Production and distribution system for plant factories

Abstract

To construct a business model in which artificial intelligence is utilized to control the harvest date of a crop in a plant factory on the basis of a state of the crop, a state of a facility environment, and market information of the crop.SOLUTION: A neural network is utilized for comparison operation to adjust an amount of artificial light and a temperature so that a deviation of a photosynthesis distribution, a deviation of a stem and leaf growth distribution, a deviation of a nutrient solution concentration, a deviation of an accumulated temperature, and a deviation of an accumulated solar radiation amount, which are indicators of a quality state of a crop, fall within acceptable deviation ranges at the optimal harvest time.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a production and distribution system for a plant factory that controls the optimum harvesting time using a crop cultivation management system by analyzing market information data so that the optimum harvesting time occurs when crop prices are rising. [Background technology]

[0002] Prices of fruits and vegetables fluctuate greatly depending on market demand and stock levels. In order to supply them at a stable high price, it is necessary to ensure that the harvest time is appropriately timed to match market demand and stock levels.

[0003] Because plant factories manage the cultivation conditions, it is possible to time the harvest, but to quickly and appropriately analyze the vast amount of data on this distribution route and control the production of living organisms (crops) at the right time requires using computers to process predictive information based on the experienced judgment of skilled workers, so there is a need for the development of a production and distribution system that utilizes artificial intelligence. (Patent Document 1) [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2019-128703 Summary of the Invention [Problem to be solved by the invention]

[0005] Conventional technology involves predicting days when there will be a shortage of outdoor-grown plants in the market, and increasing at least one of the set values ​​of light intensity, light exposure time, temperature, CO2 concentration, and air volume to coincide with the peak shipping times from the plant factory.

[0006] However, the calculation method for predicting days when distribution will be insufficient is based on the contents of adverse weather conditions, and is a sudden factor that can be dealt with by inputting interrupt information without relying on learning functions. Also, it is estimated that there is a limit to how much can be done from the time a weather issue is detected, even if measures are taken to promote shipping conditions.

[0007] Furthermore, the degree of ripeness of crops up until the time of shipping is determined by the color of the fruit, but analyzing the color of a large number of fruits takes time, and there is also variation in the degree of ripeness even within a plant factory, so it is estimated that there is a limit to the accuracy of the analysis.

[0008] In the present invention, the overall degree is estimated from the viewpoint of the degree of change in deviation from a large amount of data, and this method of judgment is based on the empirical rules of an expert. By utilizing a neural network, an overall judgment can be made based on an analysis of the crop condition, an analysis of the plant factory environment, and an analysis of market demand.

[0009] Furthermore, new information is generated about the price, demand, quality, etc. of each product as it flows through the distribution route. This information can lead to price fluctuations, but robust information management is required to capture accurate information. This can be addressed by adding a public key encryption function to the open API (Application Programming Interface) function, and by building a rational distribution system, the source of information can be identified and the accuracy can be improved, allowing the correct shipping time to be determined and production can be scheduled. The goal is to build a so-called business model. [Means for solving the problem]

[0010] The first aspect of the present invention is achieved by the following technical means.

[0011] A crop cultivation management system 100 detects the growth level of crops and controls the optimum harvesting time by controlling the amount of artificial light. A crop market management system 200 estimates fluctuations in crop prices from crop market information data 210 and calculates the optimum harvesting time. At the calculated optimum harvesting time, a neural network is used in a comparison calculation to adjust the amount of artificial light and temperature so that the amount of photosynthesis and deviation, the amount of stem and leaf elongation and deviation, the solution concentration deviation, the accumulated temperature deviation, and the accumulated amount of sunlight and deviation, which are indicators of the quality state of the crop, fall within the values ​​and deviation values ​​of the applicable range, thereby automatically controlling the harvesting day and submitting information to the market information data 210.

[0012] The second aspect of the present invention is achieved by the following technical means.

[0013] The crop cultivation management system 100 measures the distribution area of ​​disease occurrence in crops within a plant factory and the amount of carbon dioxide absorption, which is an index of crop growth, and uses a neural network to compare and collate the detected values ​​with standard values, corrects the harvesting date, and automatically submits the information to the market information database 210.

[0014] The third aspect of the present invention is achieved by the following technical means.

[0015] A user of the plant factory has a user account in the production and distribution system, obtains a public key from the production and distribution system, assigns IDs to the individual crops and crop groups to be distributed, generates a token, which is a string of characters encrypted with the public key, from the crop information acquired by the cultivation management system 100, and makes the information publicly available as an API (application programming interface) in the crop market management system 200 as a set including the harvest of the individual crops and the crop groups, the IDs assigned to the individual crops and the crop groups, and the token encrypted with the public key.

[0016] The fourth aspect of the present invention is achieved by the following technical means.

[0017] Plant factory users distribute IDs assigned to individual crops or groups of crops and tokens encrypted with a public key to relevant parties in the distribution system, and can add or change information generated during distribution using the public key. However, they cannot change items related to production, and there is a deadline set for each item determined by the plant factory users, after which information cannot be added or changed. Effect of the Invention

[0018] The first invention makes it possible to stabilize the supply of crops by predicting periods of high market demand and adjusting the harvesting period of crops.

[0019] The second invention makes it possible to estimate the actual supply volume and timing by determining the defect rate and growth level of crops, and by transmitting market information at an early stage, appropriate inventory adjustments and sales price responses can be made.

[0020] The third and fourth aspects of the present invention make it possible to exchange information in a common format data while preventing data leakage and external tampering, thereby providing high security and increasing the reliability of crop traceability. [Brief description of the drawings]

[0021] [Figure 1] FIG. 1 shows the transition of basic data registered in a database for representative items in the cultivation management system and crop market management system of the present invention, and the transition of data calculated using a neural network from the acquired data. [Diagram 2] Information data and crop flow chart of the present invention [Diagram 3] A block control diagram that uses the neural network of the present invention to determine the degree of crop growth based on changes in the crop and facility environment, estimates price fluctuations based on market information, and determines the harvest date. [Figure 4] Token issuing and information updating mechanism of the present invention [Diagram 5] Diagram of the resource authentication and database registration system of the present invention [Figure 6] Overview of the plant factory of the present invention and environmental map [Figure 7] An environmental map showing the change in photosynthetic distribution in the plant factory of the present invention. [Figure 8] FIG. 1 is a perspective view of a plant diagnostic device in a plant factory according to the present invention; [Figure 9] FIG. 1 is a perspective view of a cart equipped with a harvesting device of a plant factory according to the present invention; [Figure 10] FIG. 1 is a perspective view of a cart equipped with a pest control device for a plant factory according to the present invention; [Figure 11] Schematic diagram of the plant factory of the present invention [Figure 12] Positional relationship diagram of the cart and plants in the plant factory of the present invention [Figure 13] Schematic diagram of the opening and closing configuration of the ceiling window of the plant factory of the present invention [Figure 14] Schematic diagram of the fine mist ejection means of the plant factory of the present invention. [Figure 15] Schematic diagram of the shading curtain configuration of the plant factory of the present invention DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] The present invention will now be described with reference to the embodiments shown in the drawings.

[0023] The plant factory and related site diagrams, control diagrams, and block diagrams shown in Figs. 1 to 15 show an example of this embodiment.

[0024] The background of the present invention will now be described.

[0025] A feature of plant factories is that they can control the environment inside the factory, thereby controlling the growth conditions and adjusting the harvest time. This function allows them to ensure stable high prices by supplying products when there is a shortage in the market.

[0026] To do this, it is necessary to predict market trends and the harvest time. However, even if the harvest time is adjusted, changing the harvest time by seven days is not possible with just a few hours of adjustment control. By making predictions as early as possible, it is possible to respond, and by obtaining a large amount of reliable data from various sources and analyzing the information, it is possible to make the most of the advantages of a plant factory.

[0027] The structure of the plant factory of the present invention will be described.

