Method and apparatus for generating an economic feasibility analysis report based on ai-based automatic matching between farmers and investors

Abstract

The method for generating an economic analysis report based on AI-based automatic matching between farmers and investors according to the present invention may include, in a method performed in one or more processors, the steps of inputting agricultural data and investment data into an artificial intelligence model to predict agricultural productivity data, converting the predicted agricultural productivity data into profit units and applying them to an economic analysis engine to calculate an economic evaluation indicator, using the economic evaluation indicator to derive at least one matching candidate group by comparing the target conditions of farmland and investors, and automatically generating an economic analysis report corresponding to at least one matching candidate group.

Description

Technology Field

[0001] The present invention relates to a method and apparatus for generating an economic feasibility analysis report based on AI-based automatic matching between farmers and investors. More specifically, it relates to a method and apparatus for automatically generating a report based on the results of calculating economic feasibility indicators by performing automatic matching between two parties based on farmland information of farmers aiming for solar power commercialization and investment conditions of investors, and by integrating and analyzing expected returns.

[0002] This invention is a result carried out with support from the 'Eco-Startup Support Program' promoted by the Ministry of Environment and operated by the Korea Environmental Industry & Technology Institute. (Project No.: RT2025021707) Background Technology

[0003] In accordance with policies to expand renewable energy to achieve the 2050 carbon neutrality goal, an energy transition utilizing solar power generation is being promoted in rural areas. In particular, agrivoltaics, which combines crop cultivation with power generation, is attracting attention as an alternative for increasing land-use efficiency; however, its expansion is being limited due to factors such as reduced crop growth caused by shading, the burden of initial investment costs, and fluctuations in power generation. Furthermore, it is currently difficult for farmers to quantitatively verify their net profit by combining agricultural and power generation revenues.

[0004] Private investors require financial indicators such as IRR, NPV, and Payback Period to assess the profitability of agri-photovoltaic projects, but currently rely on manual analysis or professional consulting, resulting in excessive time and cost. Furthermore, there is a lack of an automated economic verification system capable of reflecting complex factors in real time, such as the volatility of SMP and REC prices and the uncertainty of agricultural productivity.

[0005] The current structure connecting farmers and investors is centered around construction companies or brokers, leading to severe information asymmetry and opacity in profit distribution. Farmers find it difficult to objectively analyze the conditions of their farmland and crop characteristics, while investors struggle to quantitatively verify the actual profitability of a project. Consequently, business credibility is low, and a significant number of agro-solar projects are halted in the early stages.

[0006] Therefore, intelligent platform technology is required that can simultaneously perform automatic matching and economic analysis based on data that farmers and investors can trust. Prior art literature

[0007] Patent Publication No. 10-2025-011277 (Published July 24, 2025) Patent Publication No. 10-2025-0090545 (Published June 20, 2025) The problem to be solved

[0008] In resolving the aforementioned problems, the present invention aims to provide an intelligent platform technology capable of simultaneously performing automatic matching and economic analysis based on data that farmers and investors can trust.

[0009] The problems that the present invention aims to solve are not limited to those mentioned above, and other problems not mentioned will be clearly understood by a person skilled in the art to which the present invention belongs from the description below. means of solving the problem

[0010] The method for generating an economic analysis report based on AI-based automatic matching between farmers and investors according to the present invention may include, in a method performed in one or more processors, the steps of inputting agricultural data and investment data into an artificial intelligence model to predict agricultural productivity data, converting the predicted performance data into profit units and applying them to an economic analysis engine to calculate an economic evaluation indicator; deriving at least one matching candidate group by comparing the target conditions of farmland and investors using the economic evaluation indicator; and automatically generating an economic analysis report corresponding to at least one matching candidate group.

[0011] According to one embodiment, the step of calculating an economic evaluation indicator may include: a step of predicting power generation, crop growth rate, and yield using agricultural data including farmland coordinates, area, cultivated crop, soil characteristics, shading allowance, temperature, solar radiation, wind speed, precipitation, System Marginal Price (SMP), Renewable Energy Certificate (REC) price, and charge / discharge efficiency and battery degradation coefficient of an energy storage device; a step of converting the predicted power generation, crop growth rate, and yield into a profit item indicator based on unit price information; and a step of calculating an economic evaluation indicator by combining investment data including the investor's target internal rate of return (IRR), net present value (NPV), payback period, investment amount, risk tolerance, and investment type with the profit item indicator.

[0012] According to one embodiment, the artificial intelligence model predicts power generation using a time-series prediction-based deep learning model from agricultural data, and predicts crop growth rate and yield using a regression analysis-based machine learning model, and the economic analysis engine calculates total revenue by integrating the results predicted by the artificial intelligence model, and can automatically calculate financial indicators by inputting the total revenue into a financial model along with initial investment costs, operating costs, subsidies, and price fluctuation variables.

[0013] According to one embodiment, the step of deriving a plurality of matching candidate groups may include: a step of setting power generation revenue and agricultural revenue as objective functions and imposing an investor's target rate of return as a constraint, using an economic evaluation indicator as input; a step of generating a plurality of candidate solution sets by executing a multi-objective optimization model that considers the conditions of farmland and the conditions of the investor based on the objective function; a step of calculating a similarity score by analyzing the correlation between the conditions of farmland and the conditions of the investor included in the plurality of candidate solution sets; and a step of deriving a matching candidate group by selecting at least one combination in which the degree of matching fit is greater than or equal to a threshold based on the similarity score.

[0014] According to one embodiment, the step of automatically generating an economic analysis report may include: a step of classifying a report type based on economic evaluation indicators corresponding to at least one matching candidate group; a step of automatically configuring customized items according to the report type, such that a report for farmers includes expected power generation revenue, agricultural revenue, and total revenue estimates; a report for investors includes internal rate of return (IRR), net present value (NPV), payback period, and sensitivity analysis results; and a report for policy agencies includes CO2 reduction amount, RE100 contribution, and ESG evaluation indicators; and a step of saving the automatically configured report in the form of an electronic document.

[0015] According to one embodiment, the method may include the steps of: calculating business enhancement items corresponding to farmers, investors, and policy institutions based on economic evaluation indicators included in an economic analysis report; calculating customized indicators for each business enhancement item; and generating an evaluation report that quantitatively evaluates the economic efficiency and social contribution of an agricultural solar power business by synthesizing the calculated customized indicators.

