Greenhouse modular combination screening and economic calculation system based on AI algorithm

Abstract

The application discloses a greenhouse modular combination screening and economic calculation system based on an AI algorithm, and through a multi-dimensional environment data acquisition module, historical meteorological data are automatically captured based on user input addresses and standardized environment characteristic vectors are output; an AI double-layer nested optimization engine is used to construct a candidate module set based on the vectors, to establish a next-stage state model and a multi-stage total cost minimization problem, and to solve and output an optimal decision sequence; and an economic calculation and full life cycle evaluation module is used to re-run the state model to simulate indoor microclimate by the hour, to construct a crop yield prediction model to dynamically calculate yield, to obtain a full life cycle cost based on costs in each stage, and to calculate annual income, net present value, internal rate of return and investment return period in combination with yield and local market prices, so that the problems of poor environmental adaptability and large economic benefit deviation caused by the fact that existing greenhouse design relies on experience for type selection are solved, and intelligent decision-making and precise quantification of a construction scheme are realized.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of agricultural construction engineering and artificial intelligence, specifically involving a greenhouse modular combination screening and economic calculation system based on AI algorithms. Background Technology

[0002] The Venlo-style large-scale multi-span intelligent glass greenhouse has widely adopted a modular prefabrication and on-site assembly construction mode, which can realize standardized prefabrication in the factory and rapid on-site assembly, greatly shortening the construction cycle, reducing weather and on-site environmental interference, and significantly improving construction efficiency.

[0003] Prefabricated modules include basic civil engineering, electromechanical equipment, water supply and drainage, HVAC and temperature control, and routine planting and operation and maintenance of fruits and vegetables. In the early stages of greenhouse construction, there is a lack of systematic and intelligent selection methods and no standard for module combination. The matching of equipment and product selection for each professional module relies on manual experience, and it is impossible to quickly complete the quantitative calculation of investment costs, payback period and planting economic benefits for different regional conditions.

[0004] With the industry trend of cost reduction and efficiency improvement and precise investment, how to better combine the mechanical equipment and product selection of various sectors, and calculate the cost recovery in the later stage, has become an important issue in the current environment of cost reduction and efficiency improvement. Summary of the Invention

[0005] To address the aforementioned technical problems, the present invention provides a modular greenhouse combination screening and economic calculation system based on AI algorithms, comprising:

[0006] The multidimensional environmental data acquisition and feature extraction module automatically captures historical daily geographic and meteorological data within a preset time limit based on the address information input by the user, performs data cleaning and feature engineering, and outputs standardized environmental feature vectors. The system includes a standardized module resource library and constraint mapping module, a configured AI double-layer nested optimization engine, a candidate module set built based on standardized environmental feature vectors, a next-stage state model and a multi-stage total cost minimization problem established based on the candidate module set, and the optimal decision sequence output by solving the multi-stage total cost minimization problem. The economic calculation and life cycle assessment module re-runs the next stage state model for each scheme in the optimal decision sequence, simulating hourly indoor microclimate data throughout the entire greenhouse operation cycle, generating a complete time series, constructing a crop yield prediction model, dynamically simulating crop yield under different module combinations, obtaining the life cycle cost based on the cost of each stage calculated based on the optimal decision sequence, and calculating annual income, net present value, internal rate of return, and investment return period based on crop yield, life cycle cost, and local market agricultural product prices.

[0007] Preferably, the configuration AI dual-layer nested optimization engine includes an environment-module adaptation mapping layer and a dynamic programming optimization layer; The environment-module adaptability mapping layer calculates the adaptability of modules in the environment based on the greenhouse module parameters stored in the standardized module resource library and combines them with standardized environmental feature vectors to construct a candidate module set. Based on the candidate module set, the dynamic programming optimization layer establishes the state model for the next stage and the multi-stage total cost minimization problem. It uses backward induction to solve the multi-stage total cost minimization problem, selects the decision variable that minimizes the total cost as the optimal decision for the stage, and iterates until all stages of the greenhouse construction stage sequence are completed. It outputs the optimal decision sequence and forms the Pareto front curve of cost-energy consumption.

[0008] Preferably, the adaptability of the module to the environment is calculated using a fuzzy membership function, as shown in the following formula:

[0009] In the formula, This represents the current environmental characteristic value. For module Environmental features in the standardized environmental feature vector The optimal working center point, The tolerance radius is... These are the curvature control parameters.

[0010] Preferably, the crop yield prediction model formula is as follows:

[0011] In the formula, This represents the theoretical maximum yield of this crop variety under ideal environmental conditions. The number of stages in the crop growth cycle; This is the growth stage; In the first The average difference between the actual indoor temperature and the optimal temperature for the crop during each growth stage; This refers to the temperature deviation tolerance. Gaussian impact factor; This is the light influence coefficient; This refers to the daily cumulative light intensity calculated based on the complete time series. This represents the optimal daily cumulative light intensity.

