Cadmium telluride power generation glass and photovoltaic energy storage integrated agricultural greenhouse system

Abstract

The invention belongs to the technical field of energy conservation and environmental protection, and discloses a cadmium telluride power generation glass and photovoltaic energy storage integrated agricultural greenhouse system. The assembly design module is used for dividing cadmium telluride power generation glass into N subunits, and a pi-type thermocouple is integrated on a back plate of each subunit; the crop detection module is used for acquiring and analyzing crop states through subunits and generating crop labels according to the crop states; according to the dynamic shading module, each subunit is connected with a thermoelectric module, the thermoelectric modules are directly connected with an electrochromic glass shading plate, and the fog state light transmittance of the shading plate is dynamically adjusted according to crop labels; according to the intelligent energy storage module, each subunit integrates a cadmium telluride power generation layer, a thermoelectric conversion layer and a super capacitor, and an intelligent charging and discharging strategy is generated according to the illumination intensity. The energy utilization module is used for obtaining a composite adjustment strategy according to the crop label and the energy storage level corresponding to the intelligent charging and discharging strategy; through the integrated design, collaborative optimization of the power generation efficiency, the energy utilization rate and the crop stress damage rate is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy saving and environmental protection, more specifically, the present application relates to a cadmium telluride power generation glass and photovoltaic energy storage integrated agricultural greenhouse system. BACKGROUND

[0002] Traditional agricultural greenhouses face problems such as high energy consumption, extensive light and heat regulation, and photovoltaic module hot spot effect, and urgently need a green solution that takes into account "energy production" and "agricultural production".

[0003] Chinese patent application CN119944840A discloses a two-stage optimization scheduling method and device for an agricultural greenhouse photovoltaic power generation energy supply system: the agricultural greenhouse is equipped with a photovoltaic device and an energy storage device, the method comprises: based on the operation data of each device of the agricultural greenhouse, a photovoltaic and energy storage joint scheduling model of the agricultural greenhouse in the day-ahead optimization scheduling stage is established; the photovoltaic and energy storage joint scheduling model is solved to obtain the scheduling strategy of the agricultural greenhouse; based on the scheduling strategy and the operation data of the distribution network to which the agricultural greenhouse is connected, an optimization scheduling model of the distribution network in the intra-day optimization scheduling stage is established; the optimization scheduling model is solved to determine whether there is voltage out-of-limit at each node of the distribution network; if so, reactive device optimization is performed first, and then it is determined whether there is still voltage out-of-limit at each node of the distribution network; if there is still voltage out-of-limit, the active resources of the energy storage device and / or the photovoltaic device are adjusted until there is no voltage out-of-limit at each node of the distribution network. The invention establishes a scheduling model for the day-ahead stage and the intra-day stage, ensures full and reasonable use of photovoltaic power generation, fully mobilizes reactive device resources and energy storage devices and / or photovoltaic devices, ensures that the node voltage does not exceed the limit, and enables the distribution network to operate safely.

[0004] Although the above method can meet most scenarios, research and practical application of the above method and prior art have found that the above method and prior art at least have the following defects:

[0005] The above method only optimizes energy distribution and transmission from a macroscopic perspective, simply considers the limitations of power grid operation and energy comprehensive utilization, and cannot accurately adapt to the complex light and heat requirements of crops in each growth stage in agricultural production, cannot effectively respond to local hot spot effects, and cannot achieve efficient coordination of power generation and planting.

[0006] In view of this, the present application proposes a cadmium telluride power generation glass and photovoltaic energy storage integrated agricultural greenhouse system to solve the above problems. SUMMARY

[0007] In order to overcome the above-mentioned defects of the prior art and achieve the above-mentioned purpose, the present application provides the following technical solution: a cadmium telluride power generation glass and photovoltaic energy storage integrated agricultural greenhouse system, comprising:

[0008] Component design module: divide the cadmium telluride power generation glass into N sub-units and physically isolate them; integrate a π-type thermocouple on the back plate of each sub-unit;

[0009] Crop detection module: obtain the crop state through sub-unit collection and analysis, use the crop state as the input of the state analysis model, and obtain the crop label;

[0010] Dynamic shading module: each sub-unit is connected to a thermoelectric module, which is directly connected to an electrochromic glass shading panel; the haze transmittance of the shading panel is dynamically adjusted according to the crop label;

[0011] Intelligent energy storage module: each sub-unit integrates a cadmium telluride power generation layer, a thermoelectric conversion layer, and a supercapacitor, and generates an intelligent charging and discharging strategy according to the light intensity;

[0012] Energy utilization module: obtain a composite adjustment strategy according to the corresponding energy storage level of the crop label and the intelligent charging and discharging strategy.

[0013] Further, the crop state includes leaf temperature, chlorophyll fluorescence parameter, chlorophyll content, stem diameter, image feature, environmental temperature, environmental humidity, light intensity, and CO2 concentration;

[0014] The crop label includes growth stage label, physiological health label, environmental stress label, and energy demand label;

[0015] The method for obtaining the chlorophyll fluorescence parameter includes:

[0016] Apply a preset modulated light to the plant leaves, measure the initial fluorescence Fo, then apply a saturation pulse light, measure the maximum fluorescence Fm, calculate the difference between Fm and Fo to obtain the variable fluorescence Fv, and calculate the ratio of Fv to Fm to obtain the chlorophyll fluorescence parameter.

[0017] Further, the method for dynamically adjusting the haze transmittance of the shading panel according to the crop label includes:

[0018] Analyze the crop state to obtain growth stage influence factors, physiological state influence factors, and environmental stress influence factors;

[0019] Based on the Farquhar model, calculate the photosynthetic rate according to the maximum photosynthetic rate, light quantum efficiency, crop respiration rate, and light intensity; calculate the energy required for crop photosynthesis according to the photosynthetic rate, leaf area, and energy conversion coefficient of photosynthetic products;

[0020] An energy balance equation is constructed to consider transmittance, light intensity, energy required for crop photosynthesis, and charging requirements of the energy storage system. Coupling constraints are obtained based on the energy balance equation. The physical boundary of transmittance is obtained based on the minimum and maximum transmittance of cadmium telluride power generation glass. The physiological boundary is calculated based on light intensity and light saturation point. The coupling constraints, physical boundary, and physiological boundary are integrated into a compensation function.

