Photovoltaic power generation system

Abstract

[Problem] Photovoltaic power generation systems installed on farmland require panel control for uniformly growing crops on the entirety of the farmland by suppressing non-uniform growth, but existing methods have proven to be insufficient. [Solution] This photovoltaic power generation system 100 comprises: a frame 130 for supporting a solar panel 110 in an angle adjustable manner; a drive device 120 for adjusting the angle; and a control device 200 for controlling the drive device 120. A growth homogenization information generation unit 210 generates angle information, for homogenizing the growth of crops, on the basis of a plurality of pieces of information such as the growth state of the crops, weather information, schedule information, and user designated information. Furthermore, highly accurate and adaptive panel control is achieved by accumulating history data and optimizing control based on history, or adopting a configuration including a training means for training a machine learning model on the basis of a teacher data set, and an angle information generation means for generating angle information by using the model.

Description

Solar power generation system

[0001] The present invention relates to a solar power generation system installed on farmland, and more particularly to a technology for controlling the angle of solar panels.

[0002] In recent years, agricultural solar power generation (also known as solar sharing), which utilizes agricultural land to generate solar power, has been attracting attention. In agricultural solar power generation, solar panels are installed on farmland, with the aim of achieving both solar power generation and crop cultivation.

[0003] In conventional agricultural solar power generation systems, a technology that automatically controls the angle of solar panels by tracking the movement of the sun is generally well known. Also, a technology that adjusts the angle of the panels to allow agricultural machinery such as work vehicles to pass through during agricultural work has also been proposed (see, for example, JP 2015-126683 A (Patent Document 1)).

[0004] On the other hand, it has been pointed out that the more densely solar panels are installed on farmland, the more uneven the growing environment for crops becomes, with problems such as a lack of sunlight for crops due to the shadows cast by the solar panels, rainwater concentrating in certain locations, and uneven wind and temperature environments.

[0005] To address these issues, a technology has recently been proposed that provides artificial light to crops by providing lighting units on the backside of solar panels. For example, Japanese Patent Publication No. 7584774 (Patent Document 2) discloses a plant cultivation system in which lighting units are provided on the backside of multiple solar panels that cover farmland like a roof, and the lighting units use the electricity generated by the panels to provide light to plants. The document also describes a schedule control function for controlling the panels, a function for specifying specific areas of farmland to adjust the light irradiation and panel angle, and a function for adjusting the panel angle based on weather sensors.

[0006] However, the above-mentioned conventional technologies did not provide sufficient specific disclosure of technology that integrates a variety of information about the entire farmland or specific areas (e.g., user-specified information, drone / sensor data for understanding growth conditions, detailed weather information, data on environmental conditions within the farmland, etc.) to finely control the angle of the solar panels and more precisely uniformize and optimize the growth environment for agricultural crops.

[0007] Therefore, there is a need for a new solar power generation system that can precisely and flexibly uniformly grow crops on farmland while maintaining solar power generation efficiency.

[0008] JP 2015-126683 A Patent No. 7584774 A

[0009] In agricultural solar power generation, the following problems arise when solar panels are densely placed on farmland.

[0010] First, the shadows cast by solar panels are unevenly distributed across farmland, resulting in insufficient sunlight in some areas, which can lead to uneven crop growth. This can result in unstable growth conditions across the entire farmland and reduced yields.

[0011] Second, at night or during rainfall, raindrops from horizontally arranged solar panels tend to fall in specific areas, providing excess moisture to those areas while leaving other areas deficient. This uneven moisture distribution can lead to uneven crop growth.

[0012] Third, the presence of solar panels and other structures can cause localized changes in the temperature environment of farmland, with daytime temperature increases and nighttime cooldowns occurring unevenly in specific areas, resulting in different effects on crops depending on the location.

[0013] Fourth, installed structures such as solar panels and mounting frames can block or deflect wind flow over farmland, reducing ventilation in some areas and potentially causing adverse effects on crop growth, such as the occurrence of disease.

[0014] As such, in agricultural solar power generation systems, multiple environmental factors such as solar radiation, raindrops, moisture distribution, temperature environment, and wind environment are intertwined, creating an uneven growing environment for crops. However, no conventional technology existed that could simultaneously consider these multiple environmental factors, perform an integrated analysis of them, and then flexibly and precisely control the angle of the solar panels to achieve uniform growth.

[0015] The present invention aims to solve these problems and to provide a solar power generation system that enables the uniform growth environment of crops by controlling the angle of solar panels on agricultural land.

[0016] In order to solve the above problems, the present invention has the following features: The present invention is a solar power generation system installed on farmland, comprising a plurality of solar panels, a mounting base that supports the solar panels so that the angles thereof can be adjusted, a drive device that adjusts the angles of the solar panels, a growth uniformity information generation unit that generates angle information for the solar panels to uniformize the growth of crops in the farmland based on information regarding the growth state of the crops in the farmland, and a control device that controls the drive device to adjust the angle of the solar panels based on the angle information generated by the growth uniformity information generation unit.

[0017] Preferably, the growth uniformity information generation unit generates solar panel angle information for providing solar radiation to an area within the farmland specified by a user, forming a shadow, concentrating raindrops, avoiding raindrop concentration, or improving ventilation.

[0018] Preferably, the growth uniformity information generation unit identifies areas where there is a bias in the growth state based on information about the crop growth status obtained from a drone or a sensor installed in the farmland, and generates solar panel angle information for providing sunlight to the area, forming a shadow, concentrating raindrops, avoiding raindrop concentration, or improving ventilation.

[0019] For example, the information on the growth status of the crop may include at least one of a vegetation index including NDVI (Normalized Difference Vegetation Index), soil moisture content, temperature, humidity, and solar radiation.

[0020] Preferably, the growth uniformity information generation unit generates solar panel angle information for providing solar radiation, forming shadows, concentrating raindrops, avoiding raindrop concentration, or improving ventilation for a specific farmland area based on schedule information relating to a predetermined time or time period.

[0021] Preferably, the growth uniformity information generating unit generates angle information that takes into consideration the safety of the solar panels and the stability of the farmland environment, based on meteorological information regarding wind speed, rainfall, snowfall, or frost.

[0022] Preferably, the growth equalization information generation unit generates solar panel angle information for equalizing the farmland environment based on bias in the falling position of raindrops, uneven temperature distribution within the farmland, or bias in wind flow.

[0023] Preferably, the growth uniformity information generation unit uses at least two or more pieces of information from the user's area designation information, information on the crop growth status, schedule information, weather information, and information on the environmental conditions within the farmland, weights them, and generates solar panel angle information based on the weights.

[0024] For example, weighting may be based on the reliability, accuracy, freshness or urgency of each piece of information.

[0025] Preferably, the growth uniformity information generation unit uses at least two or more of the following information: area designation information by the user, information on the crop growth status, schedule information, weather information, and information on the environmental conditions within the farmland, and generates solar panel angle information based on a predetermined priority order for these information.

[0026] Preferably, the growth equalization information generating unit generates the solar panel angle information by using a predetermined priority for each piece of information in addition to weighting the pieces of information.

[0027] Preferably, the growth uniformity information generating unit has a function of correlating the generated solar panel angle information with the crop growth status based on the angle information and accumulating the information as history information.

[0028] Preferably, the growth uniformity information generation unit generates angle information for the solar panel by selecting and reusing angle information that has produced good growth results in the past based on accumulated history information.

[0029] Preferably, the solar power generation system includes a learning means for learning a machine learning model based on training data including previously generated solar panel angle information and correspondence between the angle information and the crop growth status based on the angle information, an angle information generation means for generating solar panel angle information using the trained model to uniformize the growth of crops in farmland, and a growth uniformization information generation unit for issuing commands to a control device to control the angle of the solar panel based on the angle information generated by the angle information generation means.

[0030] Preferably, the learning means and the angle information generating means are provided in an external processing device separate from the solar power generation system, and the growth equalization information generating unit receives angle information transmitted from the external processing device and outputs it to the control device.

[0031] Preferably, the angle information generating means predicts a score regarding the uniformity of crop growth, and determines the angle information of the solar panel based on the score.

[0032] Preferably, the angle information generating means selects the angle that outputs the highest score.

[0033] Preferably, the training data or input information to the trained model includes NDVI (Normalized Difference Vegetation Index), soil moisture, temperature, humidity, or solar radiation.

[0034] Preferably, the angle information generating means outputs angle information of the solar panel directly in response to the input information.

[0035] Preferably, the learning means re-learns the machine learning model at regular intervals using the history information.

[0036] According to the present invention, a solar power generation system can be provided that achieves uniform crop growth across the entire farmland or a specific area by adjusting the angle of solar panels installed within the farmland according to the growth state of the crops, environmental conditions, etc.

[0037] According to claim 1, by configuring the angle control based on the growth state of the crop, it is possible to suppress growth deviations, standardize quality, and stabilize crop yields compared to conventional power generation optimization or simple time-of-day-based control.

[0038] According to claim 2, light, shadow, raindrops, and wind can be controlled for specific areas of farmland designated by the user, ensuring harmony with crops that require individual management and special farm work.

[0039] According to claims 3 and 4, deviations in growth status can be accurately detected based on objective growth status information collected by drones and sensors, enabling data-driven local control.

[0040] In claim 5, since control is performed according to schedule information, it is possible to perform environmental control that is consistent with agricultural work time management and worker intervention plans.

[0041] According to claim 6, a configuration is realized that can proactively respond to natural environmental factors such as wind, frost, and rain, thereby contributing to the safety of the panel and the preservation of the crop environment.

[0042] According to claim 7, it becomes possible to actively correct the imbalances in temperature, wind and water within the farmland, and the spatial uniformity of the growing environment is physically ensured.

[0043] In claims 8 and 9, a configuration in which a plurality of pieces of information are weighted and an integrated judgment is made makes it possible to dynamically optimize control according to the reliability, freshness, and urgency of each information source.

[0044] According to claims 10 and 11, priority control according to operational policies and urgency, or hybrid control of weight and priority, can be realized, enabling stable control even in situations where multiple objectives compete.

[0045] According to claim 12, by associating the control history with the growth results and managing the history, a foundation is established for utilizing past results in future control improvements.

[0046] According to claim 13, since "angle settings that have produced good results" can be reused based on historical information, optimization based on experience can be achieved step by step without using machine learning.

[0047] According to claim 14, by introducing a machine learning model that learns historical information and controls the angle, the complex relationship between numerous environmental factors and growth results can be modeled, enabling precise predictive control.

[0048] According to claim 15, by configuring the system so that the learning and inference processing can be outsourced to an external processing device, the processing load on the photovoltaic power generation system itself is reduced, and scalability and operational flexibility are greatly improved.

[0049] According to claims 16 and 17, by using a "growth uniformity score" as the output of the learning model and selecting an angle setting with a high score, selective control based on quantitative comparative evaluation becomes possible.

[0050] In claim 18, by using environmental indices such as NDVI, soil moisture, and temperature as input information, multidimensional parameters that actually affect crops are included in the learning target, thereby improving control accuracy.

[0051] According to claim 19, by configuring the AI ​​model to directly output angle information, end-to-end real-time control becomes possible, improving responsiveness and operational efficiency.

[0052] In claim 20, by configuring the model to be re-learned using historical information, continuous optimal control is realized that is adaptable to long-term changes such as seasonal fluctuations and soil changes.

[0053] This series of claims of the present invention realizes a multi-stage control system that intelligently performs physical control of light, shadow, water, and wind in agriculture based on user input, sensor data, history information, and AI learning. Unlike conventional simple angle control that follows power generation, this system provides control technology that can autonomously create an "optimal environment" according to the farmland environment and growth goals, and has innovative technical significance in the field of agriculture x renewable energy.

[0054] The objects, features, configurations, operations, and effects of the present invention and embodiments of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings.

[0055] FIG. 1 is a block diagram showing the configuration of a photovoltaic power generation system 100 according to the present invention. FIG. 2 is a flowchart showing the operation of an area designation information processing unit 211 according to the present invention. FIG. 3 is a flowchart showing the operation of a growth monitoring unit 214 according to the present invention. FIG. 4 is a flowchart showing the operation of a weather information processing unit 217 according to the present invention. FIG. 5 is a flowchart showing the operation of a reservation schedule management unit 220 according to the present invention. FIG. 6 is a flowchart showing the operation of an environment equalization processing unit 222 according to the present invention. FIG. 7 is a flowchart showing the angle information generation algorithm of the growth equalization information generation unit 210. FIG. 8 is a flowchart showing detailed processing of step S601 (data acquisition / preprocessing). FIG. 9 is a flowchart showing detailed processing of step S602 (dynamic weighting processing). FIG. 10 is a flowchart showing detailed processing of step S603 (competition processing and integration processing). FIG. 11 is a flowchart showing detailed processing of step S604 (angle trajectory calculation). FIG. 12 is a flowchart showing trajectory calculation processing for angle information. FIG. 13 is a flowchart showing the operation of a control device 200 that has received angle information. FIG. 14 is a flowchart showing the overall processing flow of a machine learning control process according to the second embodiment of the present invention. FIG. 15 is a flowchart of angle control using history information in a third embodiment of the present invention. (Drawing of the basic application) FIG. 16 is a block diagram showing the functional configuration of a solar power generation system 1 according to one embodiment of the present invention. (Drawing of the basic application) FIG. 17 is a flowchart showing the operation of the control device 2. (Drawing of the basic application) FIG. 18 is a flowchart showing the operation of the control device 2 based on the weather information detection unit 5. (Drawing of the basic application) FIG. 19 is a plan view showing an example of the arrangement of the solar power generation system 1 on farmland. (Drawing of the basic application) FIG. 20 is a plan view and a side view showing the state of an agricultural work vehicle when working. (Drawing of the basic application) FIG. 21 is a perspective view showing the state of an agricultural work vehicle when working. (Drawing of the basic application) FIG. 22 is a diagram showing the angle of the cantilevered solar panel 4. (Drawing of the basic application) FIG. 23 is a diagram showing the angle of the cantilevered solar panel 4 during agricultural work and in each time period (tracking mode and full light mode).Figure 24(a) (Figure of the basic application) explains how shadows and light are created by adjusting the angle of the solar panel 4 in accordance with the growth of crops, and Figure 24(b) is a 3D simulation of how shadows are created in Figure 24. Figure 25 (Figure of the basic application) is a flowchart when each mode is integrated.