[0012] Preferably, the life-cycle cost formula is as follows:

[0013] In the formula, For initial investment costs, Annual operating costs is the discount rate.

[0014] This invention provides an AI-based modular combination screening and economic calculation system for greenhouses. Through a multi-dimensional environmental data acquisition module, historical meteorological data is automatically retrieved based on user-input addresses, and a standardized environmental feature vector is output. An AI double-layer nested optimization engine constructs a candidate module set based on this vector, establishes a state model for the next stage, and solves a multi-stage total cost minimization problem, outputting the optimal decision sequence. An economic calculation and life-cycle assessment module re-runs the state model for the optimal solution, simulating hourly indoor microclimates, constructing a crop yield prediction model to dynamically calculate yield, obtaining the life-cycle cost based on the costs at each stage, and calculating annual income, net present value, internal rate of return, and investment return period by combining yield and local market prices. This solves the problems of poor environmental adaptability and large economic benefit deviations caused by existing greenhouse designs relying on experience-based selection, achieving intelligent decision-making and precise quantification of construction schemes. Detailed Implementation

[0015] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0016] This invention provides an AI-based modular combination screening and economic calculation system for greenhouses, comprising: The multidimensional environmental data acquisition and feature extraction module is used to automatically capture historical daily geographic and meteorological data within a preset time limit based on the address information entered by the user in the interactive interface, and to perform data cleaning and feature engineering to output standardized environmental feature vectors.

[0017] Historical daily geographic and meteorological data include temperature (maximum / minimum), light intensity, precipitation, wind speed, and humidity.

[0018] The system comprises a standardized module resource library and a constraint mapping module, configured with an AI dual-layer nested optimization engine. The environment-module adaptability mapping layer calculates the adaptability of modules in the environment based on greenhouse module parameters stored in the standardized module resource library and standardized environmental feature vectors, constructing a candidate module set. The dynamic programming optimization layer, based on the candidate module set, establishes the next-stage state model and a multi-stage total cost minimization problem. It uses backward induction to solve the multi-stage total cost minimization problem, selects the decision variable that minimizes the total cost as the stage-optimal decision, iterates until all stages of the greenhouse construction stage sequence are completed, outputs the optimal decision sequence, and forms the Pareto front curve of cost-energy consumption.

[0019] The modules in the greenhouse module parameters include double-glazed windows, double-layered inflatable film, high-efficiency heat pump, natural gas boiler, wet curtain fan system, internal shading system, external shading system, and supplemental lighting system.

[0020] The first layer, the environment-module adaptability mapping layer, uses a fuzzy membership function to calculate the adaptability of a module to the environment, as shown in the following formula:

[0021] In the formula, This represents the current environmental characteristic value. For module Environmental features in the standardized environmental feature vector The optimal working center point, The tolerance radius is... For curvature control parameters, if ( (If it is a preset threshold), then the module The modules that are eliminated are the candidate modules.

[0022] The candidate module set includes the selected modules and their corresponding physical parameters (e.g., thermal resistance, power) and cost parameters (initial purchase price, annual maintenance cost).

[0023] The second-level dynamic programming optimization layer defines decision variables based on the candidate module set. Greenhouse construction phase sequence State variables Environmental feature vectors Establish the state model for the next stage. and the problem of minimizing total cost in multiple stages The method employs backward induction to solve the multi-stage total cost minimization problem. The decision variable that minimizes the total cost is selected as the optimal decision for each stage. The process is iterated until all stages of the greenhouse construction sequence are completed, and the optimal decision sequence is output, forming the Pareto front curve of cost-energy consumption.

[0024] The decision variables are derived from the candidate module set. The greenhouse construction phase sequence includes the enclosure structure, heating system, ventilation system, shading system, etc., and the state variable is the internal microclimate state of the greenhouse after the completion of each phase in the greenhouse construction phase sequence.

[0025] The formula for the multi-stage total cost minimization problem is as follows:

[0026] In the formula, For capital expenditures, the funds are directly derived from the initial purchase price of the corresponding modules in the candidate module set. Operating expenses, including energy costs (calculated based on the physical parameters of the corresponding modules in the candidate module set). This is a penalty term for environment mismatch, used to quantify the performance degradation of the module under extreme conditions.

[0027] The economic calculation and life-cycle assessment module re-runs the next-stage state model for each option in the optimal decision sequence, simulating hourly indoor microclimate data throughout the entire greenhouse operation cycle. This generates a complete time series including temperature, humidity, and light intensity, and constructs a crop yield prediction model to dynamically simulate crop yields under different module combinations. Based on the costs of each stage calculated from the optimal decision sequence, the life cycle cost (LCC) is obtained, based on crop yield. Life cycle cost (LCC) and local market agricultural product prices Calculate annual income, net present value, internal rate of return, and investment return period.