[0021] Obtain the current glass transmittance, calculate the cumulative product of various influencing factors and the environmental compensation function; calculate the product of the current glass transmittance and the cumulative product to obtain the target transmittance at time t.

[0022] Furthermore, methods for obtaining growth stage influencing factors include:

[0023] By long-term monitoring of physiological indicators of crop growth cycle, the change curves of different stages are fitted, and the slope of the stage transition is obtained by differentiating the change curves. Chlorophyll fluorescence parameters and stem diameter changes within a preset time period are monitored. When the chlorophyll fluorescence parameter decreases by R% and the stem diameter change is lower than the growth threshold, the current daily average temperature and biological lower limit temperature are obtained, and the critical point is calculated. The growth stage influencing factors are calculated based on the slope and the critical point of the key growth stage.

[0024] Furthermore, methods for obtaining factors influencing physiological states include:

[0025] Chlorophyll content was normalized; temperature stress was constructed based on the Arrhenius equation, combined with activation energy, gas constant, optimum temperature and leaf temperature; the stem diameter change rate was calculated based on the stem diameter change within a preset time period, and the change rate stress was calculated based on the optimal change rate value; physiological state influencing factors were calculated based on normalized chlorophyll content, temperature stress and change rate stress.

[0026] Methods for obtaining environmental stress influencing factors include:

[0027] The dimensionality of the stress factors is reduced by PCA, and the contribution rate is calculated as the weight of the stress factors. The stress intensity is calculated based on the measured value of the stress factors, the optimal value of the stress, and the stress threshold. The environmental stress impact factor is obtained by calculating the stress factor weight and the stress intensity.

[0028] Furthermore, the method for obtaining the charging strategy in the intelligent charging and discharging strategy includes:

[0029] Calculate the light intensity weight based on the current light intensity, the maximum light intensity and the minimum light intensity within the preset time period;

[0030] The temperature difference weight is calculated based on the maximum and minimum temperatures of the thermoelectric conversion layer within a preset time period and the current temperature of the thermoelectric conversion layer.

[0031] The total available charging power is calculated by obtaining the real-time power of the cadmium telluride power generation layer, the real-time power of the thermoelectric conversion layer, and the current load power.

[0032] When the light intensity weight is not lower than the temperature difference weight, the cadmium telluride power generation layer itself is charged; otherwise, the total available charging power is allocated to the cadmium telluride power generation layer according to the weight ratio.

[0033] The total available charging power, excluding the charging power allocated to the cadmium telluride power generation layer, is entirely allocated to the thermoelectric conversion layer.

[0034] Furthermore, the method for obtaining the discharge strategy in the intelligent charge-discharge strategy includes:

[0035] Obtain the remaining charge SOC(t), minimum charge, minimum threshold, maximum threshold, and maximum charge of the supercapacitor at time t.

[0036] The SOC state interval is defined by a piecewise function, where SOC(t) is marked as L when it is below the minimum threshold, M when it is below the maximum threshold but not below the minimum threshold, and H when it is not below the maximum threshold.

[0037] Crop tags are used as input to the demand analysis model to obtain the crop energy demand index; G priority and F priority are calculated based on the crop energy demand index, the critical load G priority weight and the non-critical load F priority weight.

[0038] The available discharge power P is calculated based on the discharge efficiency, SOC(t), supercapacitor capacity, and time interval.

[0039] If the SOC state interval is L, then P only supplies power to G;

[0040] If the SOC state interval is M, then P supplies full power to G, and also supplies power to F according to the priority ratio of G priority and F priority;

[0041] If the SOC state interval is H, then P supplies full power to both G and F.

[0042] Furthermore, image features include morphological features, color features, and texture features;

[0043] Methods for obtaining morphological features include:

[0044] Detect the blade outline, calculate the perimeter and area, and calculate the roundness based on the perimeter and area;

[0045] The crop image is binarized to obtain a binary image. The binary image is then converted into a single-pixel skeleton, and the number of branches and the length of the main stem are calculated.

[0046] Calculate plant height and leaf length based on preset calibration objects;

[0047] The roundness, number of branches, main stem length, plant height and leaf length are combined to form morphological characteristics;

[0048] Methods for obtaining color features include:

[0049] The crop image is converted from RGB space to HSV space, and the components of each dimension are extracted.

[0050] Calculate the difference and sum of the NIR and R channels respectively, and calculate the ratio of the difference to the sum to obtain the NDVI;

[0051] Calculate the difference and sum of G531 and G570 respectively, and calculate the ratio of the difference to the sum to obtain PRI;

[0052] The color features are obtained by splicing together the components of each dimension, NDVI, and PRI.

[0053] Furthermore, methods for obtaining texture features include:

[0054] The gray-level co-occurrence matrix is ​​obtained by counting the frequency of gray values ​​(i,j) at any two points in the image.

[0055] Contrast, correlation, energy, and entropy are calculated based on the gray-level co-occurrence matrix, and the matrix features are obtained by concatenating them.

[0056] For each pixel in the image, take an A×A neighborhood, compare the neighboring pixels with the center pixel, record 1 if the neighboring pixel is greater than or equal to the center value, otherwise record 0; combine the E neighboring binary values ​​in clockwise order to form an E-bit binary number, convert it to decimal and use it as the LBP value of the center pixel; calculate the histogram of the LBP values ​​of the entire image as the global feature.

[0057] Texture features are obtained by concatenating matrix features and global features.