[0028] 11 to 15 show an example of a cultivation facility, which includes a cultivation room 1, which is a greenhouse, and a shipping room 2 adjacent to the cultivation room 1. In the center of the cultivation room 1, a main passage 4 is provided through which workers, a work vehicle (work trolley) 3, or a plant diagnostic device 111 can pass, and the main passage 4 is a concrete passage with a concrete road surface. On both sides of the main passage 4, a cultivation space 6 for cultivating crops is formed by arranging a large number of cultivation beds 5, which are cultivation units, in rows. The cultivation beds 5 are cultivation beds formed of rock wool, which is a culture medium, and are configured so that nutrient solution is supplied to each cultivation bed 5 from a nutrient solution supplying device 7 in the shipping room 2. The cultivation beds 5 in FIG. 11 are suitable for harvesting by manual labor, and can be rearranged as desired. In the configuration as shown in FIG. 6, if they are arranged in the longitudinal direction of the plant factory, long-distance operation is possible and automation is easy.

[0029] In addition, entrances 8 to the cultivation room 1 equipped with opening and closing doors are provided at both ends of the main passage 4, and one of the entrances 8 can be used to access the adjacent shipping room 2. The other entrance 8 can be used to access the cultivation facility from the outside. The work vehicle 3 can be moved from the main passage 4 to a sub-passage 9 between each cultivation bed 5, and various tasks can be performed on the plants being cultivated while moving the work vehicle 3 along the cultivation beds 5 on the sub-passage 9. The sub-passage 9 is a passage formed in the front-rear direction between the left and right sides of each cultivation bed 5. The work vehicle 3 runs on rails 13 on the left and right heating pipes laid on the sub-passage 9 for heating the entire greenhouse.

[0030] The shipping room 2 is provided with the nutrient solution supplying device 7 and a sorting device 10 that sorts the harvested crops (fruits) such as tomatoes by weight, size, or grade. The sorting device 10 is a pre-treatment device that processes the cultivated crops before shipping. The sorting device 10 is configured with a sorting conveyor 11 that transports and sorts the harvested crops, and harvested product storage sections 12 for each class provided on both sides of the sorting conveyor 11, and is configured to supply the harvested crops from the sorting conveyor 11 to each harvested product storage section 12 and sort them into each class. The sorting conveyor 11 is configured to be bent in an L-shape in a plan view. Each harvested product storage section 12 is provided with a storage box for storing the harvested crops, and the harvested crops are shipped by this storage box.

[0031] A wire 15 is provided on the upper side of the cultivation bed 5 along the cultivation line, and the cultivated plant P is attracted by the wire 15, and is attracted by a string 17 that hangs down from the wire 15 via a hook 16. The hook 16 is suspended from the wire 15, and the wrapped string 17 is appropriately unwound to hang down, as is well known. After the plant grows to a predetermined height (near the hook 16), the string 17 is unwound from the hook 16 and is shifted in sequence in the direction of the arrangement of the multiple cultivated plants, i.e., in the longitudinal direction of the bed 5, to lower the height of the plant, and the plant is continuously cultivated. Therefore, for example, when cultivating tomatoes, the stems of the tomatoes will extend from the bed 5 along the string 17.

[0032] The mobile work vehicle 3 is equipped with front, rear, left and right running wheels, a work platform on which the harvesting device and pest control machine are mounted, and an electric motor that serves as a drive source for driving the running wheels 20. The running wheels 20 can move along the rails 13. The plant diagnosis device 111 also has the same running part as the running carriage and has a similar configuration.

[0033] The cultivation room 1 is framed by a steel structure, and vinyl is stretched on the four sides and the ceiling to form a greenhouse. It is rectangular in plan view, with the long sides in the east-west direction and the short sides in the north-south direction. As shown in FIG. 14, the cultivation room 1 is provided with fine mist ejection means 26 equipped with a number of fine mist ejection nozzles 25 on the four outer sides. The fine mist ejection means 26 is accompanied by a hydraulic pump and an electromagnetic valve, which will be described later. The fine mist ejection means 26 is a first fine mist ejection means 26a and a second fine mist ejection means 26b on both outer sides of the long side, and a third fine mist ejection means 26c and a fourth fine mist ejection means 26d on both outer sides of the short side. That is, in the illustrated example, the first fine mist ejection means 26a is arranged facing north (north face), the second fine mist ejection means 26b is arranged on the south side, the third fine mist ejection means 26c is arranged on the east side, and the fourth fine mist ejection means 26d is arranged on the west side. The first to fourth fine mist ejection means 26a to 26d are respectively associated with first to fourth solenoid valves 27a to 27d, and water supplied from a water tank 29 by driving a pump 28 can be independently supplied or cut off to the first to fourth fine mist ejection means 26a to 26d. Water supplied by driving the pump 28 is supplied to the first fine mist ejection means 26a via the first solenoid valve 27a, and the water is atomized and ejected from the fine mist ejection nozzles 25. Similarly, water is atomized and ejected from the fine mist ejection nozzles 25 of the second fine mist ejection means 26b via the second solenoid valve 27b, from the fine mist ejection nozzles 25 of the third fine mist ejection means 26c via the third solenoid valve 27c, and from the fine mist ejection nozzles 25 of the fourth fine mist ejection means 26d via the fourth solenoid valve 27d.

[0034] The fine mist that is ejected evaporates at the outside air temperature, and the heat of vaporization cools the air outside the vinyl of the cultivation room 1 to a temperature lower than the outside air temperature. The cooled air then enters the cultivation room 1 through the outside air inlet 30 provided on the side wall of the cultivation room 1, lowering the temperature inside the cultivation room 1. Therefore, the cultivation room 1, which has been heated by sunlight in summer, can be cooled by the so-called fine mist, improving the plant growth environment and the working environment of the workers. The outside air inlet 30 is formed on the four side walls of the cultivation room 1, and can be in the form of a window that can be opened and closed freely, or a vent hole that is formed at a low position to easily introduce cool air. In the illustrated example, a known blind screen is formed with multiple open / close plates that can be moved freely below the side wall at the height of the worker's feet.

[0035] In this embodiment, the first fine mist ejection means 26a to the fourth fine mist ejection means 26d are arranged around the four sides of the cultivation room 1, and the fine mist can be ejected independently, so fine mist cooling can be selectively applied to the necessary places as well as the entire four sides. A wind vane 31 capable of detecting wind direction and wind speed is provided, and the detected values ​​of wind direction and wind speed can be input to the control unit C, and the ON / OFF of the four first to fourth solenoid valves 27a to 27d is controlled based on the detected values. For example, when the wind speed is equal to or higher than a predetermined value and the wind direction is north, the first solenoid valve 27a is turned ON and the other control valves 27b to 27d are turned OFF, and only the first fine mist ejection means 26a is operated. With this configuration, the air that is finely misted and cooled by receiving the wind from the north can be easily introduced into the cultivation room 1. Also, the air that is finely misted and cooled at other positions is blown by the wind and is difficult to introduce into the cultivation room 1. Therefore, efficient fine mist ejection can be performed. FIG. 15 shows an example of the wind direction and the operation of the first fine mist ejection means 26a to the fourth fine mist ejection means 26d.

[0036] As shown in Fig. 13, the roof of the cultivation room 1 is configured with skylights 35a and 35b that are inclined to the left and right, and are installed at the center of the ridge so that they can be opened and closed. Opening and closing are controlled by the detection values ​​of a humidity sensor 36, a temperature sensor 37, or a solar radiation sensor 45 described later. In other words, when the humidity or temperature reaches a predetermined level or higher, the skylights 35a and 35b are opened to ventilate, and when these levels fall within a predetermined range, they are closed. The skylights 35a and 35b are opened and closed by a drive means 38 (for example, a forward / reverse drive motor).