[0056] (First embodiment) The table of contents of the description of the first embodiment is as follows: 1. Overview of the overall configuration (see FIG. 1) 2. Detailed configuration of the power generation system 2-1. Types and characteristics of the solar panels 110 2-2. Structure and installation method of the mount 130 2-3. Types and operation methods of the drive units 120 3. Detailed configuration of the control system 3-1. Internal configuration of the control unit 200 3-2. Basic functions of the growth uniformization information generation unit 210 3-3. Individual explanations of the information provision configuration units (1) Area designation information processing unit 211 (see FIG. 2) (2) Growth monitoring unit 214 (see FIG. 3) (3) Weather information processing unit 217 (see FIG. 4) (4) Reservation schedule management unit 220 (see FIG. 5) (5) Environmental uniformization processing unit 222 (see FIG. 6) 4. Detailed explanation of the common data structure (see Tables 1 to 3) 5. 5. Angle information generation algorithm of the growth equalization information generation unit 210 (see FIG. 7) 5-1. Data acquisition and preprocessing (step S601) (see FIG. 8) 5-2. Dynamic weighting process (step S602) (see FIG. 9) 5-3. Competition process and integration process (step S603) (see FIG. 10, Tables 4 and 5) 5-4. Angle trajectory calculation (step S604) (see FIG. 11, Equations 1 to 7, Tables 6 to 9, FIG. 12, Equations 8 and 9) 6. Explanation of operation of the control device 200 (see FIG. 13, Table 10) 7. Explanation of detailed operation of the drive device 120 8. Explanation of functions of the monitoring device 400 and operation to respond to abnormalities 9. Notes on installation, maintenance and operation 10. Detailed configuration of the power system 11. Explanation of detailed algorithms of each part 11-1. 11-1. Detailed algorithm of the growth monitoring unit 214 (see Equation 10 and Tables 10 to 12) 11-2. Detailed algorithm of the weather information processing unit 217 (see Tables 13 and 14) 11-3. Detailed algorithm of the environmental equalization processing unit 222 (see Tables 15 and 16)

[0057] 1. Overview of Overall Configuration A first embodiment of the present invention will now be described with reference to Fig. 1. Fig. 1 is a block diagram showing the configuration of a solar power generation system 100 according to the present invention. The solar power generation system 100 is installed on farmland and includes a power generation system, a control system, a power management unit, and a monitoring and management system.

[0058] 2. Detailed Configuration of the Power Generation System 2-1. Types and Characteristics of Solar Panels 110 The power generation system includes a plurality of solar panels 110, a drive unit 120, and a mount 130.

[0059] The solar panel 110 used in this embodiment can be a panel using any type of solar cell, such as commonly used monocrystalline silicon, polycrystalline silicon, thin-film silicon, or compound solar cells (CIS, CIGS, CdTe, etc.). There are no particular limitations on the size or shape of the panel, and they may be selected appropriately taking into consideration factors such as the placement and installation density on farmland, light transmittance and impact on crops, etc.

[0060] In particular, in the current specific embodiment, the use of a bifacial solar panel is preferred. Such a bifacial panel increases power generation efficiency by capturing sunlight not only from the front side but also from the back side, and can increase total power generation by effectively utilizing reflected light from the ground and crop surfaces. Furthermore, if the back side is transparent or translucent, there is an advantage in that power generation efficiency can be maintained while ensuring appropriate solar radiation transmission to the farmland.

[0061] Furthermore, the solar power generation system 100 of the present invention can be expected to be applicable to solar panels using new technologies in the future (for example, organic panels, perovskite panels, high-efficiency compound semiconductor panels, etc.). In other words, the present invention is not limited to the type, material, or structure of the panel, and is designed to be flexibly adaptable to future technological developments and introductions.

[0062] 2-2. Structure and Installation Method of the Mounting Frame 130 The mounting frame 130 stably supports the solar panels 110 on farmland and allows the angle of the panels 110 to be adjusted, and its structure and type are not particularly limited. For example, a fixed-foundation structure may be adopted in which a concrete foundation is installed on the ground surface and the mounting frame 130 is anchored to the foundation. It is also preferable to adopt a pile-driving structure in which piles are driven into the ground to support the mounting frame so as not to interfere with the use of the farmland. In addition, it is also possible to adopt a freestanding structure in which the mounting frame 130 can be used simply by placing it on the farmland, or a movable caster-type structure, depending on the situation.

[0063] The solar panels 110 supported by the mounting frame 130 only need to be able to change their angle in at least one axis (e.g., east-west, north-south, or any tilt axis). If more precise adjustment is required, a structure with degrees of freedom in two axes (e.g., east-west and north-south) or three or more axes (e.g., east-west, north-south, and rotation) can also be adopted. The mounting frame configuration of the present invention is not limited to these numbers of axes or degrees of freedom, and can be appropriately selected depending on the precision of adjustment of the crop growth environment, the shape of the farmland, installation density, weather conditions, etc.

[0064] 2-3. Types of Drive Device 120 and Operation Methods As the drive device 120 for adjusting the angle of the solar panel 110, any type of actuator can be used, such as an electric actuator using an electric motor, or a hydraulic, pneumatic, or electromagnetic type. For example, a rotary actuator combining an electric motor with a reduction mechanism, or a direct-acting electric linear actuator can also be used. Alternatively, a hydraulic or pneumatic actuator may be used in consideration of weather resistance and maintainability on agricultural land. The configuration and type of the drive device are not particularly limited in the present invention, and can be flexibly selected depending on operational conditions such as the system scale, panel weight, and desired operating speed and accuracy.

[0065] Furthermore, when adjusting the angle of the solar panel 110, it is desirable that the mount 130 and the drive unit 120 have sufficient strength and durability against external loads such as wind and snow. For this reason, it is desirable that the mount 130 be equipped with a wind-resistant reinforcement structure, a locking mechanism for movable parts, or a safety mechanism that automatically fixes the panel in a safe position when a certain load is applied.

[0066] 3. Detailed Configuration of the Control System 3-1. Internal Configuration of the Control Device 200 The control system includes the control device 200 and a growth uniformization information generation unit 210.

[0067] The control device 200 is a device for controlling each drive device 120 based on angle control information received from the growth uniformity information generation unit 210, and includes, for example, a processor 201 for performing arithmetic processing, a memory 202 for storing various data and programs, and a communication interface 203 for performing data communication with each unit including the drive device 120. It may also have an input / output interface 204 for linking with a display unit 212 and an operation input unit 213.

[0068] A general microcontroller, microprocessor, or a dedicated control IC or SoC (System-on-Chip) can be used as the processor 201. These processors read data obtained from various sensors, history information, user-specified information, etc. from the memory 202, and perform arithmetic processing to analyze and process them in an integrated manner and generate specific control signals for each drive device 120.

[0069] An appropriate combination of non-volatile memory (for example, flash memory or SSD) and volatile memory (for example, DRAM or SRAM) can be used as the memory 202. For example, the non-volatile memory can be configured to store control programs and setting parameters, while the volatile memory can be configured to store data that is temporarily used in arithmetic processing.

[0070] The communication interface 203 can be either a wired communication interface (e.g., Ethernet (registered trademark), CAN, RS-485, etc.) or a wireless communication interface (e.g., Wi-Fi (registered trademark), Bluetooth (registered trademark), LoRa (registered trademark), LTE (registered trademark), etc.) The control device 200 communicates data with each component, such as the drive device 120, the growth uniformity information generation unit 210, and the monitoring device 400, via the communication interface 203, and is configured to send and receive control information in real time.

[0071] The input / output interface 204 is for exchanging data with the display unit 212 and operation input unit 213, and can use, for example, HDMI (registered trademark), USB (registered trademark), or a dedicated interface. Via the input / output interface 204, the user can input setting information, set or change a control schedule, and display or check the operating status of the system.

[0072] Furthermore, to ensure stable operation, the control device 200 preferably includes a power supply unit 205. The power supply unit 205 preferably includes a DC / DC converter or an AC / DC converter that converts DC or AC power supplied from the power storage unit 320 or the grid power into a stabilized DC voltage required within the control device. If necessary, a battery backup or an uninterruptible power supply (UPS) can also be provided to maintain operation for a certain period of time even during a power outage.

[0073] 3-2. Basic Functions of the Growth Uniformization Information Generator 210 The growth uniformization information generator 210 has the function of generating angle information that indicates how the angle of each solar panel 110 should be adjusted, with the aim of uniforming the growth of crops on farmland. This angle information is generated based on a variety of information, such as the environmental conditions of the farmland, the growth state, and conditions specified by the user. Specifically, the following various types of information are analyzed in an integrated manner to determine angle control information for each panel 110:

[0074] First, it uses area designation information provided by the area designation information processing unit 211. Specifically, the user designates a specific area within the farmland, and the designated area information is reflected in the generation of angle information to create specific light irradiation and shade conditions for that area. This allows for precise control of the local amount of sunlight and shade on the crops, achieving a uniform growth environment locally.

[0075] Second, it uses information on the growth status of farmland provided by the growth monitoring unit 214. Specifically, it evaluates the growth status and identifies areas where growth is poor by analyzing aerial image data taken by the drone 215 and environmental data such as temperature, humidity, and soil moisture acquired by the environmental sensor 216. It then generates angle information for the solar panels so that the appropriate amount of light and shadows can be formed in the identified areas.

[0076] Third, it utilizes real-time weather observation data and external weather forecast information provided by the weather information processing unit 217. Specifically, it detects and predicts weather conditions that may have a negative impact on the crop growth environment in advance, based on wind speed, rainfall, snowfall, and temperature data observed locally and forecast information obtained from external services, and generates preventative and appropriate panel angle information.

[0077] Fourth, the reservation schedule management unit 220 uses schedule information provided through the storage unit 221. Specifically, predetermined panel angle information is generated based on specific dates and times set by the user. This allows stable angle control in accordance with operational plans such as regular farm work and harvest schedules.

[0078] Fifth, it uses environmental condition data provided by the environmental equalization processing unit 222. Specifically, it detects and evaluates environmental deviations within the farmland, such as the location of concentrated raindrop fall, temperature variations, and biased wind flow, and generates optimal panel angle information to mitigate these.

[0079] The information obtained from the above components (211, 214, 217, 220, 222) is processed in an integrated and dynamic manner by the growth equalization information generating unit 210 and provided to the control device 200 as optimal angle information.

[0080] As will be explained in the second embodiment, the growth uniformity information generation unit 210 also has a machine learning process that accumulates data on angle control performed in the past and the resulting crop growth conditions as history, and automatically optimizes angle control from the next time onwards based on the history data.

[0081] A detailed algorithm for processing these various pieces of information in an integrated manner, dynamically weighting each piece of information according to its importance and application situation, and then generating the final angle information for the solar panel 110 will be described later.

[0082] 3-3. Individual Description of Information Providing Components In the following, to provide a detailed description of angle information generation by the growth equalization information generating unit 210, the processing content of each component, namely, the area designation information processing unit 211, the growth monitoring unit 214, the weather information processing unit 217, the reservation schedule management unit 220, and the environment equalization processing unit 222, will be described individually. After that, the data structure of the angle information commonly provided by each component will be shown in detail, followed by a detailed description of the specific integration algorithm of the growth equalization information generating unit 210 that processes the information in an integrated manner, and then a specific calculation method for determining the angle.

[0083] (1) Area Designation Information Processing Unit 211 The area designation information processing unit 211 has a function of generating and processing information for realizing panel angle control that intentionally casts light or casts shadows on a specific area designated by a user within farmland. This area designation is effective when you want to promote or suppress the growth of a specific crop in an area within farmland, or when you want to prevent raindrops from concentrating on a specific area.

[0084] Specifically, the area designation information processing unit 211 acquires information about a specific area within farmland designated by the user using the display unit 212 and the operation input unit 213. For example, the user can designate an area of ​​a rectangle, polygon, or any other shape by touching or using the mouse on the map of farmland displayed on the display unit 212. The user can also set a purpose for a specific crop or farmland area on the display screen (for example, to provide more light, reduce solar radiation, prevent raindrops from falling, etc.).

[0085] The area designation information processing unit 211 analyzes the area designation information thus acquired from the user, generates detailed information such as the shape, area, and position information of the area, and transmits this information to the growth equalization information generation unit 210. Based on this area designation information, the growth equalization information generation unit 210 generates specific angle information for each solar panel 110 that will achieve optimal light irradiation and shadow formation for the designated area.

[0086] The area designation information processing unit 211 also has a function of saving the designated area information in the storage unit 221 (the storage unit 221 is also used as a common storage unit; the same applies below). This saved area information can be used repeatedly as needed, and can be easily reused or modified according to the growth stage of the crops and seasonal fluctuations. It is also accumulated as historical data and used for future growth management and optimization processing using machine learning.

[0087] 2 is a flowchart showing the operation of the area designation information processing unit 211 in the present invention. First, a map screen of the entire farmland is displayed on the display unit 212 (step S101). This map screen can be generated using, for example, aerial images taken by a drone, GIS (geographical information system) data, commercially available satellite image data, publicly provided map data, or any simplified farmland layout diagram.

[0088] Next, the user performs an operation to designate a specific area using the operation input unit 213 on the farmland map displayed on the display unit 212 (step S102). The designation operation is performed by touching the screen or using the mouse to surround a rectangle, polygon, or any other shape of area.

[0089] Based on the user's input, the area designation information processing unit 211 extracts detailed data such as the position information, area shape information (vertex coordinates, etc.), and area of ​​the designated area, and stores the data in the storage unit 221 (step S103).

[0090] Furthermore, the user sets a specific control purpose (for example, increasing light irradiation, creating a shadow, avoiding raindrops, adjusting the wind environment, etc.) for each specified area (step S104). This setting information is also stored in the storage unit 221.

[0091] The region designation information processing unit 211 formats the region designation data and control objective information stored in the storage unit 221 according to a common data structure described later, and provides the formatted data to the growth equalization information generating unit 210 (step S105).

[0092] Thereafter, the area designation data provided by the growth equalization information generation unit 210 is integrated with information from each of the other components and is used to generate the final angle control information for the solar panel 110.

[0093] (2) Growth monitoring unit 214 The growth monitoring unit 214 has a function of collecting information for grasping and evaluating the growth state of agricultural crops in the entire farmland or in a specific area, and providing the information to the growth uniformity information generation unit 210. This growth monitoring unit 214 includes a drone 215 and an environmental sensor 216.

[0094] The drone 215 flies over the farmland periodically or as needed, acquiring wide-area image data of the farmland using onboard imaging equipment (e.g., high-resolution camera, multispectral camera, infrared camera, etc.). The acquired image data is used to identify areas with good and poor growth conditions through analysis using vegetation indices (e.g., NDVI (Normalized Difference Vegetation Index)) or image processing using machine learning. This enables rapid and accurate identification of areas with slow growth or abnormalities.

[0095] The environmental sensors 216 are a group of sensors installed in various locations on the farmland, including, for example, temperature sensors, humidity sensors, soil moisture sensors, and solar radiation sensors. These sensors measure environmental data within the farmland in real time or periodically, and transmit the measurement data to the control device 200. This environmental data is analyzed in conjunction with image information obtained from the drone, and is used to evaluate the growth conditions in more detail and precision.

[0096] The growth monitoring unit 214 analyzes the acquired image data and environmental data to generate specific evaluation indices and status maps relating to the growth status. These evaluation indices and status maps are input to the growth uniformity information generation unit 210 and used to generate angle control information for adjusting the light irradiation and shadows in a specific area.

[0097] Furthermore, various data collected by the growth monitoring unit 214 is stored as history data in the storage unit 221 and used for subsequent analysis and machine learning processing. This makes it possible to realize predictions and control optimization based on past growth trends.

[0098] 3 is a flowchart showing the operation of the growth monitoring unit 214 in the present invention. First, the growth monitoring unit 214 acquires aerial image data of farmland using the drone 215 (step S201). This aerial image data can be a general color image, an infrared image, a multispectral image, or the like.

[0099] Next, environmental data is acquired from the environmental sensors 216 installed in the farmland (step S202). The data from the environmental sensors includes numerical data such as temperature, humidity, soil moisture, and solar radiation.

[0100] Based on the acquired aerial image data and environmental sensor data, an analysis process is performed to evaluate the growth status of crops in each area (step S203). Specifically, a vegetation index (e.g., NDVI) is calculated from the aerial image data, and the growth status of crops is determined based on the calculated value. Details of the specific vegetation index calculation method and threshold setting method will be described later. Furthermore, by analyzing each environmental sensor data, it is evaluated whether environmental conditions suitable for growth are in place.