[0058] Furthermore, methods for obtaining composite adjustment strategies include:

[0059] Obtain the power generation and thermoelectric conversion power of cadmium telluride, and calculate the power generation.

[0060] Obtain the critical load power and non-critical load power, and calculate the load power;

[0061] Obtain the charging power threshold and calculate the maximum chargeable energy storage capacity based on the charging efficiency; obtain the discharging power threshold and calculate the maximum dischargeable energy storage capacity based on the discharging efficiency.

[0062] When the power generation is greater than the sum of the load power and the maximum chargeable capacity of the energy storage, it is judged as an energy surplus; the energy storage charging and discharging module is controlled to charge at the maximum allowable power; if the local energy storage is full, the preset redundant energy storage unit is activated to send redundant energy to other units.

[0063] When the sum of the power generation capacity and the maximum discharge capacity of the energy storage is less than the load power, it is judged as insufficient energy; the power supply to non-critical loads is cut off, and the power supply sequence of critical loads is dynamically adjusted according to the crop tag requirements and the critical load weight.

[0064] For energy balance scenarios where the power generation is not greater than the sum of the load power and the maximum chargeable energy storage, and the sum of the power generation and the maximum dischargeable energy storage is not less than the load power, the power generation is allocated according to the power generation efficiency.

[0065] The remaining charge of the supercapacitor is maintained within a preset target range, and power is regulated based on a sine function.

[0066] The technical effects and advantages of the cadmium telluride photovoltaic glass and photovoltaic energy storage integrated agricultural greenhouse system of this invention are as follows:

[0067] This invention constructs a multi-dimensional collaborative intelligent system by leveraging the divisibility and material properties of cadmium telluride (CdT) photovoltaic glass. Utilizing the structural advantage of CdT glass, which can be divided into independent sub-units, it recovers the energy from heat spots generated by localized shading through thermoelectric modules and converts it into electrical energy, simultaneously avoiding crop leaf burn caused by infrared radiation, thus overcoming the traditional thermal management challenges of photovoltaic greenhouses. Based on the negative correlation between light transmittance and power generation efficiency, and combined with the dynamic control of fogged light transmittance using electrochromic shading panels, it precisely matches the red and blue light requirements of different growth stages such as seedling and flowering based on crop tags, solving the problems of "shading harming farmers" and "power generation efficiency" caused by fixed light transmittance designs. This invention addresses the contradiction of "low power efficiency" by employing a dual-weighted charging and discharging strategy based on light and temperature differences in the intelligent energy storage module. This strategy enables multi-energy complementary power generation between the cadmium telluride power generation layer and the thermoelectric conversion layer. Combined with crop demand-driven load priority management in the energy utilization module, the supercapacitor's power is maintained within the target range, dynamically ensuring power supply to critical loads. Compared to the energy waste caused by the indivisibility, fixed transmittance, and lack of thermoelectric integration in traditional crystalline silicon systems, this invention achieves synergistic optimization of power generation efficiency, energy utilization rate, and crop stress damage rate through integrated design, constructing an innovative agricultural photovoltaic paradigm of "high-efficiency power generation - precision planting - intelligent energy storage". Attached Figure Description

[0068] Figure 1 This is a schematic diagram of the integrated agricultural greenhouse system combining cadmium telluride photovoltaic glass and photovoltaic energy storage according to the present invention.

[0069] Figure 2 This is a schematic diagram of the data flow in this invention;

[0070] Figure 3 This is a schematic diagram of the cadmium telluride thin-film photovoltaic glass structure of the present invention;

[0071] Figure 4This is a schematic diagram of the method for dynamically adjusting the hazy light transmittance of the light-shielding plate according to the crop label of the present invention. Detailed Implementation

[0072] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0073] Example 1

[0074] Please see Figure 1 , Figure 2 As shown, this embodiment provides an integrated agricultural greenhouse system combining cadmium telluride photovoltaic glass and photovoltaic energy storage, including:

[0075] Component design module: The cadmium telluride power generation glass is divided into N sub-units (e.g., 36 independent sub-units) and physically isolated; a π-type thermocouple is integrated into the back panel of each sub-unit to generate direct current through the temperature difference between the hot spot area and the room temperature area;

[0076] Reference Figure 3 Cadmium telluride thin-film photovoltaic glass has a typical multilayer film structure, consisting of a front panel (glass, used in the manufacture of triple-glass BIPV products, not used in standard products), a substrate (glass), a TCO layer (SnO2:F), a CdSe layer (N junction), a CdTe layer (P junction), a back contact layer, a back electrode, an encapsulant film, and a back panel (glass).

[0077] Cadmium ionide (CdTe) solar cells are thin-film solar cells based on a heterojunction of p-type CdTe and n-type CdS / CdSe. The process involves depositing a transparent conductive film (TCO) on a glass substrate, followed by sequential deposition of cadmium sulfide / selenide (n-type) and cadmium telluride (p-type) films using near-space sublimation (CSS). CdCl₂ treatment is then used to activate the crystal structure and activate the PN junction. A metal electrode (Mo / Al / Cr) is then deposited via PVD, forming a glass / TCO / CdS(e) / CdTe / Mo / Al / Cr cell structure. The cells are then sealed using butyl tape and encapsulating film, and finally assembled into a junction box to complete the cell assembly.

[0078] By dividing cadmium telluride (CdT) photovoltaic glass into independent sub-units and physically isolating them, and combining them with π-type thermocouples integrated on the back, electricity is generated by the temperature difference between the hot spot and the room temperature zone. On the one hand, the divisible and easily integrated thermoelectric conversion structure of CdT photovoltaic glass solves the problems of reduced power generation efficiency and crop burn caused by partial shading of hot spots in the module. Compared with crystalline silicon products, its divisibility on the glass substrate and its advantages in adapting to thermoelectric integration in agricultural scenarios are irreplaceable. On the other hand, through sub-unitization, combined with subsequent regulation based on crop status, it can dynamically respond to individual crop differences and growth stage needs, breaking through the bottleneck of traditional fixed light transmission design. In balancing power generation efficiency and crop light demand, and accurately sensing stress, the unique material and structural adaptability of CdT provides core support for solving the problem of coexistence of inefficient photovoltaic greenhouse power generation and crop damage.