[0037] In FIG. 15, a plurality of shading curtains 40 are provided at a predetermined interval on the ceiling 1s of the cultivation room 1. The shading curtains 40 are wound around a winding shaft 41 to open. A curtain opening / closing motor 42 is provided at one end of the cultivation room 1, and the motor 42 is driven to simultaneously pull out the plurality of shading curtains 40 from the winding shaft 41 via a curtain opening / closing wire 43. The plurality of shading curtains 40 can shade the entire ceiling of the cultivation room 1. The curtain opening / closing motor 42 is provided with a shading curtain opening / closing sensor 44 that detects the opening / closing degree (opening / closing amount) of the shading curtain 40 based on the rotation position of the motor 42. A solar radiation sensor 45 is provided on the south side outside the cultivation room 1. Based on the solar radiation detected by the solar radiation sensor 45, the curtain opening / closing motor 42 is driven and controlled to change the opening / closing degree of the shading curtain 40 so that the shading curtain 40 is opened to the desired opening / closing degree, and when the solar radiation is large, the shading curtain 40 is closed widely to increase the shading rate. Also, the configuration allows fine mist cooling by the first fine mist ejection means 26a to the fourth fine mist ejection means 26d when the shade state is at a predetermined level or higher based on information detected by the shading curtain opening / closing sensor 44. The leaves of plants immediately after planting in summer are thin and soft, and are prone to leaf burn when fine mist adheres to them, but with the above configuration, fine mist cooling cannot be performed unless the plant is in a shaded state, so leaf burn can be avoided.

[0038] The dolly of the plant factory will be described. As described above, if a device 221 with a harvesting mechanism is placed on the work platform of the mobile work vehicle (work dolly) 3 as shown in Fig. 9, it becomes a harvesting dolly 220. The harvesting arm 222 cuts and captures the crops, takes them into a storage section 223, and drives to a specified position for harvesting. The appropriateness of harvesting is confirmed by checking the position in the plant factory using data calculated by the cultivation management system 100 from data from the plant diagnostic device 111, and taking action for harvesting.

[0039] 10, if a pest control tank 231, a pest control pump 232, a right group of spray nozzles 233, and a left group of spray nozzles 234 are arranged, it becomes a pest control cart 230. When crops become infected with a disease caused by pests, bacteria, or the like, a chemical is stored in the pest control tank, and disinfection measures are taken based on the data from the plant diagnosis device 111.

[0040] The pest control tank 231 is sized to roughly match the width of the travel rail 13, ensuring a large capacity and safety during travel. By making it longer in the travel direction, it can accommodate a larger capacity.

[0041] The spray nozzles are located on the left and right, and the area can be analyzed based on the data on the occurrence of pests and diseases, and the right group of spray nozzles 233 and the left group of spray nozzles 234 can be sprayed simultaneously, or only on one side. They also extend vertically, and can be used from low to high positions simultaneously, to accommodate the transfer of pests and diseases. The right group of spray nozzles 233 and the left group of spray nozzles 234 are automatic sprayers, but can also be operated manually from a remote location using a remote control, and spraying on only one side is also possible. This can be done by pressing the spray button. Furthermore, the spray pressure can be changed using the spray pressure button, and the amount of spray can be adjusted. The pest control pump 232 is equipped with a flow meter, which controls and manages the amount of spray, and the cumulative amount used is also displayed on the monitor.

[0042] Even when used as a pest control device, the satellite positioning device 119 is installed, and the remaining capacity of the pest control tank 231 can be calculated as described above, so there is a control to automatically return to the supply point before it runs out completely. The right group of spray nozzles 233 and the left group of spray nozzles 234 can be adjusted in inclination using the spray setting button. By arranging them horizontally or slightly upward, the chemical solution can be sprayed so that it easily adheres to the underside of the leaves of the crops.

[0043] 8 shows a plant diagnostic device 111, but a work platform equipped with a plant diagnostic unit may be used as the traveling vehicle, such as a work vehicle (work platform) 3. Here, a plant diagnostic device that is a dedicated traveling vehicle will be described.

[0044] A light emitter 112 with an artificial light intensity is deployed. The plant factory itself has a light emitter 50, and this light emitter 50 can also change the light intensity to respond to the growth promotion of the crops, but the light emitter 112 with an artificial light intensity deployed in the plant diagnosis device 111 is close to the crops and has a large effect on growth promotion. However, since the range of the crops is large, multiple light emitters 112 with an artificial light intensity may be deployed, and a dedicated cart for the light emitter 112 with an artificial light intensity may be deployed. The response is performed while moving these carts continuously. The camera 113 and LIDAR device 118 calculate the shape, size, and color (pigment) of the crops, the position of the crops, the swaying of the crops and leaves, and the shape of the leaves, and this information is used as measurement data of the imaging device 112 in FIG. 3.

[0045] The light 117 is for measuring photosynthesis, and is turned on at night and photographed by the camera 113, and measurements are made using chlorophyll fluorescence images. A position recognition device is also provided that recognizes the positions of the mobile work vehicle 3 and the plant diagnostic device 111 in the cultivation room 1 using a satellite positioning device 119. Then, while the mobile work vehicle 3 and the plant diagnostic device 111 are automatically driven at night, the camera 113 photographs the plants at multiple pre-set photographing positions in the cultivation room 1 to obtain measurement data on the size of the plants. Photographs are taken at all photographing positions in one night, and measurement data at each photographing position is obtained and recorded every day. The photographing area taken by the camera 113 is a pre-set area, and the width in the arrangement direction of the cultivated plants (the longitudinal direction of the cultivation bed 5) is set to be the same (or approximately the same) as the spacing between the cultivated plants, but the set area can be changed.

[0046] At each photographing position, when the measurement data value of the day is greater than or equal to the measurement data value of the previous day, the LIDAR device 118 judges the size of the growth of the crop based on the increase in the measurement data value. On the other hand, when the measurement data value of the day is less than the measurement data value of the previous day, the growth of the plant is not judged based on the decrease in the decreased measurement data value, and when the measurement data value of the next day is greater than or equal to the measurement data value of the day, the growth of the plant is judged based on the increase in the measurement data value. Therefore, the cultivation management system 100 is provided with a growth degree judgment function that executes control to judge the growth of the plant. This growth degree judgment function is a system that incorporates multiple pieces of information using a neural network described later to make a judgment.

[0047] In the conventional technology, when the growth level of the crop is higher than the desired growth level based on the determination of the growth level, the amount of nutrient solution supplied by the nutrient solution supplying device 7 is reduced, the fertilizer concentration of the nutrient solution supplied by the nutrient solution supplying device 7 is increased, or the humidity in the cultivation room 1 is decreased, thereby causing water stress. In addition, the temperature around the plant and the temperature in the entire cultivation room 1 are lowered by the temperature control device and the heating pipe 37 for heating the entire greenhouse, thereby suppressing the growth of the plant. In this way, the elongation of the crop can be suppressed to control the desired growth rate, and the sugar content of the cultivated fruit can be increased. On the other hand, when the growth rate of the crop is lower than the desired growth rate, the amount of nutrient solution supplied by the nutrient solution supplying device 7 can be increased, or the fertilizer concentration of the nutrient solution supplied by the nutrient solution supplying device 7 can be lowered from a high set concentration, thereby reducing water stress. In addition, the temperature around the plant and the temperature in the entire cultivation room 1 can be increased by the temperature control device and the heating pipe 37 for heating the entire greenhouse, thereby promoting the growth of the plant. In this way, the desired growth rate can be controlled, and the harvest of the cultivated fruit can be increased. Therefore, the determined growth rate of the plant can be applied to nutrient solution supply control and environmental control (temperature control, humidity control, etc.).

[0048] In the present invention, in addition to conventional control, light intensity control is also an important factor, and the mobile work vehicle 3 and plant diagnostic device 111 are provided with strong artificial light with adjustable light intensity. This is characterized by the ability to respond to spot applications. Nutrient solution and temperature control can also be performed in separate areas, but this is not advisable when equipment costs are taken into consideration. Differences in crop growth can be as large as a single seedling, and adjusting the light intensity in the mobile work vehicle 3 and plant diagnostic device 111, which can also respond to partial areas, is effective in improving variation in growth.

