses the improved dung beetle optimization algorithm to iteratively update the position of the dung beetle individual, calculates the objective function value corresponding to the dung beetle individual, and outputs the position of the corresponding dung beetle individual as the global optimal solution when the objective function value reaches the minimum;

[0186] Step S4: The condition generation module 6 processes the remaining resource data to obtain the resource shortage state, and processes the resource shortage state to obtain the adjustment limit conditions; the processing module 7 processes the vegetable cultivation parameters in the global optimal solution to obtain the theoretical adjustment range, and restricts the theoretical adjustment range according to the adjustment limit conditions to obtain the resource optimization adjustment range;

[0187] Step S5: The in-greenhouse adjustment module 8 dynamically adjusts the air environment and culture solution environment in the vegetable greenhouse according to the resource optimization adjustment range.

[0188] The above is only the preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A plant greenhouse intelligent monitoring and environmental optimization management system, characterized in that: include: A detection module (1) is used to detect in real time the internal environment data of the vegetable greenhouse of the polar scientific research station, the remaining resource data of the polar scientific research station, and the equipment feedback data in the vegetable greenhouse; An acquisition module (2) is used to acquire the vegetable types, vegetable growth days and vegetable growth parameters of various vegetables in the vegetable greenhouse; A health prediction module (3) is connected to the detection module (1) and the acquisition module (2), and is used to obtain a current growth index based on the vegetable growth parameters, and to organize the greenhouse environment data, the vegetable types, the vegetable growth days and the current growth index to form vegetable cultivation parameters, which are input into a pre-trained vegetable growth prediction model to predict the growth index of the vegetables in the next growth period; An establishment module (4), connected to the health prediction module (3), is used to establish an objective function according to the predicted growth index and a preset standard growth index, wherein the objective function is used to output an objective function value; A parameter optimization module (5) is connected to the detection module (1) and the establishment module (4), and is used to introduce a dung beetle optimization algorithm, improve the dung beetle optimization algorithm according to the device feedback data, take each group of vegetable cultivation parameters as a dung beetle individual, iteratively update the position of the dung beetle individual using the improved dung beetle optimization algorithm, and calculate the objective function value corresponding to the dung beetle individual, and output the corresponding position of the dung beetle individual as the global optimal solution when the objective function value is minimized; A condition generation module (6), connected to the detection module (1), configured to obtain a resource shortage state according to the remaining resource data, and obtain an adjustment restriction condition based on the resource shortage state; A processing module (7) is connected to the parameter optimization module (5) and the condition generation module (6), and is used to obtain a theoretical adjustment range according to the vegetable cultivation parameters in the global optimal solution, and to limit the theoretical adjustment range according to the adjustment restriction condition to obtain a resource optimization adjustment range; The greenhouse adjustment module (8) is connected to the processing module (7) and is used to dynamically adjust the air environment and culture solution environment in the vegetable greenhouse according to the resource optimization adjustment range.

2. The plant greenhouse intelligent monitoring and environmental optimization management system according to claim 1 is characterized by: The greenhouse environmental data includes air environment parameters and culture solution environment parameters. The air environment parameters include air temperature, air humidity, carbon dioxide concentration, and light intensity; the culture solution environment parameters include culture solution nutrient element content, culture solution pH value, culture solution conductivity, and culture solution oxygen content.

3. The plant greenhouse intelligent monitoring and environmental optimization management system according to claim 2 is characterized by: The health prediction module (3) comprises: An analysis unit (31) is used to perform a correlation analysis on each of the air environment parameters, the culture fluid environment parameters and the current growth index to obtain a correlation analysis result; A screening unit (32) is connected to the analysis unit (31) and is used to screen the air environment parameters and the culture solution environment parameters according to the correlation analysis results to obtain key environment parameters, wherein the key environment parameters include air temperature, light intensity, air humidity, carbon dioxide concentration and culture solution pH value.

4. The plant greenhouse intelligent monitoring and environmental optimization management system according to claim 3 is characterized by: The health prediction module (3) further includes: A storage unit (33) is used to store historical key environmental parameters, historical vegetable types, historical growing days, and historical growth indexes at multiple historical moments; A training unit (34) is connected to the storage unit (33) and is used to introduce an initial model, and use the historical key environmental parameters, the historical vegetable types, the historical growing days and the historical growth index at multiple historical moments as input, and use the historical growth index after the next growth time period as output, to retrain the initial model and obtain the vegetable growth prediction model.

5. The plant greenhouse intelligent monitoring and environmental optimization management system according to claim 4 is characterized by: The formula configuration of the vegetable growth prediction model is: f4(C)=k C ·C Wherein, T is used to indicate the air temperature, L is used to indicate the light intensity, H is used to indicate the air humidity, C is used to indicate the carbon dioxide concentration, pH is used to indicate the pH value of the culture solution, S is used to indicate the vegetable type, D is used to indicate the number of days for the vegetable to grow, GI is used to indicate the vegetable growth index, and GI is used to indicate the vegetable growth index. next is used to represent the predicted growth index, f1(T) is used to represent the effect function of temperature on vegetable growth, T0 is used to represent the preset suitable temperature value, σ T It is used to indicate the preset standard deviation of temperature fluctuation, f2(L) is used to indicate the effect of light intensity on vegetable growth, A L Used to indicate the preset light intensity coefficient, L max It is used to represent the preset maximum light intensity, f3(H) is used to represent the effect function of air humidity on vegetable growth, H0 is used to represent the preset humidity threshold, σ H It is used to represent the preset humidity change sensitivity value, f4(C) is used to represent the effect function of carbon dioxide concentration on vegetable growth, and k C It is used to represent the preset carbon dioxide concentration coefficient, f5(pH) is used to represent the effect of the pH value of the culture medium on the growth of vegetables, pH0 is used to represent the preset most suitable pH value, ΔpH is used to represent the preset pH adaptation range, f6(S, D) is used to represent the effect of vegetable type and growth days on the growth index, k S Used to represent the preset vegetable type coefficient, γ D is used to represent a preset growth rate coefficient, t is used to represent time, and λ and μ are used to represent a preset first time attenuation coefficient and a second time attenuation coefficient, respectively.