[0101] Next, based on the above analysis process, areas where the growth conditions are unbalanced, such as areas where the growth conditions are poor or areas where the growth conditions are too good, are identified (step S204). Data such as the location and range of these areas, the type of imbalance (e.g., undergrowth, overgrowth, etc.), and the severity are generated and stored in the storage unit 221.

[0102] Finally, the growth monitoring unit 214 sets a control objective for each identified area to improve the unevenness of the growth conditions (for example, increasing light exposure in areas with poor growth, creating shade in areas with too good growth, or avoiding raindrops, adjusting the wind environment, etc.), formats the information according to the common data structure described below, and sends it to the growth uniformity information generation unit 210 (step S205).

[0103] As a result, the growth uniformity information generation unit 210 integrates the information obtained from the growth monitoring unit 214 with information from other components and generates angle control information for the solar panel 110 that is optimal for uniforming the crop growth conditions across the entire farmland.

[0104] (3) Weather Information Processing Unit 217 The weather information processing unit 217 has the function of grasping the weather conditions of the farmland and its surrounding area in real time, acquiring short-term and long-term weather forecast information, and reflecting this information in panel angle control to equalize the growing conditions of crops in the farmland. This weather information processing unit 217 includes a weather sensor 218 and a communication unit 219.

[0105] The weather sensor 218 is placed in or near the farmland and is composed of a group of sensors including, for example, a wind speed sensor, a wind direction sensor, a temperature sensor, a humidity sensor, a rainfall sensor, a solar radiation sensor, a snow accumulation sensor, etc. These sensors measure the local weather conditions in real time and transmit the data to the control device 200, making it possible to accurately grasp the weather conditions at the site.

[0106] The communication unit 219 has a function of communicating with external weather information services (for example, the Japan Meteorological Agency, private weather information providers, or local weather observation networks for each region) via the Internet or a wireless network. This makes it possible to obtain, in real time, forecast information on weather events such as rainfall, snowfall, frost, and strong winds, as well as forecast information on extreme weather conditions that may affect agricultural work.

[0107] The weather information processing unit 217 processes the real-time observation data and forecast data in an integrated manner, and provides information for maintaining optimal environmental conditions for each panel placement area within the farmland to the growth uniformity information generating unit 210. In particular, it can generate angle information for shifting the panels to a safe position during strong winds, and provide panel angle information that prevents raindrops from concentrating during rainfall.

[0108] Furthermore, the real-time and forecasted weather data collected by the weather information processing unit 217 is stored as historical information in the storage unit 221 and is used for subsequent machine learning analysis processing. This enables more precise and optimal angle control based on data on crop growth conditions under past weather conditions.

[0109] 4 is a flowchart showing the operation of the weather information processing unit 217 in the present invention. First, the weather information processing unit 217 acquires real-time weather data using the weather sensors 218 installed in the farmland (step S301). The data acquired from the weather sensors 218 includes, for example, wind speed, wind direction, temperature, humidity, rainfall, solar radiation, and snowfall.

[0110] Next, short-term and medium-term weather forecast information is acquired from an external weather forecast service (such as a forecast from the Japan Meteorological Agency or a private weather forecast service) via the communication unit 219 (step S302). The forecast information includes, for example, rainfall, snowfall, frost, strong winds, and temperature change forecasts.

[0111] Next, the system analyzes real-time observation data (step S301) and externally acquired forecast information (step S302) to detect weather conditions that may adversely affect growth in the farmland (step S303), such as forecasts of strong winds exceeding thresholds, forecasts of rainfall or snowfall exceeding a certain amount, and extreme temperature conditions.

[0112] Next, based on the detected weather conditions, an impact assessment is carried out for each specific area within the farmland, and the specific locations and areas where problems may occur are identified (step S304). For example, rainfall concentration in a specific area, areas at risk of wind damage, areas at risk of frost or low temperature damage, etc. are mapped.

[0113] Finally, for each identified area, a specific control objective (e.g., adjusting the panel angle to protect against strong winds, adjusting the angle to prevent concentrated rainfall, adjusting solar radiation to reduce frost damage, etc.) is set to reduce or avoid the effects of weather (step S305). The set area and control objective are formatted according to a common data structure described later and transmitted to the growth uniformity information generator 210.

[0114] (4) Reservation Schedule Management Unit 220 The reservation schedule management unit 220 has a function of managing information for automatically controlling the angle of the solar panel 110 based on a predetermined schedule and providing the information to the growth uniformity information generation unit 210. This reservation schedule management unit 220 includes a storage unit 221 that stores various types of schedule information set by the user.

[0115] The user can use the display unit 212 and operation input unit 213 to set the desired angle and control policy of the solar panel 110 for a specific date, time, or time period. For example, the user can freely set a schedule to limit direct sunlight on crops during a specific time period, or to temporarily change the angle of the panel for harvesting or farm work. This setting information is stored in the memory unit 221 and can be modified or updated at any time.

[0116] The reservation schedule management unit 220 automatically notifies the growth uniformity information generation unit 210 of the relevant panel angle information as the set time approaches based on the set schedule. Based on this notification, the control device 200 controls the drive unit 120 to adjust each solar panel 110 to a specified angle. This makes it possible to avoid interference with specific agricultural work and to stabilize regular growth management and environmental management.

[0117] Furthermore, the information managed by the reservation schedule management unit 220 may be integrated with other real-time information (for example, weather conditions and growth monitoring information) to dynamically determine the optimal control policy. Specifically, when there is a conflict with other input information, the weighting of various pieces of information is adjusted according to the priority specified by the user, thereby achieving optimal control.

[0118] In addition, the results of the schedule control used by this reservation schedule management unit 220 and the resulting crop growth conditions are stored as historical data in the memory unit 221, and can therefore also be used for machine learning optimization of future panel angle control.

[0119] 5 is a flowchart showing the operation of the reservation schedule management unit 220 in the present invention. First, the user inputs schedule information (date and time, target area, angle, etc.) related to the angle adjustment of the solar panels 110 that the user wants to perform for a specific area of ​​farmland using the display unit 212 and the operation input unit 213 (step S401). This input information is stored in the storage unit 221.

[0120] The reservation schedule management unit 220 continuously compares the current time with the schedule information stored in the storage unit 221 to determine whether the set time is approaching (step S402). This comparison is performed automatically at predetermined time intervals.

[0121] Next, when it is determined that the set control implementation time has arrived or is about to arrive (step S403), the reservation schedule management unit 220 formats the control objective information for the target area and panel angle according to that schedule in accordance with a common data structure (step S404).

[0122] The formatted control objective information is then sent to the growth uniformity information generating unit 210 (step S405), where it is processed in an integrated manner together with various other information (growth status information, weather information, etc.).

[0123] When the transmission is completed, the schedule processing ends, and monitoring for the next schedule processing continues (step S406).

[0124] (5) Environmental Equalization Processing Unit 222 The environmental equalization processing unit 222 has the function of generating angle information of the solar panels 110 in order to equalize the environmental conditions within the farmland (particularly raindrops, moisture distribution, temperature environment, and wind flow) and reduce localized biases in the crop growth environment.

[0125] Specifically, the environment equalization processing unit 222 uses data obtained from the weather information processing unit 217 and the environmental sensor 216 to detect, for example, concentrated raindrops falling on a specific location during rainfall, and determines the angle of the panels so that the raindrops are dispersed evenly across the entire farmland. Furthermore, if an area with uneven soil moisture is detected within the farmland, the angle of the panels is adjusted to optimally distribute the raindrops and water supply, thereby achieving a uniform moisture environment.

[0126] The environmental equalization processor 222 also analyzes the temperature distribution data of the farmland obtained from the temperature sensor, and can perform angle control to mitigate temperature unevenness by adjusting the amount of solar radiation and the position of the panel's shadow. In particular, to reduce excessive temperature rises during the day and localized chills at night, the angle of the panels can be dynamically adjusted to equalize the temperature environment within the farmland.

[0127] Furthermore, the system detects problems caused by mounting systems or panels blocking or deflecting wind flow on farmland, and generates panel angle information to appropriately guide wind flow within the field. This reduces the risk of crop disease outbreaks caused by poor ventilation in specific areas, making it possible to stabilize the growing environment.

[0128] The angle information generated in this manner is provided to the growth uniformity information generation unit 210, and is processed in an integrated manner with other elements (e.g., growth monitoring information and user area designation information) to optimize the overall environmental conditions of the farmland.

[0129] In addition, the results of various processes performed by the environment equalization processing unit 222 and the history of panel control that has been performed are stored in the memory unit 221 and are used for further refinement and optimization using machine learning in subsequent angle control.

[0130] 6 is a flowchart showing the operation of the environment equalization processor 222 in the present invention. The environment equalization processor 222 first acquires environmental data in real time from the environmental sensors 216 and meteorological sensors 218 installed in the farmland (step S501). The environmental data includes raindrop drop positions, soil moisture distribution, temperature distribution, wind speed and direction distribution, etc.

[0131] Next, the acquired environmental data is used to evaluate the non-uniformity of environmental conditions within the farmland (e.g., uneven raindrop distribution, uneven temperature, uneven wind, etc.), and the specific locations of non-uniform areas and the extent of their influence are identified (step S502). This evaluation is performed based on preset thresholds and reference values.

[0132] Furthermore, for each identified non-uniform region, a specific control objective is set to improve or eliminate the non-uniformity (step S503). For example, a panel angle that disperses raindrops in a region where raindrops are concentrated, an angle that creates appropriate shade in a region where the temperature is extremely high, an angle that improves airflow in a region where ventilation is poor, etc. are set.

[0133] Finally, the set area and control objective are formatted according to a common data structure and sent to the growth uniformity information generator 210 (step S504), whereby they are processed in an integrated manner with other information (growth monitoring information, weather information, reservation schedule information, etc.).

[0134] 4. Detailed Description of Common Data Structure The common data structure of the present invention is described in detail below. This data structure is used uniformly when providing information from the growth monitoring unit 214, meteorological information processing unit 217, reservation schedule management unit 220, environmental equalization processing unit 222, etc. to the growth equalization information generation unit 210.

[0135] (1) Overview of the common data structure This data structure clarifies the control purpose of each area and contains information for dynamically controlling the panel angle according to the time of day and time of day.

[0136] (2) Details of the common data structure The common data items are "control purpose ID, control purpose name, panel ID, area information, data acquisition date and time, start date and time, end date and time, angle activation information, priority, and remarks," as shown in the table below. (Common data structure item details)

[0137] (3) Specific examples of angle trajectory information Example of setting angle trajectory information (in the case of increasing solar radiation) The angle trajectory information is designed to dynamically change the angle over time, so that optimal environmental conditions can be maintained according to each control objective. The above common data structure is provided to the growth uniformity information generation unit 210 from each component, thereby achieving precise and efficient environmental control.

[0138] Below is a list of control objectives used in the present invention, which is provided to clarify the control objective IDs and control objective names used within the common data structure. This list of control objectives is used to clearly indicate the objectives of controls within the specification and to enable uniform operation.

[0139] 5. Angle Information Generation Algorithm of Growth Equalization Information Generation Unit 210 A detailed algorithm for generating angle information in the growth equalization information generation unit 210 of the present invention will be described below. The growth equalization information generation unit 210 comprehensively and dynamically analyzes a variety of information provided by each component unit (area designation information processing unit 211, growth monitoring unit 214, weather information processing unit 217, reservation schedule management unit 220, and environmental equalization processing unit 222), and generates detailed angle information for controlling the angle of each solar panel 110.

[0140] This algorithm consists of the following four main steps (see Figure 7). 5-1. Step S601: Data Acquisition and Preprocessing Information provided by each component (211, 214, 217, 220, 222) is acquired in a common data structure, and the suitability of the data structure and the completeness of the data are confirmed before time synchronization processing is performed. This completes the preprocessing required to enable efficient integration processing of data from different sources. Note that time synchronization processing involves aligning the timestamps of information acquired at different times by each data source to a predetermined common reference time. This enables precise and accurate data integration.

[0141] The detailed processing flow of step S601 (data acquisition / preprocessing) will be described below (see FIG. 8).

[0142] (Step S701: Data acquisition) The growth equalization information generation unit 210 acquires data provided from each component (area designation information processing unit 211, growth monitoring unit 214, weather information processing unit 217, reservation schedule management unit 220, environmental equalization processing unit 222) in a predetermined common data structure (control objective ID, control objective name, panel ID, area information, start date and time, end date and time, angle trajectory information, priority, remarks).

[0143] (Step S702: Check data compatibility) The compatibility and completeness of each acquired data is checked based on the common data structure. Specifically, it is verified that there are no omissions or inconsistencies in each data item. If any deficiencies or abnormalities are found, the component that provided the relevant data is notified and a request is made to resend accurate information.

[0144] (Step S703: Time synchronization process) Information acquired from each data source is generated at its own timing, and therefore has a different timestamp. For this reason, when integrating and analyzing this data, a time synchronization process is performed to align it to a common reference time (for example, every 10 minutes). This prepares all data for integrated analysis.

[0145] (Step S704: Storing in temporary storage area) Data that has undergone compatibility verification and time synchronization processing is stored in a temporary storage area (memory, temporary storage database, etc.) so that it can be used in the next processing step (step S602 onwards).

[0146] 5-2. Step S602: Dynamic weighting process Each piece of acquired information is dynamically evaluated for its reliability, accuracy, freshness, urgency, etc., and weighted. This dynamic weighting enables optimal control decisions to be made in a flexible manner in response to real-time changes in the environment and fluctuations in growth conditions.

[0147] The detailed processing flow of step S602 (dynamic weighting processing) will be described below (see FIG. 9).

[0148] (Step S801: Reading Data) Each piece of data acquired from each component (area designation, growth monitoring, weather information, reservation schedule, environmental equalization) is read from the temporary storage area (already stored in step S704).

[0149] (Step S802: Setting weighting evaluation criteria) Evaluation criteria and scoring criteria are set to evaluate the reliability, accuracy, freshness, urgency, etc. of each piece of data. For example, the following criteria are set: Reliability (accuracy of the data acquisition device, frequency of past errors, etc.) Freshness (time elapsed since data acquisition) Urgency (degree of weather abnormalities or poor growth that must be addressed in real time)

[0150] (Step S803: Dynamic weighting calculation) Each piece of data is scored according to the set evaluation criteria, and weighting is performed dynamically based on the results. For example, the following weighting logic is applied: High weighting is given to data that is highly reliable and fresh. Higher weighting is given to data that is highly urgent (e.g., abnormal weather or worsening growth conditions). Medium weighting is given to stable data such as regular operation plan schedules.

[0151] (Step S804: Storing Weighting Results) The data for which dynamic weighting has been completed is stored again in the temporary storage area, and prepared for use in the next step (competition processing and integration processing in step S603).

[0152] 5-3. Step S603: Conflict processing and integration processing When multiple different control objectives occur in the same area at the same time, the conflicts are resolved by referring to the priority information. After conflict processing, the data is integrated and reconstructed as a control objective map organized by area and time.

[0153] The detailed processing flow of step S603 (conflict processing and integration processing) will be described below (see FIG. 10).

[0154] (Step S901: Reading Data) The dynamically weighted data stored in the temporary storage area is read out. The data is acquired in a state where it is organized by area and by common reference time.