[0049] Light control is also an immediate control. There is an appropriate range for temperature control and nutrient solution control, and even if it is strengthened beyond that range, only the stems will grow, resulting in poor yields and inferior quality. However, light control can also be used to shorten the nighttime hours when photosynthesis is resting, making it a less stressful and more efficient control. Light control can therefore adjust growth for periods of more than half a month, and can be used as a system for a new business model that controls the harvesting period to match crop supply to market needs.

[0050] Furthermore, the camera 113 and LIDAR device 118 photograph the stomata on the underside of the crop leaves, and by utilizing the neural network of Fig. 3, a comparison calculation can be performed with the data storage unit 191 to determine the size of the stomata. Meanwhile, a carbon dioxide supplying device is provided as a carbon dioxide supplying means for supplying carbon dioxide into the cultivation chamber 1, and when the photographed image is determined to show large stomata, carbon dioxide is discharged from the carbon dioxide supplying device described below, regardless of the carbon dioxide concentration in the cultivation chamber 1. This allows carbon dioxide to be supplied accurately when the stomata are wide open and photosynthesis is active, and reduces the wasteful supply of carbon dioxide compared to conventional carbon dioxide supply control based on carbon dioxide concentration.

[0051] Temperature and humidity sensors are installed in various places in the plant factory, and the plant diagnostic device 111 is also equipped with a temperature sensor 114, a humidity sensor 115, and a CO2 sensor 121 that measures the amount of carbon dioxide. There is a close relationship between the measurement of the amount of photosynthesis and the value of CO2, and the growth level is calculated comprehensively by calculating the two.

[0052] 6 is a schematic diagram of a cultivation room 1 in a plant factory proposed by the present invention. A cart runs on rails 13 for running, and the division of these rails and the division of columns 49 of the steel structure that constitutes the plant factory are used to form an environmental map area 48. In the plant factory, the environmental map area 48 is used as a square, and data for the cultivation management system 100 is input individually.

[0053] Figure 7 shows an example of how the amount of photosynthesis is displayed when it is measured. (A) and (B) were measured on different days. The shaded areas are areas where the amount of photosynthesis is outside the range of the standard values. Photosynthesis is not good if it is too high or too low. In other words, in order to control the growth rate, the amount is kept continuously within a specified range, so that the harvest time can be controlled.

[0054] In S7-1 and S7-2, the amount of photosynthesis was below the standard value, while in S7-3 and S7-4, the amount of photosynthesis was above the standard value. However, in the measurements taken the next day, the areas where the amount of photosynthesis was above the standard value were within the standard in S7-3A and S7-4A, and there was no problem. This may be a temporary increase in growth, and is excluded if measurements are not taken continuously.

[0055] However, in the S7-1A and S7-2A systems, the amount of photosynthesis is continuously measured below the standard value, so improvements are made by controlling the amount of artificial light and using conventional technologies to adjust the nutrient solution and temperature, without placing stress on the crops.

[0056] In addition, the environmental conditions in the areas of the plant factory can be managed as data by registering the area display and data input in the environmental map as shown in Figure 7. This environmental map has a lot of data for each area, but the way it is displayed can be changed by programming. For example, the amount of photosynthesis, the growth rate of the crop, the area temperature, humidity, the average amount of light in the area, the temperature of the crop leaves, the color of the fruit of the crop, the concentration of the nutrient solution, the yield, images of the fruit, the wind speed in the area, the amount of a specified gas component, etc. can be displayed. It is also possible to set a period and display the accumulated amount, average value, and deviation.

[0057] Z1-1 in Fig. 1 is a diagram showing changes in selling prices in the crop market management system 200. The horizontal axis indicates the date, which is the change around the harvest season.

[0058] S01 is the reference data registered in the cloud server 470. As S01 indicates, the selling price of crops varies depending on the date. This price fluctuation is due to the number of crops in stock at the market. Generally, crops such as outdoor fruits and vegetables are influenced by the season. Therefore, they tend to follow a similar trend every year. In addition, although the data does not take into account the occurrence of sudden bad weather that could result in a natural disaster, it does include the impact of bad weather conditions that occur at a similar level every year.

[0059] If the current date is the starting date of S04, fluctuations in selling prices like S02 are detected from the market information data 210. If the date and time of acquisition of the data of S02, the source of acquisition of the data, the number of pieces of data, and, although not described, the average change in the data used as the reference data of S01, the temperature, etc. are within the deviation range of the reference data, it becomes possible to compare the similarities between S02 and S01.

[0060] The method for harmonizing this data is to overlay the S02 and S01 curves up to this date using a convolution integral, and if the degree of change in state falls within a certain deviation range from S01, then this deviation is taken into account and an estimated calculation is performed from S01 to S03 to estimate the future selling price.

[0061] Z1-2 in Fig. 1 is a diagram showing changes in market demand in the crop market management system 200. The horizontal axis shows the date, which is the change around the harvest season.

[0062] S06 is reference data registered in the cloud server 470. As S06 indicates, the demand for the crop varies depending on the date. This demand is due to the number of crops in stock in the market, but it also tends to move differently from the number of crops in stock and the selling price due to seasonal popularity of the crop.

[0063] When the current date is the starting date of S04, a fluctuation in the demand level like S07 is detected from the market information data 210. The criteria for matching the similarity between S07 and S06 are the same as those for Z1-1.

[0064] The method for integrating this data is to overlay the curves of S07 and S06 up to this date using a convolution integral, and if the degree of change in state falls within a certain deviation range from S06, then this deviation is taken into account and an estimate is made from S06 to S08 to estimate future demand.

[0065] In the comparison calculation of Z1-1 and Z1-2, it is sufficient to detect the degree of variation in the numerical values, such as simple deviation comparison, least squares method, or component analysis method, and it is sufficient to be able to perform an estimation calculation of the future value using the difference in the degree of comparison.

[0066] Z1-1 and Z1-2 are calculated from the crop market management system 200, and the basis of this data is preregistered data, but as shown in the information route diagram from the plant production factory to the distribution organization in Figure 2, the system is capable of acquiring the current market from all related distribution organizations. Distribution route information will be explained later in the explanation of Figure 2.

[0067] Z1-4 in FIG. 1 is a diagram showing the change in the amount of photosynthesis of crops in a plant factory in the cultivation management system 100. In this diagram, the expression "photosynthesis distribution" is used to indicate data for a specific area. The horizontal axis indicates the date, and indicates the range in which the amount of photosynthesis meets the standard value. Upper and lower limits of the standard value are set for the photosynthesis distribution, and if it is within this range, it is determined that the amount of photosynthesis is appropriate. The amount of photosynthesis may be the average value of the entire plant factory, or it is possible to determine and divide the area. If the amount of photosynthesis is low, the growth of the crop will be delayed, and if it is too high, it will cause excessive growth, which will affect the sugar content of the crop and may cause the quality of the individual crop to deteriorate, so it is important that it is within the appropriate range. This amount of photosynthesis is a mechanism for measuring the amount of photosynthesis in the plant diagnostic device 111.

[0068] The photosynthetic distribution data for S14 is dated before the period of S04 in Z1-1 and Z1-2. The data is within the standard value in the sections S15 and S16, and growth is judged by integrating the length of this section and the amount of photosynthesis.

[0069] The acceptable range for photosynthesis is between the upper and lower limits of the standard value for photosynthetic distribution, but large deviations from this range can cause stress, so a state where the photosynthesis is between the upper and lower limits and has a small deviation is considered to be good.