[0016] According to one embodiment, the apparatus may include a memory for storing one or more instructions and a processor for executing one or more instructions stored in the memory, wherein the processor may provide a method for generating an economic analysis report based on AI-based farmer-investor automatic matching by executing one or more instructions.

[0017] The above economic analysis engine can calculate total revenue using the following mathematical formula 1, which is generated using solar power generation revenue, agricultural revenue, and energy storage system operation revenue.

[0018] [Mathematical Formula 1]

[0019]

[0020] Here, R pv is solar power generation revenue, R agri is agricultural revenue, R ess is energy storage system operation revenue, C opex represents operating costs.

[0021] The above economic analysis engine can calculate the internal rate of return (IRR) based on the above total revenue using the following mathematical formula 2.

[0022] [Mathematical Formula 2]

[0023]

[0024] Here, CF t represents the net cash flow of the t-th year, C0 represents the initial investment cost, and n represents the total investment period.

[0025] The above economic analysis engine can calculate the above Net Present Value (NPV) using the following Equation 3.

[0026] [Mathematical Formula 3]

[0027]

[0028] Here, r represents the discount rate.

[0029] The above multi-objective optimization model can calculate the Pareto optimal solution set using the following mathematical formula 4 with the power generation revenue (f1), agricultural revenue (f2), and the investor's target rate of return (f3).

[0030] [Mathematical Formula 4]

[0031]

[0032] Here, Satisfies.

[0033] The step of deriving the above matching candidate group is performed based on K-Nearest Neighbor (KNN), and the above similarity score is calculated by the following mathematical formula 5.

[0034] [Mathematical Formula 5]

[0035]

[0036] Here, x ij is the j-th attribute value of candidate site i, y j is the investor preference attribute value, S i represents the similarity score.

[0037] It may include a step of correcting prediction accuracy by reusing actual operation result data included in the above economic analysis report as training data for the above artificial intelligence model.

[0038] It may include a step of verifying the CO2 reduction amount, RE100 contribution, and ESG evaluation indicators included in the above economic analysis report by linking them with an external certification system. Effects of the invention

[0039] As described above, according to the present invention, by analyzing data of farmers and investors based on artificial intelligence (AI) and performing automatic matching, it is possible to break away from the existing structure of dependence on brokers and enable transparent and objective investment linkages. Accordingly, farmers can easily find the optimal investor suited to their farmland conditions, and investors can make highly reliable decisions based on quantified data.

[0040] In addition, the present invention automatically calculates economic indicators such as IRR, NPV, and Payback by integrating and analyzing power generation revenue, agricultural revenue, and energy storage system (ESS) revenue, so farmers and investors can verify the profitability and feasibility of a business in real time without separate professional analysis. This significantly reduces analysis costs and time, and improves the accuracy of investment decisions.

[0041] Furthermore, the report generation module according to the present invention automatically provides customized economic analysis reports for each stakeholder, such as farmers, investors, and policy institutions, thereby presenting the results of a project in a visual and easy-to-understand manner. This enhances the transparency of the project's progress and can be utilized as supporting data for policy support.

[0042] Furthermore, the present invention automatically calculates the investment amount, profit distribution ratio, and contract period based on matching results to generate a draft contract, and simplifies the contract process by linking with an electronic signature system. As a result, legal consultation costs and negotiation time can be reduced, and the speed of implementation and success rate of agricultural solar power projects can be simultaneously increased. Brief explanation of the drawing

[0043] FIG. 1 is a diagram illustrating the characteristics of generating an economic analysis report based on AI-based farm-investor automatic matching performed by a computing device according to one embodiment. FIG. 2 is a diagram showing the configuration of a computing device according to one embodiment. FIG. 3 is a flowchart illustrating a method for generating an economic analysis report based on AI-based automatic matching between farmers and investors according to one embodiment. FIG. 4 is a diagram illustrating the characteristics of generating an economic analysis report that includes different contents depending on the type according to one embodiment. FIG. 5 is a diagram illustrating the process of generating a draft contract and electronically signing based on the calculation of contract conditions according to one embodiment. Specific details for implementing the invention

[0044] In the present invention, the attached drawings may be illustrated with exaggerated expressions to distinguish it from the prior art, ensure clarity, and facilitate the understanding of the technology. Furthermore, the terms described below are defined considering their functions in the present invention; since these terms may vary depending on the intentions or conventions of the user or operator, their definitions should be based on the technical content throughout this specification. Meanwhile, the embodiments are merely exemplary details of the components presented in the claims of the present invention and do not limit the scope of the rights of the present invention; the scope of rights should be interpreted based on the technical concept throughout the specification of the present invention.

[0045] Throughout the specification, when a configuration is described as "including" a configuration, this means that, unless specifically stated otherwise, it does not exclude other configurations but may include additional configurations.

[0046] Furthermore, when it is said that one configuration is "connected," "connected," or "combined" with another configuration, this means that it is not only "directly connected," "directly connected," or "directly combined," but also that there may be cases where it is "connected with another configuration interposed," "connected with another configuration interposed," or "combined with another configuration interposed." On the other hand, when it is said that one configuration is "directly connected," "directly connected," or "directly combined" with another configuration, it should be understood that there is no other configuration in between.

[0047] In addition, when directional terms such as "front," "back," "up," "down," "left," "right," "first end," "other end," and "both ends" are used, they are used exemplarily in relation to the orientation of the disclosed drawings and should not be interpreted restrictively, and when terms such as "first" and "second" are used, they are terms used to distinguish each configuration and should not be interpreted restrictively.

[0048] In order to more clearly explain the features of the embodiments of the present invention, detailed descriptions of matters widely known to those skilled in the art to which the following embodiments pertain are omitted. Additionally, detailed descriptions of parts in the drawings that are unrelated to the description of the embodiments are omitted.

[0049] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings.

[0050] FIG. 1 is a diagram illustrating the characteristics of generating an economic analysis report based on AI-based farm-investor automatic matching performed by a computing device according to one embodiment.

[0051] According to one embodiment, the computing device (100) may include a processor (110), memory (120), bus (130), communication interface (140), storage (150), and computer program (151) as a core control module of the present invention.