[0155] (Step S902: Detection of conflicting control objectives) Check whether multiple different control objectives exist in the same area and at the same time to detect conflicting situations. For example, this applies when multiple objectives exist simultaneously, such as "increasing solar radiation" and "dispersing raindrops."

[0156] (Step S903: Conflict processing (processing based on priority)) If a conflict is detected, the conflict is resolved based on the priority and weighted score assigned to each data. Control objectives with higher priority are selected as a priority, and those with lower priority are deleted or postponed for later processing. For example, conflict processing is performed according to the following rules: If priorities are clearly set, they are followed. If priorities are the same, the one with the higher weighted score is given priority.

[0157] (Step S904: Creating a post-integration control objective map) After the conflict processing is complete, the remaining optimal control objective data is organized by region and time, and a post-integration "control objective map" is generated. This map clearly indicates what control should be performed in each region according to the time, and is used in the next step (angular trajectory calculation in step S604). An example of a specific format for the control objective map is shown below. Below is a specific example of a time series of control objective maps.

[0158] The control objective map shown here is organized so that only one control objective remains for the same region and the same reference time through the conflict processing in step S603. Specifically, when multiple control objectives conflict at the same time in each region, only one control objective with the highest priority is selected based on the priority of each control objective and a dynamically calculated weighted score. As a result, the control objective map does not have multiple control objectives simultaneously in the same region and at the same time.

[0159] (Step S905: Storing Integrated Data in Temporary Storage Area) The created integrated control objective map is stored again in the temporary storage area, and made available for use in step S604.

[0160] 5-4. Step S604: Angular trajectory calculation Based on the created control objective map, the dynamic angular trajectory of the panel for each region and time is calculated while taking into account the sun's position calculation (azimuth angle and elevation angle). By doing so, the panel angle can be continuously changed to achieve a specified control objective (such as adjusting the amount of solar radiation or dispersing raindrops).

[0161] Through the above four steps, the growth uniformity information generating unit 210 efficiently integrates complex and diverse information, generates precise and optimal panel control angle information, and provides it to the control device 200.

[0162] The detailed processing flow of step S604 (angular trajectory calculation) will be described below (see FIG. 11).

[0163] (Step S1001: Reading out integrated data) The integrated control objective map (created in step S905) stored in the temporary storage area is read out.

[0164] (Step S1002: Obtaining solar position information) For each reference time, the azimuth and elevation angle of the sun are calculated based on the geographical position (latitude and longitude) of the farmland. This calculation is performed using known astronomical calculation formulas or existing software libraries.

[0165] (Step S1003: Calculate reference angle based on control objective) A reference panel angle is set according to the control objective (for example, increasing solar radiation, dispersing raindrops, improving the wind environment, etc.) written in the control objective map. For example, the following settings are made: In the case of increased solar radiation: The panel angle is set to be perpendicular or at an optimal angle to the sun so that the maximum amount of sunlight reaches the crops. In the case of decreased solar radiation: The panel is set at an angle that intentionally blocks sunlight to prevent excessive solar radiation from reaching the crops. Specifically, by tilting the panel with respect to the sunlight and creating a shadow by the panel in a specific area on the farmland, heat stress on the crops and uneven growth due to excessive solar radiation are prevented. In the case of raindrop dispersion: The angle is tilted so that raindrops do not concentrate in a specific location. In the case of improving the wind environment: The panel angle is adjusted to improve ventilation.

[0166] Solar panel mounting frames can be driven by one axis, two axes, or three axes, allowing for angle adjustment with different degrees of freedom. With each drive method, panel angles are usually adjusted based on the solar position (elevation angle α, azimuth angle β) when sufficient solar radiation is available, such as on clear days. On the other hand, when control is performed with the primary objective of environmental uniformity (dispersing raindrops and improving wind environments), such as during rainfall or strong winds, it is possible to directly set a specific angle according to the area or situation, regardless of the solar position. Below, we will explain the specific angle setting methods for each drive method. The symbols used in the formulas are as follows: α: solar elevation angle β: solar azimuth angle (south as the base, with east as positive and west as negative) γ,γ ｅｌ , γ ａｚ , γ ｒｏｌｌ : Additional adjustment angle preset for environmental equalization, etc.

[0167] (1) Single-axis drive system Single-axis drive allows rotation on the east-west axis or the north-south axis, and the panel angle is determined in the following way: East-west axis drive: Normally, the panel tilt angle θ is set based on the solar elevation angle α. If the objective is to equalize the environment, the set angle γ for each region is used directly, regardless of α. ・North-south axis drive: Normally, the panel angle θ is set based on the solar azimuth angle β. When the objective is to uniformize the environment, the set angle γ for each region is directly used without depending on β.

[0168] (2) In the case of two-axis drive system, the elevation axis (θ ｅｌ ) and the azimuthal axis (θ ａｚ ) and adjust the two axes. Normally, set as follows: Elevation direction: ・Azimuth direction: When the main purpose is to equalize the environment, it is not dependent on the solar position (α, β) but on the optimal angle (γ ｅｌ ,γ ａｚ ) to set the

[0169] (3) In the case of a three-axis drive system, the elevation axis (θ ｅｌ ), azimuthal axis (θ ａｚ ), tilt axis (roll axis: θ ｒｏｌｌ) and adjust the three axes. Normally, set as follows: Elevation angle: ・Azimuth direction: ・Tilt axis (roll axis) direction: If the main purpose is to equalize the environment, the optimum angle (γ ｅｌ , γ ａｚ , γ ｒｏｌｌ ) can be set directly.

[0170] For example, when implementing control aimed at creating a uniform environment during rainfall or strong winds, the panel angles are set directly by referencing the "angle setting table" and "control rules" that are preset for each area. Specifically, the following methods can be adopted.

[0171] For example, if a particular area of ​​farmland is experiencing a lack of rain, the angle of the panel is determined by referring to the following "angle setting table" in order to efficiently supply rain to that area when it rains.

[0172] In addition, in order to improve the wind environment during strong winds and prevent the occurrence of disease, the following control rules can be set.

[0173] In the examples shown in the "Angle Setting Table" and "Control Rules" above, for the sake of simplicity, only the setting angle (γ) for one-axis drive is listed, but similar settings are possible for two-axis and three-axis drive systems.

[0174] In the case of two-axis drive, it is possible to set independent angles in the elevation and azimuth directions for each region or panel. Specifically, an angle setting table can be created in the following format:

[0175] In the case of three-axis drive, the angle of the tilt axis (roll axis) can also be set. Specifically, it can be set in the following format.

[0176] In this way, as the number of drive axes increases, multi-axis settings are possible for more flexible and precise control, and by appropriately setting each axis, the farm environment can be optimized.

[0177] In this way, by defining an angle setting table and control rules in advance for each specific region or panel group, quick and accurate panel control becomes possible.

[0178] The growth uniformity information generator 210 references the angle setting table and control rules to determine the optimal panel angle in real time depending on the weather conditions in each area. This method makes it possible to uniformize the environmental conditions across the entire farmland and optimize the crop growth environment.

[0179] (Step S1004: Angular trajectory calculation) The sun position information calculated for each reference time (step S1002) is combined with the reference angle based on the control objective (step S1003) to calculate the specific trajectory of the panel angle for each time and region (step S1004). In step S1004, a trajectory calculation is performed to continuously change the angle information set for each region and each panel at the reference time between times for actual panel drive. However, when the goal is to standardize the environment, such as during rain or strong winds, control is adopted to maintain the specified fixed angle without interpolation between times.

[0180] The specific processing is performed in the following steps S1100 to S1105 (see FIG. 12): (Step S1101: Reading Reference Time Data) The angle information and control objective for each reference time corresponding to each panel ID are obtained from the control objective map.

[0181] (Step S1102: Processing branching based on control objective) Based on the acquired data, it is determined whether the control objective is a "objective based on the solar position" such as adjusting the amount of solar radiation, or whether the "primary objective is to equalize the environment, such as rainfall or strong winds." If the primary objective is adjusting the amount of solar radiation, proceed to step S1103 (angle trajectory interpolation calculation). If the primary objective is to equalize the environment, such as rainfall or strong winds, proceed to step S1104 (fixed angle setting processing).

[0182] (Step S1103: Angle trajectory interpolation calculation) In the case of adjusting the amount of solar radiation, etc., linear or nonlinear (spline) interpolation calculation is performed based on the acquired angle data to realize a smooth change in the panel angle between adjacent reference times. ｉ , t ｉ＋１ The set angles at θ ｉ， θ ｉ＋１ Then, at a certain time t (t ｉ <t<t ｉ＋１ The angle θ(t) of the point is calculated using the following formula (point gradient format): In the case of nonlinear (spline) interpolation, a known spline interpolation formula is used.

[0183] (Step S1104: Fixed angle setting process) When the purpose is to equalize the environment, such as rainfall or strong wind, the angle trajectory is set so that the set angle information is not interpolated between times but is kept constant (fixed) throughout the reference time period. Specifically, the angle is maintained as follows between the specified reference times.

[0184] (Step S1105: Generate angle trajectory data) The results calculated and set by the processing in step S1103 or S1104 are organized as time-series data and generated as continuous angle trajectory data for each panel ID. This data is used by the driver 120 to precisely control the panel.

[0185] The above process generates an angle trajectory of the panel optimized for each control objective, enabling accurate and effective growth uniformity control.

[0186] (Step S1005: Storing Angle Trajectory Data) The calculated angle trajectory data is stored in a temporary storage area and prepared for transmission to the control device 200.

[0187] 6. Description of Operation of Control Device 200 Next, a detailed description will be given of the operation of the control device 200. The control device 200 has the function of receiving time-series angle trajectory data generated by the growth uniformity information generation unit 210, and actually outputting an angle adjustment command to the drive device 120 of the solar panel 110, thereby accurately and safely controlling the panel to a predetermined angle.

[0188] Specific operations are performed by the following steps S1201 to S1205 (see FIG. 13). (Step S1201: Receiving angle trajectory data) The control device 200 receives the angle trajectory data for each panel output from the growth equalization information generation unit 210. This data is provided as time-series data including information such as the setting date and time and the setting angle for each region ID and panel ID.

[0189] (Step S1202: Command scheduling process) Based on the received angle trajectory data, a drive schedule is generated in a storage area inside the control device 200. The schedule is obtained by organizing the set angle information for each panel ID according to time for each control period (for example, one minute intervals).

[0190] For example, in the case of three axes, it is generated in the following format. If the set angle is fixed for purposes such as environmental uniformity, the same angle will be recorded repeatedly during a specified time period. In the case of single-axis and dual-axis measurements, the set angle is also set in advance using the date, time, and panel ID.

[0191] (Step S1203: Command Generation Process) For each control cycle, angle setting information corresponding to the current time is read from the schedule and generated as a control command to the drive unit 120. At this time, the angle information is encoded as an electrical signal (digital control signal) that can be received by the drive unit 120. For example, if the drive unit 120 uses an industrial communication protocol such as CAN communication or RS-485 communication, the command is encoded according to that protocol.

[0192] (Step S1204: Command transmission process) The generated command is transmitted to the drive units 120 via the communication unit. The transmission timing is in accordance with a predetermined control cycle, and the latest angle command is transmitted to each drive unit 120, for example, every minute. After the command is transmitted, the drive unit 120 receives the command and accurately operates the panel to the specified angle.

[0193] (Step S1205: Operation Check Process and Safety Process) After the drive unit 120 operates, the control unit 200 receives the operating status of the panel angle from sensor feedback such as an angle sensor or encoder. It checks whether the error from the set angle is within a certain tolerance (for example, within ±1°). If the error exceeds the tolerance or if an abnormality in the data from the angle sensor is detected, the control unit 200 immediately issues a command to a safe angle (for example, horizontal or vertical position) to ensure a safe state. Similarly, if an emergency condition such as strong winds or abnormal weather is detected, the control unit 200 switches the control command to a safe angle.

[0194] By periodically repeating the above operations, the control device 200 accurately and safely transmits the optimal angle information from the growth uniformity information generation unit 210 to the drive device 120, thereby achieving reliable angle adjustment of the panel 110. In this way, the entire system can uniformize the farmland environment and promote and stabilize crop growth.

[0195] 7. Detailed Operation of the Drive Device 120 The drive device 120 is a mechanism that actually adjusts the angle of the solar panel 110 based on an angle control command from the control device 200. This drive device 120 includes an actuator for changing the angle of the panel, and specifically, any type of actuator can be used, such as an electric motor, a hydraulic actuator, a pneumatic actuator, or a linear actuator.

[0196] The drive unit 120 normally operates according to the following procedure: (Step 1) Receive a communication command from the control unit 200. (Step 2) Based on the received command, obtain the specified angle information (set angle for each axis) and determine the target angle. (Step 3) Calculate the angle difference between the current panel angle and the target angle. (Step 4) Drive the actuator based on the calculated angle difference. The actuator is driven continuously or in stages until the set angle is reached. (Step 5) After changing the angle, confirm that the target angle has been reached using the actuator or angle sensor, and feed the result back to the control unit 200.

[0197] Furthermore, it is desirable that a safety function be built into the drive unit 120. For example, it is recommended that the drive unit 120 have a function that, when it detects a strong wind or an overload, receives an emergency stop signal from the control unit 200, stops the operation of the panel, and moves and fixes the panel to a safe position (a predefined position such as horizontal or vertical).

[0198] 8. Description of Functions of the Monitoring Device 400 and Operation to Respond to Abnormalities The monitoring and management system includes the monitoring device 400. The monitoring device 400 is a terminal that comprehensively displays and monitors the operating status of the photovoltaic power generation system 100, the farmland environment, the growth status, etc., and can be operated on-site or remotely.

[0199] The monitoring device 400 is a device that displays and monitors the operating status, growth environment, and various sensor data of the photovoltaic power generation system 100 in real time. This monitoring device 400 specifically displays the following data, enabling remote monitoring: - Display of current angle and historical angle of each panel 110 - Display of images captured by a drone and vegetation index maps (NDVI, etc.) - Display of real-time weather information (wind speed, temperature, rainfall, etc.) from weather sensors - Display of soil moisture, temperature, and humidity information from environmental sensors - Display of alerts and warning notifications when abnormalities are detected (for example, forecasts of strong winds or heavy rain, sensor abnormalities, etc.)

[0200] Furthermore, when an abnormal state is detected, the monitoring device 400 executes the following process: (Step 1) Detects the abnormal state (exceeding a threshold value in sensor data, communication abnormality, etc.) (Step 2) Notifies local and remote managers of the abnormal state by displaying the screen and providing an audio alert (Step 3) Sends an emergency control command (command to move panels to a safe position) to the control device 200 depending on the situation (Step 4) After recovery measures are taken, displays and records confirmation of return to normal state

[0201] 9. Installation and Maintenance Operational Precautions Proper installation and maintenance of various devices and sensors is important for the effective operation of this solar power generation system 100. The following guidelines are recommended. ・Installation of solar panels: It is recommended that panels be installed in a location that is less affected by shadows and obstacles, and that they be cleaned regularly to maintain power generation efficiency. ・Maintenance of the mounting and drive unit: The moving parts of the drive unit 120 should be inspected regularly, and lubricated and worn parts replaced. ・Operation and maintenance of drones and environmental sensors: Drones should be inspected and calibrated regularly to maintain the accuracy of aerial image data. It is recommended that environmental sensors be calibrated at least once a year to maintain the reliability of the acquired data. ・Regular inspection of the entire system: Sensors, communication equipment, and power management systems should be checked regularly to prevent communication outages and power abnormalities. ・Establishment of an emergency response system: It is recommended that administrators be trained on operating procedures and recovery methods to ensure rapid response in the event of an emergency.