6. The plant greenhouse intelligent monitoring and environmental optimization management system according to claim 1, characterized in that: The equipment feedback data includes a plurality of irrigation equipment operating parameters, temperature and humidity adjustment equipment operating parameters, and lighting equipment operating parameters.

7. The plant greenhouse intelligent monitoring and environmental optimization management system according to claim 6, characterized in that: The parameter optimization module (5) comprises: A first calculation unit (51) is used to calculate the damage degree of the irrigation equipment, the damage degree of the temperature and humidity equipment, and the damage degree of the lighting equipment according to the operating parameters of each of the irrigation equipment, the operating parameters of each of the temperature and humidity adjustment equipment, and the operating parameters of each of the lighting equipment; A second calculation unit (52) is connected to the first calculation unit (51) and is used to input the damage degree of the irrigation equipment, the damage degree of the temperature and humidity equipment, and the damage degree of the lighting equipment into a preset first improved formula and a second improved formula respectively, and to improve the theft behavior probability and the environmental disturbance intensity in the dung beetle optimization algorithm to obtain an optimized theft behavior probability and an optimized environmental disturbance intensity; An algorithm improvement unit (53) is connected to the second calculation unit (52) and is used to improve the dung beetle optimization algorithm according to the optimized theft behavior probability and the optimized environmental disturbance intensity to obtain the improved dung beetle optimization algorithm.

8. The plant greenhouse intelligent monitoring and environmental optimization management system according to claim 7 is characterized by: The first improved formula is configured as: Among them, P opt It is used to represent the optimized stealing probability, P c Used to represent the initial probability of the theft behavior, k i Used to indicate the damage attenuation coefficient of irrigation equipment, D i Used to indicate the degree of damage to irrigation equipment, D t Used to indicate the degree of damage to temperature and humidity equipment, D t0 Used to indicate the preset temperature and humidity equipment damage threshold, λ t The sensitivity coefficient used to represent the impact of damage to the preset temperature and humidity equipment, D l Used to indicate the degree of damage to the lighting equipment, v l It is used to indicate the impact intensity of lighting equipment damage, γ is used to indicate the adjustment coefficient of the probability of theft, α is used to indicate the preset third time attenuation coefficient, and t is used to indicate time.

9. The plant greenhouse intelligent monitoring and environmental optimization management system according to claim 8, characterized in that: The second improved formula is configured as: Among them, I opt It is used to represent the disturbance intensity of the optimized environment, Γ(α1, D i ·θ i ) is used to represent the Gamma function, α1 is used to represent the shape parameter of the Gamma function, θ i It is used to represent the preset irrigation equipment scale parameter, ζ(β,D t ) is used to represent the Riemann Zeta function, β is used to represent the shape parameter of the Riemann Zeta function, Erf(γ1D l ) is used to represent the error function, γ1 is used to represent the sensitivity parameter of the error function, P e It is used to represent the environmental disturbance intensity in the dung beetle optimization algorithm, δ is used to represent the adjustment factor of the logarithmic function, and λ1 is used to represent the preset time attenuation factor.

10. A plant greenhouse intelligent monitoring and environmental optimization management method, applied to the plant greenhouse intelligent monitoring and environmental optimization management system according to any one of claims 1 to 9, characterized in that: include: Step S1, the detection module (1) detects in real time the internal environment data of the vegetable greenhouse of the polar scientific research station, the remaining resource data of the polar scientific research station and the equipment feedback data in the vegetable greenhouse; the acquisition module (2) acquires the vegetable types, vegetable growth days and vegetable growth parameters of various vegetables in the vegetable greenhouse; Step S2, the health prediction module (3) obtains the current growth index according to the vegetable growth parameters, and organizes the greenhouse environment data, the vegetable type, the vegetable growth days and the current growth index to form vegetable cultivation parameters, which are input into the pre-trained vegetable growth prediction model to predict the predicted growth index of the vegetable in the next growth period; Step S3, the parameter optimization module (5) introduces a dung beetle optimization algorithm, improves the dung beetle optimization algorithm according to the device feedback data, takes each group of the vegetable cultivation parameters as a dung beetle individual, uses the improved dung beetle optimization algorithm to iteratively update the position of the dung beetle individual, and calculates the objective function value corresponding to the dung beetle individual, and outputs the corresponding position of the dung beetle individual as the global optimal solution when the objective function value is minimized; Step S4, the condition generation module (6) obtains a resource shortage state according to the remaining resource data, and obtains an adjustment restriction condition based on the resource shortage state; the processing module (7) obtains a theoretical adjustment range according to the vegetable cultivation parameters in the global optimal solution, and limits the theoretical adjustment range according to the adjustment restriction condition to obtain a resource optimization adjustment range; Step S5, the greenhouse adjustment module (8) dynamically adjusts the air environment and culture solution environment in the vegetable greenhouse according to the resource optimization adjustment range.

Images