[0079] Crop detection module: This module acquires crop status through sub-unit data collection and analysis. The crop status is then used as input to the status analysis model to obtain crop tags. Crop status includes leaf temperature, chlorophyll fluorescence parameters, chlorophyll content, stem diameter, crop image, ambient temperature, ambient humidity, light intensity, and CO2 concentration. Crop tags include growth stage tags (e.g., seedling stage, flowering stage, and fruiting stage), physiological health tags (e.g., healthy and photoinhibition), environmental stress tags (e.g., high temperature stress, drought stress, and carbon starvation), and energy requirement tags (e.g., high energy requirement and energy surplus).

[0080] By deploying various sensors within the independent sub-units of cadmium telluride (CTD) power generation glass, and collecting multi-dimensional crop status data such as leaf temperature and chlorophyll fluorescence to generate tags indicating growth stage and physiological health, a real-time response system of "crop status - sub-unit light transmittance - power generation efficiency" can be constructed, leveraging the unique glass-based separability and dynamic transmittance control potential of CTD materials. Compared to crystalline silicon products, this approach utilizes the structural characteristic of CTD being separable into N independent sub-units to achieve precise perception and differentiated control of individual crop phenotypes, for example, avoiding the limitations of crystalline silicon. The dilemma of "shading harms farmers" or "inefficient power generation" caused by fixed light transmittance can be overcome by integrating sensors at the sub-unit level to capture crop heat stress and differences in growth stages that traditional greenhouses cannot identify. By combining the negative correlation between cadmium telluride light transmittance and power generation efficiency, the light transmittance strategy of each sub-unit can be dynamically optimized. While solving the problem of local shading and heat spot burns, the material-level light transmittance-power generation synergy can break through the bottleneck of agricultural photovoltaic application caused by the indivisibility and fixed light transmittance of crystalline silicon products, and achieve a precise balance between power generation efficiency and crop growth needs.

[0081] Leaf temperature was acquired using an infrared thermal imager. A statistical method was used to fit the functional relationship between SPAD value and chlorophyll content. Chlorophyll content was obtained by acquiring SPAD value using a spectrophotometer. Stem diameter was acquired using a micro-variable monitoring sensor. Crop images were acquired using a visible light camera. Ambient temperature was acquired using a temperature sensor. Ambient humidity was acquired using a humidity sensor. Light intensity was acquired using a photometer. CO2 concentration was acquired using a non-dispersive infrared gas analyzer.

[0082] Methods for obtaining chlorophyll fluorescence parameters include:

[0083] Preset modulation light is applied to plant leaves to measure the initial fluorescence Fo. Then, saturated pulse light is applied to measure the maximum fluorescence Fm. The difference between Fm and Fo is calculated to obtain the variable fluorescence Fv. The ratio of Fv to Fm is calculated to obtain the chlorophyll fluorescence parameter.

[0084] Methods for obtaining image features include:

[0085] Image features include morphological features, color features, and texture features;

[0086] Detect the blade outline, calculate the perimeter and area, and calculate the roundness based on the perimeter and area;

[0087] The crop image is binarized to obtain a binary image. The binary image is then converted into a single-pixel skeleton, and the number of branches and the length of the main stem are calculated.

[0088] Calculate plant height and leaf length based on a preset calibration object (such as a checkerboard pattern of known size);

[0089] The roundness, number of branches, main stem length, plant height and leaf length are combined to form morphological characteristics;

[0090] The crop image is converted from RGB space to HSV space, and the components of each dimension are extracted.

[0091] Calculate the difference and sum of the NIR and R channels respectively, and calculate the ratio of the difference to the sum to obtain the NDVI;

[0092] Calculate the difference and sum of G531 and G570 respectively, and calculate the ratio of the difference to the sum to obtain PRI;

[0093] The color features are obtained by concatenating the components of each dimension, NDVI, and PRI.

[0094] The gray-level co-occurrence matrix is ​​obtained by counting the frequency of gray values ​​(i,j) at any two points in the image.

[0095] Contrast, correlation, energy, and entropy are calculated based on the gray-level co-occurrence matrix, and the matrix features are obtained by concatenating them.

[0096] For each pixel in the image, take an A×A neighborhood, compare the neighboring pixels with the center pixel, record 1 if the neighboring pixel is greater than or equal to the center value, otherwise record 0; combine the E neighboring binary values ​​in clockwise order to form an E-bit binary number, convert it to decimal and use it as the LBP value of the center pixel; calculate the histogram of the LBP values ​​of the entire image as the global feature.

[0097] Texture features are obtained by concatenating matrix features and global features.

[0098] Training methods for state analysis models include:

[0099] W sets of crop training data were collected in advance, including crop status and crop labels;

[0100] Crop training data is used as input to the state analysis model, and crop labels are used as output. With the goal of minimizing the error between the output crop labels and the actual crop labels, the network parameters of the state analysis model are optimized using a natural heuristic optimization algorithm. The network parameters that minimize the error between the crop labels output by the state analysis model and the actual crop labels are obtained. The state analysis model constructed with the corresponding network parameters is used as the trained state analysis model.