[0202] 10. Detailed Configuration of the Power System The power management unit 300 includes a power conversion unit 310, a power storage unit 320, and a power supply unit 330. The power conversion unit 310 rectifies and converts the power generated by the solar panel 110 into a format usable by each device. The power storage unit 320 is a power storage unit that stores the converted power and is used for control at night or under various weather conditions. The power supply unit 330 stably supplies power to the control device 200, the drive unit 120, and other control system components. Note that, although it is assumed here that the power stored in the power storage unit 320 is used for control and drive, the power management unit 300 may also use power supplied from the grid to supply the power required for control and drive.

[0203] 11. Explanation of detailed algorithms of each part 11-1. Detailed algorithm of growth monitoring unit 214 The growth monitoring unit 214 has a function of collecting information for grasping and evaluating the growth state of agricultural crops in the entire farmland or in a specific area, and providing the information to the growth uniformity information generation unit 210. This growth monitoring unit 214 includes a drone 215 and an environmental sensor 216.

[0204] First, the growth monitoring unit 214 acquires aerial image data of the farmland using a drone 215. This aerial image data can be a general color image, an infrared image, a multispectral image, etc. In addition, environmental data such as temperature, humidity, soil moisture, and solar radiation is acquired from an environmental sensor 216 installed in the farmland.

[0205] Next, an analysis process is performed to evaluate the growth status of crops in each area based on the acquired aerial image data and environmental sensor data. Specifically, the Normalized Difference Vegetation Index (NDVI) is calculated from the aerial image data using the following formula (1): Here, NIR indicates the reflectance of near-infrared light, and RED indicates the reflectance of red light.

[0206] The calculated NDVI value is used to evaluate the growth condition, for example, using the criteria in Table 1 below.

[0207] The environmental data obtained from the environmental sensor 216 is evaluated according to the criteria shown in Table 2 below to determine the stress of the growth environment.

[0208] These image analysis results and environmental assessment results are integrated to identify areas where the growth conditions are unbalanced, such as areas where the growth conditions are poor, areas where the growth conditions are too good, areas where the growth conditions are appropriate but there is environmental stress, etc. These areas are stored in the memory unit 221 together with information such as their location, range, type, and severity.

[0209] In order to improve the imbalance in the growth conditions in each identified area, the growth monitoring unit 214 sets detailed control objectives such as the following, formats the information according to a common data structure, and transmits it to the growth uniformity information generation unit 210.

[0210] (1) Control of poor growth areas: In areas where growth is poor, the angle of the solar panels is adjusted to extend the duration of light exposure in order to increase the amount of sunlight reaching the crops. In addition, the supply of raindrops is secured as necessary, and improvements to environmental conditions (temperature and humidity) are also considered at the same time.

[0211] (2) Control of overgrowth areas: In areas where overgrowth is assessed, solar panels are adjusted to intentionally create shade to reduce the amount of sunlight reaching the crops. This adjustment reduces the risk of quality deterioration and lodging due to overgrowth.

[0212] (3) Control of areas with excess moisture In areas where the soil moisture is assessed to be excessive, the panel angle is adjusted to prevent raindrops from falling in a concentrated manner, thereby dispersing the rainwater evenly.

[0213] (4) Control in areas with a lack of moisture In areas where soil moisture is lacking, the angle of the panels is adjusted to guide and concentrate raindrops, thereby increasing the moisture supply.

[0214] (5) Control of areas where improvement of the wind environment is required In areas where ventilation is poor and the risk of disease occurrence is assessed to be high, the angle of the panels is adjusted to promote airflow and improve the wind environment.

[0215] (6) Control of areas where temperature improvement is required In areas where daytime temperatures rise excessively and crops are subjected to heat stress, the angle of the panels is adjusted to create appropriate shade, optimizing the temperature environment.

[0216] These control objectives and data for each region are structured in a common format, efficiently provided to the growth equalization information generating unit 210 (described later), and reflected in the final control algorithm.

[0217] 11-2. Detailed Algorithm of the Weather Information Processing Unit 217 The detailed algorithm of the weather information processing unit 217 is explained below. (1) Method of Obtaining Weather Data Weather sensors 218 are installed at multiple locations within the farmland and measure wind speed and direction, temperature, humidity, rainfall, solar radiation, snowfall, etc. in real time. The measurement frequency can be set to a fixed interval, such as every 5 minutes or 10 minutes. In addition, the communication unit 219 obtains the latest short-term and medium-term forecast data from the Japan Meteorological Agency and private weather forecast services via the Internet.

[0218] (2) Setting thresholds for evaluating weather conditions The acquired weather data is evaluated based on the following thresholds.

[0219] (3) Area-specific impact assessment algorithm: Weather sensor data and forecast data are used to assess the weather impact for each area within farmland. For example, areas of farmland at high risk of wind damage are identified based on wind direction and speed data. Rainfall and snowfall data are also used to predict moisture concentration and poor drainage in specific areas. These assessment results are linked to GIS coordinate data and displayed on a map, visually indicating specific areas.

[0220] (4) Method for Setting Control Objectives The following control objectives are set according to the evaluation results for each identified area. These control objectives are formatted according to a common data structure and sent to the growth equalization information generator 210 .

[0221] 11-3. Detailed Algorithm of the Environmental Equalization Processor 222 The detailed algorithm of the environmental equalization processor 222 is explained below. (1) Environmental Data Acquisition Method Environmental data is collected in real time from multiple locations within the farmland using the environmental sensor 216 and meteorological sensor 218. These sensors measure raindrop drop positions, soil moisture distribution, temperature distribution, and wind speed and direction distribution, and update the data periodically (for example, every 5 minutes).

[0222] (2) Evaluation criteria for heterogeneity Based on the collected environmental data, heterogeneity is evaluated according to the following criteria.

[0223] (3) Identifying specific heterogeneous areas Using heterogeneity data determined based on the evaluation criteria for each environmental element, area mapping is performed in conjunction with GIS coordinates. This mapping visually identifies the location and extent of specific problem areas.

[0224] (4) Specific Method of Panel Angle Adjustment In order to improve the non-uniformity in each identified region, the following specific panel angle adjustment control is carried out.

[0225] These control objective and area data are structured into a common format and sent to the growth equalization information generation unit 210 .

[0226] According to the first embodiment of the present invention, by adopting a new configuration in which the angle of the solar panels is dynamically controlled not only for the purpose of optimizing power generation but also for the purpose of uniform growth of agricultural crops, the following remarkable effects can be obtained.

[0227] (1) Mitigating spatial heterogeneity in agricultural land environments By understanding environmental differences and biases in growth conditions within agricultural land from multiple perspectives, such as weather sensors, environmental sensors, growth monitoring units (NDVI analysis), area designation information, and reservation schedule information, and then finely adjusting the angle of each panel accordingly, it is possible to correct localized excess or lack of solar radiation, poor ventilation, uneven moisture distribution, and temperature variations.

[0228] (2) Flexible angle control to meet various control objectives The angle control is flexibly configured to meet multiple objectives, such as "adjusting the amount of sunlight," "forming shadows," "dispersing raindrops," "improving the wind environment," and "aligning with work schedules." It is possible to generate angle trajectories optimized for specific objectives, enabling the realization of complex growth control that was difficult to achieve with conventional fixed control.

[0229] (3) Achieving local optimal control by specifying an area Users can pinpoint the amount of light exposure and shadows, avoid weather effects, and perform other operations on any area of ​​farmland specified by touch operation, enabling precise control according to the growth stage of the crop and differences in variety.

[0230] (4) Dynamic decision-making through real-time data integration Weather information, environmental conditions, growth conditions, user schedules, etc. are integrated into a common data structure, and a design is made to make comprehensive angle control decisions based on priorities and weighting, enabling dynamic decision-making according to the situation.

[0231] (5) Computationally consistent angular trajectory generation Depending on the drive method (single-axis, two-axis, or three-axis), the angle relative to the sun's position and environmental target is clearly defined using a formula, and interpolation or fixed control is also selectable, enabling precise control while maintaining high consistency between the control system and physical structure.

[0232] In this way, the first embodiment of the present invention redefines solar panels on farmland not as conventional power generation devices but as "environmental control devices" that adjust the growing environment of crops, and has the excellent effect of making a significant contribution to the advancement of agriculture and the uniformity of yields and quality through a system that integrates sensing, control, and scheduling.

[0233] (Second Embodiment) A second embodiment of the present invention will be described below with reference to FIG. 14. The table of contents of the description of the second embodiment is as follows. Note that the second embodiment utilizes the various sensors and configuration of the first embodiment as they are, and performs angle control by a machine learning model using collected data through communication with an external device. 1. Overview of the Second Embodiment 2. Detailed Configuration of Collected Data 2-1. Date and Time Data 2-2. Weather Sensor Data 2-3. Environmental Sensor Data 2-4. Growth Monitoring Data (NDVI) 2-5. Reservation Schedule Information 2-6. Area Designation Information (User-Designated Information) 2-7. Environmental Uniformization Processing Data 2-8. Actual Control Angle History Data 2-9. Growth Monitoring Result Data 3. Ranking Processing of Collected Data 3-1. NDVI Uniformity Evaluation (First Evaluation Index) 3-2. Growth Status Evaluation (Second Evaluation Index) 3-3. Overall Uniformity Evaluation (Ranking Processing) 3-4. Implementing ranking 4. Building a teacher dataset 4-1. Method for building a time-series teacher dataset 4-2. Configuration of a teacher dataset (1) Configuration of input data (2) Configuration of output data 4-3. Specific examples of teacher datasets 5. Learning a machine learning model using teacher data 5-1. Preprocessing of learning data 5-2. Selection of a machine learning model 5-3. Setting input and output of the learning model 5-4. Model learning process 5-5. Performance evaluation and optimization of the learning model 5-6. Storage and saving of trained model 6. Real-time prediction and angle control decision 6-1. Acquisition of real-time data 6-2. Generation and evaluation of angle candidates 6-3. Selection of optimal angle candidate 6-4. Execution of control using the selected angle 6-5. Collection of new data and addition to teacher data 7. Configuration for outsourcing machine learning processing to an external device 7-1. Contents of processing to be outsourced to an external device 7-2. Communication processing between the control device 200 and an external device 7-3. Advantages of external equipment consignment configuration

[0234] 1. Overview of the Second Embodiment The second embodiment of the present invention aims to continuously collect various data described in the first embodiment (weather information, growth monitoring information, reservation schedule information, user-specified information, environmental equalization processing data, angle control history, etc.), build a machine learning model for quantitatively evaluating the growth uniformity of crops based on the data, and optimally control the angles of the solar panels 110 in real time.

[0235] Specifically, the collected data is acquired over a fixed period (e.g., one week) in chronological order, overlapping on a daily basis, and each data set is evaluated through a ranking process to create a training data set. This training data set is used to train and optimize a machine learning model on an external dedicated device (e.g., a cloud server or a high-performance computer) to generate a model that can accurately predict crop growth uniformity.

[0236] FIG. 14 is a flowchart showing the overall processing flow of the machine learning control process according to the second embodiment of the present invention. As shown in this figure, in this embodiment, various sensor data, control history, growth monitoring results, etc. are first acquired over a fixed period (e.g., one week) in a manner similar to that of the first embodiment (step S1301). Based on the acquired data, a ranking process is performed to quantitatively evaluate growth uniformity (step S1302). A training dataset is constructed (step S1303), which inputs index data, angle history data, and monitoring data, including the evaluation results, and outputs an evaluation rank. The constructed training dataset is then preprocessed (e.g., normalized and encoded) (step S1304), an appropriate machine learning model (e.g., a neural network) is selected (step S1305), and a training process is performed (step S1306). The trained model is stored in an external device (step S1307). The control device 200 transmits the latest data acquired in real time (step S1308), and the external device predicts growth uniformity scores for multiple angle candidates (steps S1309-S1310). Based on the results, an optimal angle control plan is selected (S1311), and the control device 200 controls the panel angle (S1312). Finally, the growth results are monitored again and fed back as new training data (S1313). In this way, in this embodiment, the on-site control device 200 and the external learning and inference device work cooperatively to continuously and autonomously optimize growth uniformity. This will be explained in detail below.

[0237] 2. Detailed Configuration of Collected Data In the second embodiment of the present invention, the various data already described in the first embodiment are collected over a predetermined period (e.g., one or two weeks) to generate a database for constructing training data. The specific data to be collected is clearly defined below using the same terms as in the first embodiment.

[0238] 2-1. Date and time data This is information on the date and time when various data was acquired, and is used to manage the control history in chronological order.

[0239] 2-2. Weather Sensor Data This is real-time weather data acquired from the weather sensor 218 installed in the weather information processing unit 217, and includes, for example, the following: Wind speed (m / s) Wind direction (degrees) Temperature (°C) Humidity (%) Rainfall (mm / h) Solar radiation (W / m 2 ) Snow depth (cm) These data are obtained at specified intervals (for example, every 10 minutes).

[0240] 2-3. Environmental Sensor Data This data is acquired from the environmental sensor 216 installed in the growth monitoring unit 214, and specifically includes the following: Soil moisture content (volume water content %) Ground surface temperature (°C) Soil temperature (°C) Air temperature and humidity at each point within the farmland These data are also collected at predetermined intervals (for example, every 10 minutes).

[0241] 2-4. Growth monitoring data (NDVI) This is the vegetation index (NDVI) calculated by the growth monitoring unit 214 based on aerial image data from the drone 215. The NDVI value is calculated using the above-mentioned formula (1). The NDVI data is measured and analyzed, for example, once a day or every few days, and recorded as an index of the growth status of each area of ​​farmland.

[0242] 2-5. Reservation Schedule Information This is data stored in the storage unit 221 by the reservation schedule management unit 220, and includes preset angle information, date and time information, target area, etc. based on a schedule specified by the user.

[0243] 2-6. Area Designation Information (User Designation Information) This is area designation data acquired by the area designation information processing unit 211 from the user, and includes specific area coordinates (GIS coordinates), shape information (polygon, rectangle, etc.), and control objectives set by the user (for example, increased light irradiation, shadow formation, raindrop dispersion, wind environment adjustment, etc.).

[0244] 2-7. Environmental Equalization Processing Data Environmental data collected by the environmental equalization processing unit 222 and data for environmental equalization control, including the following: Raindrop concentration area information Temperature unevenness information Wind deviation information These are evaluated and recorded at predetermined time intervals.

[0245] 2-8. History data of actually controlled angles This is history data of angle control of the solar panel 110 actually performed by the control device 200 and the drive device 120 described in the first embodiment. This data specifically includes the following information: Date and time of control Panel identification information (panel ID) Angle setting information for one, two or three axes (elevation angle θ ｅｌ , azimuth θ ａｚ , inclination angle θ ｒｏｌｌ etc.) Actual operation completion angle in response to control command (sensor feedback value)

[0246] Growth monitoring result data This is the evaluation result of the growth status analyzed by the growth monitoring unit 214 based on the NDVI values ​​etc. collected in (4) above. Specifically, the NDVI values ​​are evaluated for each region, and the data includes region information classified as poor growth region, good growth region, excessive growth region, etc.