[0070] Z1-5 in FIG. 1 is a diagram showing the change in the amount of growth of crops in a plant factory in the cultivation management system 100. In this diagram, the expression "crop growth distribution" is used to show data in a specific area. The horizontal axis shows the date, and indicates the range in which the growth distribution meets the standard value. Upper and lower limits of the standard value are set for the amount of growth (distribution), and if it is within this range, it is judged to be in proper growth. The growth distribution may be the average value of the entire plant factory, or it is possible to determine and divide the area and judge it. In addition, although it is called the growth distribution, the growth rate of the growth point of each crop is measured, and the proportion of the total number that is in the growth rate range of the standard range is shown. There are many crops, and it is not possible to measure all samples. Therefore, a representative sample is extracted and calculated, but continuous observation of the same sample is inaccurate. Therefore, since calculations are performed using different representative samples depending on the date, the growth rate distribution expresses the overall trend in the sense that the same sample is not used continuously.

[0071] If the elongation distribution is too small, crop growth will be delayed, and if it is too much, it will cause excessive growth, which will affect the sugar content of the crop and may even lead to a deterioration in the quality of the individual crop, so it is important that it is within an appropriate range.

[0072] The elongation distribution is determined by an imaging device or by measurement values ​​using a laser measuring instrument, measuring the same area as in the previous step, and calculating the deviation of the elongation values ​​of each measurement value.

[0073] There are upper and lower limits to the amount of growth of crops, and these ranges are considered appropriate, so an average judgment can be made on the amount; however, it is desirable for the entire crop to grow uniformly, and deviations in the amount of growth are also important factors, so both sets of data are important.

[0074] The date scale of the crop elongation distribution data for S17 is earlier than the period of S04 for Z1-1 and Z1-2. The data is within the standard value in the sections S18 and S19, and growth is judged by integrating the length and elongation distribution of this section.

[0075] Z1-6 in Fig. 1 is a diagram showing the change in nutrient solution concentration in the plant factory in the cultivation management system 100. The horizontal axis shows the date, and is a confirmation of whether the nutrient solution concentration meets the standard value. Since the concentration is continuously detected and controlled in the plant factory, it is unlikely that this will be a problem, but equipment malfunctions are anticipated and the impact on growth when the standard is not met is also considered.

[0076] Z1-3 in Fig. 1 is a diagram showing changes in the growth degree of crops in a plant factory in the cultivation management system 100. This growth degree is a diagram showing the achievement rate of the current growth in the time until appropriate harvesting, based on the appropriate time from the seedling state to the harvest time, looking at the current photosynthesis distribution and the crop growth distribution tendency.

[0077] The starting point of S04 is the growth degree determined by Z1-4 and Z1-5. This growth degree can be calculated using conventional calculation methods, but currently, conditions such as the occurrence of pests and diseases are also taken into account, and the environment and impact of the plant factory must also be taken into account, so it is not possible to analyze using simple calculation methods. Therefore, it is better to use the neural network shown in Figure 3 to include a learning function while performing data analysis (highly repeating comparison and collation of predictions and learning), thereby making more accurate predictions. The use of neural networks is explained in Figure 15.

[0078] S09 in Z1-3 is the value of the change in the standard growth rate, and is the core judgment criterion in the cultivation management system 100. The growth rate S10 is calculated from the photosynthesis distribution in Z1-4 and the crop elongation distribution in Z1-5. It is found that this growth rate is delayed compared to the growth rate S09, which is the standard.

[0079] The current state of the plant factory can be detected by the cultivation management system 100. At the same time, data from the crop market management system 200 is used to estimate changes in the selling price of Z1-1 and changes in the demand for Z1-2, as described above. In order to build a business model that achieves the objective of the present invention, "ensuring stable high prices by supplying when market inventory is low," this is achieved by matching the delayed growth rate of S10 with S09 at the time of S04.

[0080] However, in order to improve the current growth conditions, the photosynthesis distribution and crop elongation distribution must be continuously kept within the standard values. If the nutrient solution concentration is the standard, other factors include measures to improve the environmental temperature, humidity, and amount of sunlight in the plant factory.

[0081] Z1-7 in Figure 1 is a prediction of the future schedules of S04 and S05. S21 is the accumulated temperature in the registered future plant factory.

[0082] In addition, Z1-8 is used to estimate the cumulative amount of sunlight. S23 is the cumulative amount of sunlight in the registered future plant factory.

[0083] Thus, if we go by the usual trend, it is predicted that the weather will be somewhat worse from now on, and that it will be difficult to increase both the temperature and the amount of sunlight. As a result, the cultivation management system 100 predicts that S11 of Z1-3 is the future growth state. In other words, it is predicted that harvesting will be delayed during the period S05 from the start of the period when demand in the market increases and the selling price becomes high. This delay will result in crops being released into the market at a time when demand is declining, which may result in an increase in inventory and a fall in prices.

[0084] In the business model of the present invention, the plant factory is a place for controlled crop production, but unlike machines, it is not capable of instantaneous production. Since crops are living creatures, the plant factory must be able to quickly predict and assess their condition, respond instantly to improve crop production, and conform to market needs, thereby preventing sudden changes in sales prices due to product shortages and building a system that enables stable, high-price sales.

[0085] As a remedy for this, the growth rate can be increased to S12, which improves the delay in growth rate S13 and the number of days S05, making it possible to harvest on the scheduled harvest date.

[0086] This improvement in harvest date is the cumulative temperature of Z1-7 and the cumulative amount of sunlight of Z1-8 described above, and requires changing S21 to S22 and S23 to S24. The solution to this problem is to increase the amount of light by using artificial light.

[0087] Z1-9 in Figure 1 shows the change in the amount of artificial light in the plant factory. At the start of S04, when the future harvest is predicted, the amount of artificial light is S25, but in order to accelerate the harvest period and match S12, which improves the delay in growth, the amount of artificial light is increased to S26 during the period S28. Note that LED lights are effective in terms of light amount and economics as artificial light. These can be installed on the ceiling of the plant factory, but they can also be installed on the plant diagnostic device 111 or a cart that performs various detections. Note that if the artificial light is incandescent or infrared, it will also increase the temperature at the same time, and if heaters in the plant factory are relied upon, it can also be a somewhat assisted control. In other words, controlling the amount of light inevitably affects the temperature, so it is necessary to control the temperature at the same time. For example, if the amount of light is LED lighting, there will be no significant increase in temperature, but if the light is incandescent, fluorescent, or infrared, the effect on temperature change will be greater in that order. In conventional technology, this temperature rise is not considered to be active temperature management, so it is not considered control management, and some controls only involve light intensity management. However, if we look at it as plant growth management, we believe that temperature management is also inevitably carried out, although the effect may vary.

[0088] In this way, the sense of balance between the amount of light and the temperature has traditionally been managed by an experienced person through intuition, but the present invention makes it possible to achieve the same control balance as that experienced person.

[0089] Temperature control in a plant factory is easy to manage by controlling the introduction of outside air. This can be done by opening and closing the skylights in the ceiling, controlling the fans, or controlling the blackout curtains. Each control can be done using conventional feedback control, but because the temperature, amount of light, and resulting growth rate are estimated from multiple data, they can be reliably handled by artificial intelligence control using the neural network shown in Figure 3.

[0090] In addition, when using neural networks, there are methods such as S28, which gradually change the amount of light and temperature. Since these are plants, sudden changes can cause stress and lead to unpredictable results. The method of making gradual changes while observing the growth status is based on the wisdom and know-how of experienced people, and can be dealt with using artificial intelligence.

[0091] Figure 2 is a diagram of the information route from the plant production factory to the distribution organization. The plant factory has a skylight 31 for temperature control within the plant factory, and temperature control is performed by adjusting the opening degree based on the internal and external temperatures. Although not shown in the figure, temperature control is also possible by providing fans in each section to forcibly generate air. There is also lighting equipment 32, which can adjust the amount of light by using LED lights, etc. Adjustments can be made from the amount of light S25 required for people to work to the amount of light S26 that promotes plant growth.

[0092] The adjustment and control of the light intensity, temperature, and humidity in the plant factory are performed using the methods and configurations described above with reference to FIGS.

[0093] The plant diagnostic device 111 is equipped with an imaging device, which can capture the growth of the plant's growing point and calculate the growth rate. In addition, by imaging the fruit, not only the size but also the color can be determined. If the crop is a tomato, the red pigment can be detected and the growth rate can be directly determined by comparing it with the white and blue pigment rates. Furthermore, damage caused by pests can be confirmed by checking the pigments in the leaves and stems.