[0101] By deploying sensors within independent sub-units of cadmium telluride (CdT) power-generating glass to acquire chlorophyll fluorescence parameters and image features, and leveraging the unique glass-based divisibility and dynamic transmittance control capabilities of CdT, a real-time mapping system of "crop physiological state - phenotypic characteristics - sub-unit transmittance - power generation efficiency" can be constructed. Compared to crystalline silicon products, this approach utilizes the structural characteristic of CdT that allows for the division into N independent sub-units to accurately capture crop photosynthetic system stress and physiological health status through chlorophyll fluorescence parameters. Combined with phenotypic information such as plant height and branch number extracted from image features, growth stages can be determined. Furthermore, the transmittance and power generation efficiency of CdT can be leveraged... The negative correlation characteristic transforms the photosynthetic energy demand reflected by fluorescence parameters and the morphological stress characterized by image features into sub-unit-level transmittance control commands. While solving the contradiction between "light demand and power generation efficiency" caused by the fixed transmittance of traditional crystalline silicon, it uses a phenotypic sensor integrated on a glass substrate to perceive the differences in individual crop heat stress and growth stage in real time. This breaks through the bottleneck of precise control in agricultural photovoltaic scenarios caused by the indivisibility, fixed transmittance, and incompatibility of multi-dimensional sensors of crystalline silicon products. With the material-level transmittance-power generation-phenotypic perception synergy, it achieves a multi-objective balance of avoiding hot spot damage, matching crop growth needs, and optimizing power generation efficiency.

[0102] Dynamic shading module: Each sub-unit is connected to a thermoelectric module, which is directly connected to an electrochromic glass shading plate. The hazy light transmittance of the shading plate is dynamically adjusted according to the crop label.

[0103] Reference Figure 4Methods for dynamically adjusting the hazy light transmittance of the shade plate based on the crop label include:

[0104] By long-term monitoring of physiological indicators of crop growth cycle, the change curves of different stages are fitted, and the slope of the stage transition is obtained by differentiating the change curves. Chlorophyll fluorescence parameters and stem diameter changes within a preset time period are monitored. When the chlorophyll fluorescence parameter decreases by R% and the stem diameter change is lower than the growth threshold, the current daily average temperature and biological lower limit temperature are obtained, and the critical point is calculated. The growth stage influencing factors are calculated based on the slope and the critical point of the key growth stage.

[0105] Chlorophyll content was normalized; temperature stress was constructed based on the Arrhenius equation, combined with activation energy, gas constant, optimum temperature and leaf temperature; the stem diameter change rate was calculated based on the stem diameter change within a preset time period, and the change rate stress was calculated based on the optimal change rate value; physiological state influencing factors were calculated based on normalized chlorophyll content, temperature stress and change rate stress.

[0106] The stress factors are reduced in dimensionality using PCA, and the contribution rate is calculated as the stress factor weight. The stress intensity is calculated based on the measured value of the stress factor, the optimal stress value, and the stress threshold. The environmental stress impact factor is obtained based on the stress factor weight and stress intensity.

[0107] Based on the Farquhar model, a photosynthetic energy model of photosynthetic rate with respect to light intensity was constructed according to the maximum photosynthetic rate, photon efficiency, and crop respiration rate; the energy required for crop photosynthesis was calculated based on the photosynthetic rate, leaf area, and energy conversion coefficient of photosynthetic products.

[0108] An energy balance equation is constructed regarding transmittance, light intensity, energy required for crop photosynthesis, and charging requirements of the energy storage system. Specifically, the product of transmittance and light intensity must not be less than the sum of the energy required for crop photosynthesis and the charging requirements of the energy storage system. Coupling constraints are calculated based on these three factors. The physical boundary of transmittance is obtained based on the minimum and maximum transmittance of cadmium telluride photovoltaic glass. A physiological boundary is calculated based on light intensity and the light saturation point; when the light intensity exceeds the light saturation point, transmittance needs to be reduced to avoid light inhibition. The coupling constraints, physical boundaries, and physiological boundaries are then integrated into a compensation function.

[0109] Obtain the current glass transmittance, calculate the cumulative product of various influencing factors and the environmental compensation function; calculate the product of the current glass transmittance and the cumulative product to obtain the target transmittance at time t.

[0110] By connecting thermoelectric modules to independent sub-units of cadmium telluride (CdT) power-generating glass and directly linking them to electrochromic glass shading panels, and leveraging the unique glass-based divisibility and dynamic transmittance control potential of CdT, a closed-loop response system of "thermoelectric conversion - phenotypic sensing - electrochromism" is constructed. Compared to crystalline silicon products, this system utilizes the structural characteristic of CdT being divisible into N independent sub-units to convert the temperature difference of hot spots generated by local shading into direct current through thermoelectric modules. Furthermore, it dynamically calculates growth stages, physiological states, and environmental stress factors based on crop tags, combined with Farqu... The energy balance equation constructed by the HAR model transforms the negative correlation between cadmium telluride transmittance and power generation efficiency into an advantage. By adjusting the hazy transmittance in real time through an electrochromic shading plate, it avoids the dilemma of "power generation-growth" caused by the fixed transmittance of crystalline silicon. At the same time, it utilizes the thermoelectric-electrochromic integrated structure on the glass substrate to achieve multi-objective synergy of hot spot energy recovery, precise matching of crop light demand, and optimization of power generation efficiency. This breaks through the bottleneck of agricultural photovoltaic application caused by the indivisibility, fixed transmittance, and incompatibility of thermoelectric conversion structure of crystalline silicon products.

[0111] Intelligent energy storage module: Each sub-unit integrates a cadmium telluride power generation layer, a thermoelectric conversion layer, and a supercapacitor. It generates an intelligent charging and discharging strategy based on the light intensity to drive the cadmium telluride power generation layer and the thermoelectric conversion layer to generate electricity, and stores the electricity through the supercapacitor.

[0112] Methods for generating intelligent charge / discharge strategies include:

[0113] Charging strategy:

[0114] The light intensity weight is calculated based on the current light intensity, the maximum light intensity and the minimum light intensity within a preset time period; wherein, the current light intensity is limited to between the maximum light intensity and the minimum light intensity.