[0247] The above data is recorded in a set over a predetermined period (for example, one or two weeks) with each item associated with the data. Such data sets are then collected over a period of time and used as a training data set for the next step, the ranking process.

[0248] 3. Ranking Process of Collected Data In the second embodiment of the present invention, the data collected in the above section 1 is used to perform ranking process to evaluate the uniformity of the crop growth conditions associated with angle control of the solar panel 110. The ranking process is performed by using the NDVI value acquired by the growth monitoring unit 214 as the main index, and by also comprehensively evaluating data from the environmental sensor 216 and the weather sensor 218 as necessary.

[0249] The specific ranking process is performed, for example, in the following steps. 3-1. NDVI uniformity evaluation (first evaluation index) The collected NDVI data is used to evaluate the uniformity of growth. Specifically, the entire farmland is divided into multiple evaluation areas, the average NDVI value for each area is calculated, and the standard deviation σ of NDVI for all areas is calculated. The smaller the NDVI standard deviation σ, the more uniform the growth, and the larger the standard deviation σ, the more uneven the growth. The value of this standard deviation σ is used as the first evaluation index (NDVI uniformity index).

[0250] 3-2. Growth Status Evaluation (Second Evaluation Index) Furthermore, the growth monitoring unit 214 identifies areas where the NDVI value is below a certain threshold (poor growth areas) and areas where growth is excessive above a certain threshold, and calculates the area ratio of these areas. This area ratio is used as the second evaluation index (growth unevenness index). A specific example is shown below. The smaller the area ratio of the undergrown and overgrown regions, the higher the uniformity.

[0251] 3-3. Overall uniformity evaluation (ranking process) The first evaluation index (NDVI uniformity index: standard deviation σ) and the second evaluation index (growth unevenness index: area ratio) are evaluated comprehensively to calculate an overall growth uniformity evaluation score. For example, the final evaluation score is obtained using the following integrated evaluation formula: An example of evaluation is shown below.

[0252] 3-4. Ranking The calculated overall evaluation score will be used to rank the following: By ranking in this way, the growth uniformity of each data set collected for each period is quantitatively evaluated and organized into a form that can be used as training data.

[0253] By carrying out the above process at predetermined intervals (for example, every week or every two weeks), the collected data is subjected to a ranking process and is completed as training data.

[0254] 4. Construction of a Teacher Data Set 4-1. Method for constructing a time-series teacher data set In steps 1 and 2 above, each data set for a predetermined period (e.g., one week) is evaluated by a ranking process, and then, in the second embodiment of the present invention, a teacher data set is constructed using these data.

[0255] Specifically, by acquiring data consecutively with a one-day shift in the data acquisition period, each data set is generated so that it overlaps in chronological order. For example, a training data set can be constructed using the following method: - Data set 1: One week from June 1st to June 7th - Data set 2: One week from June 2nd to June 8th - Data set 3: One week from June 3rd to June 9th Collecting data from consecutive and overlapping periods in this way enables highly accurate learning in chronological order.

[0256] 4-2. Structure of the training dataset The training dataset is specifically defined by the following structure: (1) Structure of input data - Date and time data (start and end dates of the period) - Weather sensor data (wind speed, temperature, rainfall, etc.) - Environmental sensor data (soil moisture, temperature, etc.) - Growth monitoring data (NDVI value for each area) - Reservation schedule information (control schedule specified by the user) - Area designation information (area coordinates and control purpose specified by the user) - Environmental equalization processing data (raindrop concentration areas, temperature unevenness, wind deviation, etc.) - Actual controlled angle data (panel angle history during the control period) (2) Structure of output data - Overall evaluation score and rank (A to E) calculated by the ranking process

[0257] 4-3. Specific examples of training datasets Specific examples of training datasets are shown below.

[0258] By collecting and accumulating multiple such training datasets over successive and overlapping periods, a database that precisely captures the effects of angle control on crop growth conditions is completed. This database is then used as input and output data for subsequent machine learning model training.

[0259] Although a one-week period has been used as an example for convenience in this specification, in practice it is possible to select two weeks or another period setting, and the period setting is selected appropriately depending on the actual crop growth cycle and operating environment.

[0260] 5. Learning of a machine learning model using training data In the second embodiment of the present invention, a machine learning model is trained using the training data set constructed in the above section 3. Through this training, a model is constructed that can quantitatively evaluate and predict the uniformity of growth conditions from the collected various indices and angle control history.

[0261] Specifically, the machine learning model is trained using the following procedure: 5-1. Preprocessing of training data The input data extracted from the training dataset (date and time information, weather data, environmental data, NDVI data, reservation schedule information, area designation information, environmental uniformity data, and angle control history data) is converted and normalized into a format suitable for training. For example, numerical data is standardized (mean value 0, standard deviation 1), and categorical data is processed using one-hot encoding or similar.

[0262] 5-2. Selection of Machine Learning Model The machine learning algorithm used in this embodiment can be, for example, the following methods: Deep Neural Network (DNN) Recurrent Neural Network (RNN, LSTM, etc.) Convolutional Neural Network (CNN) Decision tree model (Random Forest, XGBoost, etc.) For example, considering that growth conditions change over time, it is desirable to employ an RNN that includes LSTM, which is suitable for time-series data.

[0263] 5-3. Setting the input and output of the learning model The following data is set in the input layer of the learning model: Date and time (period information) Weather sensor data (wind speed, temperature, rainfall, etc.) Environmental sensor data (soil moisture, temperature, etc.) NDVI value (growth status of each area) Reservation schedule information and area designation information Environmental equalization processing data Control angle history data (angle information for axes 1 to 3) In addition, the overall evaluation score or rank (A to E) obtained from the ranking process is set in the output layer.

[0264] 5-4. Model learning process Supervised learning is carried out using the input and output data set above. For the learning process, it is desirable to divide the learning data into training data and validation data, and apply cross validation to prevent overfitting of the model. As an example of a specific learning algorithm, the general structure when using a deep neural network is shown below. - Input layer: Each data item is numerically encoded - Hidden layers: Multiple layer neural network (ReLU, sigmoid function, tanh, etc. are used as activation functions) - Output layer: Linear output when predicting the overall evaluation score, and a softmax function is used to output the probability of each rank when performing rank classification

[0265] 5-5. Performance evaluation and optimization of learning models After learning, the predictive accuracy of the model is evaluated. Specifically, the error between the prediction results for the validation data and the actual evaluation rank is calculated, and the accuracy is quantitatively evaluated (for example, mean squared error (MSE), mean absolute error (MAE), classification accuracy (Accuracy), F1 score, etc.). Based on the evaluation results, the model parameters (number of layers, number of units, learning rate, etc.) are adjusted, and accuracy is optimized by re-learning. This process is repeated until a machine learning model with sufficient predictive accuracy for practical use is completed.

[0266] 5-6. Storage and saving of trained model The trained final machine learning model is stored and saved in a database within the control device 200 or an external database. This model will be used later as a prediction model when performing real-time growth uniformity control. Next, the real-time prediction and angle control decision process in the second embodiment will be described in detail.

[0267] 6. Real-time prediction and angle control determination In the second embodiment of the present invention, the machine learning model constructed and trained in 4. is used to perform real-time prediction of crop growth uniformity and angle control of the solar panel 110 during actual operation.

[0268] The specific procedures for real-time prediction and angle control are as follows: 6-1. Acquisition of real-time data During actual operation, the following data is acquired at predetermined intervals (for example, every 10 minutes): Weather sensor data (wind speed, temperature, rainfall, etc.) Environmental sensor data (soil moisture, temperature, etc.) Latest growth monitoring data (most recent NDVI value) Current reservation schedule information and user-specified information Environmental equalization processing data (raindrop concentration, temperature unevenness, wind deviation, etc.)

[0269] 6-2. Generation and evaluation of angle candidates The control device 200 uses real-time acquired data as input to virtually generate multiple angle control candidates, and evaluates the growth uniformity of each candidate using a prediction model. For example, candidate angles are generated in the following ways: - Generate multiple candidates that differ by ± several degrees based on the current angle setting (e.g., create three to five candidates with elevation angle in ±2° increments) - In the case of two-axis or three-axis control, generate multiple candidates with elevation angle, azimuth angle, and tilt angle adjusted independently

[0270] Specific angle candidates and their evaluation examples are shown below.

[0271] 6-3. Selection of the optimal angle candidate The control device 200 compares the predicted evaluation scores or ranks of each candidate and selects the candidate predicted to have the highest growth uniformity. For example, in the above evaluation example, candidate 1 with the highest evaluation score (83 points, rank A) is selected.

[0272] In selecting candidates, it is also possible to take into account real-time weather conditions and urgency (such as forecasts of strong winds or heavy rain) and apply special rules for safety or environmental protection.

[0273] 6-4. Execution of control based on selected angle Once the optimal angle candidate has been selected, the control device 200 sends a control command to the drive device 120 instructing that angle, and actually adjusts the angle of the solar panel 110. The drive device 120 receives this command, accurately adjusts the panel to the set angle, and sends feedback from the angle sensor back to the control device 200. The control device 200 receives this feedback and verifies that the error between the instructed angle and the actual angle is within an acceptable range.

[0274] 6-5. Collection of new data and addition to training data After control is performed, growth result data such as NDVI is collected again after a certain period of time (for example, one week) has passed by the growth monitoring unit 214. This data undergoes a ranking process and is added to the database as new training data. This new training data is used to periodically re-learn, and the machine learning model is continuously updated and improved.

[0275] By continuously carrying out the above real-time prediction and angle control process, it is possible to constantly optimize the crop growth environment and improve growth uniformity.

[0276] 7. Configuration for Outsourcing Machine Learning Processing to an External Device In the second embodiment of the present invention, it is assumed that there may be cases where it is difficult for the control device 200 to directly execute preprocessing or machine learning model learning processing due to hardware constraints (processing power, memory capacity, etc.). In such cases, it is desirable to employ a configuration in which the processing is outsourced to an external device as described below, and the control device 200 transmits and receives information necessary for control via communication.

[0277] 7-1. Processing to be outsourced to external devices Specifically, an external processing device (such as a cloud server or dedicated computer) will be responsible for the following machine learning-related processes: (1) Preprocessing of collected data (2) Construction of training dataset (3) Training and optimization of machine learning models (4) Storage and saving of trained models

[0278] 7-2. Communication processing between the control device 200 and external device The control device 200 transmits various sensor data (weather data, environmental data, growth monitoring data, etc.) and control history information to the external device in real time or periodically via the communication interface 203. The external device predicts growth uniformity in real time using a machine learning model based on this received data and returns the results to the control device 200.

[0279] 7-3. Advantages of external device outsourcing configuration The control device 200 executes actual angle control based on the optimal angle control information received from the external device. By outsourcing the processing to the external device in this way, the burden on the control device 200 is reduced and a high level of prediction accuracy can be achieved. Furthermore, the data and models stored in the external device are regularly updated and improved, providing a mechanism for always performing the latest and optimal control.

[0280] As described above, the second embodiment of the present invention has a configuration that can be fully implemented in a practical operating environment while taking into consideration hardware restrictions.

[0281] According to the second embodiment of the present invention, in a complex environment that includes many factors, such as the growth status of crops in an agricultural environment, weather conditions and soil conditions that affect growth, and the control history of solar panels, a machine learning model is constructed using past control results and their growth results, and this model is utilized in real time, thereby achieving the following excellent effects:

[0282] (1) High-precision control of growth uniformity Conventionally, it has been difficult for human-configured rule-based control to respond flexibly and quantitatively to complex fluctuating factors in the growth environment. In this embodiment, a machine learning model built based on past performance data can predictively determine the optimal panel angle, thereby achieving high-precision growth uniformity.

[0283] (2) Autonomous and continuous improvement of accuracy. Monitoring information on growth results is periodically added as training data, and the model is continuously retrained. This allows the control accuracy to automatically evolve according to the season, crop, and characteristics of each field, and the effectiveness is maintained and improved even in long-term operation.

[0284] (3) Lightweight control device and practical system implementation By outsourcing the learning process and inference process to an external device (cloud or high-performance server), the control device 200 does not require advanced computing power, and the system can be implemented with lightweight, low-cost hardware. This significantly reduces the introduction cost and maintenance burden in actual operation.

[0285] (4) Flexible response to individual environments Even in multiple fields with different weather conditions, land shapes, and growth histories, it is possible to control the angle optimally for each field, thereby realizing individual optimization for regional and varietal differences, contributing to the stabilization of farming operations and the improvement of yields and quality.

[0286] As described above, the second embodiment of the present invention is distinct from conventional simple schedule control and environmentally responsive control, and has the remarkable effect of realizing practical and sustainable smart farming in the agricultural field through data-driven intelligent control.

[0287] (Third embodiment) In the second embodiment, various sensor information, monitoring results, schedule information, etc. are processed in an integrated manner, and angle control is optimized by a machine learning model using an external device as needed. However, the introduction of AI or an external learning device is not necessarily appropriate in all operating environments, and it is possible to achieve useful feedback control that contributes to uniform growth using a relatively simple control strategy.

[0288] In this embodiment, the growth uniformity information generator stores past angle control information and the crop growth results based on that information (NDVI values, environmental sensor values, etc.) as history information, and reuses the stored history for future control. Here, automatic feature extraction and prediction processing using machine learning are not performed, but angle control is optimized by explicit condition judgment such as rule-based or threshold comparison.

[0289] The growth uniformity information generator or control device 200 is configured to record, in association with each other, historical information related to solar panel angle control and the monitoring results of crop growth conditions obtained as a result of that control. Specifically, the following time-series and area-specific linked information is accumulated as a database: Date and time of control implementation Identification information for the target panel or farmland area (panel ID, area ID, etc.) Angle information output at that time (elevation angle, azimuth angle, roll angle, etc.) Growth monitoring data obtained within a specified period thereafter (e.g., 3 days to 1 week) (NDVI, temperature / moisture sensor information, etc.) Weather and environmental conditions at the time of monitoring (as needed) These are recorded, for example, in the following correspondence table structure: In this way, a structural one-to-one record is made of the actual growth results that a given angle setting led to. This makes it possible to search for and reuse "reproducible, good angle settings" in a specific environment from past history for subsequent control decisions. Furthermore, because there is a time lag between the angle control of a given panel and the monitoring results, defining rules in the system, such as matching data after a certain period of time (e.g., 72 hours), can increase the reliability of the history.

[0290] For example, the following processing flow is envisioned: The angle information from the previous or previous few control attempts and the resulting crop growth conditions (e.g., NDVI above a threshold) are recorded as a history for each panel or region. If the same or similar environmental conditions (temperature, rainfall, soil moisture, etc.) are predicted for the next control, reference is made to past control results showing "good growth conditions" and the corresponding angle information is prioritized. On the other hand, if a specific angle control in the past history corresponds to "poor growth," avoid that setting and switch to another history or a predetermined alternative setting. By adopting a conditional reuse strategy based on history in this way, the system's decision-making process is simple, yet it is possible to gradually improve control accuracy over long-term operation.

[0291] The information recorded as history data includes the following correspondences: - Control date and time and panel ID - Contents of angle control performed (elevation angle, azimuth angle, roll angle, etc.) - Weather and environmental sensor information for the relevant area (temperature, humidity, soil moisture, etc.) - Growth monitoring results for the relevant area (NDVI, etc.) - "Growth condition good / bad" index automatically evaluated by the system By storing this information in chronological order, the control device 200 or the growth uniformity information generation unit can explicitly compare the relationship between environmental conditions and past controls and results, and select a control policy using simple comparison logic.