[0028] The formula for the crop yield prediction model is as follows:

[0029] In the formula, This represents the theoretical maximum yield of this crop variety under ideal environmental conditions. The number of stages in the crop growth cycle; This is the growth stage; In the first The average difference between the actual indoor temperature and the optimal temperature for the crop during each growth stage; This refers to the temperature deviation tolerance. Gaussian impact factor; This is the light influence coefficient; This refers to the daily cumulative light intensity calculated based on the complete time series. This represents the optimal daily cumulative light intensity.

[0030] The formula for total lifecycle cost is as follows:

[0031] In the formula, For initial investment costs, Annual operating costs is the discount rate.

[0032] The formula for annual income is as follows:

[0033] The net present value formula is as follows:

[0034] The formula for the internal rate of return (IRR) investment return period is as follows:

[0035] The number of years at the above value is taken as the investment return period for the internal rate of return.

[0036] The beneficial effects of the embodiments of the present invention are as follows: Environmental interference resistance: By calculating the fuzzy membership degree of the modules, the system can automatically identify and exclude module combinations that may fail under extreme weather conditions (such as high humidity and extreme cold), ensuring the robustness of the solution.

[0037] Enhanced optimization: By adopting a double-layer nested optimization, it avoids the local optima that traditional single-layer algorithms are prone to getting stuck in, and can find the best balance between total lifecycle cost and device performance.

[0038] Accurate economic calculations: The introduction of a nonlinear third-order coupled model solves the problem of excessive errors in traditional linear estimation when predicting crop yields, providing a more scientific basis for investment decisions.

Claims

1. An AI algorithm-based greenhouse modular combination screening and economic calculation system, characterized in that, include: The multidimensional environmental data acquisition and feature extraction module automatically captures historical daily geographic and meteorological data within a preset time limit based on the address information input by the user, performs data cleaning and feature engineering, and outputs standardized environmental feature vectors. The system includes a standardized module resource library and constraint mapping module, a configured AI double-layer nested optimization engine, a candidate module set built based on standardized environmental feature vectors, a next-stage state model and a multi-stage total cost minimization problem established based on the candidate module set, and the optimal decision sequence output by solving the multi-stage total cost minimization problem. The economic calculation and life cycle assessment module re-runs the next stage state model for each scheme in the optimal decision sequence, simulating hourly indoor microclimate data throughout the entire greenhouse operation cycle, generating a complete time series, constructing a crop yield prediction model, dynamically simulating crop yield under different module combinations, obtaining the life cycle cost based on the cost of each stage calculated based on the optimal decision sequence, and calculating annual income, net present value, internal rate of return, and investment return period based on crop yield, life cycle cost, and local market agricultural product prices.

2. The AI algorithm based greenhouse modular combined screening and economic calculation system according to claim 1, characterized in that, The configured AI dual-layer nested optimization engine includes an environment-module adaptation mapping layer and a dynamic programming optimization layer; The environment-module adaptability mapping layer calculates the adaptability of modules in the environment based on the greenhouse module parameters stored in the standardized module resource library and combines them with standardized environmental feature vectors to construct a candidate module set. Based on the candidate module set, the dynamic programming optimization layer establishes the state model for the next stage and the multi-stage total cost minimization problem. It uses backward induction to solve the multi-stage total cost minimization problem, selects the decision variable that minimizes the total cost as the optimal decision for the stage, and iterates until all stages of the greenhouse construction stage sequence are completed. It outputs the optimal decision sequence and forms the Pareto front curve of cost-energy consumption.

3. The AI algorithm based greenhouse modular combined screening and economic calculation system according to claim 1, wherein, The adaptability of the module to the environment is calculated using a fuzzy membership function, as shown in the following formula: In the formula, This represents the current environmental characteristic value. For module Environmental features in the standardized environmental feature vector The optimal working center point, The tolerance radius is... These are the curvature control parameters.

4. The modular combination screening and economic calculation system for greenhouses based on AI algorithms as described in claim 1, characterized in that, The formula for the crop yield prediction model is as follows: In the formula, This represents the theoretical maximum yield of this crop variety under ideal environmental conditions. The number of stages in the crop growth cycle; This is the growth stage; In the first The average difference between the actual indoor temperature and the optimal temperature for the crop during each growth stage; This refers to the temperature deviation tolerance. Gaussian impact factor; This is the light influence coefficient; This refers to the daily cumulative light intensity calculated based on the complete time series. This represents the optimal daily cumulative light intensity.

5. The modular combination screening and economic calculation system for greenhouses based on AI algorithms as described in claim 1, characterized in that, The formula for the total lifecycle cost is as follows: In the formula, For initial investment costs, Annual operating costs is the discount rate.

Images