[0094] Although the degree of growth can be observed using imaging devices, it is difficult to capture images of everything in a plant factory, and the parts of the crops are in their natural positions, and the amount of light from outside is also involved, so accurate growth cannot be obtained using imaging devices alone. Therefore, it is necessary to calculate the amount of photosynthesis at night, which can be calculated stably.

[0095] Furthermore, in order to measure the entire area in a short period of time, the amount of carbon dioxide (hereinafter CO2) can also be measured, and this is provided in the plant diagnostic device 111.

[0096] The plant factory is managed by a user's computer terminal 80. The flow indicated by the dashed dotted line in FIG. 2 is the flow of information. Information on the plant factory in the user's computer terminal 80 is automatically input as a cultivation management system 100 after information acquired from a plant diagnostic device 111 is confirmed by the user and data is calculated. Representative data is photosynthesis distribution data, which is divided into many areas in the plant factory or divided by period, etc., and used as photosynthesis management data. The concepts of such areas and periods are similar, and include light quantity management data and crop growth rate management data.

[0097] In addition, crops to be shipped are randomly sampled and measured to obtain sugar content data and taste management data. Taste Meter (registered trademark) has its own index according to the development manufacturer, and although it is only reference data, it can be used as is or partially utilized to make a comprehensive judgment based on sugar content, moisture, weight, and color.

[0098] The data of the cultivation management system 100 is the quality itself, and this data needs to be communicated accurately to the customers who eat the crops, so a robust system is required to prevent the data from being tampered with. The data communicated to the customers may be limited, but each crop or group of crops is linked to an individual ID to prevent rewriting. The data of the cultivation management system 100 will be entered into this individual ID, and when checking the data of the cultivation management system 100, adding data, or in some cases changing data, it can be handled by limiting it to specific people. This system uses a token, which is an authentication key, and constitutes a feature of the business model of the present invention.

[0099] The distribution machine A410 may be, for example, a large-scale distribution and sales organization such as a national agricultural cooperative. Its role is to create aggregates of a single crop into large sections, classify the crops by product variety or quality, and other data from the cultivation management system 100 and the crop market management system 200, and determine the transaction price, which is the basis of the sales price, which is the value of the product, while obtaining data approval from the user 400.

[0100] Since the distribution machine A410 handles a large amount of crops, it is possible to combine and manage crop IDs as crop group IDs in small cardboard boxes, pallets, and even larger shipping units. The method of this integration is within the scope permitted by the user 400 of the crop producer's plant factory, and the system is designed so that the contents of the crop group created by the distribution machine A410 by combining crops can be understood, and the system allows the search for crop IDs in the created crop group ID, and conversely, the search for crop IDs allows the search for related crop group IDs. Each of these crop IDs is provided with a token 405, so that the crop 90, crop ID 404, and token 405 are managed together.

[0101] One method of managing the crops as a whole is to attach an electronic tag to the crop 90 and register the crop ID 404 and token 405 on this electronic tag. However, since this can be costly, for crops where each individual crop is not expensive, electronic tags are used to manage groups of crops in small cardboard boxes or pallets, and individual crops are handled with simple labels that only indicate the crop ID, and the actual crop ID 404 and token 405 are handled online and physically separated from the crop 90.

[0102] In addition, it is possible to simplify the process by managing crop groups by small cardboard boxes or pallets with electronic tags, treating any crop data as representative data for individual crops, and treating the crop ID as the crop group ID. In this example, this is effective for citrus fruits, which are small in size and shipped in large quantities.

[0103] In this way, the cultivation management system 100 is programmed to accommodate a variety of forms, and while it has a minimum crop ID for one individual crop, it is also possible to integrate these into a different crop group ID on a cardboard box basis.Similarly, the crop group IDs per cardboard box can be further integrated, creating a system in which data is integrated into crop IDs per pallet or even larger shipping category units.

[0104] The distribution organization Bα421 and the distribution organization Bβ422 can be transportation organizations. Here, the management conditions for transportation are additionally registered to the crop ID. Traceability can be strengthened by registering the management temperature during transportation, the travel distance, and the travel route.

[0105] Distribution institutions Cα and Cβ correspond to intermediate wholesalers and warehouses of transport institutions in the region, and perform the final division to distribution institution Dα 441, which corresponds to a retail store. Distribution institutions Cα and Cβ divide the crop groups transported by the transport institution in pallets into cardboard boxes or further into smaller units, and in some cases also temporarily store them in a warehouse. They expand the crop group ID into crop ID data, and in the case of temporary storage, they register the storage time and storage conditions. By managing this data, producers 400 and distribution institutions A to B can check whether there are any problems with distribution.

[0106] The distribution agency Dα441, which corresponds to a retail store, develops the crop group ID as an individual crop ID and discloses the data to consumers.

[0107] The user 400 is aware of all distribution institutions from distribution institution A to distribution institution D, and distribution institutions that the user 400 does not recognize cannot check the data of the cultivation management system 100. Only institutions specified by the user 400 can respond. The characteristic of this advantage is the restriction of authentication. If data is disclosed to an unspecified number of distribution institutions and data is disclosed to a large number of related parties or data changes are permitted, there is a risk of creating unexpected false data. Therefore, if the user were to set up a system to confirm whether or not to permit each data change, the user would have to be on duty 24 hours a day.

[0108] The system that solves this problem is one of the features of the present invention. If the crop ID 404 and token 405 can be authenticated, the authorized person can freely add subsequent data and use previous data. Therefore, data can be used by specific people without the need for approval from the user 400 each time, enabling work that is not time-based.

[0109] Up until now, the use of data from the cultivation management system 100 has been mainly explained, but for the crop market management system 200, the information path is the opposite flow. In the crop information data 210F, the desired selling price, or so-called purchase price, the current stock quantity, the actual number of items received this time, and the number of items reserved for the next time are input. This input information is divided into two routes: one where the information flows one step at a time to a higher-level distribution institution, and the other where it is directly communicated to the cloud server 470, thereby ensuring the speed and accuracy of the information.

[0110] The stock and incoming quantities in this distribution route include losses in the distribution route, and the figures differ between the actual shipment numbers and the actual sales numbers, so it is necessary to take such losses into account, and taking this actual situation into account, crop information data 210F is converted to crop information data 210D and 210E, and data is added or corrected as 210B, and is transmitted to cloud server 470 as final market information data 210A. In this way, the system adds market information to the initial market information data 210 basic data and returns it as market information data 210A, and market trends can be analyzed by analyzing this difference.

[0111] The above describes the flow of crop information, and information is managed at each terminal via the Internet, but actual crop distribution management is as shown in the lower part of Fig. 2. In some cases, the plant diagnostic device 111 has a harvesting function, and in some cases, a dedicated harvesting robot cart is used, but the sugar content data, Taste Meter (registered trademark) management data, nutrient solution management data, photosynthesis management data, light quantity management data, and growth rate management data acquired from the harvested crop 90 are registered as information in the crop ID 404. As described above, the crop ID 404, token 405, and crop 90 are a set, and further crops are collected to form crop group 91, but the crop ID 404 and token 405 are data that form crop group 91, and the crop ID 404 and token 405 themselves do not change.

[0112] Distribution organization B often transports crops by truck 70, and data from the refrigerated storage of this truck 70 can also be incorporated into the crop ID 404. Distribution organization Dα 441, like retail store 71, sells individual crops, but can extract information from crop ID 404 to promote sales.

[0113] FIG. 3 explains a method for analyzing data from the cultivation management system 100 and the crop market management system 200 in the present invention by using a neural network, which is an artificial intelligence, to quickly and accurately calculate large amounts of data while also performing deep learning.