[0115] The temperature difference weight is calculated based on the maximum and minimum temperatures of the thermoelectric conversion layer within a preset time period and the current temperature of the thermoelectric conversion layer; wherein, the current temperature of the thermoelectric conversion layer is limited to between the maximum and minimum temperatures;

[0116] The total available charging power is calculated by obtaining the real-time power of the cadmium telluride power generation layer, the real-time power of the thermoelectric conversion layer, and the current load power.

[0117] When the light intensity weight is not lower than the temperature difference weight, the cadmium telluride power generation layer itself is charged; otherwise, the total available charging power is allocated to the cadmium telluride power generation layer according to the weight ratio.

[0118] The total available charging power, excluding the charging power allocated to the cadmium telluride power generation layer, is entirely allocated to the thermoelectric conversion layer.

[0119] Discharge strategy:

[0120] Obtain the remaining supercapacitor charge SOC(t), minimum charge value, minimum threshold, maximum threshold, and maximum charge value at time t; where SOC(t) is updated based on charging power, discharging power, capacitor charge, and charge / discharge cycle efficiency;

[0121] The SOC state interval is defined by a piecewise function, where SOC(t) is marked as L when it is below the minimum threshold, M when it is below the maximum threshold but not below the minimum threshold, and H when it is not below the maximum threshold.

[0122] Crop tags are used as input to the demand analysis model to obtain the crop energy demand index. The priority of critical loads is calculated based on the crop energy demand index and the priority weight w1 of critical loads, and the priority of non-critical loads is calculated based on the crop energy demand index and the priority weight w2 of non-critical loads. Among them, critical loads and non-critical loads are defined in stages according to the growth stage tags. Equipment that must be used in each stage (such as seedling raising equipment, irrigation equipment, etc.) is defined as critical loads, and other equipment is defined as non-critical loads (such as pollination auxiliary equipment, etc. in the seedling stage).

[0123] Training methods for demand analysis models include:

[0124] K sets of training data were collected in advance, including crop labels and crop energy demand index;

[0125] The training data is used as the input to the demand analysis model, and the crop energy demand index is used as the output of the demand analysis model. With the goal of minimizing the error between the output crop energy demand index and the actual crop energy demand index, the network parameters of the demand analysis model are optimized by a natural heuristic optimization algorithm. The network parameters that minimize the error between the crop energy demand index output by the demand analysis model and the actual crop energy demand index are obtained. The demand analysis model constructed with the corresponding network parameters is used as the trained demand analysis model.

[0126] The sum of w1 and w2 is 1; w1 is obtained by defining an objective function, such as maximizing the power supply satisfaction rate of critical loads while ensuring crop survival, or the load weight combination with the highest energy utilization rate, and using a natural heuristic optimization algorithm for optimization. The reward function can also be calculated based on effective power utilization, total power generation, target power, charging power, and load unmet requirements; w1 is dynamically adjusted based on the current w1, learning rate, reward function, and the gradient of the reward function with respect to w1.

[0127] The available discharge power P is calculated based on the discharge efficiency, SOC(t), supercapacitor capacity, and time interval.

[0128] If the SOC state interval is L, then P only supplies power to G;

[0129] If the SOC state interval is M, then P supplies full power to G, and also supplies power to F according to the priority ratio of G priority and F priority;

[0130] If the SOC state interval is H, then P supplies full power to both G and F.

[0131] By integrating a cadmium telluride (CdT) power generation layer, a thermoelectric conversion layer, and a supercapacitor into an independent subunit of cadmium telluride (CdT) power generation glass, a smart energy storage module with multi-energy synergy of "light-heat-storage" is constructed. Leveraging the unique glass-based divisibility and dynamic transmittance control characteristics of CdT materials, the module can dynamically allocate the charging power of the CdT power generation layer and the thermoelectric conversion layer through light intensity and temperature difference weights. Furthermore, it can achieve phased intelligent discharge based on the priority of critical / non-critical loads defined by crop tags, combined with the state of charge (SOC). This module transforms the negative correlation between CdT transmittance and power generation efficiency into a dynamic control advantage. Through a dual-weighted charge-discharge strategy based on light and temperature difference, it ensures power supply to critical loads at different growth stages of crops while recovering hot spot energy through the thermoelectric conversion layer and storing it through the supercapacitor. This avoids the energy distribution imbalance between power generation and growth caused by the indivisibility and fixed transmittance of crystalline silicon. Through an integrated design of materials, structure, and algorithms, it achieves synergistic optimization of power generation efficiency, crop demand, and energy storage management in agricultural photovoltaic scenarios.

[0132] Energy Utilization Module: Based on the energy storage level corresponding to crop tags and smart charging / discharging strategies, a composite adjustment strategy is obtained.

[0133] Methods for obtaining composite adjustment strategies include:

[0134] Obtain the power generation and thermoelectric conversion power of cadmium telluride, and calculate the power generation.

[0135] Obtain the critical load power and non-critical load power, and calculate the load power;

[0136] Obtain the charging power threshold and calculate the maximum chargeable energy storage capacity based on the charging efficiency; obtain the discharging power threshold and calculate the maximum dischargeable energy storage capacity based on the discharging efficiency.

[0137] When the power generation is greater than the sum of the load power and the maximum chargeable capacity of the energy storage, it is judged as an energy surplus.

[0138] Control the energy storage charging and discharging module to charge at the maximum allowable power;

[0139] If the local energy storage is full, activate the preset redundant energy storage unit to send redundant energy to other areas;

[0140] When the sum of the power generation and the maximum discharge capacity of the energy storage is less than the load power, it is judged as insufficient energy;

[0141] Cut off power to non-critical loads and dynamically adjust the power supply sequence of critical loads according to crop tag requirements and critical load weights (prioritizing the equipment with the highest critical load weight);

[0142] For other energy balance scenarios, namely, scenarios where the power generation is not greater than the sum of the load power and the maximum chargeable energy storage, and scenarios where the sum of the power generation and the maximum dischargeable energy storage is not less than the load power, the power generation is allocated according to the power generation efficiency.