[0052] According to one embodiment, the computing device (100) executes an artificial intelligence model and an economic analysis engine to integrate and analyze agricultural data and investment data, calculates economic evaluation indicators, and derives a group of matching candidates that meet the conditions of farmers and investors through a matching algorithm. Additionally, based on the derived matching results, an economic analysis report can be automatically generated and stored or provided to a farmer terminal (200) and an investor terminal (300) via a network.

[0053] According to one embodiment, the farmer terminal (200) may be implemented as a smartphone, tablet, or agricultural management device. The farmer terminal (200) inputs or transmits in real time the location, area, cultivated crops, shading allowance, soil composition, weather sensor data, etc. of the self-owned farmland, and can receive and verify investment proposals recommended from AI analysis results or reports on expected development and agricultural profits.

[0054] According to one embodiment, the investor terminal (300) may be implemented as an investment portfolio management system or a general client terminal (PC, mobile, etc.). The investor may set conditions such as a target internal rate of return (IRR), investment amount, risk tolerance, and payback period, and receive an economic analysis report and optimal matching candidate sites from the computing device (100). Through this, the investor can compare various farmland conditions in real time and comprehensively determine the investment rate of return and risk level.

[0055] According to one embodiment, the communication network (400) may be composed of a wired or wireless network, an internet network, or a cloud network. Here, the communication network (400) performs data transmission and reception between a farmer terminal (200), an investor terminal (300), and a computing device (100), and reliably transmits agricultural data, investment data, and analysis result data. In addition, the latest data of the economic analysis report can be reflected through real-time updates.

[0056] According to one embodiment, a computing device (100) collects agricultural data and investment data input through a farmer terminal (200) and an investor terminal (300), and preprocesses them into the form of input data for an artificial intelligence model.

[0057] Here, agricultural data may include the coordinates, area, cultivated crops, soil characteristics, allowable shading range, temperature, solar radiation, wind speed, precipitation, System Marginal Price (SMP), Renewable Energy Certificate (REC) price, and the charge / discharge efficiency and degradation factor of the Energy Storage System (ESS).

[0058] According to one embodiment, investment data may include the investor's target internal rate of return (IRR), net present value (NPV), investment payback period, investment amount, risk tolerance, and investment type.

[0059] According to one embodiment, the processor of the computing device (100) can predict power generation using a time series prediction-based deep learning model and predict crop growth rate and yield using a regression-based machine learning model. Here, the predicted power generation revenue and agricultural revenue are converted into revenue items in monetary units using unit price information, and the economic analysis engine can automatically calculate financial indicators such as internal rate of return (IRR), net present value (NPV), payback period, and debt repayment ratio (DSCR) based on this.

[0060] According to one embodiment, a computing device (100) can take a calculated economic evaluation index as input, set the power generation profit, agricultural profit, and the investor's target rate of return as objective functions, and apply a multi-objective evolutionary algorithm (NSGA-II) to search for a non-dominant solution set on the Pareto front. For the searched candidate solutions, a similarity analysis is performed considering the farmer's shading tolerance, profit retention rate, and the investor's risk preference, and as a result, the farmland-investor combination with the highest matching suitability is selected as the final matching candidate group.

[0061] According to one embodiment, a computing device (100) can automatically generate an economic analysis report corresponding to a derived matching candidate group, wherein the contents of the report may be configured differently depending on the user type, the report for farmers includes estimated power generation revenue, agricultural revenue, and total revenue estimates, the report for investors includes IRR, NPV, payback period, and sensitivity analysis results, and the report for policy agencies includes policy-linked data such as CO2 reduction amount, RE100 contribution, and ESG evaluation indicators. Such reports may be stored in the form of electronic documents or displayed on a cloud dashboard.

[0062] Accordingly, the system of the present invention illustrated in Fig. 1 provides a sustainable agro-solar business platform that balances the profit structure of farmers and investors through an AI-based multi-purpose optimization algorithm and automatically visualizes the results of economic analysis, thereby enabling income stability for farmers, investment efficiency for investors, and objective evaluation of carbon reduction and rural income growth performance for policy agencies.

[0063] FIG. 2 is a diagram showing the configuration of a computing device according to one embodiment.

[0064] According to one embodiment, a computing device (100) may be provided that includes a memory (120) for storing one or more instructions and a processor (110) for executing one or more instructions stored in the memory (120), wherein the processor (110) performs a method for generating an economic analysis report based on AI-based farmer-investor automatic matching by executing one or more instructions.

[0065] According to one embodiment, the computing device (100) may include a processor (110), memory (120), a bus (130), a communication interface (140), and storage (150).

[0066] According to one embodiment, the processor (110) is a central processing unit (CPU) that performs overall control of the present invention and is responsible for AI model training and inference, execution of an economic analysis engine, execution of a matching algorithm, automatic generation of reports, etc. In particular, the processor (110) can control a series of processes to receive agricultural data and investment data, predict performance, calculate economic evaluation indicators, and derive matching results by loading and executing a computer program (151) stored in memory (120) or storage (150).

[0067] According to one embodiment, the memory (120) may be configured as a volatile or non-volatile storage and temporarily stores data during the execution of an artificial intelligence model, an optimization algorithm, and a database module executed on the processor (110). Additionally, a computer program (151) for implementing the method of the present invention is loaded into the memory (120) to manage the weights of the AI ​​model, input feature data, prediction result values, etc., in real time.

[0068] According to one embodiment, the bus (130) is an internal connection means that controls the flow of data between the processor (110), memory (120), communication interface (140), and storage (150). The bus (130) manages the transmission and reception of instructions, data, and control signals, and enables all components within the system to interact with each other so that data is processed efficiently.

[0069] According to one embodiment, the communication interface (140) transmits and receives data to and from a farmer terminal (200), an investor terminal (300), and an external database (such as a Meteorological Agency API, a Power Exchange API, etc.) via a wired or wireless network. Through this, the computing device (100) collects agricultural data, investment conditions, weather and power information in real time, and transmits the analysis results to an external terminal to enable report sharing and system control.

[0070] According to one embodiment, the storage (150) is a large-capacity non-volatile storage device that permanently stores economic analysis reports, investment records, farmland data, algorithm setting values, statistical data for policy agencies, etc. Additionally, the storage (150) stores a computer program (151) for performing an embodiment of the present invention and can be loaded and executed by a processor (110).