[0114] As an artificial intelligence system for a plant factory, it uses information from both the crop state detection unit 181, which is detection data for the crop itself, and the facility environment state detection unit 182, which is detection data for the plant factory environment. The data from each measuring device to obtain both pieces of information is not directly related, and is used to make predictions about future harvest times from this large amount of data based on the so-called producer's experience.

[0115] The plant diagnostic device 111 may be equipped with an imaging device 112 and a CO2 sensor 121 for detecting the amount of carbon dioxide, or a dedicated cart may be equipped with the imaging device 112 and the CO2 sensor 121 for detecting the amount of carbon dioxide.

[0116] The imaging device 112 uses a CCD camera to capture the shape and color, and a LIDAR to measure three-dimensional distance and movement. In particular, for fruit size and maturity, the data from both the CCD camera and LIDAR can be overlaid to verify the ratio of pigments to shape and their positions against past data, making it possible to non-destructively estimate the degree of ripeness of the crop.

[0117] Furthermore, the amount of photosynthesis is detected by measuring the amount of photosynthesis of plants at night using LEDs in the plant diagnostic device 111. The amount of photosynthesis is basic data for verifying the growth state of plants, and if there are many crops with a high amount of photosynthesis within a specified range, photosynthesis is being promoted stably. By importing the crops within this specified range into a work map and observing their distribution, it becomes possible to see on the map where growth is good and where growth is poor.

[0118] At the same time, images are taken focusing on the areas where the growth is taking place, and the color of the leaves is also captured. Based on the leaf shape and color, it is possible to determine whether the leaf is wilted or has been caused by pests or disease.

[0119] These image data are a combination of data from both the CCD camera and LIDAR, but the parts of the plant that are used and the reference data are different, and each data is calculated so that they influence each other. Newly measured data is also added to the deviation from the base data. The data measured in this way influence each other to calculate the deviation in photosynthesis distribution 110, the deviation in the elongation rate of stems and leaves 120, and the deviation in the distribution of pest and disease occurrence 170.

[0120] The plant factory also includes an indoor temperature sensor 141, an indoor humidity sensor 142, a light intensity sensor 151, an anemometer 143, a power meter 152, an indoor gas analyzer 143, a power meter 152, an indoor gas analyzer 143, and a nutrient solution concentration meter 131. The facility environment state detection unit 182 also performs calculations in which the measured sensor values ​​affect each other, and calculates the deviation 140 of the accumulated temperature, the deviation 150 of the accumulated amount of sunlight, and the deviation 130 of the nutrient solution concentration.

[0121] The judgment of each of these deviations is a unique phenomenon in a plant factory. Normally, the desired data value itself is used, but when there are many deliverables such as crops in a plant factory, it is better to use the deviation, which is processed data, and it is a measure to perform a non-dimensional state comparison. The comparison judgment up to this point is the first state change 180, and the image distribution deviation analysis and decision unit 183 holds it as the currently obtained data.

[0122] On the other hand, the data storage unit 191 stores basic data to be compared with the image distribution deviation analysis and determination unit 183. The data storage unit 191 may store single data from each sensor, and can change the standard in accordance with the data format of the image distribution deviation analysis and determination unit 183. In this embodiment, the deviation value is non-dimensional, so the data storage unit 191 also performs similar processing.

[0123] The image distribution deviation analysis, the comparison of the data in the determination unit 183 and the data in the data storage unit 191, and the evaluation and judgment are performed by a predetermined calculation formula, and the outputs of this calculation, that is, the crop growth determination 121, the artificial light intensity determination 151, the defect rate determination 171, and the misrecognition estimation 192, are calculations in which the results of the respective determinations affect each other, and there is a cyclic calculation unit, and a method of determining and converging one of the priority output units is adopted to form a second state change 190.

[0124] In this way, the system has the ability to infer phenomena of different dimensions, extract highly relevant states, and determine decision points while mutually utilizing the values ​​of multiple sensors, and also has a learning function that registers the data itself as comparison data, making it possible to respond by using a neural network, an artificial intelligence system, as shown in Figure 3.

[0125] In addition to the calculations of the cultivation management system 100, this system also uses crop market information data 210 from the crop market management system 200 to determine the crop growth level 121, the amount of artificial light emitted 151, the defective product rate 171, and the estimation of misrecognition 192, and the harvest date is determined by the harvesting work date determination unit 160.

[0126] In the defective product rate determination 171, the disease occurrence distribution area of ​​the crops in the plant factory and the amount of carbon dioxide absorption as an index of the result of the growth of the crops are calculated. If there are many defective product rate determinations and they affect the shipping volume, the harvest date determination must be corrected in the harvest work date determination unit 160. At the same time, by automatically submitting information to the market information data 210, it is possible to quickly respond and stabilize the selling price by adjusting the inventory in the market.

[0127] Similarly, unforeseen events may occur in the distribution organization in Fig. 2. There may be cases where the demand for a product increases rapidly, resulting in a shortage of inventory, or where a product is lost due to a large loss in the distribution channel. This information is also added to the crop ID as market information data by distribution organizations, including retailers, and is then fed back and calculated in the crop market management system 200. Taking into account the state of the cultivation management system 100, if the degree of growth cannot be ensured even with the control of the amount of artificial light and the temperature control by conventional technology, the harvesting operation date determination unit 160 calculates the number of days until the unsatisfied harvest date and the harvest amount as correction values.

[0128] In this way, by utilizing the neural network function for the output information in the harvest work plan 520 and the market information data issue 530, related information can be weighted for judgment, and rapid output processing can be performed, including correction of control of the harvest work day.

[0129] The harvesting procedure at harvest time, the travel route 521 and work time 522 of the harvesting robot cart are determined by simultaneously processing data from the cultivation management system 100 and data from the crop market management system 200, and calculating and correcting the next data from the cultivation management system 100. Therefore, the amount of data processed, the inference logic from the data, and the reliability can be said to be similar to a process that utilizes the know-how of an experienced person, and the use of a neural network is required.

[0130] In this way, the crop cultivation management system 100 detects the growth level of the crop and controls the optimum harvesting time by controlling the amount of artificial light, and the crop market management system 200 estimates the time when crop prices will tend to rise from crop market information data 210, and uses information calculated for the optimum harvesting time to adjust the amount of artificial light and temperature using a neural network for comparison calculations so that the photosynthesis distribution amount and deviation, stem and leaf elongation distribution amount and deviation, solution concentration deviation, accumulated temperature deviation, and accumulated solar radiation deviation, which are indicators of the crop's quality state, fall within the deviation range of application, thereby automatically controlling the harvesting day and submitting information to the market information data 210.

[0131] Each item may be a value itself, for example an average value, or the deviation of each value may be prioritized, or both the value and the deviation may be used. In other words, in a plant factory, there are many objects to be measured, and crops are selected and measured, and it is better to select and detect different crops each day rather than measuring the same ones continuously. In such measurements, it is considered more effective to determine the range of the standard value, determine that the standard value is satisfied if it is within that range, and place emphasis on controlling the growth variation based on the deviation. Controlling based on the deviation of this deviation, the value that is the coefficient of variation against the standard value, is controlling the value itself, and controlling the growth rate comprehensively. However, if there are many values, calculations are also required to divide the data by the time of data acquisition and the data area, so using a neural network to repeatedly search for related items, collate, compare, add data, and learn is a method with fewer errors in calculation processing.

[0132] The measurements so far have been good crop values, but unfortunately there have been losses. The crop cultivation management system 100 measures the distribution area of ​​disease occurrence in crops in the plant factory and the amount of carbon dioxide absorption, which is an index of crop growth, and includes these calculated values, uses a neural network in a comparative calculation, corrects the control of the harvest work day, and automatically submits information to the market information data 210. It was explained that including this results in a more accurate harvest amount.

[0133] In Fig. 3, a basic map for storing harvest data is created as the output map control 523, so that data that will be generated later can be added. Also, at the time when the harvest date is estimated, crop ID and token issuance 407 is activated, and reservations are made for crop ID 404, information token 405, and tag issuance 409. Furthermore, market information data issuance 530 is created. Desired price 531, shipping quantity 532, distribution route selection 533, and user information issuance 534 are performed. Since such data that is highly feasible for the market is prepared before harvest, retailers feel reassured and reservations for harvest are made before harvest, and a stable supply is secured for both producers and retailers, so a business route and business model that guarantees prices are built.