[0143] The remaining charge of the supercapacitor is maintained within a preset target range, and power is regulated based on a sine function.

[0144] By constructing a collaborative energy utilization module integrating power generation, load, and energy storage using independent sub-units of cadmium telluride (CdT) photovoltaic glass, and leveraging the unique glass-based divisibility and dynamic transmittance control capabilities of CdT, a composite strategy response to crop tags and energy storage levels is achieved. This module utilizes the structural characteristic of CdT being divisible into N independent sub-units to integrate CdT power generation and thermoelectric conversion power in real time. When energy is excessive, it prioritizes fully charging local supercapacitors and activating redundant energy storage for external transmission. Conversely, when energy is insufficient, it dynamically cuts off non-critical loads and adjusts the power supply sequence of critical loads based on crop tags. This module transforms the negative correlation between CdT transmittance and power generation efficiency into a dynamic control advantage. It smooths power fluctuations using a sine function and maintains supercapacitor charge within the target range. This overcomes the bottleneck of extensive energy distribution in agricultural photovoltaic scenarios caused by the indivisibility, fixed transmittance, and incompatibility with thermoelectric conversion structures of crystalline silicon products. Through material-level multi-energy collaborative design, it achieves the triple goals of optimized power generation efficiency, crop demand matching, and extended energy storage lifespan.

[0145] Example 2

[0146] This embodiment provides a quantum dot-based dynamic spectral reconstruction method applied to Embodiment 1, comprising the following steps:

[0147] A red and blue quantum dot modulation layer is embedded in a cadmium telluride glass substrate.

[0148] A growth stage spectral library was constructed based on historical data, containing spectral formulations and quantum dot excitation voltages corresponding to different growth stages of crops.

[0149] Based on crop growth stage labels and influencing factors of growth stage, the quantum dot excitation voltage is adjusted in real time.

[0150] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0151] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An integrated agricultural greenhouse system combining cadmium telluride photovoltaic glass and photovoltaic energy storage, characterized in that: include: Component design module: Divide the cadmium telluride power-generating glass into N sub-units and physically isolate them; A π-type thermocouple is integrated into the backplane of each subunit. Crop detection module: It collects and analyzes crop status through sub-units, and uses the crop status as input to the status analysis model to obtain crop labels; Dynamic shading module: Each sub-unit is connected to a thermoelectric module, which is directly connected to an electrochromic glass shading plate. The hazy light transmittance of the shading plate is dynamically adjusted according to the crop label. Intelligent energy storage module: Each sub-unit integrates a cadmium telluride power generation layer, a thermoelectric conversion layer, and a supercapacitor, and generates an intelligent charging and discharging strategy based on the light intensity; Energy Utilization Module: Based on the energy storage level corresponding to crop tags and smart charging / discharging strategies, a composite adjustment strategy is obtained.

2. The integrated agricultural greenhouse system combining cadmium telluride photovoltaic glass and photovoltaic energy storage according to claim 1, characterized in that, Crop status includes leaf temperature, chlorophyll fluorescence parameters, chlorophyll content, stem diameter, image features, ambient temperature, ambient humidity, light intensity, and CO2 concentration; Crop labels include growth stage labels, physiological health labels, environmental stress labels, and energy requirement labels; Methods for obtaining chlorophyll fluorescence parameters include: Preset modulation light is applied to plant leaves to measure the initial fluorescence Fo. Then, saturated pulse light is applied to measure the maximum fluorescence Fm. The difference between Fm and Fo is calculated to obtain the variable fluorescence Fv. The ratio of Fv to Fm is calculated to obtain the chlorophyll fluorescence parameter.

3. The integrated agricultural greenhouse system combining cadmium telluride photovoltaic glass and photovoltaic energy storage according to claim 2, characterized in that, Methods for dynamically adjusting the hazy light transmittance of the shade plate based on the crop label include: The crop status was analyzed to obtain factors influencing growth stage, physiological state, and environmental stress. Based on the Farquhar model, the photosynthetic rate is calculated according to the maximum photosynthetic rate, photon efficiency, crop respiration rate, and light intensity; the energy required for crop photosynthesis is calculated according to the photosynthetic rate, leaf area, and energy conversion coefficient of photosynthetic products. An energy balance equation is constructed to consider transmittance, light intensity, energy required for crop photosynthesis, and charging requirements of the energy storage system. Coupling constraints are obtained based on the energy balance equation. The physical boundary of transmittance is obtained based on the minimum and maximum transmittance of cadmium telluride power generation glass. The physiological boundary is calculated based on light intensity and light saturation point. The coupling constraints, physical boundary, and physiological boundary are integrated into a compensation function. Obtain the current glass transmittance, calculate the cumulative product of various influencing factors and the environmental compensation function; calculate the product of the current glass transmittance and the cumulative product to obtain the target transmittance at time t.

4. The integrated agricultural greenhouse system combining cadmium telluride photovoltaic glass and photovoltaic energy storage according to claim 3, characterized in that, Methods for obtaining growth stage influencing factors include: By long-term monitoring of physiological indicators of crop growth cycle, the change curves of different stages are fitted, and the slope of the stage transition is obtained by differentiating the change curves. Chlorophyll fluorescence parameters and stem diameter changes within a preset time period are monitored. When the chlorophyll fluorescence parameter decreases by R% and the stem diameter change is lower than the growth threshold, the current daily average temperature and biological lower limit temperature are obtained, and the critical point is calculated. The growth stage influencing factors are calculated based on the slope and the critical point of the key growth stage.