[0071] Accordingly, the computing device (100) illustrated in FIG. 2 functions as an intelligent integrated computing device that serves as the central control unit of an AI-based farm-investor automatic matching system, analyzes collected data, converts prediction results into financial indicators for economic evaluation and report generation, and ultimately provides them to users (farmers, investors, and policy institutions).

[0072] FIG. 3 is a flowchart illustrating a method for generating an economic analysis report based on AI-based automatic matching between farmers and investors according to one embodiment.

[0073] The method for generating an economic analysis report based on AI-based automatic matching between a farmer and an investor according to the present invention may include the step (S10) of inputting agricultural data and investment data into an artificial intelligence model to predict agricultural productivity data, converting the predicted performance data into profit units, and applying them to an economic analysis engine to calculate an economic evaluation indicator; the step (S20) of deriving at least one matching candidate group by comparing the target conditions of the farmland and the investor using the economic evaluation indicator; and the step (S30) of automatically generating an economic analysis report corresponding to at least one matching candidate group.

[0074] According to one embodiment, agricultural productivity data may include predicted crop growth rates, yields, and power generation results using variables such as the location and area of ​​farmland, cultivated crops, soil composition, shading allowance, temperature, solar radiation, wind speed, and precipitation. An artificial intelligence model can quantitatively analyze the complex profit structure of the agriculture and energy sectors by simultaneously predicting crop growth patterns, harvest timing, and energy generation efficiency based on this agricultural productivity data. Here, the economic analysis engine refers to an integrated computation module that receives data such as power generation, crop yields, and energy storage device operational efficiency predicted by the artificial intelligence model, converts this into financial variables of revenue, cost, and risk, and then automatically calculates economic evaluation indicators such as the Internal Rate of Return (IRR), Net Present Value (NPV), Payback Period, and Debt Service Coverage Ratio (DSCR).

[0075] In particular, the economic analysis engine may include scenario analysis, sensitivity analysis, or the application of probabilistic variables to account for the uncertainty of input data, and the results are converted into quantitative economic indicators for investment decisions and policy verification. In other words, the economic analysis engine is a core analysis algorithm that integrates agriculture, power generation, and energy data to quantify profitability, stability, and efficiency.

[0076] Here, economic evaluation indicators refer to quantitative financial indicators produced as a result of the economic analysis engine. Specifically, they may include the Internal Rate of Return (IRR), Net Present Value (NPV), Payback Period, Debt Service Coverage Ratio (DSCR), Total Revenue, Break-Even Point (BEP), Carbon Reduction, and ESG contribution. These indicators are calculated by reflecting various factors such as power generation revenue, agricultural revenue, ESS operating revenue, investment amount, operating costs, and government subsidies, and are used as benchmark values ​​to evaluate economic feasibility from the perspectives of farmers, investors, and policy institutions.

[0077] Here, the target condition refers to the economic and technical target values ​​presented by farmers and investors in the matching algorithm. The farmer's target condition may include the minimum conservation rate of agricultural income, the allowable shading ratio, or long-term management stability indicators, while the investor's target condition may include the target internal rate of return (IRR), payback period, investment amount, risk tolerance, and preferred investment type (stable, balanced, high-return, etc.). In other words, the target condition serves as a constraint parameter that each entity must satisfy, and the matching candidate group is determined by comparing and contrasting it with economic evaluation indicators.

[0078] Here, the matching candidate pool refers to the optimal combinations among multiple farmland-investor pairs derived by comparing economic evaluation indicators with target conditions, where the comprehensive score for economic feasibility, risk balance, shading tolerance, and ESG contribution exceeds a critical threshold. This candidate pool is derived by finding the Pareto equilibrium point between power generation revenue and agricultural revenue using multi-objective optimization algorithms (such as NSGA-II and MOGA), and then evaluating the suitability between investor propensity and farm conditions using a K-Nearest Neighbor (KNN)-based similarity analysis algorithm. Consequently, the matching candidate pool constitutes a list of realistic and economically viable investment targets derived through AI-based data optimization.

[0079] According to one embodiment, the step (S10) of calculating an economic evaluation indicator may include: a step of predicting power generation, crop growth rate and yield using agricultural data including the coordinates, area, cultivated crop, soil characteristics, shading allowable range, temperature, solar radiation, wind speed, precipitation, System Marginal Price (SMP), Renewable Energy Certificate (REC) price, and charge / discharge efficiency and battery degradation coefficient of an energy storage device; a step of converting the predicted power generation, crop growth rate and yield into a profit item indicator based on unit price information; and a step of calculating an economic evaluation indicator by combining investment data including the investor's target internal rate of return (IRR), net present value (NPV), payback period, investment amount, risk tolerance, and investment type with the profit item indicator.

[0080] According to one embodiment, the step (S10) of calculating economic evaluation indicators as described above is characterized by predicting productivity and power generation efficiency at the actual farmland level by comprehensively considering physical variables related to agriculture, power generation, and energy storage systems (ESS).

[0081] According to one embodiment, the artificial intelligence model predicts changes in power generation based on weather conditions using a time series prediction model (LSTM, GRU, etc.) and estimates crop growth rates and expected yields using a regression-based machine learning model (Random Forest, XGBoost, etc.). In addition, it can simulate fluctuations in energy revenue resulting from ESS operation by reflecting the charge / discharge efficiency and battery degradation coefficient of the energy storage device.

[0082] The predicted power generation, crop growth rate, and yield are converted into monetary profit units based on unit price information for each item (e.g., electricity sales unit price, crop unit price, REC unit price, etc.). The converted profit item indicators are classified into power generation profit, agricultural profit, and ESS operation profit to constitute total profit. Subsequently, the above total profit and investment data (investment amount, target rate of return, payback period, risk tolerance, etc.) are input into an economic analysis engine, and financial-based economic evaluation indicators such as Internal Rate of Return (IRR), Net Present Value (NPV), and Payback Period are automatically calculated.

[0083] In other words, the economic feasibility evaluation indicator calculation step (S10) of the present invention enables realistic and highly reliable investment feasibility analysis beyond the existing single-profit-centered evaluation by implementing an AI-based economic feasibility analysis structure capable of integrating and calculating agricultural productivity, energy generation efficiency, and investors' financial goals within the same model.

[0084] The above economic analysis engine can calculate total revenue using the following mathematical formula 1, which is generated using solar power generation revenue, agricultural revenue, and energy storage system operation revenue.