[0134] The configuration of token authentication is explained in Fig. 4. When user information issuance 534 is performed, a crop ID 404 and an information token 405 are prepared, but in reality, a complex authentication path is followed.

[0135] First, a public key and a private key are generated 460. This is a task that mechanically performs the first authentication of the user 400. A public key 403 and a resource ID 452 are sent to the related distribution institutions 410-440 and the user 400. This is a public key, and in order to open the key, it is necessary to verify it with a token 405 issued by the user 400 via a token issuance 407.

[0136] The resource ID 452 is an image of a large frame, and is the basis of the crop ID. It is registered as the crop ID upon authentication by the user 400, and serves as a container for various information.

[0137] Regarding the token, the public key 403 may be considered as a physical key, but it is in a locked state. The secret number that unlocks this locked state is the token. The private key 403A is a key with a time limit, and the format of the additional information, etc. is also determined. The time limit and the format of the additional data correspond to each public key 403. The public key 403 is unlocked by the token, and it is verified that the added or updated data is within the time limit and that the data is entered with the expected content. Unless this decryption and verification 465 are established, the system does not allow the data update work for the user 400 to be performed.

[0138] The characteristics of the public key 403 and the private key 403A will be described. The public key 403 is automatically distributed to related parties uniformly, but the resource ID 452 can be changed depending on the other party. This is a feature of the proposed invention, but as mentioned above, the decryption with the private key 403A can be limited to a certain period, and the user 400 can decide the period. For example, the distribution institution A 410 can only update data for a maximum of one week after harvesting, so the token is limited to a certain period. The subsequent distribution institutions Bα 421 and Bβ 422 cannot update data after the distribution institution A 410 updates the data or after the date registered in advance by the user 400. By sending this information management system by the resource information transmission 453, information is added or updated to the crop ID. The token 405 + update information 461 is sent to the cloud server 470, and decryption and verification 465 are confirmed. At this stage, the data is accepted and stored in the data pool 463 in FIG. 5.

[0139] The tokens are made to have a time limit in order to prevent data tampering by a third party that the user 400 is not aware of, and to confirm by transmitting the data that the shipped crops are certainly flowing into the distribution REIT. Problems such as data tampering are likely to occur in an environment where data can be corrected over a long period of time. Simple data input errors can be corrected immediately, but updating data after time has passed may involve some kind of fraud. Therefore, actions such as updating data after time has passed cannot be performed, and the business model system proposed here updates data in real time.

[0140] However, while there is a time limit for updating data using token 405, there is plenty of time for database registration 404, and user 400 can check the updated data when he has time and respond with update permission 466. In other words, it is a system tailored to the work of user 400, and only data that is quick and reliable is accepted, and data registration is done without fail while checking, in a business model that prioritizes producers. However, this is not information management that is difficult for distribution institutions to use. It is not necessary to obtain approval from user 400 every time information is updated, and if the data can be decrypted with token 405, acceptance is complete, making it a hassle-free, rational, and safe system.

[0141] FIG. 5 shows the overall flow of the data update system. The overlapping points of the functions described in FIG. 4 are shown in the figure. The resource API 450 contains the resource information 453 in FIG. 4. The information in the resource API 450 includes information that has been received beyond the expiration date of the token 405 and other information that the user did not expect. These pieces of information are stopped by the resource authentication 451 and do not become the update data 462. This is the content that is judged as inappropriate by the control of expiration response of data update in the present invention. However, in some cases, appropriate updates may have been neglected, so as a rescue measure, re-authentication is performed by the user authentication 401. The user authentication 401 confirms the information that could not be resource authenticated by sending the information data 402 to the user 400. If it is judged that there is no problem, the decrypted data 406 for the token 405 is sent, the resource authentication 451 is released, and the update data 462 is sent to the data pool 463. The update data is registered in the database 464 through the user authentication 401. In this way, the system is designed so that any data that has passed its expiration date or that has not been authenticated for data update can be handled through user authentication 401.

[0142] The resource API 450 uses a programming language and data format that allows information to be shared. Since various software is used by distribution institutions, the data formats may not be the same, but the data formats can be shared by using the API.

[0143] This resource system also allows a distribution institution to set a temporary resource ID and issue a token even when the institution is offline and cannot connect to the Internet. Both can be handled by using the settings of resource authentication 451 and temporary storage data pool 463. The data stored in the temporary area is linked to the official resource ID when the user returns online and a formal resource ID is issued.

[0144] In this way, a user of a plant factory has a user account in the production and distribution system, obtains a public key from the production and distribution system, assigns IDs to the individual crops and crop groups to be distributed, generates a token which is a string of characters encrypted with the public key from the crop information acquired by the cultivation management system 100, and makes the information public as an API in the crop market management system 200 as a set consisting of the harvest of the individual crops and the crop groups, the IDs assigned to the individual crops and the crop groups, and the token encrypted with the public key.

[0145] In addition, plant factory users can distribute IDs assigned to individual crops and groups of crops, and tokens encrypted with a public key to relevant parties in the distribution system. They can use the public key to add or change information generated during distribution, but cannot change items related to production. In addition, a deadline is set for each item determined by the plant factory users, and information cannot be added or changed after that date, providing strong security.

[0146] The information from this system is used to continuously control future shipments by adding data with each harvest and learning from it; it is a system that can build a business model for plant factories that includes distribution, using artificial intelligence to predict market demand and control the harvest time. [Explanation of symbols]

[0147] 100 Cultivation Management System 110 Photosynthetic distribution deviation 111 Plant diagnostic equipment 120 Deviation in the growth rate of stems and leaves 121 Determination of crop growth 151 Determination of the amount of artificial light 170 Distribution deviation of pests and diseases 200 Crop Market Management System 210 Market Information Data 404 Crop ID 405 Token 410 Distribution Institution A

Claims

1. A crop cultivation management system (100) that detects the growth level of crops and controls the optimum harvesting time by controlling the amount of artificial light, A crop market management system (200) that estimates fluctuations in crop prices from crop market information data (210) calculates the optimal harvest time; At the calculated optimum harvest time, the amount of photosynthesis and its deviation, the amount of stem and leaf growth and its deviation, the deviation of the solution concentration, the deviation of the accumulated temperature, and the deviation of the accumulated amount of sunlight, which are indicators of the quality of the crop, are calculated so that they fall within the applicable range of values ​​and deviation values. A production and distribution system for a plant factory that uses a neural network for comparison calculations to adjust the amount of artificial light and temperature, thereby automatically controlling harvesting days and submitting information to market information data (210).

2. A crop cultivation management system (100) measures the distribution area of ​​disease occurrence of crops in a plant factory and the amount of carbon dioxide absorption as an index of crop growth, The detected value is compared with the reference value and collated using a neural network, and the harvest date is automatically corrected and the information is submitted to the market information data (210). The production and distribution system of claim 1.

3. A user of the plant factory has a user account in the production and distribution system, acquires a public key from the production and distribution system, assigns an ID to each crop and each group of crops to be distributed, and generates a token, which is a character string obtained by encrypting crop information acquired in the cultivation management system (100) with the public key; The yields of individual crops and groups of crops; IDs assigned to individual crops and crop groups; A set of public key encrypted tokens The information is made public as an API (Application Programming Interface) in the crop market management system (200). The production and distribution system for a plant factory according to claim 1 or 2.

4. The plant factory user distributes IDs assigned to individual crops and groups of crops and tokens encrypted with a public key to related parties in the distribution system, and can add or change information generated during distribution using the public key. However, items related to production cannot be changed, and a deadline is set for each item determined by the plant factory user, and information cannot be added or changed after that deadline. The production and distribution system for a plant factory according to claim 3.

Images