5. The integrated agricultural greenhouse system combining cadmium telluride photovoltaic glass and photovoltaic energy storage according to claim 3, characterized in that, Methods for obtaining factors influencing physiological states include: Chlorophyll content was normalized; temperature stress was constructed based on the Arrhenius equation, combined with activation energy, gas constant, optimum temperature and leaf temperature; the stem diameter change rate was calculated based on the stem diameter change within a preset time period, and the change rate stress was calculated based on the optimal change rate value; physiological state influencing factors were calculated based on normalized chlorophyll content, temperature stress and change rate stress. Methods for obtaining environmental stress influencing factors include: The dimensionality of the stress factors is reduced by PCA, and the contribution rate is calculated as the weight of the stress factors. The stress intensity is calculated based on the measured value of the stress factors, the optimal value of the stress, and the stress threshold. The environmental stress impact factor is obtained by calculating the stress factor weight and the stress intensity.

6. The integrated agricultural greenhouse system combining cadmium telluride photovoltaic glass and photovoltaic energy storage according to claim 1, characterized in that, Methods for obtaining the charging strategy in the intelligent charging and discharging strategy include: Calculate the light intensity weight based on the current light intensity, the maximum light intensity and the minimum light intensity within the preset time period; The temperature difference weight is calculated based on the maximum and minimum temperatures of the thermoelectric conversion layer within a preset time period and the current temperature of the thermoelectric conversion layer. The total available charging power is calculated by obtaining the real-time power of the cadmium telluride power generation layer, the real-time power of the thermoelectric conversion layer, and the current load power. When the light intensity weight is not lower than the temperature difference weight, the cadmium telluride power generation layer itself is charged; otherwise, the total available charging power is allocated to the cadmium telluride power generation layer according to the weight ratio. The total available charging power, excluding the charging power allocated to the cadmium telluride power generation layer, is entirely allocated to the thermoelectric conversion layer.

7. The integrated agricultural greenhouse system combining cadmium telluride photovoltaic glass and photovoltaic energy storage according to claim 1, characterized in that, Methods for obtaining the discharge strategy in the intelligent charge-discharge strategy include: Obtain the remaining charge SOC(t), minimum charge, minimum threshold, maximum threshold, and maximum charge of the supercapacitor at time t. The SOC state interval is defined by a piecewise function, where SOC(t) is marked as L when it is below the minimum threshold, M when it is below the maximum threshold but not below the minimum threshold, and H when it is not below the maximum threshold. Crop tags are used as input to the demand analysis model to obtain the crop energy demand index; G priority and F priority are calculated based on the crop energy demand index, the critical load G priority weight and the non-critical load F priority weight. The available discharge power P is calculated based on the discharge efficiency, SOC(t), supercapacitor capacity, and time interval. If the SOC state interval is L, then P only supplies power to G; If the SOC state interval is M, then P supplies full power to G, and also supplies power to F according to the priority ratio of G priority and F priority; If the SOC state interval is H, then P supplies full power to both G and F.

8. The integrated agricultural greenhouse system combining cadmium telluride photovoltaic glass and photovoltaic energy storage according to claim 1, characterized in that, Image features include morphological features, color features, and texture features; Methods for obtaining morphological features include: Detect the blade outline, calculate the perimeter and area, and calculate the roundness based on the perimeter and area; The crop image is binarized to obtain a binary image. The binary image is then converted into a single-pixel skeleton, and the number of branches and the length of the main stem are calculated. Calculate plant height and leaf length based on preset calibration objects; The roundness, number of branches, main stem length, plant height and leaf length are combined to form morphological characteristics; Methods for obtaining color features include: The crop image is converted from RGB space to HSV space, and the components of each dimension are extracted. Calculate the difference and sum of the NIR and R channels respectively, and calculate the ratio of the difference to the sum to obtain the NDVI; Calculate the difference and sum of G531 and G570 respectively, and calculate the ratio of the difference to the sum to obtain PRI; The color features are obtained by splicing together the components of each dimension, NDVI, and PRI.

9. The integrated agricultural greenhouse system combining cadmium telluride photovoltaic glass and photovoltaic energy storage according to claim 8, characterized in that, Methods for obtaining texture features include: The gray-level co-occurrence matrix is ​​obtained by counting the frequency of gray values ​​(i,j) at any two points in the image. Contrast, correlation, energy, and entropy are calculated based on the gray-level co-occurrence matrix, and the matrix features are obtained by concatenating them. For each pixel in the image, take an A×A neighborhood, compare the neighboring pixels with the center pixel, record 1 if the neighboring pixel is greater than or equal to the center value, otherwise record 0; combine the E neighboring binary values ​​in clockwise order to form an E-bit binary number, convert it to decimal and use it as the LBP value of the center pixel; calculate the histogram of the LBP values ​​of the entire image as the global feature. Texture features are obtained by concatenating matrix features and global features.

10. The cadmium telluride photovoltaic glass and photovoltaic energy storage integrated agricultural greenhouse system according to claim 1, characterized in that, Methods for obtaining composite adjustment strategies include: Obtain the power generation and thermoelectric conversion power of cadmium telluride, and calculate the power generation. Obtain the critical load power and non-critical load power, and calculate the load power; Obtain the charging power threshold and calculate the maximum chargeable energy storage capacity based on the charging efficiency; obtain the discharging power threshold and calculate the maximum dischargeable energy storage capacity based on the discharging efficiency. When the power generation is greater than the sum of the load power and the maximum chargeable capacity of the energy storage, it is judged as an energy surplus; the energy storage charging and discharging module is controlled to charge at the maximum allowable power; if the local energy storage is full, the preset redundant energy storage unit is activated to send redundant energy to other units. When the sum of the power generation capacity and the maximum discharge capacity of the energy storage is less than the load power, it is judged as insufficient energy; the power supply to non-critical loads is cut off, and the power supply sequence of critical loads is dynamically adjusted according to the crop tag requirements and the critical load weight. For energy balance scenarios where the power generation is not greater than the sum of the load power and the maximum chargeable energy storage, and the sum of the power generation and the maximum dischargeable energy storage is not less than the load power, the power generation is allocated according to the power generation efficiency. The remaining charge of the supercapacitor is maintained within a preset target range, and power is regulated based on a sine function.

Images