[0085]

[0086] Here, R pv is solar power generation revenue, R agri is agricultural revenue, R ess is energy storage system operation revenue, Copex represents operating costs.

[0087] According to one embodiment, heterogeneous revenue sources of power generation, agriculture, and ESS are integrated into a single cash flow framework, and the total return R which becomes the input to the subsequent financial model (IRR / NPV) total Existing evaluations of agricultural solar power systems are generally limited to qualitative evaluations centered on generation revenue (LCOE, single revenue) or agricultural loss correction, but the present invention can calculate total revenue that can be directly input into a financial model by integrating agriculture, generation, and ESS into the same monetary unit.

[0088] In addition, by summing and correcting interaction items such as ESS scheduling and crop shading effects, overestimation or underestimation of a single sector is prevented, and total revenue becomes a common input for sensitivity and scenario analysis (SMP fluctuations, changes in shading rates, precipitation deviations, etc.) to quickly reflect policy and market changes.

[0089] According to one embodiment, the economic analysis engine can calculate the internal rate of return (IRR) based on the total revenue using the following mathematical formula 2.

[0090]

[0091] Here, CF t represents the net cash flow of the t-th year, C0 represents the initial investment cost, and n represents the total investment period.

[0092] According to one embodiment, in a business requiring agreement between farmers and investors by Equation 2, negotiation of terms can be facilitated through a single, intuitive indicator (IRR), and R integrated by Equation 1 total CF reflecting the OPEX and CAPEX structure t Since the IRR is calculated, the true profitability of the combined farming, power generation, and ESS model can be determined.

[0093] According to one embodiment, if SMP / REC fluctuations, yield deviations, and ESS efficiency degradation are reflected in ±Δ scenarios, it becomes easier to reduce variance and recognize downside risks compared to an IRR that considers only a single sector. In other words, it can be seen that even with conservative estimation, it is possible to determine early whether the target IRR is met.

[0094] The above economic analysis engine can calculate the above Net Present Value (NPV) using the following Equation 3.

[0095]

[0096] Here, r represents the discount rate.

[0097] The above multi-objective optimization model can calculate the Pareto optimal solution set using the following mathematical formula 4 with the power generation revenue (f1), agricultural revenue (f), and the investor's target rate of return (f3).

[0098] According to one embodiment, investment efficiency can be maximized by distinguishing cash flows of different sizes and durations among candidate sites with similar IRRs and selecting projects with a large NPV. Simulation results show that even with the same IRR, the NPV tends to be higher in cases that include ESS operation optimization, which is attributed to the effect of improved cash flow timing (night sales and peak substitution).

[0099] According to one embodiment, the artificial intelligence model predicts power generation using a time-series prediction-based deep learning model from agricultural data, and predicts crop growth rate and yield using a regression analysis-based machine learning model, and the economic analysis engine calculates total revenue by integrating the results predicted by the artificial intelligence model, and can automatically calculate financial indicators by inputting the total revenue into a financial model along with initial investment costs, operating costs, subsidies, and price fluctuation variables.

[0100] The artificial intelligence model according to the present invention goes beyond a simple prediction level and has a complex prediction structure that simultaneously reflects physical constraints and economic variables of power generation, agriculture, and energy storage systems (ESS). For example, power generation minimizes prediction errors through LSTM-based time series prediction and an ARIMA-XGBoost hybrid model, and agricultural productivity is calculated using a physical-data fusion method that combines the output indicators of a Random Forest model and a DSSAT / PROMET growth model.

[0101] In addition, the ESS operation schedule derives optimal profit scheduling based on Mixed Integer Programming (MILP) by simultaneously considering charge / discharge efficiency, SOC / SoH constraints, and degradation costs. The predicted results for power generation, agriculture, and ESS are integrated into a total profit model, and financial indicators such as IRR, NPV, Payback, and DSCR are automatically calculated.

[0102] Furthermore, it is differentiated from conventional simple regression-based economic analysis in that it includes a scenario-specific sensitivity analysis function based on SMP / REC price fluctuations and changes in subsidy policies, enabling the quantitative comparison of investment risks.

[0103] According to one embodiment, the step (S20) of deriving a plurality of matching candidate groups may include: a step of setting power generation revenue and agricultural revenue as objective functions and imposing an investor's target rate of return as a constraint, using an economic evaluation indicator as input; a step of generating a plurality of candidate solution sets by executing a multi-objective optimization model that considers the conditions of farmland and the conditions of the investor based on the objective function; a step of calculating a similarity score by analyzing the correlation between the conditions of farmland and the conditions of the investor included in the plurality of candidate solution sets; and a step of deriving a matching candidate group by selecting at least one combination in which the degree of matching fit is greater than or equal to a threshold based on the similarity score.

[0104] The present invention goes beyond simple condition filtering by additionally performing a K-Nearest Neighbor (KNN)-based similarity matching algorithm to calculate a quantified matching score by comparing actual investment performance data under similar past weather, crop, and soil conditions. In this process, only candidates with a matching score above a threshold are classified as the final candidate group, and the mutual compatibility between the candidate sites and investors is automatically evaluated. This enables quantitative matching based on profitability, risk, and suitability, rather than simple recommendations centered on data retrieval.

[0105]

[0106] Here, Satisfies.

[0107] According to one embodiment, according to Equation 4, it is classified as a valid candidate only when it secures a return rate greater than the investor's expected rate of return while the payback period is below a threshold value. This objective function structure is characterized by not relying on a single indicator (e.g., power generation or simple IRR) and simultaneously considering the trade-offs between power generation efficiency, agricultural sustainability, and investment profitability.

[0108] According to a simulation in one embodiment, f1, f2, and f3 were evaluated in parallel for a total of 120 candidate farmlands using multi-objective optimization (NSGA-II). As a result, compared to the conventional single IRR-centered analysis, an average IRR improvement of 8.4%, a 12.3% reduction in the agricultural profit loss rate, and a 2.1-fold increase in the number of effective solutions within the Pareto front were confirmed.

[0109] In other words, the multi-objective optimization based on Equation 4 of the present invention moves away from conventional single-indicator-centered design and can efficiently explore the balance point between maintaining the real income of farm households and guaranteeing profitability for investors. This is significant in that it resolves the issue of economic imbalance among stakeholders, which was the biggest problem in the expansion of agricultural solar power, based on data.