[0292] FIG. 15 is a flowchart of angle control using history information in the third embodiment of the present invention. This flow structured the process in which the growth uniformity information generation unit 210 references accumulated history information and selects or corrects angle information based on past control results. (S1301) Acquires current weather information, growth monitoring information, farmland environmental sensor information, etc. (S1302) Searches for past history entries similar to the acquired current information. The search is performed using the panel ID, weather conditions (temperature, precipitation, etc.), time (season), target area, etc. as matching keys. (S1303) From the searched history, records with a growth evaluation label of "good" are preferentially extracted. If an evaluation label based on an NDVI threshold, etc. is not explicitly specified, a simple evaluation is performed using a score based on the NDVI value or sensor value. (S1304) If a corresponding history exists, the angle information used in that control is adopted as a candidate for reuse. (S1305) If there are multiple candidates, the candidates are scored according to the degree of match to the matching conditions, closeness over time, etc., and the angle information from the history with the highest degree of match is adopted. (S1306) If there is no relevant history or the degree of match is low, the control is switched to a predetermined default angle or control based on another information source (user specification, schedule, weather forecast, etc.). (S1307) The determined angle information is sent to the control device 200, and the driver 120 adjusts the panel angle. (S1308) Thereafter, the NDVI and environmental data acquired as a result of the control are recorded in association with the control and accumulated as a new history entry.

[0293] As described above, in this embodiment, by repeatedly storing and reusing historical information, the system gradually accumulates angle control that is considered optimal for each environmental condition, making it possible to gradually improve the accuracy of control. With this configuration, feedback-type control based on past performance can be realized without using AI.

[0294] This embodiment has the advantage of being easy to implement in terms of cost and operation, since it has a lighter configuration than the introduction of AI and does not require communication with external devices or learning processing. Also, from the perspective of utilizing past performance, it is useful as a practical application method that reduces the complexity of implementation while maintaining the basic concept of data-driven control.

[0295] (Modifications) In the above first and second embodiments, the following five types of information providing components have been described as examples of information sources referenced by the growth uniformity information generation unit: - An area designation information processing unit that processes area information explicitly designated by the user - A growth monitoring unit that monitors the growth state of farmland using drones or sensors - A reservation schedule management unit that manages control based on date and time designations by the user - A weather information processing unit that processes weather conditions based on weather sensors and external forecast information - An environment uniformity processing unit that uniforms environmental conditions such as raindrop distribution, temperature unevenness, and wind deviation However, the present invention is not limited to a configuration that uses all five of these components, and angle information may be generated using only one of them, or a combination of two, three, or four of them may be used.

[0296] Furthermore, it is also possible to adopt a configuration that uses information sources other than the above five as additional information, such as: Measurement results of the nutritional status in the soil; A characteristic database for each crop variety; Topographical and geological information; Periodic patterns of day / night and seasonal variations; Image analysis results using artificial intelligence (disease detection, etc.); and the growth uniformity information generation unit may be flexibly configured to combine any of this information to generate angle information.

[0297] In this way, the angle control according to the present invention does not require a fixed configuration or type of information provision component, but rather allows for a control design that can be flexibly selected, changed, and expanded depending on the situation, purpose, and scale of implementation. This flexibility in configuration can be applied to both the first embodiment (rule-based control) and the second embodiment (machine learning control) in the same way.

[0298] The technical essence of this invention is that the angle of solar panels installed on farmland is not simply controlled for conventional purposes such as improving power generation efficiency or ensuring work efficiency, but rather focuses on the growth conditions of crops within the farmland and controls them with the aim of ``uniformizing'' their growth.

[0299] To achieve this "uniform growth," it is necessary to obtain information on the growth status of crops in the entire farmland or in part of it in some form. In this invention, "information on growth status" is a broad concept that includes the following: Vegetation indices such as NDVI obtained by observation means such as drones and ground sensors Measurements of temperature, humidity, soil moisture content, etc. obtained by environmental sensors Weather data and its forecast values ​​Or information that the user subjectively judges and specifies based on the status of the farmland as visually recognized on a monitor screen or through on-site visual inspection (for example, by specifying an area on the screen).

[0300] In other words, whether the information is obtained automatically or is based on the subjective judgment of the user, it should be interpreted as being "based on information regarding the developmental state of the crop."

[0301] Based on this information, the growth uniformity information generation unit determines how to adjust the angle of each solar panel to correct the growth environment within the farmland, and generates angle information based on the results of that determination.

[0302] In this way, the present invention has an essential technical concept in the linkage between two elements: "information on the growth state" and "angle control for uniformity based on that information," and allows for a flexible configuration that is not dependent on the means of acquiring the information or the subject of the decision maker (automatic / human).

[0303] Although the present invention has been described in detail above, the above description is merely illustrative of the present invention in all respects and is not intended to limit its scope. It goes without saying that various improvements and modifications can be made without departing from the scope of the present invention. Each of the constituent elements of the invention disclosed in this specification is considered to be an independent, stand-alone invention. Inventions that combine the constituent elements in any manner are also included in the present invention. The specific expressions in this specification are merely examples, and the present invention also includes those that conceptualize these exemplary expressions.

[0304] (Reference Embodiment) This application claims priority based on the domestic application shown below, and the specification of that basic application is attached below for reference. Among the technical contents described in that basic application, the control of solar panels in accordance with the agricultural land environment, particularly the control configuration based on consideration of agricultural work and weather conditions such as wind, rain, and frost, is the basis of the invention of this application and is eligible for priority.

[0305] The present invention relates to a solar power generation system equipped with movable solar panels, and more particularly to a solar power generation system for use on agricultural land.

[0306] Japanese Patent Laid-Open Publication No. 2015-126683 aims to ensure that a solar power generation system does not interfere with work, even when installed in a location where workers enter. The solar power generation system (10) described in Patent Literature 5 (the symbols in parentheses are symbols described in the patent document; the same applies below) includes a solar panel (20), a support mechanism (30) that rotatably supports the solar panel (20), a drive mechanism (40) that rotates the solar panel (20), and a control unit (70) that performs a tracking operation to control the drive mechanism (40) so that the light-receiving surface (21) of the solar panel (20) continues to face the sun. The solar power generation system (10) also includes a stop input unit (84) through which a worker inputs a stop command for the solar panel (20). When a stop command is input to the stop input unit (84), the control unit (70) stops the tracking operation and performs a stop operation to control the drive mechanism (40) so that the solar panel (20) is kept stopped at a predetermined stop position.

[0307] According to JP 2015-126683 A, by performing a stopping operation in the solar power generation system (10), when the solar power generation system (10) is installed in a work space, the solar panels (20) can be held in a position that does not interfere with the workers.

[0308] In the case of agricultural solar power generation, which generates solar power on farmland, so-called agricultural solar power generation, it is necessary to bring agricultural vehicles into the field. In this case, it is necessary to secure enough space for the agricultural vehicles to operate. Furthermore, while securing enough space for the agricultural vehicles, it is also necessary to avoid a decrease in power generation efficiency. At the same time, it is also necessary to ensure that the crops receive sufficient sunlight. Furthermore, it is necessary to deal with climatic fluctuations such as wind, rain, snow, and frost. Although these issues are extremely important in agricultural solar power generation, until now, attention has been focused only on the efficiency of solar power generation, and consideration has been given to the agriculture that takes place beneath the panels.

[0309] In the past, there were many cases where a limited number of suitable crops (such as sakaki trees) were grown under the panels and used for agricultural solar power generation. Until now, there have been few cases where wheat, rice, or large-scale vegetable cultivation was used for agricultural solar power generation.

[0310] Therefore, an object of the present invention is to provide a solar power generation system suitable for agricultural use.

[0311] In order to solve the above problems, the present invention has the following features: The present invention is a solar power generation system installed on farmland, comprising a plurality of solar panels, a mounting base for supporting the plurality of solar panels so that the angles of the plurality of solar panels can be adjusted, a drive device for adjusting the angles of the plurality of solar panels, and a control device for controlling the drive device.

[0312] The control device controls the drive device to set the solar panels vertically when an agricultural work vehicle enters the farmland, controls the drive device to adjust the angle of the multiple solar panels so that they face the sun in the first mode that prioritizes power generation, and controls the drive device to adjust the angle of the multiple solar panels so that they face the sun in the second mode that prioritizes sunlight hitting the farmland, thereby minimizing the shadows cast on the farmland.

[0313] This allows agricultural vehicles to enter farmland during agricultural work. In the first mode, power generation is prioritized, improving power generation efficiency. In the second mode, sunlight exposure to crops is prioritized, promoting crop growth.

[0314] Preferably, the plurality of solar panels are capable of generating electricity on both the front and back surfaces.

[0315] By making it possible to generate power not only on the front surface but also on the back surface, it becomes possible for the solar panel to generate power even in the second mode.

[0316] Preferably, the mount supports the plurality of solar panels so that the panels are rotatable around an axis.

[0317] Preferably, the axis of rotation is located to one side of the solar panel or offset from the center of the solar panel.

[0318] Preferably, in the first and second modes, the control device controls the drive devices so that the plurality of solar panels are at a predetermined angle in accordance with a predetermined time period.

[0319] Setting a predetermined angle depending on the time of day simplifies control and also minimizes the power required for driving.

[0320] Preferably, the angles of the multiple solar panels in the first mode and the angles of the multiple solar panels in the second mode are reversed.

[0321] By reversing the angles, it is possible to increase the power generation efficiency in the first mode while providing solar radiation to the farmland in the second mode.

[0322] Preferably, the solar power generation system further includes a weather information detection unit for obtaining weather information, and the control device controls the drive device to adjust the angles of the multiple solar panels based on the weather information obtained by the weather information detection unit.

[0323] Preferably, when the weather information detection unit detects a strong wind and its direction, the control device controls the drive devices so that the plurality of solar panels are horizontal.

[0324] If you keep it horizontal during strong winds, you can prevent the solar panels from being blown away by the wind.

[0325] Preferably, when rain or snow is detected by the weather information detection unit, the control device controls the drive devices so that the plurality of solar panels are vertical.

[0326] If you make it vertical during rain, you can water the farmland evenly, and if you make it vertical during snow, you can prevent it from being crushed by the weight of the snow.

[0327] Preferably, when the weather information detection unit detects that frost is forecast, the control device controls the drive devices so that the plurality of solar panels are horizontal.

[0328] Leveling when frost is forecast may prevent frost from falling on the crop.

[0329] Preferably, the mounting base comprises a plurality of posts and a plurality of shafts, and the plurality of solar panels are attached to the shafts.

[0330] Preferably, the shaft is rotatably supported on a support.

[0331] Preferably, the base has a bearing attached to a support and a slewing drive, and the shaft is made rotatable by passing the shaft through a worm wheel in the bearing and slewing drive and rotating the worm wheel.

[0332] Such a simple configuration allows the solar panel to be rotated, thereby reducing costs.

[0333] The present invention also provides a solar power generation system to be installed on agricultural land, comprising a plurality of solar panels, a mounting base for supporting the plurality of solar panels so that the angles thereof can be adjusted, a drive device for adjusting the angles of the plurality of solar panels, and a control device for controlling the drive device, wherein the control device controls the drive device for each solar panel to adjust the angle of the plurality of solar panels so that the angle of the plurality of solar panels is oriented to minimize the shadow cast on the agricultural land when priority is given to the growth of crops on the agricultural land, and controls the drive device to adjust the angle of the plurality of solar panels so that the panels are oriented toward the sun when priority is not given to the growth of crops on the agricultural land.

[0334] This allows the areas that are growing well to be shaded to prioritize power generation, while areas that are not growing well can be exposed to as much sunlight as possible to promote growth.In addition, it is expected that the amount of power generated will increase as the reflected light from the solar panel hits surrounding solar panels.

[0335] As described above, according to the present invention, a solar power generation system suitable for agricultural use can be obtained.

[0336] The objects, features, configurations, operations, and effects of the present invention and embodiments of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings.

[0337] FIG. 16 is a block diagram showing the functional configuration of a solar power generation system 1 according to one embodiment of the present invention. FIG. 17 is a flowchart showing the operation of the control device 2. FIG. 18 is a flowchart showing the operation of the control device 2 based on the weather information detection unit 5. FIG. 19 is a plan view showing an example of the layout of the solar power generation system 1 on farmland. FIG. 20 is a plan view and a side view showing the state of an agricultural work vehicle working. FIG. 21 is a perspective view showing the state of an agricultural work vehicle working. FIG. 22 is a diagram showing the angle of the cantilevered solar panel 4. FIG. 23 is a diagram showing the angle of the cantilevered solar panel 4 during agricultural work and in each time period (tracking mode and full light mode). FIG. 24(a) explains how shadows and light are generated depending on the angle of the solar panel 4 in accordance with crop growth, and FIG. 24(b) is a 3D simulation of the shadow generation in FIG. 24. FIG. 25 is a flowchart when each mode is integrated.

[0338] 16, a solar power generation system 1 according to one embodiment of the present invention includes a control device 2, an input device 2a, a plurality of drive devices 3, a plurality of solar panels 4, and a weather information detection unit 5. In FIG. 19, a plurality of vertical posts 9 are erected on farmland, and shafts 8 are provided between the posts 9. A plurality of bearings 6 (the bearings 6 shown in FIG. 19 are an example of resin bearings) are attached to the upper part of a portion of the post 9. In addition, a slewing drive 7 is attached to the upper part of a portion of the post 9 (the slewing drive 7 shown in FIG. 19 is an example).

[0339] A shaft 8 is attached to the center of the bearing 6. The outer periphery of the bearing 6 is attached to a support 9, which allows the shaft 8 to rotate. A screw drive 7 is provided between the two bearings 6. The screw drive 7 is generally composed of a motor, a worm, and a worm wheel, and is a mechanism in which the worm wheel rotates when the worm is rotated by the motor (see, for example, Utility Model Registration Publication No. 3201640).

[0340] The case of the screw drive 7 is attached to a support 9. A shaft 8 is attached to the worm wheel. A motor that can rotate to a specified angle, such as a servo motor or stepping motor, is used as the motor. In this way, the screw drive 7 and bearing 6 can rotate the shaft 8 to a desired angle.

[0341] The driving device 3 is composed of a slew drive 7 and devices required to operate it.

[0342] It should be noted that various other well-known mechanisms can be used as the drive device for rotating the solar panel 4.

[0343] One side of the solar panel 4 is attached to the shaft 8. In other words, the solar panel 4 is attached to the shaft 8 in a cantilevered manner.

[0344] As shown in Figure 19, when installing solar panels 4 on farmland, support posts 9 and axles 8 are installed on the farmland to ensure a width D that allows agricultural work vehicles to pass through. Furthermore, spaces 10 are provided on both sides of the farmland so that agricultural work vehicles can change direction. As such, the configuration shown in Figure 19 is the basic configuration of this embodiment, but it is of course only one example and does not limit the present invention.

[0345] An example of size is shown in Figure 20. If the farmland is approximately 20 meters by 50 meters in size (Figure 20(a)), and the width of the farm vehicle is approximately 2 meters, the distance between the support posts 9 and the axles 8 is set to 5 meters or more (Figure 20(b)). In this way, three rows of solar panels can be placed on the farmland, as shown in Figure 20(a).