[0110] The step of deriving the above matching candidate group is performed based on K-Nearest Neighbor (KNN), and the above similarity score is calculated by the following mathematical formula 5.

[0111]

[0112] Here, x ij is the j-th attribute value of candidate site i, y j is the investor preference attribute value, S i represents the similarity score. This formula operates based on the K-Nearest Neighbor (KNN) algorithm and calculates similarity using the Euclidean distance in the attribute space between candidate sites and investors. Consequently, S i The closer the value is to 1, the higher the fit with the investor's preferred conditions.

[0113] According to the simulation results of one embodiment, when actually implemented, a similarity score was calculated based on m=10 major attributes (temperature, solar radiation, shading rate, IRR, Payback, risk score, etc.). As a result of the simulation, matching combinations with S ≥ 0.85 or higher accounted for approximately 18% of all candidates, and the standard deviation of the IRR for that group was 1.7%, and the Payback deviation was 0.9 years or less, showing a 2.3-fold improvement in the stability of profit prediction compared to the dissimilar matching group (S < 0.7). This demonstrates that the propensity fit between investment and farmland can be quantitatively reflected more than simple condition filtering methods (e.g., area, region-based).

[0114] In addition, this method can select the most practical and feasible candidates from the Pareto solution sets generated by NSGA-II, and it was confirmed to be excellent in terms of computational efficiency (reduction in the number of searches by 42%) and accuracy (actual profit prediction error within ±4%).

[0115] According to one embodiment, the method may include a step of correcting prediction accuracy by reusing actual operation result data included in the economic analysis report as training data for the artificial intelligence model.

[0116] According to one embodiment, a self-learning loop of the model can be formed by periodically collecting empirical data such as actual power generation, agricultural yield, and ESS operation efficiency included in the economic analysis report and re-feeding it as an input dataset of the artificial intelligence model.

[0117] At this time, the error (Δe) between past prediction results and actual operating results is measured, and the prediction accuracy can be continuously improved by applying a hybrid correction algorithm based on LSTM or XGBoost to items where the error exceeds a threshold (e.g., ±5%).

[0118] For example, in a demonstration in the Jeollabuk-do region, the power generation prediction error decreased from 7.8% to 3.2% and the crop growth rate prediction error decreased from 10.5% to 4.1% after one year of operation, resulting in an improvement of approximately 9.4%p in total revenue estimation accuracy. This data feedback structure has a technical effect of improving prediction stability and reliability in real time compared to a single learning model.

[0119] According to one embodiment, the method may include a step of verifying the CO2 reduction amount, RE100 contribution, and ESG evaluation indicators included in the economic analysis report by linking them with an external certification system.

[0120] According to one embodiment, the CO2 reduction amount, RE100 contribution, and ESG evaluation indicators included in the economic analysis report can be automatically verified through API integration with an external certification system or a blockchain-based verification module.

[0121] According to one embodiment, CO2 reduction is calculated by mapping power generation data with national carbon emission factors (based on 0.5 kg / kWh), and RE100 contribution is verified by linking with the Korea Power Exchange's REC database to determine the renewable energy usage ratio at the corporate level. Additionally, the ESG evaluation quantifies the effect of improving the ESG score through integration with external certification bodies (e.g., Carbon Neutrality Certification Institute, CDP, etc.), and the verification results are displayed as <Accredited Verification Completed> metadata within the report.

[0122] According to one embodiment, based on a simulation for policy agencies, when the interlocking structure is applied, the CO2 reduction verification time is reduced from an average of 5 days to 1 day, and the error in calculating the ESG score of RE100 participating companies converges to 0.2 points or less. Accordingly, the present invention provides a real-time verifiable ESG data infrastructure, thereby having the effect of simultaneously securing policy transparency and market credibility.

[0123] According to one embodiment, the step (S30) of automatically generating an economic analysis report may include: a step of classifying a report type based on an economic evaluation indicator corresponding to at least one matching candidate group; a step of automatically configuring customized items according to the report type, such that a report for farmers includes expected power generation revenue, agricultural revenue, and total revenue estimates, a report for investors includes internal rate of return (IRR), net present value (NPV), payback period, and sensitivity analysis results, and a report for policy agencies includes CO2 reduction amount, RE100 contribution, and ESG evaluation indicators; and a step of saving the automatically configured report in the form of an electronic document.

[0124] According to one embodiment, the method may include the steps of: calculating business enhancement items corresponding to farmers, investors, and policy institutions based on economic evaluation indicators included in an economic analysis report; calculating customized indicators for each business enhancement item; and generating an evaluation report that quantitatively evaluates the economic efficiency and social contribution of an agricultural solar power business by synthesizing the calculated customized indicators.

[0125] According to one embodiment, the economic evaluation indicators included in the economic analysis report are

[0126] It includes multiple key data such as IRR, NPV, Payback, agricultural loss rate, CO2 reduction amount, and RE100 contribution, and the computing device (100) can calculate each business enhancement item based on this.

[0127] For example, items for farm households include "agricultural income stabilization rate," "subsidized income ratio," and "farmland use efficiency"; items for investors include "risk-adjusted ROI," "sensitivity volatility index," and "market return rate"; and items for policy institutions may include social contribution indicators such as "carbon reduction rate," "contribution to ESG improvement," and "rural job creation rate."

[0128] According to one embodiment, each item is standardized into a score in the range of 0 to 1 by an artificial intelligence-based normalization model, and then a customized indicator is calculated by applying weights based on weighted regression or AHP (Analytic Hierarchy Process). The importance (weight) of this customized indicator is set differently depending on the user's interests (farmers, investors, policy institutions), for example, weights such as "production stability index (0.4)" for farmers, "ROI achievement probability (0.5)" for investors, and "ESG improvement effect (0.6)" for policy institutions may be applied.

[0129] According to one embodiment, the calculated customized indicators can be combined by an integrated analysis module and expressed as the Agri-Solar Impact Index (ASII). This index is a composite indicator that simultaneously considers economic efficiency and social contribution, and can generally be calculated using the following mathematical formula 6.

[0130]

[0131] Here, E eco is an economic efficiency indicator (e.g., IRR, NPV), E env is an environmental contribution indicator (e.g., CO2 reduction, RE100 participation rate), E soc represents social contribution indicators (e.g., farm income growth rate, job creation effect), and ω1, ω, and ω3 are weights set according to the user type. Unlike existing simple profitability-centered evaluation methods, this integrated index-based analysis provides the technical effect of quantitatively evaluating the comprehensive feasibility of a project by reflecting agricultural sustainability, carbon reduction, and social ripple effects.

[0132] FIG. 4 is a diagram illustrating the characteristics of generating an economic analysis report that includes different contents depending on the type according to one embodiment.

[0133] As illustrated in FIG. 4, the computing device (100) takes result data produced from the economic analysis engine and the matching algorithm module as input and automatically generates a customized report corresponding to each user type (farmer, investor, policy agency).

[0134] According to one embodiment, a computing device (100) first integrates key financial indicators (IRR, NPV, Payback, etc.) and environmental and policy indicators (CO2 reduction amount, RE100 contribution) based on the results of power generation, agriculture, and ESS revenue analysis, and transmits them to a report generation module. At this time, each data is managed in a JSON-based structure, and included items are automatically filtered according to the user type.

[0135] According to one embodiment, the computing device (100) selectively generates one of a report for farmers, an investor, and a policy agency based on the input analysis results.

[0136] Here, the report for farmers focuses on farm management stability and long-term income structure, and includes agricultural loss rates (%), conservation revenue (KRW), and total integrated revenue (KRW / year) for power generation, agriculture, and ESS. Additionally, it presents a long-term profit curve following payback in graph form, allowing farmers to intuitively verify the actual profit structure following the inflow of electricity sales revenue and subsidies. This report includes correlation graphs between shading rates and yields by crop and predicted agricultural loss rates, which serve as a basis for evaluating acceptability and sustainability in actual farm settings.

[0137] According to one embodiment, the investor report focuses on financial performance and risk analysis. The report includes key financial indicators such as IRR, NPV, and Payback, along with the results of a sensitivity analysis simulating the impact of weather fluctuations and SMP / REC price fluctuations on yield.

[0138] In addition, a graph showing the distribution of expected returns based on the investor's risk propensity (conservative, neutral, aggressive) and risk assessment indicators (σ, VAR, etc.) are visualized and presented. According to this embodiment, the investor report automatically displays whether investment conditions are met, and if the IRR or Payback criteria are not met, an automatic warning message is inserted, thereby preventing unnecessary negotiation or condition review costs in advance.

[0139] For example, in Example 1, when both the target IRR (12%) and Payback (7 years or less) of Investor B were met, a suitability judgment was made and the contract conditions (investment amount of 150 million won, distribution ratio of 60:40) were automatically calculated; conversely, in Example 2, an IRR of 11% and a Payback of 8.2 years were analyzed, and a suitability judgment was indicated. Such an automatic judgment function significantly improves the efficiency of data-driven investment decision-making.

[0140] According to one embodiment, a report for policy agencies comprehensively presents environmental and social ripple effects. In this embodiment, the system calculates CO2 reductions from power generation data and automatically calculates RE100 contributions and ESG improvement effects by linking REC information from the power exchange with the status of RE100 participation. The report includes a distribution map of changes in farm income by region, the contribution of renewable energy generation, and the results of an ROI (Return on Incentive) analysis relative to subsidy input, which are utilized as quantitative grounds for policy formulation.

[0141] For example, in the analysis of the Jeollabuk-do region in Example 3, an annual CO2 reduction of 480 tons, an improvement of +18% in average farm income, and a total ESS revenue of approximately 720 million won were reported, and the data was converted into an automated report for policy agencies and used to establish subsidy allocation policies and RE100 corporate linkage strategies.

[0142] According to one embodiment, the present invention implements a decision-oriented economic analysis report automation system through customized analysis and visualization for each user, rather than merely listing revenue data. In particular, by quantifying the balance of interests between farmers and investors, it is possible to create a transparent and verifiable data-based agri-photovoltaic ecosystem, and by providing quantitative ESG and carbon neutrality verification data to policy institutions in real time, it can simultaneously enhance the implementation capability of rural energy transition and policy credibility.

[0143] FIG. 5 is a diagram illustrating the process of generating a draft contract and electronically signing based on the calculation of contract conditions according to one embodiment.

[0144] According to one embodiment, the computing device (100) takes the result of a matching algorithm, information on the optimal farmland candidate site, and data on whether investor conditions are met as input, automatically calculates contract conditions, performs a contract simulation based thereon, automatically generates a draft contract, and performs the electronic signature and storage steps in a batch.

[0145] According to one embodiment, a computing device (100) receives farmer-investor matching pair data received from a matching algorithm module, evaluates the conditions of both parties, and automatically calculates key items such as investment amount, distribution rate, and contract period. At this time, the computing device (100) searches for a combination that simultaneously satisfies the investor's expected rate of return and the farmer's allowable loss rate based on the calculated IRR, NPV, and Payback.

[0146] For example, if the computing device (100) satisfies the conditions of IRR ≥ 12%, Payback ≤ 7 years, and α ≤ 10%, the system can automatically derive a basic contract proposal of "investment of 150 million won, distribution ratio of 60:40, and contract period of 7 years." This eliminates the repetitive calculation and condition adjustment procedures in the existing manual negotiation process, thereby reducing the average contract negotiation time by more than 70%.

[0147] According to one embodiment, a computing device (100) simulates a cash flow scenario during the investment and operation period based on conditions calculated in the contract simulation stage. In this embodiment, probabilistic financial forecasting can be performed by linking with an AI-based profit prediction model to reflect exogenous variables such as fluctuations in SMP and REC, weather conditions, and ESS efficiency degradation. Through this, the user can visually check the risk-return trade-off curve, the payback point distribution graph, etc. The results of this simulation are directly reflected in the contract creation stage to ensure the reasonableness of the contract conditions.

[0148] According to one embodiment, an automatic contract generation module automatically inserts calculated contract condition data into a standard contract template to immediately generate a draft contract in PDF or electronic document format. This template includes information on farmers, investors, project location, facility capacity, investment amount, distribution structure, duration, subsidy processing method, etc., and, if necessary, is linked with an electronic signature module via API to be converted into a legally binding electronic document.

[0149] In addition, the contract includes risk summary sentences and economic feasibility verification results automatically analyzed by AI, enabling contracting parties to sign with a clear understanding of the technical basis.