[0346] Figure 21 shows how an agricultural work vehicle moves between the panels. As shown in Figure 22, the cantilevered solar panel 4 can be adjusted to various angles between horizontal and vertical. Here, it is shown as being ±45 degrees, but other angles are also possible. As the solar panel 4, a panel with power generation portions on both sides is used, but a single-sided panel is also acceptable. When a double-sided panel is used, power generation by reflected light from the other solar panel 4 can also be expected.

[0347] In Figure 22, the length of the support pole 9 buried in the ground is 2.5 meters, but this is just one example. The length of the support pole 9 protruding from the ground is 0.8 meters, but this is just one example. The width of the solar panel 4 is 2.2 meters, but this is just one example. The height of the support pole 9 protruding from the ground and the solar panel 4 when vertical is 3.2 meters, but this is just one example. The height of the solar panel 4 when horizontal is 1 meter (when the front side of the panel is facing up) or 0.8 meters (when the back side of the panel is facing up), but this is just one example. If the solar panel 4 is buried in the ground at roughly these ratios, it can be firmly fixed.

[0348] The operation of the control device 2 will be described with reference to Fig. 17. The control device 2 is instructed by the input device 2a to select either the tracking mode, full light mode or V mode (S101).

[0349] The tracking mode is a mode (first mode) that prioritizes power generation, and is a mode in which the solar panel 4 is moved to track the movement of the sun. In this embodiment, in the tracking mode, the solar panel 4 is fixed at a predetermined angle during a predetermined time period, but the time period and angle may be set more precisely, the position of the sun may be detected by a sensor and the solar panel 4 may be moved constantly, or the solar panel 4 may be moved at a pre-programmed angle based on the date, time, and latitude and longitude.

[0350] When agricultural work time is specified by the input device 2a, the control device 2 sets the solar panel 4 vertically (S102) and drives the drive device 3 to rotate the thread drive 7 by the required angle so that the solar panel 4 is vertical (S107).

[0351] When not performing agricultural work, the control device 2, in tracking mode, sets the solar panel 4 to be horizontal from 4 PM to 8 AM (S103), sets the solar panel 4 to be at +45 degrees from 8 AM to 11 AM (S104), sets the solar panel 4 to be horizontal from 11 AM to 2 PM (S105), and sets the solar panel 4 to be at -45 degrees from 2 PM to 4 PM (S106). In response to this, the control device 2 drives the drive device 3 so that the solar panel 4 is at the desired angle (S107).

[0352] When not performing agricultural work, in full light mode, the control device 2 sets the solar panel 4 to be horizontal from 4 PM to 8 AM (S103), sets the solar panel 4 to be at -45 degrees from 8 AM to 11 AM (S104), sets the solar panel 4 to be vertical from 11 AM to 2 PM (S105), and sets the solar panel 4 to be at +45 degrees from 2 PM to 4 PM (S106). In response to this, the control device 2 drives the drive device 3 so that the solar panel 4 is at the desired angle (S107).

[0353] The full light mode is a mode (second mode) that prioritizes sunlight falling on farmland, and is a mode in which the multiple solar panels 4 are moved in a direction that minimizes the shadows cast on the farmland.

[0354] 23 and 22, at the time of the vernal or autumnal equinox, the right side of the page is east and the left side is west. In Fig. 23, the column marked "Tracking" indicates the angle of the solar panel 4 in tracking mode. In Fig. 23, the column marked "FULL LIGHT" indicates the angle of the solar panel 4 in full light mode.

[0355] As can be seen from the shadows in Figure 23, in tracking mode, shadows are cast on the farmland, but the angle is set to prioritize power generation efficiency. On the other hand, in full light mode, shadows cast on the farmland are minimized so that light can reach the crops. In this way, full light mode is a mode that prioritizes cultivation over power generation by minimizing the shadow cast by the solar panel 4 as much as possible.

[0356] V-mode is a mode in which an operator observes the growth of crops or determines the growth level of the crops through image diagnosis or automatic determination using various sensors, and then orients the solar panels 4 at an angle that prioritizes power generation in areas where the crops are growing and shade is appropriate, and orients the solar panels 4 at an angle that prioritizes sunlight for the crops in areas where sunlight is required. For example, as shown in Figures 24(a) and (b) , in the afternoon, the solar panels 4 on the west side of the area where shade is appropriate are oriented at -45 degrees, and the solar panels 4 on the west side of the area where sunlight is appropriate are oriented at +45 degrees. In such a case, the solar panels 4 are oriented symmetrically like a V, which is why it is referred to as V-mode. However, this is merely an example of angles, so it would be better to refer to it as the third mode.

[0357] 18, the angle of the solar panel 4 can be changed depending on the weather. The weather information detection unit 5 is a device that can obtain weather information from weather sensors installed in farmland or through communications. In this embodiment, the weather information detection unit 5 can obtain information on wind, rain, snow accumulation, frost, etc.

[0358] The control device 2 drives the drive device 3 automatically or manually based on information from the weather information detection unit 5 (S200). Automatically changing the angle based on weather information is referred to as the weather mode (fourth mode). In parallel with the operation shown in FIG. 17 , the control device 2 obtains weather information from the weather information detection unit 5 and adjusts the solar panel 4 to a horizontal position when a strong wind with a predetermined wind speed or direction is detected (S201). Similarly, the control device 2 adjusts the solar panel 4 to a horizontal position when a frost forecast is issued (S202). Similarly, the control device 2 adjusts the solar panel 4 to a vertical position when a predetermined amount of rainfall or snowfall is detected (S203).

[0359] In this way, the control device 2 adjusts the angle of the solar panel 4 depending on weather conditions to prevent the panel from receiving more force than necessary from strong winds, to prevent the ground from getting too cold when there is frost, to allow rainwater to reach crops when it rains, and to prevent snow from accumulating on the solar panel 4 when it snows.

[0360] As described above, according to the embodiment of the present invention, it is possible to set the solar panel 4 vertically during agricultural work and to allow agricultural vehicles to enter farmland. Furthermore, when it is desired to increase power generation efficiency, the tracking mode can be used to orient the solar panel 4 in an appropriate direction. On the other hand, when it is desired to expose crops to sunlight, the full light mode can be used to minimize the casting of shadows on the farmland.

[0361] Furthermore, by adjusting the orientation of the solar panels 4 based on weather information, it is possible to reduce the load on the solar panels 4 and provide an appropriate natural environment for the crops. Note that moving the solar panels 4 based on weather information is an option and is not a required configuration.

[0362] Therefore, the embodiment of the present invention is a solar power generation system suitable for agricultural use.

[0363] In the above embodiment, a support 9 and an axis 8 are used as a mount for supporting the solar panel 4 so that the angle thereof can be adjusted, but the mount is not limited to the configuration presented above as long as it can support the angles of multiple solar panels 4 so that the angle thereof can be adjusted.

[0364] In the above embodiment, the rotation axis is installed on only one side of the solar panel 4, but this is not limited to this. Of course, the axis may be located at the center of the solar panel 4. Furthermore, the axis may be located at a position offset from the center of the solar panel 4. An axis located at an offset position is referred to as an eccentric type.

[0365] The tracking mode (first mode), full light mode (second mode), V mode (third mode), and weather mode (fourth mode) in the above-described embodiments, as well as the mode for keeping the solar panel 4 vertical during agricultural work, can be integrated into an automatic operation mode (operation mode). Figure 25 shows a flowchart when these modes are integrated.

[0366] 25, first, the control device 2 determines whether the mode is manual (maintenance mode) (S301) or automatic (operation mode) (S302). In the manual (maintenance mode), the control device 2 controls the drive device 3 in accordance with an operation instruction from the input device 2a so as to fine-tune the panel angle during maintenance inspection, etc. (S107).

[0367] In the automatic (operation mode), the control device 2 controls the drive device 3 in accordance with the set mode (S303 to S307).

[0368] The control device 2 sets the solar panel 4 vertically during agricultural work (S303, S107). In the weather mode, the control device 2 controls the drive device 3 in accordance with Fig. 18, for example, to set the solar panel 4 horizontally during strong winds (S304, S107).

[0369] In V mode, the control device 2 controls the drive device 3 to move the solar panels 4 automatically or manually, in accordance with S108 in Figure 17, based on the growth of the plants, so that areas that can be shaded are given priority for power generation and areas that require sunlight are given priority for sunlight (S305, S107).

[0370] In the full light mode, the control device 2 controls the drive device 3 in accordance with S103 to S106 in FIG. 17 so that priority is given to exposing the crops to light (S306, S107).

[0371] In the tracking mode, the control device 2 controls the drive device 3 in accordance with S103 to S106 in FIG. 17 so that the solar panel 4 faces the sun (S307, S107).

[0372] The gist of the present invention is that movable panels installed within farmland locally adjust the crop growth environment and create a uniform cultivation environment. Therefore, the panels do not necessarily need to have a power-generating function. For example, a similar effect can be achieved by movably supporting non-power-generating shielding panels for environmental control purposes such as shading, blocking rain, and ventilating, and adjusting the panel angle using a similar control mechanism and information processing system. That is, the present invention can also be understood as an agricultural environmental control system installed on farmland, comprising: a plurality of movable panels; a frame for supporting the movable panels so that their angles can be adjusted; a drive unit for adjusting the angle of the movable panels; a growth uniformity information generation unit for generating angle information for the movable panels to uniformize the growth of crops within the farmland based on information regarding the growth status of the crops within the farmland; and a control unit for controlling the drive unit to adjust the angle of the movable panels based on the angle information generated by the growth uniformity information generation unit.

[0373] Although the present invention has been described in detail above, the above description is merely illustrative of the present invention in all respects and is not intended to limit its scope. It goes without saying that various improvements and modifications can be made without departing from the scope of the present invention. Each of the constituent elements of the invention disclosed in this specification is considered to be an independent, stand-alone invention. Inventions that combine the constituent elements in any manner are also included in the present invention. The specific expressions in this specification are merely examples, and the present invention also includes those that conceptualize these exemplary expressions.

[0374] INDUSTRIAL APPLICABILITY The present invention is a solar power generation system that enables the uniform growth environment of crops and is industrially applicable.

[0375] 100 Solar power generation system 110 Solar panel 120 Drive unit 130 Stand 200 Control device 201 Processor 202 Memory 203 Communication interface 204 Input / output interface 205 Power supply unit 210 Growth uniformization information generation unit 211 Area designation information processing unit 212 Display unit 213 Operation input unit 214 Growth monitoring unit 215 Drone 216 Environmental sensor 217 Weather information processing unit 218 Weather sensor 219 Communication unit 220 Reservation schedule management unit 221 Memory unit 222 Environmental uniformization processing unit 300 Power management unit 310 Power conversion unit 320 Power storage unit 330 Power supply unit 400 Monitoring device

Claims

1. A solar power generation system installed on agricultural land, comprising: a plurality of solar panels; a mounting base that supports the solar panels so that the angle can be adjusted; a drive unit that adjusts the angle of the solar panels; a growth uniformity information generation unit that generates angle information for the solar panels to uniformize the growth of crops in the agricultural land based on information regarding the growth state of the crops in the agricultural land; and a control unit that controls the drive unit to adjust the angle of the solar panels based on the angle information generated by the growth uniformity information generation unit.

2. A solar power generation system as described in claim 1, characterized in that the growth uniformity information generation unit generates solar panel angle information for providing solar radiation to an area, forming a shadow, concentrating raindrops, avoiding the concentration of raindrops, or improving ventilation, based on area information within the farmland specified by a user.

3. A solar power generation system as described in claim 1, characterized in that the growth uniformity information generation unit identifies areas where there is a bias in the growth state based on information about the crop growth status obtained from a drone or a sensor installed in the farmland, and generates solar panel angle information to provide sunlight to those areas, create shadows, concentrate raindrops, avoid raindrop concentration, or improve ventilation.

4. A solar power generation system according to claim 3, characterized in that the information relating to the crop growth status includes at least one of a vegetation index including NDVI (Normalized Difference Vegetation Index), soil moisture content, temperature, humidity, or solar radiation.

5. A solar power generation system as described in claim 1, characterized in that the growth uniformity information generation unit generates solar panel angle information for providing solar radiation, forming shadows, concentrating raindrops, avoiding raindrop concentration, or improving ventilation for a specific agricultural land area based on schedule information relating to a predetermined time or time period.

6. A solar power generation system as described in claim 1, characterized in that the growth uniformity information generation unit generates angle information that takes into account the safety of the solar panels and the stability of the agricultural land environment, based on meteorological information regarding wind speed, rainfall, snowfall, or frost.

7. A solar power generation system as described in claim 1, characterized in that the growth uniformity information generation unit generates solar panel angle information for uniforming the farmland environment based on bias in the falling position of raindrops, uneven temperature distribution within the farmland, or bias in wind flow.

8. A solar power generation system as described in claim 1, characterized in that the growth uniformity information generation unit uses at least two or more pieces of information from among area designation information by a user, information on the growth status of crops, schedule information, weather information, and information on the environmental conditions within the farmland, weights them, and generates solar panel angle information based on the weights.

9. A solar power generation system according to claim 8, characterized in that the weighting is performed based on the reliability, accuracy, freshness or urgency of each piece of information.

10. A solar power generation system as described in claim 1, characterized in that the growth uniformity information generation unit uses at least two or more pieces of information from among area designation information by a user, information on the growth status of crops, schedule information, weather information, and information on the environmental conditions within the farmland, and generates solar panel angle information based on a predetermined priority order for these pieces of information.

11. A solar power generation system as described in claim 8, characterized in that the growth equalization information generation unit generates solar panel angle information using a predetermined priority for each piece of information in addition to weighting multiple pieces of information.

12. A solar power generation system as described in claim 1, characterized in that the growth uniformity information generation unit has a function of correlating the generated solar panel angle information with the crop growth status based on the angle information and storing it as history information.

13. A solar power generation system as described in claim 12, characterized in that the growth uniformity information generation unit generates angle information for the solar panel by selecting and reusing angle information that has previously produced good growth results based on accumulated history information.

14. A solar power generation system as claimed in claim 1, characterized in that the solar power generation system comprises: a learning means for learning a machine learning model based on training data including previously generated solar panel angle information and correspondence between the angle information and the growth status of crops based on the angle information; an angle information generation means for generating solar panel angle information for uniforming the growth of crops in farmland using the trained model; and the growth uniformity information generation unit issues a command to a control device based on the angle information generated by the angle information generation means to control the angle of the solar panels.

15. A solar power generation system as described in claim 14, characterized in that the learning means and the angle information generating means are provided in an external processing device separate from the solar power generation system, and the growth equalization information generating unit receives angle information transmitted from the external processing device and outputs it to the control device.

16. A solar power generation system according to claim 14, characterized in that the angle information generating means predicts a score relating to the uniformity of crop growth and determines angle information for the solar panel based on the score.

17. A solar power generation system according to claim 16, characterized in that the angle information generating means selects the angle that outputs the highest score.

18. A solar power generation system as described in claim 14, characterized in that the training data or input information to the trained model includes NDVI (Normalized Difference Vegetation Index), soil moisture, temperature, humidity, or solar radiation.

19. A solar power generation system according to claim 14, characterized in that the angle information generating means outputs angle information of the solar panel directly in response to input information.

20. A solar power generation system according to claim 14, characterized in that the learning means re-learns the machine learning model at regular intervals using historical information.

Images