[0150] According to one embodiment, when a contract is established, the computing device (100) can generate a contract hash value in a blockchain-based storage module and record it as a tamper-proof electronic signature log. The signed contract and hash data are encrypted and stored in an IPFS (InterPlanetary File System)-based distributed storage or cloud storage (e.g., AWS S3). This prevents tampering with the contract and allows it to be used as legal evidence in the event of a subsequent dispute.

[0151] In particular, the present invention has the advantage of structurally resolving risks such as contract delays, document forgery, and discrepancies in conditions that occurred in existing agricultural solar power businesses by integrating contract automation and electronic authentication functions.

[0152] According to one embodiment, a computer program stored on a computer-readable recording medium can be provided to execute a method for generating an economic analysis report based on AI-based farm-investor automatic matching on a computer.

[0153] In addition, according to one embodiment, the method according to the various embodiments disclosed herein may be provided by being included in a computer program product. The computer program product may be traded between a seller and a buyer as a product. The computer program product may be distributed in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)), or distributed online (e.g., download or upload) through an application store (e.g., Play Store™) or directly between two user devices (e.g., smartphones). In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily created in a device-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0154] As described above, the present invention has been explained with reference to the embodiments illustrated in the drawings, but this is merely illustrative, and it should be understood that various modifications and equivalent alternative embodiments are possible based on the ordinary knowledge of the art to which the art belongs. Accordingly, the true technical scope of protection of the present invention is defined by the claims described below and should be determined based on the specific details of the invention described above.

Claims

Claim 1 A method performed on one or more processors of a computing device comprises: a step of inputting agricultural data and investment data into an artificial intelligence model to predict agricultural productivity data, and converting the predicted productivity data into profit units and applying them to an economic analysis engine to calculate an economic evaluation index; a step of deriving at least one matching candidate group by comparing the conditions of farmland and investors using the economic evaluation index; and a step of automatically generating an economic analysis report corresponding to the at least one matching candidate group; wherein the step of deriving the matching candidate group comprises: a step of setting power generation revenue and agricultural revenue as objective functions with the economic evaluation index as input and assigning the investor's target rate of return as a constraint; a step of generating a plurality of candidate solution sets by executing a multi-objective optimization model that considers the conditions of the farmland and the conditions of the investor based on the objective function; and a step of calculating a similarity score by analyzing the correlation between the conditions of the farmland and the conditions of the investor included in the plurality of candidate solution sets.The method comprises the step of selecting at least one combination whose matching fitness is greater than or equal to a threshold based on the similarity score and deriving it as a matching candidate group, wherein the processor performs a K-Nearest Neighbor (KNN)-based similarity matching algorithm to compare actual investment performance data of similar past weather, crop, and soil conditions to calculate a quantified matching score, classifies only candidates whose matching score is greater than or equal to a threshold as a final candidate group, evaluates the mutual compatibility between the candidate site and the investor, wherein the processor calculates a Pareto optimal solution set for power generation revenue, agricultural revenue, and the investor's target rate of return according to the multi-objective optimization model, classifies only those cases where the investment payback period is less than or equal to a threshold value while securing a rate of return greater than or equal to the investor's expected rate of return as a final candidate group, and is configured to calculate the similarity score according to the following mathematical formula when performing the K-Nearest Neighbor (KNN)-based similarity matching algorithm, [Mathematical Formula]; (here, x ij is the j-th attribute value of candidate site i, y j is the investor preference attribute value, S i represents the similarity score.), method for generating an economic analysis report based on AI-based automatic farm-investor matching. Claim 2 A method for generating an economic analysis report based on AI-based automatic farmer-investor matching, characterized in that, in claim 1, the artificial intelligence model predicts power generation using a time-series prediction-based deep learning model and predicts crop productivity indicators using a regression analysis-based machine learning model from the agricultural data, and the economic analysis engine calculates total revenue by integrating the results predicted by the artificial intelligence model and automatically calculates financial indicators by inputting the total revenue into a financial model along with initial investment costs, operating costs, subsidies, and price fluctuation variables. Claim 3 A method for generating an economic analysis report based on AI-based automatic matching between farmers and investors, characterized in that, in claim 1, the step of calculating the economic evaluation indicator comprises: a step of predicting power generation, crop growth rate, and yield using agricultural data including the coordinates, area, cultivated crop, soil characteristics, shading allowable range, temperature, solar radiation, wind speed, precipitation, System Marginal Price (SMP), Renewable Energy Certificate (REC) price, and charge / discharge efficiency and battery degradation coefficient of the energy storage device; a step of converting the predicted power generation, crop growth rate, and yield into a profit item indicator based on unit price information; and a step of calculating the economic evaluation indicator by combining investment data including the investor's target internal rate of return (IRR), net present value (NPV), payback period, investment amount, risk tolerance, and investment type with the profit item indicator. Claim 4 delete Claim 5 A method for generating an economic analysis report based on AI-based automatic matching between farmers and investors, characterized in that, in claim 1, the step of automatically generating the economic analysis report comprises: a step of classifying a report type based on an economic evaluation indicator corresponding to at least one matching candidate group; a step of automatically configuring customized items according to the report type, such that a report for farmers includes expected power generation revenue, agricultural revenue, and total revenue estimates; a report for investors includes internal rate of return (IRR), net present value (NPV), payback period, and sensitivity analysis results; and a report for policy agencies includes CO2 reduction amount, RE100 contribution, and ESG evaluation indicators; and a step of saving the automatically configured report in an electronic document format. Claim 6 A method for generating an economic analysis report based on AI-based automatic matching between farmers and investors, characterized in that, in claim 1, it comprises: a step of calculating business enhancement items corresponding to farmers, investors, and policy institutions based on economic evaluation indicators included in the economic analysis report; a step of calculating customized indicators for each of the business enhancement items; and a step of generating an evaluation report that quantitatively evaluates the economic efficiency and social contribution of an agricultural solar power business by synthesizing the calculated customized indicators. Claim 7 An apparatus for generating an economic analysis report based on AI-based farmer-investor automatic matching, comprising: a memory for storing one or more instructions in claim 1; and a processor for executing one or more instructions stored in the memory, wherein the processor performs the method of claim 1 by executing one or more instructions.

Images